diff options
544 files changed, 25547 insertions, 10119 deletions
diff --git a/Documentation/ABI/testing/ima_policy b/Documentation/ABI/testing/ima_policy index 6cd6dae..9869466 100644 --- a/Documentation/ABI/testing/ima_policy +++ b/Documentation/ABI/testing/ima_policy @@ -12,11 +12,14 @@ Description: then closing the file. The new policy takes effect after the file ima/policy is closed. + IMA appraisal, if configured, uses these file measurements + for local measurement appraisal. + rule format: action [condition ...] - action: measure | dont_measure + action: measure | dont_measure | appraise | dont_appraise | audit condition:= base | lsm - base: [[func=] [mask=] [fsmagic=] [uid=]] + base: [[func=] [mask=] [fsmagic=] [uid=] [fowner]] lsm: [[subj_user=] [subj_role=] [subj_type=] [obj_user=] [obj_role=] [obj_type=]] @@ -24,36 +27,50 @@ Description: mask:= [MAY_READ] [MAY_WRITE] [MAY_APPEND] [MAY_EXEC] fsmagic:= hex value uid:= decimal value + fowner:=decimal value lsm: are LSM specific default policy: # PROC_SUPER_MAGIC dont_measure fsmagic=0x9fa0 + dont_appraise fsmagic=0x9fa0 # SYSFS_MAGIC dont_measure fsmagic=0x62656572 + dont_appraise fsmagic=0x62656572 # DEBUGFS_MAGIC dont_measure fsmagic=0x64626720 + dont_appraise fsmagic=0x64626720 # TMPFS_MAGIC dont_measure fsmagic=0x01021994 + dont_appraise fsmagic=0x01021994 + # RAMFS_MAGIC + dont_measure fsmagic=0x858458f6 + dont_appraise fsmagic=0x858458f6 # SECURITYFS_MAGIC dont_measure fsmagic=0x73636673 + dont_appraise fsmagic=0x73636673 measure func=BPRM_CHECK measure func=FILE_MMAP mask=MAY_EXEC measure func=FILE_CHECK mask=MAY_READ uid=0 + appraise fowner=0 The default policy measures all executables in bprm_check, all files mmapped executable in file_mmap, and all files - open for read by root in do_filp_open. + open for read by root in do_filp_open. The default appraisal + policy appraises all files owned by root. Examples of LSM specific definitions: SELinux: # SELINUX_MAGIC - dont_measure fsmagic=0xF97CFF8C + dont_measure fsmagic=0xf97cff8c + dont_appraise fsmagic=0xf97cff8c dont_measure obj_type=var_log_t + dont_appraise obj_type=var_log_t dont_measure obj_type=auditd_log_t + dont_appraise obj_type=auditd_log_t measure subj_user=system_u func=FILE_CHECK mask=MAY_READ measure subj_role=system_r func=FILE_CHECK mask=MAY_READ diff --git a/Documentation/ABI/testing/sysfs-devices-system-cpu b/Documentation/ABI/testing/sysfs-devices-system-cpu index 5dab364..6943133 100644 --- a/Documentation/ABI/testing/sysfs-devices-system-cpu +++ b/Documentation/ABI/testing/sysfs-devices-system-cpu @@ -176,3 +176,14 @@ Description: Disable L3 cache indices All AMD processors with L3 caches provide this functionality. For details, see BKDGs at http://developer.amd.com/documentation/guides/Pages/default.aspx + + +What: /sys/devices/system/cpu/cpufreq/boost +Date: August 2012 +Contact: Linux kernel mailing list <linux-kernel@vger.kernel.org> +Description: Processor frequency boosting control + + This switch controls the boost setting for the whole system. + Boosting allows the CPU and the firmware to run at a frequency + beyound it's nominal limit. + More details can be found in Documentation/cpu-freq/boost.txt diff --git a/Documentation/ABI/testing/sysfs-driver-ppi b/Documentation/ABI/testing/sysfs-driver-ppi new file mode 100644 index 0000000..97a003e --- /dev/null +++ b/Documentation/ABI/testing/sysfs-driver-ppi @@ -0,0 +1,70 @@ +What: /sys/devices/pnp0/<bus-num>/ppi/ +Date: August 2012 +Kernel Version: 3.6 +Contact: xiaoyan.zhang@intel.com +Description: + This folder includes the attributes related with PPI (Physical + Presence Interface). Only if TPM is supported by BIOS, this + folder makes sence. The folder path can be got by command + 'find /sys/ -name 'pcrs''. For the detail information of PPI, + please refer to the PPI specification from + http://www.trustedcomputinggroup.org/ + +What: /sys/devices/pnp0/<bus-num>/ppi/version +Date: August 2012 +Contact: xiaoyan.zhang@intel.com +Description: + This attribute shows the version of the PPI supported by the + platform. + This file is readonly. + +What: /sys/devices/pnp0/<bus-num>/ppi/request +Date: August 2012 +Contact: xiaoyan.zhang@intel.com +Description: + This attribute shows the request for an operation to be + executed in the pre-OS environment. It is the only input from + the OS to the pre-OS environment. The request should be an + integer value range from 1 to 160, and 0 means no request. + This file can be read and written. + +What: /sys/devices/pnp0/00:<bus-num>/ppi/response +Date: August 2012 +Contact: xiaoyan.zhang@intel.com +Description: + This attribute shows the response to the most recent operation + request it acted upon. The format is "<request> <response num> + : <response description>". + This file is readonly. + +What: /sys/devices/pnp0/<bus-num>/ppi/transition_action +Date: August 2012 +Contact: xiaoyan.zhang@intel.com +Description: + This attribute shows the platform-specific action that should + take place in order to transition to the BIOS for execution of + a requested operation. The format is "<action num>: <action + description>". + This file is readonly. + +What: /sys/devices/pnp0/<bus-num>/ppi/tcg_operations +Date: August 2012 +Contact: xiaoyan.zhang@intel.com +Description: + This attribute shows whether it is allowed to request an + operation to be executed in the pre-OS environment by the BIOS + for the requests defined by TCG, i.e. requests from 1 to 22. + The format is "<request> <status num>: <status description>". + This attribute is only supported by PPI version 1.2+. + This file is readonly. + +What: /sys/devices/pnp0/<bus-num>/ppi/vs_operations +Date: August 2012 +Contact: xiaoyan.zhang@intel.com +Description: + This attribute shows whether it is allowed to request an + operation to be executed in the pre-OS environment by the BIOS + for the verdor specific requests, i.e. requests from 128 to + 255. The format is same with tcg_operations. This attribute + is also only supported by PPI version 1.2+. + This file is readonly. diff --git a/Documentation/cpu-freq/boost.txt b/Documentation/cpu-freq/boost.txt new file mode 100644 index 0000000..9b4edfc --- /dev/null +++ b/Documentation/cpu-freq/boost.txt @@ -0,0 +1,93 @@ +Processor boosting control + + - information for users - + +Quick guide for the impatient: +-------------------- +/sys/devices/system/cpu/cpufreq/boost +controls the boost setting for the whole system. You can read and write +that file with either "0" (boosting disabled) or "1" (boosting allowed). +Reading or writing 1 does not mean that the system is boosting at this +very moment, but only that the CPU _may_ raise the frequency at it's +discretion. +-------------------- + +Introduction +------------- +Some CPUs support a functionality to raise the operating frequency of +some cores in a multi-core package if certain conditions apply, mostly +if the whole chip is not fully utilized and below it's intended thermal +budget. This is done without operating system control by a combination +of hardware and firmware. +On Intel CPUs this is called "Turbo Boost", AMD calls it "Turbo-Core", +in technical documentation "Core performance boost". In Linux we use +the term "boost" for convenience. + +Rationale for disable switch +---------------------------- + +Though the idea is to just give better performance without any user +intervention, sometimes the need arises to disable this functionality. +Most systems offer a switch in the (BIOS) firmware to disable the +functionality at all, but a more fine-grained and dynamic control would +be desirable: +1. While running benchmarks, reproducible results are important. Since + the boosting functionality depends on the load of the whole package, + single thread performance can vary. By explicitly disabling the boost + functionality at least for the benchmark's run-time the system will run + at a fixed frequency and results are reproducible again. +2. To examine the impact of the boosting functionality it is helpful + to do tests with and without boosting. +3. Boosting means overclocking the processor, though under controlled + conditions. By raising the frequency and the voltage the processor + will consume more power than without the boosting, which may be + undesirable for instance for mobile users. Disabling boosting may + save power here, though this depends on the workload. + + +User controlled switch +---------------------- + +To allow the user to toggle the boosting functionality, the acpi-cpufreq +driver exports a sysfs knob to disable it. There is a file: +/sys/devices/system/cpu/cpufreq/boost +which can either read "0" (boosting disabled) or "1" (boosting enabled). +Reading the file is always supported, even if the processor does not +support boosting. In this case the file will be read-only and always +reads as "0". Explicitly changing the permissions and writing to that +file anyway will return EINVAL. + +On supported CPUs one can write either a "0" or a "1" into this file. +This will either disable the boost functionality on all cores in the +whole system (0) or will allow the hardware to boost at will (1). + +Writing a "1" does not explicitly boost the system, but just allows the +CPU (and the firmware) to boost at their discretion. Some implementations +take external factors like the chip's temperature into account, so +boosting once does not necessarily mean that it will occur every time +even using the exact same software setup. + + +AMD legacy cpb switch +--------------------- +The AMD powernow-k8 driver used to support a very similar switch to +disable or enable the "Core Performance Boost" feature of some AMD CPUs. +This switch was instantiated in each CPU's cpufreq directory +(/sys/devices/system/cpu[0-9]*/cpufreq) and was called "cpb". +Though the per CPU existence hints at a more fine grained control, the +actual implementation only supported a system-global switch semantics, +which was simply reflected into each CPU's file. Writing a 0 or 1 into it +would pull the other CPUs to the same state. +For compatibility reasons this file and its behavior is still supported +on AMD CPUs, though it is now protected by a config switch +(X86_ACPI_CPUFREQ_CPB). On Intel CPUs this file will never be created, +even with the config option set. +This functionality is considered legacy and will be removed in some future +kernel version. + +More fine grained boosting control +---------------------------------- + +Technically it is possible to switch the boosting functionality at least +on a per package basis, for some CPUs even per core. Currently the driver +does not support it, but this may be implemented in the future. diff --git a/Documentation/cpuidle/sysfs.txt b/Documentation/cpuidle/sysfs.txt index 9d28a34..b6f44f4 100644 --- a/Documentation/cpuidle/sysfs.txt +++ b/Documentation/cpuidle/sysfs.txt @@ -76,9 +76,17 @@ total 0 * desc : Small description about the idle state (string) -* disable : Option to disable this idle state (bool) +* disable : Option to disable this idle state (bool) -> see note below * latency : Latency to exit out of this idle state (in microseconds) * name : Name of the idle state (string) * power : Power consumed while in this idle state (in milliwatts) * time : Total time spent in this idle state (in microseconds) * usage : Number of times this state was entered (count) + +Note: +The behavior and the effect of the disable variable depends on the +implementation of a particular governor. In the ladder governor, for +example, it is not coherent, i.e. if one is disabling a light state, +then all deeper states are disabled as well, but the disable variable +does not reflect it. Likewise, if one enables a deep state but a lighter +state still is disabled, then this has no effect. diff --git a/Documentation/devicetree/bindings/arm/calxeda/combophy.txt b/Documentation/devicetree/bindings/arm/calxeda/combophy.txt new file mode 100644 index 0000000..6622bdb --- /dev/null +++ b/Documentation/devicetree/bindings/arm/calxeda/combophy.txt @@ -0,0 +1,17 @@ +Calxeda Highbank Combination Phys for SATA + +Properties: +- compatible : Should be "calxeda,hb-combophy" +- #phy-cells: Should be 1. +- reg : Address and size for Combination Phy registers. +- phydev: device ID for programming the combophy. + +Example: + + combophy5: combo-phy@fff5d000 { + compatible = "calxeda,hb-combophy"; + #phy-cells = <1>; + reg = <0xfff5d000 0x1000>; + phydev = <31>; + }; + diff --git a/Documentation/devicetree/bindings/ata/ahci-platform.txt b/Documentation/devicetree/bindings/ata/ahci-platform.txt index 8bb8a76..b519f9b 100644 --- a/Documentation/devicetree/bindings/ata/ahci-platform.txt +++ b/Documentation/devicetree/bindings/ata/ahci-platform.txt @@ -8,9 +8,18 @@ Required properties: - interrupts : <interrupt mapping for SATA IRQ> - reg : <registers mapping> +Optional properties: +- calxeda,port-phys: phandle-combophy and lane assignment, which maps each + SATA port to a combophy and a lane within that + combophy +- dma-coherent : Present if dma operations are coherent + Example: sata@ffe08000 { compatible = "calxeda,hb-ahci"; reg = <0xffe08000 0x1000>; interrupts = <115>; + calxeda,port-phys = <&combophy5 0 &combophy0 0 &combophy0 1 + &combophy0 2 &combophy0 3>; + }; diff --git a/Documentation/devicetree/bindings/ata/pata-arasan.txt b/Documentation/devicetree/bindings/ata/pata-arasan.txt new file mode 100644 index 0000000..95ec7f8 --- /dev/null +++ b/Documentation/devicetree/bindings/ata/pata-arasan.txt @@ -0,0 +1,17 @@ +* ARASAN PATA COMPACT FLASH CONTROLLER + +Required properties: +- compatible: "arasan,cf-spear1340" +- reg: Address range of the CF registers +- interrupt-parent: Should be the phandle for the interrupt controller + that services interrupts for this device +- interrupt: Should contain the CF interrupt number + +Example: + + cf@fc000000 { + compatible = "arasan,cf-spear1340"; + reg = <0xfc000000 0x1000>; + interrupt-parent = <&vic1>; + interrupts = <12>; + }; diff --git a/Documentation/devicetree/bindings/cpufreq/cpufreq-cpu0.txt b/Documentation/devicetree/bindings/cpufreq/cpufreq-cpu0.txt new file mode 100644 index 0000000..4416ccc --- /dev/null +++ b/Documentation/devicetree/bindings/cpufreq/cpufreq-cpu0.txt @@ -0,0 +1,55 @@ +Generic CPU0 cpufreq driver + +It is a generic cpufreq driver for CPU0 frequency management. It +supports both uniprocessor (UP) and symmetric multiprocessor (SMP) +systems which share clock and voltage across all CPUs. + +Both required and optional properties listed below must be defined +under node /cpus/cpu@0. + +Required properties: +- operating-points: Refer to Documentation/devicetree/bindings/power/opp.txt + for details + +Optional properties: +- clock-latency: Specify the possible maximum transition latency for clock, + in unit of nanoseconds. +- voltage-tolerance: Specify the CPU voltage tolerance in percentage. + +Examples: + +cpus { + #address-cells = <1>; + #size-cells = <0>; + + cpu@0 { + compatible = "arm,cortex-a9"; + reg = <0>; + next-level-cache = <&L2>; + operating-points = < + /* kHz uV */ + 792000 1100000 + 396000 950000 + 198000 850000 + >; + transition-latency = <61036>; /* two CLK32 periods */ + }; + + cpu@1 { + compatible = "arm,cortex-a9"; + reg = <1>; + next-level-cache = <&L2>; + }; + + cpu@2 { + compatible = "arm,cortex-a9"; + reg = <2>; + next-level-cache = <&L2>; + }; + + cpu@3 { + compatible = "arm,cortex-a9"; + reg = <3>; + next-level-cache = <&L2>; + }; +}; diff --git a/Documentation/devicetree/bindings/dma/arm-pl330.txt b/Documentation/devicetree/bindings/dma/arm-pl330.txt index a4cd273..36e27d5 100644 --- a/Documentation/devicetree/bindings/dma/arm-pl330.txt +++ b/Documentation/devicetree/bindings/dma/arm-pl330.txt @@ -9,6 +9,9 @@ Required properties: region. - interrupts: interrupt number to the cpu. +Optional properties: +- dma-coherent : Present if dma operations are coherent + Example: pdma0: pdma@12680000 { diff --git a/Documentation/devicetree/bindings/net/calxeda-xgmac.txt b/Documentation/devicetree/bindings/net/calxeda-xgmac.txt index 411727a..c8ae996 100644 --- a/Documentation/devicetree/bindings/net/calxeda-xgmac.txt +++ b/Documentation/devicetree/bindings/net/calxeda-xgmac.txt @@ -6,6 +6,9 @@ Required properties: - interrupts : Should contain 3 xgmac interrupts. The 1st is main interrupt. The 2nd is pwr mgt interrupt. The 3rd is low power state interrupt. +Optional properties: +- dma-coherent : Present if dma operations are coherent + Example: ethernet@fff50000 { diff --git a/Documentation/devicetree/bindings/power/opp.txt b/Documentation/devicetree/bindings/power/opp.txt new file mode 100644 index 0000000..74499e5 --- /dev/null +++ b/Documentation/devicetree/bindings/power/opp.txt @@ -0,0 +1,25 @@ +* Generic OPP Interface + +SoCs have a standard set of tuples consisting of frequency and +voltage pairs that the device will support per voltage domain. These +are called Operating Performance Points or OPPs. + +Properties: +- operating-points: An array of 2-tuples items, and each item consists + of frequency and voltage like <freq-kHz vol-uV>. + freq: clock frequency in kHz + vol: voltage in microvolt + +Examples: + +cpu@0 { + compatible = "arm,cortex-a9"; + reg = <0>; + next-level-cache = <&L2>; + operating-points = < + /* kHz uV */ + 792000 1100000 + 396000 950000 + 198000 850000 + >; +}; diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt index df43807..f777fa9 100644 --- a/Documentation/kernel-parameters.txt +++ b/Documentation/kernel-parameters.txt @@ -1051,6 +1051,14 @@ bytes respectively. Such letter suffixes can also be entirely omitted. ihash_entries= [KNL] Set number of hash buckets for inode cache. + ima_appraise= [IMA] appraise integrity measurements + Format: { "off" | "enforce" | "fix" } + default: "enforce" + + ima_appraise_tcb [IMA] + The builtin appraise policy appraises all files + owned by uid=0. + ima_audit= [IMA] Format: { "0" | "1" } 0 -- integrity auditing messages. (Default) @@ -1350,6 +1358,9 @@ bytes respectively. Such letter suffixes can also be entirely omitted. * nohrst, nosrst, norst: suppress hard, soft and both resets. + * rstonce: only attempt one reset during + hot-unplug link recovery + * dump_id: dump IDENTIFY data. If there are multiple matching configurations changing diff --git a/Documentation/scsi/ChangeLog.megaraid_sas b/Documentation/scsi/ChangeLog.megaraid_sas index 80441ab..3a30794 100644 --- a/Documentation/scsi/ChangeLog.megaraid_sas +++ b/Documentation/scsi/ChangeLog.megaraid_sas @@ -1,3 +1,13 @@ +Release Date : Tue. Jun 17, 2012 17:00:00 PST 2012 - + (emaild-id:megaraidlinux@lsi.com) + Adam Radford/Kashyap Desai +Current Version : 00.00.06.18-rc1 +Old Version : 00.00.06.15-rc1 + 1. Fix Copyright dates. + 2. Add throttlequeuedepth module parameter. + 3. Add resetwaittime module parameter. + 4. Move poll_aen_lock initializer. +------------------------------------------------------------------------------- Release Date : Mon. Mar 19, 2012 17:00:00 PST 2012 - (emaild-id:megaraidlinux@lsi.com) Adam Radford diff --git a/Documentation/scsi/LICENSE.qla2xxx b/Documentation/scsi/LICENSE.qla2xxx index ce0fdf3..27a91cf 100644 --- a/Documentation/scsi/LICENSE.qla2xxx +++ b/Documentation/scsi/LICENSE.qla2xxx @@ -1,4 +1,4 @@ -Copyright (c) 2003-2011 QLogic Corporation +Copyright (c) 2003-2012 QLogic Corporation QLogic Linux FC-FCoE Driver This program includes a device driver for Linux 3.x. diff --git a/Documentation/scsi/LICENSE.qla4xxx b/Documentation/scsi/LICENSE.qla4xxx index ab89959..78c169f 100644 --- a/Documentation/scsi/LICENSE.qla4xxx +++ b/Documentation/scsi/LICENSE.qla4xxx @@ -1,4 +1,4 @@ -Copyright (c) 2003-2011 QLogic Corporation +Copyright (c) 2003-2012 QLogic Corporation QLogic Linux iSCSI Driver This program includes a device driver for Linux 3.x. diff --git a/Documentation/scsi/st.txt b/Documentation/scsi/st.txt index 685bf35..f346abb 100644 --- a/Documentation/scsi/st.txt +++ b/Documentation/scsi/st.txt @@ -112,10 +112,8 @@ attempted). MINOR NUMBERS -The tape driver currently supports 128 drives by default. This number -can be increased by editing st.h and recompiling the driver if -necessary. The upper limit is 2^17 drives if 4 modes for each drive -are used. +The tape driver currently supports up to 2^17 drives if 4 modes for +each drive are used. The minor numbers consist of the following bit fields: diff --git a/Documentation/security/Smack.txt b/Documentation/security/Smack.txt index a416479..8a177e4 100644 --- a/Documentation/security/Smack.txt +++ b/Documentation/security/Smack.txt @@ -28,12 +28,11 @@ Smack kernels use the CIPSO IP option. Some network configurations are intolerant of IP options and can impede access to systems that use them as Smack does. -The current git repositories for Smack user space are: +The current git repository for Smack user space is: - git@gitorious.org:meego-platform-security/smackutil.git - git@gitorious.org:meego-platform-security/libsmack.git + git://github.com/smack-team/smack.git -These should make and install on most modern distributions. +This should make and install on most modern distributions. There are three commands included in smackutil: smackload - properly formats data for writing to /smack/load @@ -194,6 +193,9 @@ onlycap these capabilities are effective at for processes with any label. The value is set by writing the desired label to the file or cleared by writing "-" to the file. +revoke-subject + Writing a Smack label here sets the access to '-' for all access + rules with that subject label. You can add access rules in /etc/smack/accesses. They take the form: diff --git a/MAINTAINERS b/MAINTAINERS index 7833639..cb009e4 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1650,7 +1650,6 @@ F: drivers/bcma/ F: include/linux/bcma/ BROCADE BFA FC SCSI DRIVER -M: Jing Huang <huangj@brocade.com> M: Krishna C Gudipati <kgudipat@brocade.com> L: linux-scsi@vger.kernel.org S: Supported @@ -3438,6 +3437,13 @@ L: netdev@vger.kernel.org S: Supported F: drivers/net/ethernet/ibm/ibmveth.* +IBM Power Virtual SCSI/FC Device Drivers +M: Robert Jennings <rcj@linux.vnet.ibm.com> +L: linux-scsi@vger.kernel.org +S: Supported +F: drivers/scsi/ibmvscsi/ +X: drivers/scsi/ibmvscsi/ibmvstgt.c + IBM ServeRAID RAID DRIVER P: Jack Hammer M: Dave Jeffery <ipslinux@adaptec.com> diff --git a/arch/alpha/kernel/osf_sys.c b/arch/alpha/kernel/osf_sys.c index 9503a4b..63e77e3 100644 --- a/arch/alpha/kernel/osf_sys.c +++ b/arch/alpha/kernel/osf_sys.c @@ -145,27 +145,24 @@ SYSCALL_DEFINE4(osf_getdirentries, unsigned int, fd, long __user *, basep) { int error; - struct file *file; + struct fd arg = fdget(fd); struct osf_dirent_callback buf; - error = -EBADF; - file = fget(fd); - if (!file) - goto out; + if (!arg.file) + return -EBADF; buf.dirent = dirent; buf.basep = basep; buf.count = count; buf.error = 0; - error = vfs_readdir(file, osf_filldir, &buf); + error = vfs_readdir(arg.file, osf_filldir, &buf); if (error >= 0) error = buf.error; if (count != buf.count) error = count - buf.count; - fput(file); - out: + fdput(arg); return error; } diff --git a/arch/arm/boot/dts/highbank.dts b/arch/arm/boot/dts/highbank.dts index 9fecf1a..0c6fc34 100644 --- a/arch/arm/boot/dts/highbank.dts +++ b/arch/arm/boot/dts/highbank.dts @@ -121,6 +121,10 @@ compatible = "calxeda,hb-ahci"; reg = <0xffe08000 0x10000>; interrupts = <0 83 4>; + calxeda,port-phys = <&combophy5 0 &combophy0 0 + &combophy0 1 &combophy0 2 + &combophy0 3>; + dma-coherent; }; sdhci@ffe0e000 { @@ -306,5 +310,19 @@ reg = <0xfff51000 0x1000>; interrupts = <0 80 4 0 81 4 0 82 4>; }; + + combophy0: combo-phy@fff58000 { + compatible = "calxeda,hb-combophy"; + #phy-cells = <1>; + reg = <0xfff58000 0x1000>; + phydev = <5>; + }; + + combophy5: combo-phy@fff5d000 { + compatible = "calxeda,hb-combophy"; + #phy-cells = <1>; + reg = <0xfff5d000 0x1000>; + phydev = <31>; + }; }; }; diff --git a/arch/arm/configs/sam9_l9260_defconfig b/arch/arm/configs/sam9_l9260_defconfig index ecf2531..b4384af 100644 --- a/arch/arm/configs/sam9_l9260_defconfig +++ b/arch/arm/configs/sam9_l9260_defconfig @@ -39,7 +39,7 @@ CONFIG_MTD_NAND=y CONFIG_MTD_NAND_ATMEL=y CONFIG_MTD_NAND_PLATFORM=y CONFIG_MTD_UBI=y -CONFIG_MTD_UBI_BEB_RESERVE=3 +CONFIG_MTD_UBI_BEB_LIMIT=25 CONFIG_MTD_UBI_GLUEBI=y CONFIG_BLK_DEV_LOOP=y CONFIG_BLK_DEV_RAM=y diff --git a/arch/arm/include/asm/barrier.h b/arch/arm/include/asm/barrier.h index 0511238..8dcd9c7 100644 --- a/arch/arm/include/asm/barrier.h +++ b/arch/arm/include/asm/barrier.h @@ -44,10 +44,9 @@ #define rmb() dsb() #define wmb() mb() #else -#include <asm/memory.h> -#define mb() do { if (arch_is_coherent()) dmb(); else barrier(); } while (0) -#define rmb() do { if (arch_is_coherent()) dmb(); else barrier(); } while (0) -#define wmb() do { if (arch_is_coherent()) dmb(); else barrier(); } while (0) +#define mb() barrier() +#define rmb() barrier() +#define wmb() barrier() #endif #ifndef CONFIG_SMP diff --git a/arch/arm/include/asm/dma-mapping.h b/arch/arm/include/asm/dma-mapping.h index 5c44dcb..2300484 100644 --- a/arch/arm/include/asm/dma-mapping.h +++ b/arch/arm/include/asm/dma-mapping.h @@ -13,6 +13,7 @@ #define DMA_ERROR_CODE (~0) extern struct dma_map_ops arm_dma_ops; +extern struct dma_map_ops arm_coherent_dma_ops; static inline struct dma_map_ops *get_dma_ops(struct device *dev) { diff --git a/arch/arm/include/asm/memory.h b/arch/arm/include/asm/memory.h index 5f6ddcc..73cf03a 100644 --- a/arch/arm/include/asm/memory.h +++ b/arch/arm/include/asm/memory.h @@ -275,14 +275,6 @@ static inline __deprecated void *bus_to_virt(unsigned long x) #define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT) #define virt_addr_valid(kaddr) ((unsigned long)(kaddr) >= PAGE_OFFSET && (unsigned long)(kaddr) < (unsigned long)high_memory) -/* - * Optional coherency support. Currently used only by selected - * Intel XSC3-based systems. - */ -#ifndef arch_is_coherent -#define arch_is_coherent() 0 -#endif - #endif #include <asm-generic/memory_model.h> diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c index aa4ffe6..dea7a92 100644 --- a/arch/arm/kernel/smp.c +++ b/arch/arm/kernel/smp.c @@ -24,6 +24,7 @@ #include <linux/percpu.h> #include <linux/clockchips.h> #include <linux/completion.h> +#include <linux/cpufreq.h> #include <linux/atomic.h> #include <asm/smp.h> @@ -650,3 +651,56 @@ int setup_profiling_timer(unsigned int multiplier) { return -EINVAL; } + +#ifdef CONFIG_CPU_FREQ + +static DEFINE_PER_CPU(unsigned long, l_p_j_ref); +static DEFINE_PER_CPU(unsigned long, l_p_j_ref_freq); +static unsigned long global_l_p_j_ref; +static unsigned long global_l_p_j_ref_freq; + +static int cpufreq_callback(struct notifier_block *nb, + unsigned long val, void *data) +{ + struct cpufreq_freqs *freq = data; + int cpu = freq->cpu; + + if (freq->flags & CPUFREQ_CONST_LOOPS) + return NOTIFY_OK; + + if (!per_cpu(l_p_j_ref, cpu)) { + per_cpu(l_p_j_ref, cpu) = + per_cpu(cpu_data, cpu).loops_per_jiffy; + per_cpu(l_p_j_ref_freq, cpu) = freq->old; + if (!global_l_p_j_ref) { + global_l_p_j_ref = loops_per_jiffy; + global_l_p_j_ref_freq = freq->old; + } + } + + if ((val == CPUFREQ_PRECHANGE && freq->old < freq->new) || + (val == CPUFREQ_POSTCHANGE && freq->old > freq->new) || + (val == CPUFREQ_RESUMECHANGE || val == CPUFREQ_SUSPENDCHANGE)) { + loops_per_jiffy = cpufreq_scale(global_l_p_j_ref, + global_l_p_j_ref_freq, + freq->new); + per_cpu(cpu_data, cpu).loops_per_jiffy = + cpufreq_scale(per_cpu(l_p_j_ref, cpu), + per_cpu(l_p_j_ref_freq, cpu), + freq->new); + } + return NOTIFY_OK; +} + +static struct notifier_block cpufreq_notifier = { + .notifier_call = cpufreq_callback, +}; + +static int __init register_cpufreq_notifier(void) +{ + return cpufreq_register_notifier(&cpufreq_notifier, + CPUFREQ_TRANSITION_NOTIFIER); +} +core_initcall(register_cpufreq_notifier); + +#endif diff --git a/arch/arm/mach-highbank/highbank.c b/arch/arm/mach-highbank/highbank.c index af1da34..40e36a5 100644 --- a/arch/arm/mach-highbank/highbank.c +++ b/arch/arm/mach-highbank/highbank.c @@ -15,6 +15,7 @@ */ #include <linux/clk.h> #include <linux/clkdev.h> +#include <linux/dma-mapping.h> #include <linux/io.h> #include <linux/irq.h> #include <linux/irqdomain.h> @@ -23,6 +24,7 @@ #include <linux/of_platform.h> #include <linux/of_address.h> #include <linux/smp.h> +#include <linux/amba/bus.h> #include <asm/cacheflush.h> #include <asm/smp_plat.h> @@ -149,11 +151,61 @@ static void highbank_power_off(void) cpu_do_idle(); } +static int highbank_platform_notifier(struct notifier_block *nb, + unsigned long event, void *__dev) +{ + struct resource *res; + int reg = -1; + struct device *dev = __dev; + + if (event != BUS_NOTIFY_ADD_DEVICE) + return NOTIFY_DONE; + + if (of_device_is_compatible(dev->of_node, "calxeda,hb-ahci")) + reg = 0xc; + else if (of_device_is_compatible(dev->of_node, "calxeda,hb-sdhci")) + reg = 0x18; + else if (of_device_is_compatible(dev->of_node, "arm,pl330")) + reg = 0x20; + else if (of_device_is_compatible(dev->of_node, "calxeda,hb-xgmac")) { + res = platform_get_resource(to_platform_device(dev), + IORESOURCE_MEM, 0); + if (res) { + if (res->start == 0xfff50000) + reg = 0; + else if (res->start == 0xfff51000) + reg = 4; + } + } + + if (reg < 0) + return NOTIFY_DONE; + + if (of_property_read_bool(dev->of_node, "dma-coherent")) { + writel(0xff31, sregs_base + reg); + set_dma_ops(dev, &arm_coherent_dma_ops); + } else + writel(0, sregs_base + reg); + + return NOTIFY_OK; +} + +static struct notifier_block highbank_amba_nb = { + .notifier_call = highbank_platform_notifier, +}; + +static struct notifier_block highbank_platform_nb = { + .notifier_call = highbank_platform_notifier, +}; + static void __init highbank_init(void) { pm_power_off = highbank_power_off; highbank_pm_init(); + bus_register_notifier(&platform_bus_type, &highbank_platform_nb); + bus_register_notifier(&amba_bustype, &highbank_amba_nb); + of_platform_populate(NULL, of_default_bus_match_table, NULL, NULL); } diff --git a/arch/arm/mach-shmobile/Makefile b/arch/arm/mach-shmobile/Makefile index 0df5ae6..fe2c97c 100644 --- a/arch/arm/mach-shmobile/Makefile +++ b/arch/arm/mach-shmobile/Makefile @@ -3,7 +3,7 @@ # # Common objects -obj-y := timer.o console.o clock.o common.o +obj-y := timer.o console.o clock.o # CPU objects obj-$(CONFIG_ARCH_SH7367) += setup-sh7367.o clock-sh7367.o intc-sh7367.o diff --git a/arch/arm/mach-shmobile/board-ap4evb.c b/arch/arm/mach-shmobile/board-ap4evb.c index bc3b5da..790dc68 100644 --- a/arch/arm/mach-shmobile/board-ap4evb.c +++ b/arch/arm/mach-shmobile/board-ap4evb.c @@ -1231,6 +1231,15 @@ static struct i2c_board_info i2c1_devices[] = { #define USCCR1 IOMEM(0xE6058144) static void __init ap4evb_init(void) { + struct pm_domain_device domain_devices[] = { + { "A4LC", &lcdc1_device, }, + { "A4LC", &lcdc_device, }, + { "A4MP", &fsi_device, }, + { "A3SP", &sh_mmcif_device, }, + { "A3SP", &sdhi0_device, }, + { "A3SP", &sdhi1_device, }, + { "A4R", &ceu_device, }, + }; u32 srcr4; struct clk *clk; @@ -1463,14 +1472,8 @@ static void __init ap4evb_init(void) platform_add_devices(ap4evb_devices, ARRAY_SIZE(ap4evb_devices)); - rmobile_add_device_to_domain(&sh7372_pd_a4lc, &lcdc1_device); - rmobile_add_device_to_domain(&sh7372_pd_a4lc, &lcdc_device); - rmobile_add_device_to_domain(&sh7372_pd_a4mp, &fsi_device); - - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sh_mmcif_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sdhi0_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sdhi1_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &ceu_device); + rmobile_add_devices_to_domains(domain_devices, + ARRAY_SIZE(domain_devices)); hdmi_init_pm_clock(); fsi_init_pm_clock(); @@ -1485,6 +1488,6 @@ MACHINE_START(AP4EVB, "ap4evb") .init_irq = sh7372_init_irq, .handle_irq = shmobile_handle_irq_intc, .init_machine = ap4evb_init, - .init_late = shmobile_init_late, + .init_late = sh7372_pm_init_late, .timer = &shmobile_timer, MACHINE_END diff --git a/arch/arm/mach-shmobile/board-armadillo800eva.c b/arch/arm/mach-shmobile/board-armadillo800eva.c index c6593d3..2912eab 100644 --- a/arch/arm/mach-shmobile/board-armadillo800eva.c +++ b/arch/arm/mach-shmobile/board-armadillo800eva.c @@ -1209,10 +1209,10 @@ static void __init eva_init(void) eva_clock_init(); - rmobile_add_device_to_domain(&r8a7740_pd_a4lc, &lcdc0_device); - rmobile_add_device_to_domain(&r8a7740_pd_a4lc, &hdmi_lcdc_device); + rmobile_add_device_to_domain("A4LC", &lcdc0_device); + rmobile_add_device_to_domain("A4LC", &hdmi_lcdc_device); if (usb) - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, usb); + rmobile_add_device_to_domain("A3SP", usb); } static void __init eva_earlytimer_init(void) diff --git a/arch/arm/mach-shmobile/board-mackerel.c b/arch/arm/mach-shmobile/board-mackerel.c index 62783b5..0c27c81 100644 --- a/arch/arm/mach-shmobile/board-mackerel.c +++ b/arch/arm/mach-shmobile/board-mackerel.c @@ -1412,6 +1412,22 @@ static struct i2c_board_info i2c1_devices[] = { #define USCCR1 IOMEM(0xE6058144) static void __init mackerel_init(void) { + struct pm_domain_device domain_devices[] = { + { "A4LC", &lcdc_device, }, + { "A4LC", &hdmi_lcdc_device, }, + { "A4LC", &meram_device, }, + { "A4MP", &fsi_device, }, + { "A3SP", &usbhs0_device, }, + { "A3SP", &usbhs1_device, }, + { "A3SP", &nand_flash_device, }, + { "A3SP", &sh_mmcif_device, }, + { "A3SP", &sdhi0_device, }, +#if !defined(CONFIG_MMC_SH_MMCIF) && !defined(CONFIG_MMC_SH_MMCIF_MODULE) + { "A3SP", &sdhi1_device, }, +#endif + { "A3SP", &sdhi2_device, }, + { "A4R", &ceu_device, }, + }; u32 srcr4; struct clk *clk; @@ -1626,20 +1642,8 @@ static void __init mackerel_init(void) platform_add_devices(mackerel_devices, ARRAY_SIZE(mackerel_devices)); - rmobile_add_device_to_domain(&sh7372_pd_a4lc, &lcdc_device); - rmobile_add_device_to_domain(&sh7372_pd_a4lc, &hdmi_lcdc_device); - rmobile_add_device_to_domain(&sh7372_pd_a4lc, &meram_device); - rmobile_add_device_to_domain(&sh7372_pd_a4mp, &fsi_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &usbhs0_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &usbhs1_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &nand_flash_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sh_mmcif_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sdhi0_device); -#if !defined(CONFIG_MMC_SH_MMCIF) && !defined(CONFIG_MMC_SH_MMCIF_MODULE) - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sdhi1_device); -#endif - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &sdhi2_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &ceu_device); + rmobile_add_devices_to_domains(domain_devices, + ARRAY_SIZE(domain_devices)); hdmi_init_pm_clock(); sh7372_pm_init(); @@ -1653,6 +1657,6 @@ MACHINE_START(MACKEREL, "mackerel") .init_irq = sh7372_init_irq, .handle_irq = shmobile_handle_irq_intc, .init_machine = mackerel_init, - .init_late = shmobile_init_late, + .init_late = sh7372_pm_init_late, .timer = &shmobile_timer, MACHINE_END diff --git a/arch/arm/mach-shmobile/common.c b/arch/arm/mach-shmobile/common.c deleted file mode 100644 index 608aba9..0000000 --- a/arch/arm/mach-shmobile/common.c +++ /dev/null @@ -1,24 +0,0 @@ -/* - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; version 2 of the License. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA - * - */ -#include <linux/kernel.h> -#include <linux/init.h> -#include <mach/common.h> - -void __init shmobile_init_late(void) -{ - shmobile_suspend_init(); - shmobile_cpuidle_init(); -} diff --git a/arch/arm/mach-shmobile/cpuidle.c b/arch/arm/mach-shmobile/cpuidle.c index 7b541e91..9e05026 100644 --- a/arch/arm/mach-shmobile/cpuidle.c +++ b/arch/arm/mach-shmobile/cpuidle.c @@ -16,51 +16,38 @@ #include <asm/cpuidle.h> #include <asm/io.h> -static void shmobile_enter_wfi(void) +int shmobile_enter_wfi(struct cpuidle_device *dev, struct cpuidle_driver *drv, + int index) { cpu_do_idle(); -} - -void (*shmobile_cpuidle_modes[CPUIDLE_STATE_MAX])(void) = { - shmobile_enter_wfi, /* regular sleep mode */ -}; - -static int shmobile_cpuidle_enter(struct cpuidle_device *dev, - struct cpuidle_driver *drv, - int index) -{ - shmobile_cpuidle_modes[index](); - - return index; + return 0; } static struct cpuidle_device shmobile_cpuidle_dev; -static struct cpuidle_driver shmobile_cpuidle_driver = { +static struct cpuidle_driver shmobile_cpuidle_default_driver = { .name = "shmobile_cpuidle", .owner = THIS_MODULE, .en_core_tk_irqen = 1, .states[0] = ARM_CPUIDLE_WFI_STATE, + .states[0].enter = shmobile_enter_wfi, .safe_state_index = 0, /* C1 */ .state_count = 1, }; -void (*shmobile_cpuidle_setup)(struct cpuidle_driver *drv); +static struct cpuidle_driver *cpuidle_drv = &shmobile_cpuidle_default_driver; + +void shmobile_cpuidle_set_driver(struct cpuidle_driver *drv) +{ + cpuidle_drv = drv; +} int shmobile_cpuidle_init(void) { struct cpuidle_device *dev = &shmobile_cpuidle_dev; - struct cpuidle_driver *drv = &shmobile_cpuidle_driver; - int i; - - for (i = 0; i < CPUIDLE_STATE_MAX; i++) - drv->states[i].enter = shmobile_cpuidle_enter; - - if (shmobile_cpuidle_setup) - shmobile_cpuidle_setup(drv); - cpuidle_register_driver(drv); + cpuidle_register_driver(cpuidle_drv); - dev->state_count = drv->state_count; + dev->state_count = cpuidle_drv->state_count; cpuidle_register_device(dev); return 0; diff --git a/arch/arm/mach-shmobile/include/mach/common.h b/arch/arm/mach-shmobile/include/mach/common.h index f80f9c5..ed77ab8 100644 --- a/arch/arm/mach-shmobile/include/mach/common.h +++ b/arch/arm/mach-shmobile/include/mach/common.h @@ -13,8 +13,10 @@ extern int shmobile_clk_init(void); extern void shmobile_handle_irq_intc(struct pt_regs *); extern struct platform_suspend_ops shmobile_suspend_ops; struct cpuidle_driver; -extern void (*shmobile_cpuidle_modes[])(void); -extern void (*shmobile_cpuidle_setup)(struct cpuidle_driver *drv); +struct cpuidle_device; +extern int shmobile_enter_wfi(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index); +extern void shmobile_cpuidle_set_driver(struct cpuidle_driver *drv); extern void sh7367_init_irq(void); extern void sh7367_map_io(void); @@ -75,8 +77,6 @@ extern void r8a7740_meram_workaround(void); extern void r8a7779_register_twd(void); -extern void shmobile_init_late(void); - #ifdef CONFIG_SUSPEND int shmobile_suspend_init(void); #else @@ -100,4 +100,10 @@ static inline int shmobile_cpu_is_dead(unsigned int cpu) { return 1; } extern void shmobile_smp_init_cpus(unsigned int ncores); +static inline void shmobile_init_late(void) +{ + shmobile_suspend_init(); + shmobile_cpuidle_init(); +} + #endif /* __ARCH_MACH_COMMON_H */ diff --git a/arch/arm/mach-shmobile/include/mach/pm-rmobile.h b/arch/arm/mach-shmobile/include/mach/pm-rmobile.h index 5a40284..690553a 100644 --- a/arch/arm/mach-shmobile/include/mach/pm-rmobile.h +++ b/arch/arm/mach-shmobile/include/mach/pm-rmobile.h @@ -12,6 +12,8 @@ #include <linux/pm_domain.h> +#define DEFAULT_DEV_LATENCY_NS 250000 + struct platform_device; struct rmobile_pm_domain { @@ -29,16 +31,33 @@ struct rmobile_pm_domain *to_rmobile_pd(struct generic_pm_domain *d) return container_of(d, struct rmobile_pm_domain, genpd); } +struct pm_domain_device { + const char *domain_name; + struct platform_device *pdev; +}; + #ifdef CONFIG_PM -extern void rmobile_init_pm_domain(struct rmobile_pm_domain *rmobile_pd); -extern void rmobile_add_device_to_domain(struct rmobile_pm_domain *rmobile_pd, - struct platform_device *pdev); -extern void rmobile_pm_add_subdomain(struct rmobile_pm_domain *rmobile_pd, - struct rmobile_pm_domain *rmobile_sd); +extern void rmobile_init_domains(struct rmobile_pm_domain domains[], int num); +extern void rmobile_add_device_to_domain_td(const char *domain_name, + struct platform_device *pdev, + struct gpd_timing_data *td); + +static inline void rmobile_add_device_to_domain(const char *domain_name, + struct platform_device *pdev) +{ + rmobile_add_device_to_domain_td(domain_name, pdev, NULL); +} + +extern void rmobile_add_devices_to_domains(struct pm_domain_device data[], + int size); #else -#define rmobile_init_pm_domain(pd) do { } while (0) -#define rmobile_add_device_to_domain(pd, pdev) do { } while (0) -#define rmobile_pm_add_subdomain(pd, sd) do { } while (0) + +#define rmobile_init_domains(domains, num) do { } while (0) +#define rmobile_add_device_to_domain_td(name, pdev, td) do { } while (0) +#define rmobile_add_device_to_domain(name, pdev) do { } while (0) + +static inline void rmobile_add_devices_to_domains(struct pm_domain_device d[], + int size) {} #endif /* CONFIG_PM */ #endif /* PM_RMOBILE_H */ diff --git a/arch/arm/mach-shmobile/include/mach/r8a7740.h b/arch/arm/mach-shmobile/include/mach/r8a7740.h index 7143147..59d252f 100644 --- a/arch/arm/mach-shmobile/include/mach/r8a7740.h +++ b/arch/arm/mach-shmobile/include/mach/r8a7740.h @@ -607,9 +607,9 @@ enum { }; #ifdef CONFIG_PM -extern struct rmobile_pm_domain r8a7740_pd_a4s; -extern struct rmobile_pm_domain r8a7740_pd_a3sp; -extern struct rmobile_pm_domain r8a7740_pd_a4lc; +extern void __init r8a7740_init_pm_domains(void); +#else +static inline void r8a7740_init_pm_domains(void) {} #endif /* CONFIG_PM */ #endif /* __ASM_R8A7740_H__ */ diff --git a/arch/arm/mach-shmobile/include/mach/r8a7779.h b/arch/arm/mach-shmobile/include/mach/r8a7779.h index f504c5e..499f52d 100644 --- a/arch/arm/mach-shmobile/include/mach/r8a7779.h +++ b/arch/arm/mach-shmobile/include/mach/r8a7779.h @@ -347,17 +347,9 @@ extern int r8a7779_sysc_power_down(struct r8a7779_pm_ch *r8a7779_ch); extern int r8a7779_sysc_power_up(struct r8a7779_pm_ch *r8a7779_ch); #ifdef CONFIG_PM -extern struct r8a7779_pm_domain r8a7779_sh4a; -extern struct r8a7779_pm_domain r8a7779_sgx; -extern struct r8a7779_pm_domain r8a7779_vdp1; -extern struct r8a7779_pm_domain r8a7779_impx3; - -extern void r8a7779_init_pm_domain(struct r8a7779_pm_domain *r8a7779_pd); -extern void r8a7779_add_device_to_domain(struct r8a7779_pm_domain *r8a7779_pd, - struct platform_device *pdev); +extern void __init r8a7779_init_pm_domains(void); #else -#define r8a7779_init_pm_domain(pd) do { } while (0) -#define r8a7779_add_device_to_domain(pd, pdev) do { } while (0) +static inline void r8a7779_init_pm_domains(void) {} #endif /* CONFIG_PM */ extern struct smp_operations r8a7779_smp_ops; diff --git a/arch/arm/mach-shmobile/include/mach/sh7372.h b/arch/arm/mach-shmobile/include/mach/sh7372.h index b59048e..eb98b45 100644 --- a/arch/arm/mach-shmobile/include/mach/sh7372.h +++ b/arch/arm/mach-shmobile/include/mach/sh7372.h @@ -478,21 +478,17 @@ extern struct clk sh7372_fsibck_clk; extern struct clk sh7372_fsidiva_clk; extern struct clk sh7372_fsidivb_clk; -#ifdef CONFIG_PM -extern struct rmobile_pm_domain sh7372_pd_a4lc; -extern struct rmobile_pm_domain sh7372_pd_a4mp; -extern struct rmobile_pm_domain sh7372_pd_d4; -extern struct rmobile_pm_domain sh7372_pd_a4r; -extern struct rmobile_pm_domain sh7372_pd_a3rv; -extern struct rmobile_pm_domain sh7372_pd_a3ri; -extern struct rmobile_pm_domain sh7372_pd_a4s; -extern struct rmobile_pm_domain sh7372_pd_a3sp; -extern struct rmobile_pm_domain sh7372_pd_a3sg; -#endif /* CONFIG_PM */ - extern void sh7372_intcs_suspend(void); extern void sh7372_intcs_resume(void); extern void sh7372_intca_suspend(void); extern void sh7372_intca_resume(void); +#ifdef CONFIG_PM +extern void __init sh7372_init_pm_domains(void); +#else +static inline void sh7372_init_pm_domains(void) {} +#endif + +extern void __init sh7372_pm_init_late(void); + #endif /* __ASM_SH7372_H__ */ diff --git a/arch/arm/mach-shmobile/pm-r8a7740.c b/arch/arm/mach-shmobile/pm-r8a7740.c index 893504d..21e5316d 100644 --- a/arch/arm/mach-shmobile/pm-r8a7740.c +++ b/arch/arm/mach-shmobile/pm-r8a7740.c @@ -21,14 +21,6 @@ static int r8a7740_pd_a4s_suspend(void) return -EBUSY; } -struct rmobile_pm_domain r8a7740_pd_a4s = { - .genpd.name = "A4S", - .bit_shift = 10, - .gov = &pm_domain_always_on_gov, - .no_debug = true, - .suspend = r8a7740_pd_a4s_suspend, -}; - static int r8a7740_pd_a3sp_suspend(void) { /* @@ -38,17 +30,31 @@ static int r8a7740_pd_a3sp_suspend(void) return console_suspend_enabled ? 0 : -EBUSY; } -struct rmobile_pm_domain r8a7740_pd_a3sp = { - .genpd.name = "A3SP", - .bit_shift = 11, - .gov = &pm_domain_always_on_gov, - .no_debug = true, - .suspend = r8a7740_pd_a3sp_suspend, +static struct rmobile_pm_domain r8a7740_pm_domains[] = { + { + .genpd.name = "A4S", + .bit_shift = 10, + .gov = &pm_domain_always_on_gov, + .no_debug = true, + .suspend = r8a7740_pd_a4s_suspend, + }, + { + .genpd.name = "A3SP", + .bit_shift = 11, + .gov = &pm_domain_always_on_gov, + .no_debug = true, + .suspend = r8a7740_pd_a3sp_suspend, + }, + { + .genpd.name = "A4LC", + .bit_shift = 1, + }, }; -struct rmobile_pm_domain r8a7740_pd_a4lc = { - .genpd.name = "A4LC", - .bit_shift = 1, -}; +void __init r8a7740_init_pm_domains(void) +{ + rmobile_init_domains(r8a7740_pm_domains, ARRAY_SIZE(r8a7740_pm_domains)); + pm_genpd_add_subdomain_names("A4S", "A3SP"); +} #endif /* CONFIG_PM */ diff --git a/arch/arm/mach-shmobile/pm-r8a7779.c b/arch/arm/mach-shmobile/pm-r8a7779.c index a18a4ae..d50a8e9 100644 --- a/arch/arm/mach-shmobile/pm-r8a7779.c +++ b/arch/arm/mach-shmobile/pm-r8a7779.c @@ -183,7 +183,7 @@ static bool pd_active_wakeup(struct device *dev) return true; } -void r8a7779_init_pm_domain(struct r8a7779_pm_domain *r8a7779_pd) +static void r8a7779_init_pm_domain(struct r8a7779_pm_domain *r8a7779_pd) { struct generic_pm_domain *genpd = &r8a7779_pd->genpd; @@ -199,43 +199,44 @@ void r8a7779_init_pm_domain(struct r8a7779_pm_domain *r8a7779_pd) pd_power_up(&r8a7779_pd->genpd); } -void r8a7779_add_device_to_domain(struct r8a7779_pm_domain *r8a7779_pd, - struct platform_device *pdev) -{ - struct device *dev = &pdev->dev; - - pm_genpd_add_device(&r8a7779_pd->genpd, dev); - if (pm_clk_no_clocks(dev)) - pm_clk_add(dev, NULL); -} - -struct r8a7779_pm_domain r8a7779_sh4a = { - .ch = { - .chan_offs = 0x80, /* PWRSR1 .. PWRER1 */ - .isr_bit = 16, /* SH4A */ - } -}; - -struct r8a7779_pm_domain r8a7779_sgx = { - .ch = { - .chan_offs = 0xc0, /* PWRSR2 .. PWRER2 */ - .isr_bit = 20, /* SGX */ - } +static struct r8a7779_pm_domain r8a7779_pm_domains[] = { + { + .genpd.name = "SH4A", + .ch = { + .chan_offs = 0x80, /* PWRSR1 .. PWRER1 */ + .isr_bit = 16, /* SH4A */ + }, + }, + { + .genpd.name = "SGX", + .ch = { + .chan_offs = 0xc0, /* PWRSR2 .. PWRER2 */ + .isr_bit = 20, /* SGX */ + }, + }, + { + .genpd.name = "VDP1", + .ch = { + .chan_offs = 0x100, /* PWRSR3 .. PWRER3 */ + .isr_bit = 21, /* VDP */ + }, + }, + { + .genpd.name = "IMPX3", + .ch = { + .chan_offs = 0x140, /* PWRSR4 .. PWRER4 */ + .isr_bit = 24, /* IMP */ + }, + }, }; -struct r8a7779_pm_domain r8a7779_vdp1 = { - .ch = { - .chan_offs = 0x100, /* PWRSR3 .. PWRER3 */ - .isr_bit = 21, /* VDP */ - } -}; +void __init r8a7779_init_pm_domains(void) +{ + int j; -struct r8a7779_pm_domain r8a7779_impx3 = { - .ch = { - .chan_offs = 0x140, /* PWRSR4 .. PWRER4 */ - .isr_bit = 24, /* IMP */ - } -}; + for (j = 0; j < ARRAY_SIZE(r8a7779_pm_domains); j++) + r8a7779_init_pm_domain(&r8a7779_pm_domains[j]); +} #endif /* CONFIG_PM */ diff --git a/arch/arm/mach-shmobile/pm-rmobile.c b/arch/arm/mach-shmobile/pm-rmobile.c index 32e1772..1fc05d9 100644 --- a/arch/arm/mach-shmobile/pm-rmobile.c +++ b/arch/arm/mach-shmobile/pm-rmobile.c @@ -134,7 +134,7 @@ static int rmobile_pd_start_dev(struct device *dev) return ret; } -void rmobile_init_pm_domain(struct rmobile_pm_domain *rmobile_pd) +static void rmobile_init_pm_domain(struct rmobile_pm_domain *rmobile_pd) { struct generic_pm_domain *genpd = &rmobile_pd->genpd; struct dev_power_governor *gov = rmobile_pd->gov; @@ -149,19 +149,38 @@ void rmobile_init_pm_domain(struct rmobile_pm_domain *rmobile_pd) __rmobile_pd_power_up(rmobile_pd, false); } -void rmobile_add_device_to_domain(struct rmobile_pm_domain *rmobile_pd, - struct platform_device *pdev) +void rmobile_init_domains(struct rmobile_pm_domain domains[], int num) +{ + int j; + + for (j = 0; j < num; j++) + rmobile_init_pm_domain(&domains[j]); +} + +void rmobile_add_device_to_domain_td(const char *domain_name, + struct platform_device *pdev, + struct gpd_timing_data *td) { struct device *dev = &pdev->dev; - pm_genpd_add_device(&rmobile_pd->genpd, dev); + __pm_genpd_name_add_device(domain_name, dev, td); if (pm_clk_no_clocks(dev)) pm_clk_add(dev, NULL); } -void rmobile_pm_add_subdomain(struct rmobile_pm_domain *rmobile_pd, - struct rmobile_pm_domain *rmobile_sd) +void rmobile_add_devices_to_domains(struct pm_domain_device data[], + int size) { - pm_genpd_add_subdomain(&rmobile_pd->genpd, &rmobile_sd->genpd); + struct gpd_timing_data latencies = { + .stop_latency_ns = DEFAULT_DEV_LATENCY_NS, + .start_latency_ns = DEFAULT_DEV_LATENCY_NS, + .save_state_latency_ns = DEFAULT_DEV_LATENCY_NS, + .restore_state_latency_ns = DEFAULT_DEV_LATENCY_NS, + }; + int j; + + for (j = 0; j < size; j++) + rmobile_add_device_to_domain_td(data[j].domain_name, + data[j].pdev, &latencies); } #endif /* CONFIG_PM */ diff --git a/arch/arm/mach-shmobile/pm-sh7372.c b/arch/arm/mach-shmobile/pm-sh7372.c index 1621218..a0826a4 100644 --- a/arch/arm/mach-shmobile/pm-sh7372.c +++ b/arch/arm/mach-shmobile/pm-sh7372.c @@ -21,6 +21,7 @@ #include <linux/irq.h> #include <linux/bitrev.h> #include <linux/console.h> +#include <asm/cpuidle.h> #include <asm/io.h> #include <asm/tlbflush.h> #include <asm/suspend.h> @@ -72,20 +73,7 @@ #ifdef CONFIG_PM -struct rmobile_pm_domain sh7372_pd_a4lc = { - .genpd.name = "A4LC", - .bit_shift = 1, -}; - -struct rmobile_pm_domain sh7372_pd_a4mp = { - .genpd.name = "A4MP", - .bit_shift = 2, -}; - -struct rmobile_pm_domain sh7372_pd_d4 = { - .genpd.name = "D4", - .bit_shift = 3, -}; +#define PM_DOMAIN_ON_OFF_LATENCY_NS 250000 static int sh7372_a4r_pd_suspend(void) { @@ -94,39 +82,25 @@ static int sh7372_a4r_pd_suspend(void) return 0; } -struct rmobile_pm_domain sh7372_pd_a4r = { - .genpd.name = "A4R", - .bit_shift = 5, - .suspend = sh7372_a4r_pd_suspend, - .resume = sh7372_intcs_resume, -}; +static bool a4s_suspend_ready; -struct rmobile_pm_domain sh7372_pd_a3rv = { - .genpd.name = "A3RV", - .bit_shift = 6, -}; - -struct rmobile_pm_domain sh7372_pd_a3ri = { - .genpd.name = "A3RI", - .bit_shift = 8, -}; - -static int sh7372_pd_a4s_suspend(void) +static int sh7372_a4s_pd_suspend(void) { /* * The A4S domain contains the CPU core and therefore it should - * only be turned off if the CPU is in use. + * only be turned off if the CPU is not in use. This may happen + * during system suspend, when SYSC is going to be used for generating + * resume signals and a4s_suspend_ready is set to let + * sh7372_enter_suspend() know that it can turn A4S off. */ + a4s_suspend_ready = true; return -EBUSY; } -struct rmobile_pm_domain sh7372_pd_a4s = { - .genpd.name = "A4S", - .bit_shift = 10, - .gov = &pm_domain_always_on_gov, - .no_debug = true, - .suspend = sh7372_pd_a4s_suspend, -}; +static void sh7372_a4s_pd_resume(void) +{ + a4s_suspend_ready = false; +} static int sh7372_a3sp_pd_suspend(void) { @@ -137,18 +111,80 @@ static int sh7372_a3sp_pd_suspend(void) return console_suspend_enabled ? 0 : -EBUSY; } -struct rmobile_pm_domain sh7372_pd_a3sp = { - .genpd.name = "A3SP", - .bit_shift = 11, - .gov = &pm_domain_always_on_gov, - .no_debug = true, - .suspend = sh7372_a3sp_pd_suspend, +static struct rmobile_pm_domain sh7372_pm_domains[] = { + { + .genpd.name = "A4LC", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 1, + }, + { + .genpd.name = "A4MP", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 2, + }, + { + .genpd.name = "D4", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 3, + }, + { + .genpd.name = "A4R", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 5, + .suspend = sh7372_a4r_pd_suspend, + .resume = sh7372_intcs_resume, + }, + { + .genpd.name = "A3RV", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 6, + }, + { + .genpd.name = "A3RI", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 8, + }, + { + .genpd.name = "A4S", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 10, + .gov = &pm_domain_always_on_gov, + .no_debug = true, + .suspend = sh7372_a4s_pd_suspend, + .resume = sh7372_a4s_pd_resume, + }, + { + .genpd.name = "A3SP", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 11, + .gov = &pm_domain_always_on_gov, + .no_debug = true, + .suspend = sh7372_a3sp_pd_suspend, + }, + { + .genpd.name = "A3SG", + .genpd.power_on_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .genpd.power_off_latency_ns = PM_DOMAIN_ON_OFF_LATENCY_NS, + .bit_shift = 13, + }, }; -struct rmobile_pm_domain sh7372_pd_a3sg = { - .genpd.name = "A3SG", - .bit_shift = 13, -}; +void __init sh7372_init_pm_domains(void) +{ + rmobile_init_domains(sh7372_pm_domains, ARRAY_SIZE(sh7372_pm_domains)); + pm_genpd_add_subdomain_names("A4LC", "A3RV"); + pm_genpd_add_subdomain_names("A4R", "A4LC"); + pm_genpd_add_subdomain_names("A4S", "A3SG"); + pm_genpd_add_subdomain_names("A4S", "A3SP"); +} #endif /* CONFIG_PM */ @@ -304,6 +340,21 @@ static void sh7372_enter_a3sm_common(int pllc0_on) sh7372_set_reset_vector(__pa(sh7372_resume_core_standby_sysc)); sh7372_enter_sysc(pllc0_on, 1 << 12); } + +static void sh7372_enter_a4s_common(int pllc0_on) +{ + sh7372_intca_suspend(); + sh7372_set_reset_vector(SMFRAM); + sh7372_enter_sysc(pllc0_on, 1 << 10); + sh7372_intca_resume(); +} + +static void sh7372_pm_setup_smfram(void) +{ + memcpy((void *)SMFRAM, sh7372_resume_core_standby_sysc, 0x100); +} +#else +static inline void sh7372_pm_setup_smfram(void) {} #endif /* CONFIG_SUSPEND || CONFIG_CPU_IDLE */ #ifdef CONFIG_CPU_IDLE @@ -313,7 +364,8 @@ static int sh7372_do_idle_core_standby(unsigned long unused) return 0; } -static void sh7372_enter_core_standby(void) +static int sh7372_enter_core_standby(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index) { sh7372_set_reset_vector(__pa(sh7372_resume_core_standby_sysc)); @@ -324,83 +376,102 @@ static void sh7372_enter_core_standby(void) /* disable reset vector translation */ __raw_writel(0, SBAR); + + return 1; } -static void sh7372_enter_a3sm_pll_on(void) +static int sh7372_enter_a3sm_pll_on(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index) { sh7372_enter_a3sm_common(1); + return 2; } -static void sh7372_enter_a3sm_pll_off(void) +static int sh7372_enter_a3sm_pll_off(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index) { sh7372_enter_a3sm_common(0); + return 3; } -static void sh7372_cpuidle_setup(struct cpuidle_driver *drv) +static int sh7372_enter_a4s(struct cpuidle_device *dev, + struct cpuidle_driver *drv, int index) { - struct cpuidle_state *state = &drv->states[drv->state_count]; - - snprintf(state->name, CPUIDLE_NAME_LEN, "C2"); - strncpy(state->desc, "Core Standby Mode", CPUIDLE_DESC_LEN); - state->exit_latency = 10; - state->target_residency = 20 + 10; - state->flags = CPUIDLE_FLAG_TIME_VALID; - shmobile_cpuidle_modes[drv->state_count] = sh7372_enter_core_standby; - drv->state_count++; - - state = &drv->states[drv->state_count]; - snprintf(state->name, CPUIDLE_NAME_LEN, "C3"); - strncpy(state->desc, "A3SM PLL ON", CPUIDLE_DESC_LEN); - state->exit_latency = 20; - state->target_residency = 30 + 20; - state->flags = CPUIDLE_FLAG_TIME_VALID; - shmobile_cpuidle_modes[drv->state_count] = sh7372_enter_a3sm_pll_on; - drv->state_count++; - - state = &drv->states[drv->state_count]; - snprintf(state->name, CPUIDLE_NAME_LEN, "C4"); - strncpy(state->desc, "A3SM PLL OFF", CPUIDLE_DESC_LEN); - state->exit_latency = 120; - state->target_residency = 30 + 120; - state->flags = CPUIDLE_FLAG_TIME_VALID; - shmobile_cpuidle_modes[drv->state_count] = sh7372_enter_a3sm_pll_off; - drv->state_count++; + unsigned long msk, msk2; + + if (!sh7372_sysc_valid(&msk, &msk2)) + return sh7372_enter_a3sm_pll_off(dev, drv, index); + + sh7372_setup_sysc(msk, msk2); + sh7372_enter_a4s_common(0); + return 4; } +static struct cpuidle_driver sh7372_cpuidle_driver = { + .name = "sh7372_cpuidle", + .owner = THIS_MODULE, + .en_core_tk_irqen = 1, + .state_count = 5, + .safe_state_index = 0, /* C1 */ + .states[0] = ARM_CPUIDLE_WFI_STATE, + .states[0].enter = shmobile_enter_wfi, + .states[1] = { + .name = "C2", + .desc = "Core Standby Mode", + .exit_latency = 10, + .target_residency = 20 + 10, + .flags = CPUIDLE_FLAG_TIME_VALID, + .enter = sh7372_enter_core_standby, + }, + .states[2] = { + .name = "C3", + .desc = "A3SM PLL ON", + .exit_latency = 20, + .target_residency = 30 + 20, + .flags = CPUIDLE_FLAG_TIME_VALID, + .enter = sh7372_enter_a3sm_pll_on, + }, + .states[3] = { + .name = "C4", + .desc = "A3SM PLL OFF", + .exit_latency = 120, + .target_residency = 30 + 120, + .flags = CPUIDLE_FLAG_TIME_VALID, + .enter = sh7372_enter_a3sm_pll_off, + }, + .states[4] = { + .name = "C5", + .desc = "A4S PLL OFF", + .exit_latency = 240, + .target_residency = 30 + 240, + .flags = CPUIDLE_FLAG_TIME_VALID, + .enter = sh7372_enter_a4s, + .disabled = true, + }, +}; + static void sh7372_cpuidle_init(void) { - shmobile_cpuidle_setup = sh7372_cpuidle_setup; + shmobile_cpuidle_set_driver(&sh7372_cpuidle_driver); } #else static void sh7372_cpuidle_init(void) {} #endif #ifdef CONFIG_SUSPEND -static void sh7372_enter_a4s_common(int pllc0_on) -{ - sh7372_intca_suspend(); - memcpy((void *)SMFRAM, sh7372_resume_core_standby_sysc, 0x100); - sh7372_set_reset_vector(SMFRAM); - sh7372_enter_sysc(pllc0_on, 1 << 10); - sh7372_intca_resume(); -} - static int sh7372_enter_suspend(suspend_state_t suspend_state) { unsigned long msk, msk2; /* check active clocks to determine potential wakeup sources */ - if (sh7372_sysc_valid(&msk, &msk2)) { - if (!console_suspend_enabled && - sh7372_pd_a4s.genpd.status == GPD_STATE_POWER_OFF) { - /* convert INTC mask/sense to SYSC mask/sense */ - sh7372_setup_sysc(msk, msk2); - - /* enter A4S sleep with PLLC0 off */ - pr_debug("entering A4S\n"); - sh7372_enter_a4s_common(0); - return 0; - } + if (sh7372_sysc_valid(&msk, &msk2) && a4s_suspend_ready) { + /* convert INTC mask/sense to SYSC mask/sense */ + sh7372_setup_sysc(msk, msk2); + + /* enter A4S sleep with PLLC0 off */ + pr_debug("entering A4S\n"); + sh7372_enter_a4s_common(0); + return 0; } /* default to enter A3SM sleep with PLLC0 off */ @@ -426,7 +497,7 @@ static int sh7372_pm_notifier_fn(struct notifier_block *notifier, * executed during system suspend and resume, respectively, so * that those functions don't crash while accessing the INTCS. */ - pm_genpd_poweron(&sh7372_pd_a4r.genpd); + pm_genpd_name_poweron("A4R"); break; case PM_POST_SUSPEND: pm_genpd_poweroff_unused(); @@ -455,6 +526,14 @@ void __init sh7372_pm_init(void) /* do not convert A3SM, A3SP, A3SG, A4R power down into A4S */ __raw_writel(0, PDNSEL); + sh7372_pm_setup_smfram(); + sh7372_suspend_init(); sh7372_cpuidle_init(); } + +void __init sh7372_pm_init_late(void) +{ + shmobile_init_late(); + pm_genpd_name_attach_cpuidle("A4S", 4); +} diff --git a/arch/arm/mach-shmobile/setup-r8a7740.c b/arch/arm/mach-shmobile/setup-r8a7740.c index 78948a9..11bb1d9 100644 --- a/arch/arm/mach-shmobile/setup-r8a7740.c +++ b/arch/arm/mach-shmobile/setup-r8a7740.c @@ -673,12 +673,7 @@ void __init r8a7740_add_standard_devices(void) r8a7740_i2c_workaround(&i2c0_device); r8a7740_i2c_workaround(&i2c1_device); - /* PM domain */ - rmobile_init_pm_domain(&r8a7740_pd_a4s); - rmobile_init_pm_domain(&r8a7740_pd_a3sp); - rmobile_init_pm_domain(&r8a7740_pd_a4lc); - - rmobile_pm_add_subdomain(&r8a7740_pd_a4s, &r8a7740_pd_a3sp); + r8a7740_init_pm_domains(); /* add devices */ platform_add_devices(r8a7740_early_devices, @@ -688,16 +683,16 @@ void __init r8a7740_add_standard_devices(void) /* add devices to PM domain */ - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif0_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif1_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif2_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif3_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif4_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif5_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif6_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scif7_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &scifb_device); - rmobile_add_device_to_domain(&r8a7740_pd_a3sp, &i2c1_device); + rmobile_add_device_to_domain("A3SP", &scif0_device); + rmobile_add_device_to_domain("A3SP", &scif1_device); + rmobile_add_device_to_domain("A3SP", &scif2_device); + rmobile_add_device_to_domain("A3SP", &scif3_device); + rmobile_add_device_to_domain("A3SP", &scif4_device); + rmobile_add_device_to_domain("A3SP", &scif5_device); + rmobile_add_device_to_domain("A3SP", &scif6_device); + rmobile_add_device_to_domain("A3SP", &scif7_device); + rmobile_add_device_to_domain("A3SP", &scifb_device); + rmobile_add_device_to_domain("A3SP", &i2c1_device); } static void __init r8a7740_earlytimer_init(void) diff --git a/arch/arm/mach-shmobile/setup-r8a7779.c b/arch/arm/mach-shmobile/setup-r8a7779.c index e98e46f..2917668 100644 --- a/arch/arm/mach-shmobile/setup-r8a7779.c +++ b/arch/arm/mach-shmobile/setup-r8a7779.c @@ -251,10 +251,7 @@ void __init r8a7779_add_standard_devices(void) #endif r8a7779_pm_init(); - r8a7779_init_pm_domain(&r8a7779_sh4a); - r8a7779_init_pm_domain(&r8a7779_sgx); - r8a7779_init_pm_domain(&r8a7779_vdp1); - r8a7779_init_pm_domain(&r8a7779_impx3); + r8a7779_init_pm_domains(); platform_add_devices(r8a7779_early_devices, ARRAY_SIZE(r8a7779_early_devices)); diff --git a/arch/arm/mach-shmobile/setup-sh7372.c b/arch/arm/mach-shmobile/setup-sh7372.c index 838a87b..a07954f 100644 --- a/arch/arm/mach-shmobile/setup-sh7372.c +++ b/arch/arm/mach-shmobile/setup-sh7372.c @@ -1001,21 +1001,34 @@ static struct platform_device *sh7372_late_devices[] __initdata = { void __init sh7372_add_standard_devices(void) { - rmobile_init_pm_domain(&sh7372_pd_a4lc); - rmobile_init_pm_domain(&sh7372_pd_a4mp); - rmobile_init_pm_domain(&sh7372_pd_d4); - rmobile_init_pm_domain(&sh7372_pd_a4r); - rmobile_init_pm_domain(&sh7372_pd_a3rv); - rmobile_init_pm_domain(&sh7372_pd_a3ri); - rmobile_init_pm_domain(&sh7372_pd_a4s); - rmobile_init_pm_domain(&sh7372_pd_a3sp); - rmobile_init_pm_domain(&sh7372_pd_a3sg); - - rmobile_pm_add_subdomain(&sh7372_pd_a4lc, &sh7372_pd_a3rv); - rmobile_pm_add_subdomain(&sh7372_pd_a4r, &sh7372_pd_a4lc); - - rmobile_pm_add_subdomain(&sh7372_pd_a4s, &sh7372_pd_a3sg); - rmobile_pm_add_subdomain(&sh7372_pd_a4s, &sh7372_pd_a3sp); + struct pm_domain_device domain_devices[] = { + { "A3RV", &vpu_device, }, + { "A4MP", &spu0_device, }, + { "A4MP", &spu1_device, }, + { "A3SP", &scif0_device, }, + { "A3SP", &scif1_device, }, + { "A3SP", &scif2_device, }, + { "A3SP", &scif3_device, }, + { "A3SP", &scif4_device, }, + { "A3SP", &scif5_device, }, + { "A3SP", &scif6_device, }, + { "A3SP", &iic1_device, }, + { "A3SP", &dma0_device, }, + { "A3SP", &dma1_device, }, + { "A3SP", &dma2_device, }, + { "A3SP", &usb_dma0_device, }, + { "A3SP", &usb_dma1_device, }, + { "A4R", &iic0_device, }, + { "A4R", &veu0_device, }, + { "A4R", &veu1_device, }, + { "A4R", &veu2_device, }, + { "A4R", &veu3_device, }, + { "A4R", &jpu_device, }, + { "A4R", &tmu00_device, }, + { "A4R", &tmu01_device, }, + }; + + sh7372_init_pm_domains(); platform_add_devices(sh7372_early_devices, ARRAY_SIZE(sh7372_early_devices)); @@ -1023,30 +1036,8 @@ void __init sh7372_add_standard_devices(void) platform_add_devices(sh7372_late_devices, ARRAY_SIZE(sh7372_late_devices)); - rmobile_add_device_to_domain(&sh7372_pd_a3rv, &vpu_device); - rmobile_add_device_to_domain(&sh7372_pd_a4mp, &spu0_device); - rmobile_add_device_to_domain(&sh7372_pd_a4mp, &spu1_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif0_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif1_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif2_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif3_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif4_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif5_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &scif6_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &iic1_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &dma0_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &dma1_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &dma2_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &usb_dma0_device); - rmobile_add_device_to_domain(&sh7372_pd_a3sp, &usb_dma1_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &iic0_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &veu0_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &veu1_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &veu2_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &veu3_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &jpu_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &tmu00_device); - rmobile_add_device_to_domain(&sh7372_pd_a4r, &tmu01_device); + rmobile_add_devices_to_domains(domain_devices, + ARRAY_SIZE(domain_devices)); } static void __init sh7372_earlytimer_init(void) diff --git a/arch/arm/mm/dma-mapping.c b/arch/arm/mm/dma-mapping.c index 13f555d..477a2d2 100644 --- a/arch/arm/mm/dma-mapping.c +++ b/arch/arm/mm/dma-mapping.c @@ -73,11 +73,18 @@ static dma_addr_t arm_dma_map_page(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { - if (!arch_is_coherent() && !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) + if (!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) __dma_page_cpu_to_dev(page, offset, size, dir); return pfn_to_dma(dev, page_to_pfn(page)) + offset; } +static dma_addr_t arm_coherent_dma_map_page(struct device *dev, struct page *page, + unsigned long offset, size_t size, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + return pfn_to_dma(dev, page_to_pfn(page)) + offset; +} + /** * arm_dma_unmap_page - unmap a buffer previously mapped through dma_map_page() * @dev: valid struct device pointer, or NULL for ISA and EISA-like devices @@ -96,7 +103,7 @@ static void arm_dma_unmap_page(struct device *dev, dma_addr_t handle, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { - if (!arch_is_coherent() && !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) + if (!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) __dma_page_dev_to_cpu(pfn_to_page(dma_to_pfn(dev, handle)), handle & ~PAGE_MASK, size, dir); } @@ -106,8 +113,7 @@ static void arm_dma_sync_single_for_cpu(struct device *dev, { unsigned int offset = handle & (PAGE_SIZE - 1); struct page *page = pfn_to_page(dma_to_pfn(dev, handle-offset)); - if (!arch_is_coherent()) - __dma_page_dev_to_cpu(page, offset, size, dir); + __dma_page_dev_to_cpu(page, offset, size, dir); } static void arm_dma_sync_single_for_device(struct device *dev, @@ -115,8 +121,7 @@ static void arm_dma_sync_single_for_device(struct device *dev, { unsigned int offset = handle & (PAGE_SIZE - 1); struct page *page = pfn_to_page(dma_to_pfn(dev, handle-offset)); - if (!arch_is_coherent()) - __dma_page_cpu_to_dev(page, offset, size, dir); + __dma_page_cpu_to_dev(page, offset, size, dir); } static int arm_dma_set_mask(struct device *dev, u64 dma_mask); @@ -138,6 +143,22 @@ struct dma_map_ops arm_dma_ops = { }; EXPORT_SYMBOL(arm_dma_ops); +static void *arm_coherent_dma_alloc(struct device *dev, size_t size, + dma_addr_t *handle, gfp_t gfp, struct dma_attrs *attrs); +static void arm_coherent_dma_free(struct device *dev, size_t size, void *cpu_addr, + dma_addr_t handle, struct dma_attrs *attrs); + +struct dma_map_ops arm_coherent_dma_ops = { + .alloc = arm_coherent_dma_alloc, + .free = arm_coherent_dma_free, + .mmap = arm_dma_mmap, + .get_sgtable = arm_dma_get_sgtable, + .map_page = arm_coherent_dma_map_page, + .map_sg = arm_dma_map_sg, + .set_dma_mask = arm_dma_set_mask, +}; +EXPORT_SYMBOL(arm_coherent_dma_ops); + static u64 get_coherent_dma_mask(struct device *dev) { u64 mask = (u64)arm_dma_limit; @@ -586,7 +607,7 @@ static void *__alloc_simple_buffer(struct device *dev, size_t size, gfp_t gfp, static void *__dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, - gfp_t gfp, pgprot_t prot, const void *caller) + gfp_t gfp, pgprot_t prot, bool is_coherent, const void *caller) { u64 mask = get_coherent_dma_mask(dev); struct page *page; @@ -619,7 +640,7 @@ static void *__dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, *handle = DMA_ERROR_CODE; size = PAGE_ALIGN(size); - if (arch_is_coherent() || nommu()) + if (is_coherent || nommu()) addr = __alloc_simple_buffer(dev, size, gfp, &page); else if (gfp & GFP_ATOMIC) addr = __alloc_from_pool(size, &page); @@ -647,7 +668,20 @@ void *arm_dma_alloc(struct device *dev, size_t size, dma_addr_t *handle, if (dma_alloc_from_coherent(dev, size, handle, &memory)) return memory; - return __dma_alloc(dev, size, handle, gfp, prot, + return __dma_alloc(dev, size, handle, gfp, prot, false, + __builtin_return_address(0)); +} + +static void *arm_coherent_dma_alloc(struct device *dev, size_t size, + dma_addr_t *handle, gfp_t gfp, struct dma_attrs *attrs) +{ + pgprot_t prot = __get_dma_pgprot(attrs, pgprot_kernel); + void *memory; + + if (dma_alloc_from_coherent(dev, size, handle, &memory)) + return memory; + + return __dma_alloc(dev, size, handle, gfp, prot, true, __builtin_return_address(0)); } @@ -684,8 +718,9 @@ int arm_dma_mmap(struct device *dev, struct vm_area_struct *vma, /* * Free a buffer as defined by the above mapping. */ -void arm_dma_free(struct device *dev, size_t size, void *cpu_addr, - dma_addr_t handle, struct dma_attrs *attrs) +static void __arm_dma_free(struct device *dev, size_t size, void *cpu_addr, + dma_addr_t handle, struct dma_attrs *attrs, + bool is_coherent) { struct page *page = pfn_to_page(dma_to_pfn(dev, handle)); @@ -694,7 +729,7 @@ void arm_dma_free(struct device *dev, size_t size, void *cpu_addr, size = PAGE_ALIGN(size); - if (arch_is_coherent() || nommu()) { + if (is_coherent || nommu()) { __dma_free_buffer(page, size); } else if (__free_from_pool(cpu_addr, size)) { return; @@ -710,6 +745,18 @@ void arm_dma_free(struct device *dev, size_t size, void *cpu_addr, } } +void arm_dma_free(struct device *dev, size_t size, void *cpu_addr, + dma_addr_t handle, struct dma_attrs *attrs) +{ + __arm_dma_free(dev, size, cpu_addr, handle, attrs, false); +} + +static void arm_coherent_dma_free(struct device *dev, size_t size, void *cpu_addr, + dma_addr_t handle, struct dma_attrs *attrs) +{ + __arm_dma_free(dev, size, cpu_addr, handle, attrs, true); +} + int arm_dma_get_sgtable(struct device *dev, struct sg_table *sgt, void *cpu_addr, dma_addr_t handle, size_t size, struct dma_attrs *attrs) @@ -1012,11 +1059,12 @@ static struct page **__iommu_alloc_buffer(struct device *dev, size_t size, gfp_t if (!pages[i]) goto error; - if (order) + if (order) { split_page(pages[i], order); - j = 1 << order; - while (--j) - pages[i + j] = pages[i] + j; + j = 1 << order; + while (--j) + pages[i + j] = pages[i] + j; + } __dma_clear_buffer(pages[i], PAGE_SIZE << order); i += 1 << order; @@ -1303,7 +1351,8 @@ static int arm_iommu_get_sgtable(struct device *dev, struct sg_table *sgt, */ static int __map_sg_chunk(struct device *dev, struct scatterlist *sg, size_t size, dma_addr_t *handle, - enum dma_data_direction dir, struct dma_attrs *attrs) + enum dma_data_direction dir, struct dma_attrs *attrs, + bool is_coherent) { struct dma_iommu_mapping *mapping = dev->archdata.mapping; dma_addr_t iova, iova_base; @@ -1322,8 +1371,8 @@ static int __map_sg_chunk(struct device *dev, struct scatterlist *sg, phys_addr_t phys = page_to_phys(sg_page(s)); unsigned int len = PAGE_ALIGN(s->offset + s->length); - if (!arch_is_coherent() && - !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) + if (!is_coherent && + !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) __dma_page_cpu_to_dev(sg_page(s), s->offset, s->length, dir); ret = iommu_map(mapping->domain, iova, phys, len, 0); @@ -1341,20 +1390,9 @@ fail: return ret; } -/** - * arm_iommu_map_sg - map a set of SG buffers for streaming mode DMA - * @dev: valid struct device pointer - * @sg: list of buffers - * @nents: number of buffers to map - * @dir: DMA transfer direction - * - * Map a set of buffers described by scatterlist in streaming mode for DMA. - * The scatter gather list elements are merged together (if possible) and - * tagged with the appropriate dma address and length. They are obtained via - * sg_dma_{address,length}. - */ -int arm_iommu_map_sg(struct device *dev, struct scatterlist *sg, int nents, - enum dma_data_direction dir, struct dma_attrs *attrs) +static int __iommu_map_sg(struct device *dev, struct scatterlist *sg, int nents, + enum dma_data_direction dir, struct dma_attrs *attrs, + bool is_coherent) { struct scatterlist *s = sg, *dma = sg, *start = sg; int i, count = 0; @@ -1370,7 +1408,7 @@ int arm_iommu_map_sg(struct device *dev, struct scatterlist *sg, int nents, if (s->offset || (size & ~PAGE_MASK) || size + s->length > max) { if (__map_sg_chunk(dev, start, size, &dma->dma_address, - dir, attrs) < 0) + dir, attrs, is_coherent) < 0) goto bad_mapping; dma->dma_address += offset; @@ -1383,7 +1421,8 @@ int arm_iommu_map_sg(struct device *dev, struct scatterlist *sg, int nents, } size += s->length; } - if (__map_sg_chunk(dev, start, size, &dma->dma_address, dir, attrs) < 0) + if (__map_sg_chunk(dev, start, size, &dma->dma_address, dir, attrs, + is_coherent) < 0) goto bad_mapping; dma->dma_address += offset; @@ -1398,17 +1437,44 @@ bad_mapping: } /** - * arm_iommu_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg + * arm_coherent_iommu_map_sg - map a set of SG buffers for streaming mode DMA * @dev: valid struct device pointer * @sg: list of buffers - * @nents: number of buffers to unmap (same as was passed to dma_map_sg) - * @dir: DMA transfer direction (same as was passed to dma_map_sg) + * @nents: number of buffers to map + * @dir: DMA transfer direction * - * Unmap a set of streaming mode DMA translations. Again, CPU access - * rules concerning calls here are the same as for dma_unmap_single(). + * Map a set of i/o coherent buffers described by scatterlist in streaming + * mode for DMA. The scatter gather list elements are merged together (if + * possible) and tagged with the appropriate dma address and length. They are + * obtained via sg_dma_{address,length}. */ -void arm_iommu_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, - enum dma_data_direction dir, struct dma_attrs *attrs) +int arm_coherent_iommu_map_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, struct dma_attrs *attrs) +{ + return __iommu_map_sg(dev, sg, nents, dir, attrs, true); +} + +/** + * arm_iommu_map_sg - map a set of SG buffers for streaming mode DMA + * @dev: valid struct device pointer + * @sg: list of buffers + * @nents: number of buffers to map + * @dir: DMA transfer direction + * + * Map a set of buffers described by scatterlist in streaming mode for DMA. + * The scatter gather list elements are merged together (if possible) and + * tagged with the appropriate dma address and length. They are obtained via + * sg_dma_{address,length}. + */ +int arm_iommu_map_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, struct dma_attrs *attrs) +{ + return __iommu_map_sg(dev, sg, nents, dir, attrs, false); +} + +static void __iommu_unmap_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, struct dma_attrs *attrs, + bool is_coherent) { struct scatterlist *s; int i; @@ -1417,7 +1483,7 @@ void arm_iommu_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, if (sg_dma_len(s)) __iommu_remove_mapping(dev, sg_dma_address(s), sg_dma_len(s)); - if (!arch_is_coherent() && + if (!is_coherent && !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) __dma_page_dev_to_cpu(sg_page(s), s->offset, s->length, dir); @@ -1425,6 +1491,38 @@ void arm_iommu_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, } /** + * arm_coherent_iommu_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg + * @dev: valid struct device pointer + * @sg: list of buffers + * @nents: number of buffers to unmap (same as was passed to dma_map_sg) + * @dir: DMA transfer direction (same as was passed to dma_map_sg) + * + * Unmap a set of streaming mode DMA translations. Again, CPU access + * rules concerning calls here are the same as for dma_unmap_single(). + */ +void arm_coherent_iommu_unmap_sg(struct device *dev, struct scatterlist *sg, + int nents, enum dma_data_direction dir, struct dma_attrs *attrs) +{ + __iommu_unmap_sg(dev, sg, nents, dir, attrs, true); +} + +/** + * arm_iommu_unmap_sg - unmap a set of SG buffers mapped by dma_map_sg + * @dev: valid struct device pointer + * @sg: list of buffers + * @nents: number of buffers to unmap (same as was passed to dma_map_sg) + * @dir: DMA transfer direction (same as was passed to dma_map_sg) + * + * Unmap a set of streaming mode DMA translations. Again, CPU access + * rules concerning calls here are the same as for dma_unmap_single(). + */ +void arm_iommu_unmap_sg(struct device *dev, struct scatterlist *sg, int nents, + enum dma_data_direction dir, struct dma_attrs *attrs) +{ + __iommu_unmap_sg(dev, sg, nents, dir, attrs, false); +} + +/** * arm_iommu_sync_sg_for_cpu * @dev: valid struct device pointer * @sg: list of buffers @@ -1438,8 +1536,7 @@ void arm_iommu_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int i; for_each_sg(sg, s, nents, i) - if (!arch_is_coherent()) - __dma_page_dev_to_cpu(sg_page(s), s->offset, s->length, dir); + __dma_page_dev_to_cpu(sg_page(s), s->offset, s->length, dir); } @@ -1457,22 +1554,21 @@ void arm_iommu_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int i; for_each_sg(sg, s, nents, i) - if (!arch_is_coherent()) - __dma_page_cpu_to_dev(sg_page(s), s->offset, s->length, dir); + __dma_page_cpu_to_dev(sg_page(s), s->offset, s->length, dir); } /** - * arm_iommu_map_page + * arm_coherent_iommu_map_page * @dev: valid struct device pointer * @page: page that buffer resides in * @offset: offset into page for start of buffer * @size: size of buffer to map * @dir: DMA transfer direction * - * IOMMU aware version of arm_dma_map_page() + * Coherent IOMMU aware version of arm_dma_map_page() */ -static dma_addr_t arm_iommu_map_page(struct device *dev, struct page *page, +static dma_addr_t arm_coherent_iommu_map_page(struct device *dev, struct page *page, unsigned long offset, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs) { @@ -1480,9 +1576,6 @@ static dma_addr_t arm_iommu_map_page(struct device *dev, struct page *page, dma_addr_t dma_addr; int ret, len = PAGE_ALIGN(size + offset); - if (!arch_is_coherent() && !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) - __dma_page_cpu_to_dev(page, offset, size, dir); - dma_addr = __alloc_iova(mapping, len); if (dma_addr == DMA_ERROR_CODE) return dma_addr; @@ -1498,6 +1591,51 @@ fail: } /** + * arm_iommu_map_page + * @dev: valid struct device pointer + * @page: page that buffer resides in + * @offset: offset into page for start of buffer + * @size: size of buffer to map + * @dir: DMA transfer direction + * + * IOMMU aware version of arm_dma_map_page() + */ +static dma_addr_t arm_iommu_map_page(struct device *dev, struct page *page, + unsigned long offset, size_t size, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + if (!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) + __dma_page_cpu_to_dev(page, offset, size, dir); + + return arm_coherent_iommu_map_page(dev, page, offset, size, dir, attrs); +} + +/** + * arm_coherent_iommu_unmap_page + * @dev: valid struct device pointer + * @handle: DMA address of buffer + * @size: size of buffer (same as passed to dma_map_page) + * @dir: DMA transfer direction (same as passed to dma_map_page) + * + * Coherent IOMMU aware version of arm_dma_unmap_page() + */ +static void arm_coherent_iommu_unmap_page(struct device *dev, dma_addr_t handle, + size_t size, enum dma_data_direction dir, + struct dma_attrs *attrs) +{ + struct dma_iommu_mapping *mapping = dev->archdata.mapping; + dma_addr_t iova = handle & PAGE_MASK; + int offset = handle & ~PAGE_MASK; + int len = PAGE_ALIGN(size + offset); + + if (!iova) + return; + + iommu_unmap(mapping->domain, iova, len); + __free_iova(mapping, iova, len); +} + +/** * arm_iommu_unmap_page * @dev: valid struct device pointer * @handle: DMA address of buffer @@ -1519,7 +1657,7 @@ static void arm_iommu_unmap_page(struct device *dev, dma_addr_t handle, if (!iova) return; - if (!arch_is_coherent() && !dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) + if (!dma_get_attr(DMA_ATTR_SKIP_CPU_SYNC, attrs)) __dma_page_dev_to_cpu(page, offset, size, dir); iommu_unmap(mapping->domain, iova, len); @@ -1537,8 +1675,7 @@ static void arm_iommu_sync_single_for_cpu(struct device *dev, if (!iova) return; - if (!arch_is_coherent()) - __dma_page_dev_to_cpu(page, offset, size, dir); + __dma_page_dev_to_cpu(page, offset, size, dir); } static void arm_iommu_sync_single_for_device(struct device *dev, @@ -1572,6 +1709,19 @@ struct dma_map_ops iommu_ops = { .sync_sg_for_device = arm_iommu_sync_sg_for_device, }; +struct dma_map_ops iommu_coherent_ops = { + .alloc = arm_iommu_alloc_attrs, + .free = arm_iommu_free_attrs, + .mmap = arm_iommu_mmap_attrs, + .get_sgtable = arm_iommu_get_sgtable, + + .map_page = arm_coherent_iommu_map_page, + .unmap_page = arm_coherent_iommu_unmap_page, + + .map_sg = arm_coherent_iommu_map_sg, + .unmap_sg = arm_coherent_iommu_unmap_sg, +}; + /** * arm_iommu_create_mapping * @bus: pointer to the bus holding the client device (for IOMMU calls) @@ -1665,7 +1815,7 @@ int arm_iommu_attach_device(struct device *dev, dev->archdata.mapping = mapping; set_dma_ops(dev, &iommu_ops); - pr_info("Attached IOMMU controller to %s device.\n", dev_name(dev)); + pr_debug("Attached IOMMU controller to %s device.\n", dev_name(dev)); return 0; } diff --git a/arch/arm/mm/mmu.c b/arch/arm/mm/mmu.c index 18144e6..941dfb9 100644 --- a/arch/arm/mm/mmu.c +++ b/arch/arm/mm/mmu.c @@ -423,17 +423,6 @@ static void __init build_mem_type_table(void) vecs_pgprot = kern_pgprot = user_pgprot = cp->pte; /* - * Enable CPU-specific coherency if supported. - * (Only available on XSC3 at the moment.) - */ - if (arch_is_coherent() && cpu_is_xsc3()) { - mem_types[MT_MEMORY].prot_sect |= PMD_SECT_S; - mem_types[MT_MEMORY].prot_pte |= L_PTE_SHARED; - mem_types[MT_MEMORY_DMA_READY].prot_pte |= L_PTE_SHARED; - mem_types[MT_MEMORY_NONCACHED].prot_sect |= PMD_SECT_S; - mem_types[MT_MEMORY_NONCACHED].prot_pte |= L_PTE_SHARED; - } - /* * ARMv6 and above have extended page tables. */ if (cpu_arch >= CPU_ARCH_ARMv6 && (cr & CR_XP)) { diff --git a/arch/ia64/include/asm/xen/interface.h b/arch/ia64/include/asm/xen/interface.h index 09d5f7f..3d52a5b 100644 --- a/arch/ia64/include/asm/xen/interface.h +++ b/arch/ia64/include/asm/xen/interface.h @@ -67,6 +67,10 @@ #define set_xen_guest_handle(hnd, val) do { (hnd).p = val; } while (0) #ifndef __ASSEMBLY__ +/* Explicitly size integers that represent pfns in the public interface + * with Xen so that we could have one ABI that works for 32 and 64 bit + * guests. */ +typedef unsigned long xen_pfn_t; /* Guest handles for primitive C types. */ __DEFINE_GUEST_HANDLE(uchar, unsigned char); __DEFINE_GUEST_HANDLE(uint, unsigned int); @@ -79,7 +83,6 @@ DEFINE_GUEST_HANDLE(void); DEFINE_GUEST_HANDLE(uint64_t); DEFINE_GUEST_HANDLE(uint32_t); -typedef unsigned long xen_pfn_t; DEFINE_GUEST_HANDLE(xen_pfn_t); #define PRI_xen_pfn "lx" #endif @@ -265,6 +268,8 @@ typedef struct xen_callback xen_callback_t; #endif /* !__ASSEMBLY__ */ +#include <asm/pvclock-abi.h> + /* Size of the shared_info area (this is not related to page size). */ #define XSI_SHIFT 14 #define XSI_SIZE (1 << XSI_SHIFT) diff --git a/arch/ia64/kernel/perfmon.c b/arch/ia64/kernel/perfmon.c index 5a5c222..f388b4e 100644 --- a/arch/ia64/kernel/perfmon.c +++ b/arch/ia64/kernel/perfmon.c @@ -2306,7 +2306,7 @@ pfm_smpl_buffer_alloc(struct task_struct *task, struct file *filp, pfm_context_t * partially initialize the vma for the sampling buffer */ vma->vm_mm = mm; - vma->vm_file = filp; + vma->vm_file = get_file(filp); vma->vm_flags = VM_READ| VM_MAYREAD |VM_RESERVED; vma->vm_page_prot = PAGE_READONLY; /* XXX may need to change */ @@ -2345,8 +2345,6 @@ pfm_smpl_buffer_alloc(struct task_struct *task, struct file *filp, pfm_context_t goto error; } - get_file(filp); - /* * now insert the vma in the vm list for the process, must be * done with mmap lock held @@ -4782,7 +4780,7 @@ recheck: asmlinkage long sys_perfmonctl (int fd, int cmd, void __user *arg, int count) { - struct file *file = NULL; + struct fd f = {NULL, 0}; pfm_context_t *ctx = NULL; unsigned long flags = 0UL; void *args_k = NULL; @@ -4879,17 +4877,17 @@ restart_args: ret = -EBADF; - file = fget(fd); - if (unlikely(file == NULL)) { + f = fdget(fd); + if (unlikely(f.file == NULL)) { DPRINT(("invalid fd %d\n", fd)); goto error_args; } - if (unlikely(PFM_IS_FILE(file) == 0)) { + if (unlikely(PFM_IS_FILE(f.file) == 0)) { DPRINT(("fd %d not related to perfmon\n", fd)); goto error_args; } - ctx = file->private_data; + ctx = f.file->private_data; if (unlikely(ctx == NULL)) { DPRINT(("no context for fd %d\n", fd)); goto error_args; @@ -4919,8 +4917,8 @@ abort_locked: if (call_made && PFM_CMD_RW_ARG(cmd) && copy_to_user(arg, args_k, base_sz*count)) ret = -EFAULT; error_args: - if (file) - fput(file); + if (f.file) + fdput(f); kfree(args_k); diff --git a/arch/parisc/hpux/fs.c b/arch/parisc/hpux/fs.c index c71eb6c..6785de7 100644 --- a/arch/parisc/hpux/fs.c +++ b/arch/parisc/hpux/fs.c @@ -109,33 +109,32 @@ Efault: int hpux_getdents(unsigned int fd, struct hpux_dirent __user *dirent, unsigned int count) { - struct file * file; + struct fd arg; struct hpux_dirent __user * lastdirent; struct getdents_callback buf; - int error = -EBADF; + int error; - file = fget(fd); - if (!file) - goto out; + arg = fdget(fd); + if (!arg.file) + return -EBADF; buf.current_dir = dirent; buf.previous = NULL; buf.count = count; buf.error = 0; - error = vfs_readdir(file, filldir, &buf); + error = vfs_readdir(arg.file, filldir, &buf); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(file->f_pos, &lastdirent->d_off)) + if (put_user(arg.file->f_pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; } - fput(file); -out: + fdput(arg); return error; } diff --git a/arch/powerpc/include/asm/systbl.h b/arch/powerpc/include/asm/systbl.h index 559ae1e..8408387 100644 --- a/arch/powerpc/include/asm/systbl.h +++ b/arch/powerpc/include/asm/systbl.h @@ -189,7 +189,7 @@ SYSCALL_SPU(getcwd) SYSCALL_SPU(capget) SYSCALL_SPU(capset) COMPAT_SYS(sigaltstack) -SYSX_SPU(sys_sendfile64,compat_sys_sendfile,sys_sendfile) +SYSX_SPU(sys_sendfile,compat_sys_sendfile_wrapper,sys_sendfile) SYSCALL(ni_syscall) SYSCALL(ni_syscall) PPC_SYS(vfork) @@ -229,7 +229,7 @@ COMPAT_SYS_SPU(sched_setaffinity) COMPAT_SYS_SPU(sched_getaffinity) SYSCALL(ni_syscall) SYSCALL(ni_syscall) -SYS32ONLY(sendfile64) +SYSX(sys_ni_syscall,compat_sys_sendfile64_wrapper,sys_sendfile64) COMPAT_SYS_SPU(io_setup) SYSCALL_SPU(io_destroy) COMPAT_SYS_SPU(io_getevents) diff --git a/arch/powerpc/include/asm/unistd.h b/arch/powerpc/include/asm/unistd.h index bd377a3..c683fa3 100644 --- a/arch/powerpc/include/asm/unistd.h +++ b/arch/powerpc/include/asm/unistd.h @@ -419,6 +419,7 @@ #define __ARCH_WANT_COMPAT_SYS_TIME #define __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND #define __ARCH_WANT_SYS_NEWFSTATAT +#define __ARCH_WANT_COMPAT_SYS_SENDFILE #endif /* diff --git a/arch/powerpc/kernel/prom_init.c b/arch/powerpc/kernel/prom_init.c index 0794a30..e144498 100644 --- a/arch/powerpc/kernel/prom_init.c +++ b/arch/powerpc/kernel/prom_init.c @@ -1624,6 +1624,63 @@ static void __init prom_instantiate_rtas(void) #ifdef CONFIG_PPC64 /* + * Allocate room for and instantiate Stored Measurement Log (SML) + */ +static void __init prom_instantiate_sml(void) +{ + phandle ibmvtpm_node; + ihandle ibmvtpm_inst; + u32 entry = 0, size = 0; + u64 base; + + prom_debug("prom_instantiate_sml: start...\n"); + + ibmvtpm_node = call_prom("finddevice", 1, 1, ADDR("/ibm,vtpm")); + prom_debug("ibmvtpm_node: %x\n", ibmvtpm_node); + if (!PHANDLE_VALID(ibmvtpm_node)) + return; + + ibmvtpm_inst = call_prom("open", 1, 1, ADDR("/ibm,vtpm")); + if (!IHANDLE_VALID(ibmvtpm_inst)) { + prom_printf("opening vtpm package failed (%x)\n", ibmvtpm_inst); + return; + } + + if (call_prom_ret("call-method", 2, 2, &size, + ADDR("sml-get-handover-size"), + ibmvtpm_inst) != 0 || size == 0) { + prom_printf("SML get handover size failed\n"); + return; + } + + base = alloc_down(size, PAGE_SIZE, 0); + if (base == 0) + prom_panic("Could not allocate memory for sml\n"); + + prom_printf("instantiating sml at 0x%x...", base); + + if (call_prom_ret("call-method", 4, 2, &entry, + ADDR("sml-handover"), + ibmvtpm_inst, size, base) != 0 || entry == 0) { + prom_printf("SML handover failed\n"); + return; + } + prom_printf(" done\n"); + + reserve_mem(base, size); + + prom_setprop(ibmvtpm_node, "/ibm,vtpm", "linux,sml-base", + &base, sizeof(base)); + prom_setprop(ibmvtpm_node, "/ibm,vtpm", "linux,sml-size", + &size, sizeof(size)); + + prom_debug("sml base = 0x%x\n", base); + prom_debug("sml size = 0x%x\n", (long)size); + + prom_debug("prom_instantiate_sml: end...\n"); +} + +/* * Allocate room for and initialize TCE tables */ static void __init prom_initialize_tce_table(void) @@ -2916,6 +2973,11 @@ unsigned long __init prom_init(unsigned long r3, unsigned long r4, prom_instantiate_opal(); #endif +#ifdef CONFIG_PPC64 + /* instantiate sml */ + prom_instantiate_sml(); +#endif + /* * On non-powermacs, put all CPUs in spin-loops. * diff --git a/arch/powerpc/kernel/sys_ppc32.c b/arch/powerpc/kernel/sys_ppc32.c index 81c5706..abd1112 100644 --- a/arch/powerpc/kernel/sys_ppc32.c +++ b/arch/powerpc/kernel/sys_ppc32.c @@ -143,48 +143,17 @@ long compat_sys_ipc(u32 call, u32 first, u32 second, u32 third, compat_uptr_t pt * proper conversion (sign extension) between the register representation of a signed int (msr in 32-bit mode) * and the register representation of a signed int (msr in 64-bit mode) is performed. */ -asmlinkage long compat_sys_sendfile(u32 out_fd, u32 in_fd, compat_off_t __user * offset, u32 count) +asmlinkage long compat_sys_sendfile_wrapper(u32 out_fd, u32 in_fd, + compat_off_t __user *offset, u32 count) { - mm_segment_t old_fs = get_fs(); - int ret; - off_t of; - off_t __user *up; - - if (offset && get_user(of, offset)) - return -EFAULT; - - /* The __user pointer cast is valid because of the set_fs() */ - set_fs(KERNEL_DS); - up = offset ? (off_t __user *) &of : NULL; - ret = sys_sendfile((int)out_fd, (int)in_fd, up, count); - set_fs(old_fs); - - if (offset && put_user(of, offset)) - return -EFAULT; - - return ret; + return compat_sys_sendfile((int)out_fd, (int)in_fd, offset, count); } -asmlinkage int compat_sys_sendfile64(int out_fd, int in_fd, compat_loff_t __user *offset, s32 count) +asmlinkage long compat_sys_sendfile64_wrapper(u32 out_fd, u32 in_fd, + compat_loff_t __user *offset, u32 count) { - mm_segment_t old_fs = get_fs(); - int ret; - loff_t lof; - loff_t __user *up; - - if (offset && get_user(lof, offset)) - return -EFAULT; - - /* The __user pointer cast is valid because of the set_fs() */ - set_fs(KERNEL_DS); - up = offset ? (loff_t __user *) &lof : NULL; - ret = sys_sendfile64(out_fd, in_fd, up, count); - set_fs(old_fs); - - if (offset && put_user(lof, offset)) - return -EFAULT; - - return ret; + return sys_sendfile((int)out_fd, (int)in_fd, + (off_t __user *)offset, count); } long compat_sys_execve(unsigned long a0, unsigned long a1, unsigned long a2, diff --git a/arch/powerpc/platforms/cell/spu_syscalls.c b/arch/powerpc/platforms/cell/spu_syscalls.c index 714bbfc..db4e638 100644 --- a/arch/powerpc/platforms/cell/spu_syscalls.c +++ b/arch/powerpc/platforms/cell/spu_syscalls.c @@ -69,8 +69,6 @@ SYSCALL_DEFINE4(spu_create, const char __user *, name, unsigned int, flags, umode_t, mode, int, neighbor_fd) { long ret; - struct file *neighbor; - int fput_needed; struct spufs_calls *calls; calls = spufs_calls_get(); @@ -78,11 +76,11 @@ SYSCALL_DEFINE4(spu_create, const char __user *, name, unsigned int, flags, return -ENOSYS; if (flags & SPU_CREATE_AFFINITY_SPU) { + struct fd neighbor = fdget(neighbor_fd); ret = -EBADF; - neighbor = fget_light(neighbor_fd, &fput_needed); - if (neighbor) { - ret = calls->create_thread(name, flags, mode, neighbor); - fput_light(neighbor, fput_needed); + if (neighbor.file) { + ret = calls->create_thread(name, flags, mode, neighbor.file); + fdput(neighbor); } } else ret = calls->create_thread(name, flags, mode, NULL); @@ -94,8 +92,7 @@ SYSCALL_DEFINE4(spu_create, const char __user *, name, unsigned int, flags, asmlinkage long sys_spu_run(int fd, __u32 __user *unpc, __u32 __user *ustatus) { long ret; - struct file *filp; - int fput_needed; + struct fd arg; struct spufs_calls *calls; calls = spufs_calls_get(); @@ -103,10 +100,10 @@ asmlinkage long sys_spu_run(int fd, __u32 __user *unpc, __u32 __user *ustatus) return -ENOSYS; ret = -EBADF; - filp = fget_light(fd, &fput_needed); - if (filp) { - ret = calls->spu_run(filp, unpc, ustatus); - fput_light(filp, fput_needed); + arg = fdget(fd); + if (arg.file) { + ret = calls->spu_run(arg.file, unpc, ustatus); + fdput(arg); } spufs_calls_put(calls); diff --git a/arch/powerpc/platforms/cell/spufs/coredump.c b/arch/powerpc/platforms/cell/spufs/coredump.c index c2c5b07..657e3f2 100644 --- a/arch/powerpc/platforms/cell/spufs/coredump.c +++ b/arch/powerpc/platforms/cell/spufs/coredump.c @@ -106,6 +106,17 @@ static int spufs_ctx_note_size(struct spu_context *ctx, int dfd) return total; } +static int match_context(const void *v, struct file *file, unsigned fd) +{ + struct spu_context *ctx; + if (file->f_op != &spufs_context_fops) + return 0; + ctx = SPUFS_I(file->f_dentry->d_inode)->i_ctx; + if (ctx->flags & SPU_CREATE_NOSCHED) + return 0; + return fd + 1; +} + /* * The additional architecture-specific notes for Cell are various * context files in the spu context. @@ -115,29 +126,18 @@ static int spufs_ctx_note_size(struct spu_context *ctx, int dfd) * internal functionality to dump them without needing to actually * open the files. */ +/* + * descriptor table is not shared, so files can't change or go away. + */ static struct spu_context *coredump_next_context(int *fd) { - struct fdtable *fdt = files_fdtable(current->files); struct file *file; - struct spu_context *ctx = NULL; - - for (; *fd < fdt->max_fds; (*fd)++) { - if (!fd_is_open(*fd, fdt)) - continue; - - file = fcheck(*fd); - - if (!file || file->f_op != &spufs_context_fops) - continue; - - ctx = SPUFS_I(file->f_dentry->d_inode)->i_ctx; - if (ctx->flags & SPU_CREATE_NOSCHED) - continue; - - break; - } - - return ctx; + int n = iterate_fd(current->files, *fd, match_context, NULL); + if (!n) + return NULL; + *fd = n - 1; + file = fcheck(*fd); + return SPUFS_I(file->f_dentry->d_inode)->i_ctx; } int spufs_coredump_extra_notes_size(void) diff --git a/arch/s390/hypfs/inode.c b/arch/s390/hypfs/inode.c index 124ec1a..06ea69b 100644 --- a/arch/s390/hypfs/inode.c +++ b/arch/s390/hypfs/inode.c @@ -72,8 +72,6 @@ static void hypfs_remove(struct dentry *dentry) struct dentry *parent; parent = dentry->d_parent; - if (!parent || !parent->d_inode) - return; mutex_lock(&parent->d_inode->i_mutex); if (hypfs_positive(dentry)) { if (S_ISDIR(dentry->d_inode->i_mode)) diff --git a/arch/sparc/include/asm/unistd.h b/arch/sparc/include/asm/unistd.h index fb26934..d9a677c 100644 --- a/arch/sparc/include/asm/unistd.h +++ b/arch/sparc/include/asm/unistd.h @@ -447,6 +447,7 @@ #else #define __ARCH_WANT_COMPAT_SYS_TIME #define __ARCH_WANT_COMPAT_SYS_RT_SIGSUSPEND +#define __ARCH_WANT_COMPAT_SYS_SENDFILE #endif /* diff --git a/arch/sparc/kernel/sys32.S b/arch/sparc/kernel/sys32.S index d97f3eb..44025f4 100644 --- a/arch/sparc/kernel/sys32.S +++ b/arch/sparc/kernel/sys32.S @@ -90,7 +90,7 @@ SIGN1(sys32_mkdir, sys_mkdir, %o1) SIGN3(sys32_futex, compat_sys_futex, %o1, %o2, %o5) SIGN1(sys32_sysfs, compat_sys_sysfs, %o0) SIGN2(sys32_sendfile, compat_sys_sendfile, %o0, %o1) -SIGN2(sys32_sendfile64, compat_sys_sendfile64, %o0, %o1) +SIGN2(sys32_sendfile64, sys_sendfile, %o0, %o1) SIGN1(sys32_prctl, sys_prctl, %o0) SIGN1(sys32_sched_rr_get_interval, compat_sys_sched_rr_get_interval, %o0) SIGN2(sys32_waitpid, sys_waitpid, %o0, %o2) diff --git a/arch/sparc/kernel/sys_sparc32.c b/arch/sparc/kernel/sys_sparc32.c index f739233..d862499 100644 --- a/arch/sparc/kernel/sys_sparc32.c +++ b/arch/sparc/kernel/sys_sparc32.c @@ -506,52 +506,6 @@ long compat_sys_fadvise64_64(int fd, advice); } -asmlinkage long compat_sys_sendfile(int out_fd, int in_fd, - compat_off_t __user *offset, - compat_size_t count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - off_t of; - - if (offset && get_user(of, offset)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_sendfile(out_fd, in_fd, - offset ? (off_t __user *) &of : NULL, - count); - set_fs(old_fs); - - if (offset && put_user(of, offset)) - return -EFAULT; - - return ret; -} - -asmlinkage long compat_sys_sendfile64(int out_fd, int in_fd, - compat_loff_t __user *offset, - compat_size_t count) -{ - mm_segment_t old_fs = get_fs(); - int ret; - loff_t lof; - - if (offset && get_user(lof, offset)) - return -EFAULT; - - set_fs(KERNEL_DS); - ret = sys_sendfile64(out_fd, in_fd, - offset ? (loff_t __user *) &lof : NULL, - count); - set_fs(old_fs); - - if (offset && put_user(lof, offset)) - return -EFAULT; - - return ret; -} - /* This is just a version for 32-bit applications which does * not force O_LARGEFILE on. */ diff --git a/arch/um/drivers/mconsole_kern.c b/arch/um/drivers/mconsole_kern.c index c17de0d..9efeb6d 100644 --- a/arch/um/drivers/mconsole_kern.c +++ b/arch/um/drivers/mconsole_kern.c @@ -21,6 +21,9 @@ #include <linux/un.h> #include <linux/workqueue.h> #include <linux/mutex.h> +#include <linux/fs.h> +#include <linux/mount.h> +#include <linux/file.h> #include <asm/uaccess.h> #include <asm/switch_to.h> @@ -118,90 +121,38 @@ void mconsole_log(struct mc_request *req) mconsole_reply(req, "", 0, 0); } -/* This is a more convoluted version of mconsole_proc, which has some stability - * problems; however, we need it fixed, because it is expected that UML users - * mount HPPFS instead of procfs on /proc. And we want mconsole_proc to still - * show the real procfs content, not the ones from hppfs.*/ -#if 0 void mconsole_proc(struct mc_request *req) { struct vfsmount *mnt = current->nsproxy->pid_ns->proc_mnt; - struct file *file; - int n; - char *ptr = req->request.data, *buf; - mm_segment_t old_fs = get_fs(); - - ptr += strlen("proc"); - ptr = skip_spaces(ptr); - - file = file_open_root(mnt->mnt_root, mnt, ptr, O_RDONLY); - if (IS_ERR(file)) { - mconsole_reply(req, "Failed to open file", 1, 0); - goto out; - } - - buf = kmalloc(PAGE_SIZE, GFP_KERNEL); - if (buf == NULL) { - mconsole_reply(req, "Failed to allocate buffer", 1, 0); - goto out_fput; - } - - if (file->f_op->read) { - do { - loff_t pos; - set_fs(KERNEL_DS); - n = vfs_read(file, buf, PAGE_SIZE - 1, &pos); - file_pos_write(file, pos); - set_fs(old_fs); - if (n >= 0) { - buf[n] = '\0'; - mconsole_reply(req, buf, 0, (n > 0)); - } - else { - mconsole_reply(req, "Read of file failed", - 1, 0); - goto out_free; - } - } while (n > 0); - } - else mconsole_reply(req, "", 0, 0); - - out_free: - kfree(buf); - out_fput: - fput(file); - out: ; -} -#endif - -void mconsole_proc(struct mc_request *req) -{ - char path[64]; char *buf; int len; - int fd; + struct file *file; int first_chunk = 1; char *ptr = req->request.data; ptr += strlen("proc"); ptr = skip_spaces(ptr); - snprintf(path, sizeof(path), "/proc/%s", ptr); - fd = sys_open(path, 0, 0); - if (fd < 0) { + file = file_open_root(mnt->mnt_root, mnt, ptr, O_RDONLY); + if (IS_ERR(file)) { mconsole_reply(req, "Failed to open file", 1, 0); - printk(KERN_ERR "open %s: %d\n",path,fd); + printk(KERN_ERR "open /proc/%s: %ld\n", ptr, PTR_ERR(file)); goto out; } buf = kmalloc(PAGE_SIZE, GFP_KERNEL); if (buf == NULL) { mconsole_reply(req, "Failed to allocate buffer", 1, 0); - goto out_close; + goto out_fput; } - for (;;) { - len = sys_read(fd, buf, PAGE_SIZE-1); + do { + loff_t pos; + mm_segment_t old_fs = get_fs(); + set_fs(KERNEL_DS); + len = vfs_read(file, buf, PAGE_SIZE - 1, &pos); + set_fs(old_fs); + file->f_pos = pos; if (len < 0) { mconsole_reply(req, "Read of file failed", 1, 0); goto out_free; @@ -211,22 +162,14 @@ void mconsole_proc(struct mc_request *req) mconsole_reply(req, "\n", 0, 1); first_chunk = 0; } - if (len == PAGE_SIZE-1) { - buf[len] = '\0'; - mconsole_reply(req, buf, 0, 1); - } else { - buf[len] = '\0'; - mconsole_reply(req, buf, 0, 0); - break; - } - } - + buf[len] = '\0'; + mconsole_reply(req, buf, 0, (len > 0)); + } while (len > 0); out_free: kfree(buf); - out_close: - sys_close(fd); - out: - /* nothing */; + out_fput: + fput(file); + out: ; } #define UML_MCONSOLE_HELPTEXT \ diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index 957ec87..fbee971 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -248,6 +248,9 @@ #define MSR_IA32_PERF_STATUS 0x00000198 #define MSR_IA32_PERF_CTL 0x00000199 +#define MSR_AMD_PSTATE_DEF_BASE 0xc0010064 +#define MSR_AMD_PERF_STATUS 0xc0010063 +#define MSR_AMD_PERF_CTL 0xc0010062 #define MSR_IA32_MPERF 0x000000e7 #define MSR_IA32_APERF 0x000000e8 diff --git a/arch/x86/include/asm/xen/interface.h b/arch/x86/include/asm/xen/interface.h index cbf0c9d..555f94d 100644 --- a/arch/x86/include/asm/xen/interface.h +++ b/arch/x86/include/asm/xen/interface.h @@ -47,6 +47,10 @@ #endif #ifndef __ASSEMBLY__ +/* Explicitly size integers that represent pfns in the public interface + * with Xen so that on ARM we can have one ABI that works for 32 and 64 + * bit guests. */ +typedef unsigned long xen_pfn_t; /* Guest handles for primitive C types. */ __DEFINE_GUEST_HANDLE(uchar, unsigned char); __DEFINE_GUEST_HANDLE(uint, unsigned int); @@ -57,6 +61,7 @@ DEFINE_GUEST_HANDLE(long); DEFINE_GUEST_HANDLE(void); DEFINE_GUEST_HANDLE(uint64_t); DEFINE_GUEST_HANDLE(uint32_t); +DEFINE_GUEST_HANDLE(xen_pfn_t); #endif #ifndef HYPERVISOR_VIRT_START @@ -121,6 +126,8 @@ struct arch_shared_info { #include "interface_64.h" #endif +#include <asm/pvclock-abi.h> + #ifndef __ASSEMBLY__ /* * The following is all CPU context. Note that the fpu_ctxt block is filled diff --git a/arch/x86/include/asm/xen/swiotlb-xen.h b/arch/x86/include/asm/xen/swiotlb-xen.h index 1be1ab7..ee52fca 100644 --- a/arch/x86/include/asm/xen/swiotlb-xen.h +++ b/arch/x86/include/asm/xen/swiotlb-xen.h @@ -5,10 +5,12 @@ extern int xen_swiotlb; extern int __init pci_xen_swiotlb_detect(void); extern void __init pci_xen_swiotlb_init(void); +extern int pci_xen_swiotlb_init_late(void); #else #define xen_swiotlb (0) static inline int __init pci_xen_swiotlb_detect(void) { return 0; } static inline void __init pci_xen_swiotlb_init(void) { } +static inline int pci_xen_swiotlb_init_late(void) { return -ENXIO; } #endif #endif /* _ASM_X86_SWIOTLB_XEN_H */ diff --git a/arch/x86/xen/apic.c b/arch/x86/xen/apic.c index ec57bd3..7005ced 100644 --- a/arch/x86/xen/apic.c +++ b/arch/x86/xen/apic.c @@ -6,8 +6,9 @@ #include <xen/xen.h> #include <xen/interface/physdev.h> +#include "xen-ops.h" -unsigned int xen_io_apic_read(unsigned apic, unsigned reg) +static unsigned int xen_io_apic_read(unsigned apic, unsigned reg) { struct physdev_apic apic_op; int ret; diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index 1fbe75a..2d932c3 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -80,6 +80,8 @@ #include "smp.h" #include "multicalls.h" +#include <xen/events.h> + EXPORT_SYMBOL_GPL(hypercall_page); DEFINE_PER_CPU(struct vcpu_info *, xen_vcpu); @@ -1288,7 +1290,6 @@ asmlinkage void __init xen_start_kernel(void) { struct physdev_set_iopl set_iopl; int rc; - pgd_t *pgd; if (!xen_start_info) return; @@ -1380,8 +1381,6 @@ asmlinkage void __init xen_start_kernel(void) acpi_numa = -1; #endif - pgd = (pgd_t *)xen_start_info->pt_base; - /* Don't do the full vcpu_info placement stuff until we have a possible map and a non-dummy shared_info. */ per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; @@ -1390,7 +1389,7 @@ asmlinkage void __init xen_start_kernel(void) early_boot_irqs_disabled = true; xen_raw_console_write("mapping kernel into physical memory\n"); - pgd = xen_setup_kernel_pagetable(pgd, xen_start_info->nr_pages); + xen_setup_kernel_pagetable((pgd_t *)xen_start_info->pt_base, xen_start_info->nr_pages); /* Allocate and initialize top and mid mfn levels for p2m structure */ xen_build_mfn_list_list(); @@ -1441,11 +1440,19 @@ asmlinkage void __init xen_start_kernel(void) const struct dom0_vga_console_info *info = (void *)((char *)xen_start_info + xen_start_info->console.dom0.info_off); + struct xen_platform_op op = { + .cmd = XENPF_firmware_info, + .interface_version = XENPF_INTERFACE_VERSION, + .u.firmware_info.type = XEN_FW_KBD_SHIFT_FLAGS, + }; xen_init_vga(info, xen_start_info->console.dom0.info_size); xen_start_info->console.domU.mfn = 0; xen_start_info->console.domU.evtchn = 0; + if (HYPERVISOR_dom0_op(&op) == 0) + boot_params.kbd_status = op.u.firmware_info.u.kbd_shift_flags; + xen_init_apic(); /* Make sure ACS will be enabled */ diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c index 7a769b7..5a16824 100644 --- a/arch/x86/xen/mmu.c +++ b/arch/x86/xen/mmu.c @@ -84,6 +84,7 @@ */ DEFINE_SPINLOCK(xen_reservation_lock); +#ifdef CONFIG_X86_32 /* * Identity map, in addition to plain kernel map. This needs to be * large enough to allocate page table pages to allocate the rest. @@ -91,7 +92,7 @@ DEFINE_SPINLOCK(xen_reservation_lock); */ #define LEVEL1_IDENT_ENTRIES (PTRS_PER_PTE * 4) static RESERVE_BRK_ARRAY(pte_t, level1_ident_pgt, LEVEL1_IDENT_ENTRIES); - +#endif #ifdef CONFIG_X86_64 /* l3 pud for userspace vsyscall mapping */ static pud_t level3_user_vsyscall[PTRS_PER_PUD] __page_aligned_bss; @@ -1176,13 +1177,6 @@ static void xen_exit_mmap(struct mm_struct *mm) static void xen_post_allocator_init(void); -static void __init xen_pagetable_init(void) -{ - paging_init(); - xen_setup_shared_info(); - xen_post_allocator_init(); -} - static __init void xen_mapping_pagetable_reserve(u64 start, u64 end) { /* reserve the range used */ @@ -1197,6 +1191,87 @@ static __init void xen_mapping_pagetable_reserve(u64 start, u64 end) } } +#ifdef CONFIG_X86_64 +static void __init xen_cleanhighmap(unsigned long vaddr, + unsigned long vaddr_end) +{ + unsigned long kernel_end = roundup((unsigned long)_brk_end, PMD_SIZE) - 1; + pmd_t *pmd = level2_kernel_pgt + pmd_index(vaddr); + + /* NOTE: The loop is more greedy than the cleanup_highmap variant. + * We include the PMD passed in on _both_ boundaries. */ + for (; vaddr <= vaddr_end && (pmd < (level2_kernel_pgt + PAGE_SIZE)); + pmd++, vaddr += PMD_SIZE) { + if (pmd_none(*pmd)) + continue; + if (vaddr < (unsigned long) _text || vaddr > kernel_end) + set_pmd(pmd, __pmd(0)); + } + /* In case we did something silly, we should crash in this function + * instead of somewhere later and be confusing. */ + xen_mc_flush(); +} +#endif +static void __init xen_pagetable_init(void) +{ +#ifdef CONFIG_X86_64 + unsigned long size; + unsigned long addr; +#endif + paging_init(); + xen_setup_shared_info(); +#ifdef CONFIG_X86_64 + if (!xen_feature(XENFEAT_auto_translated_physmap)) { + unsigned long new_mfn_list; + + size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long)); + + /* On 32-bit, we get zero so this never gets executed. */ + new_mfn_list = xen_revector_p2m_tree(); + if (new_mfn_list && new_mfn_list != xen_start_info->mfn_list) { + /* using __ka address and sticking INVALID_P2M_ENTRY! */ + memset((void *)xen_start_info->mfn_list, 0xff, size); + + /* We should be in __ka space. */ + BUG_ON(xen_start_info->mfn_list < __START_KERNEL_map); + addr = xen_start_info->mfn_list; + /* We roundup to the PMD, which means that if anybody at this stage is + * using the __ka address of xen_start_info or xen_start_info->shared_info + * they are in going to crash. Fortunatly we have already revectored + * in xen_setup_kernel_pagetable and in xen_setup_shared_info. */ + size = roundup(size, PMD_SIZE); + xen_cleanhighmap(addr, addr + size); + + size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long)); + memblock_free(__pa(xen_start_info->mfn_list), size); + /* And revector! Bye bye old array */ + xen_start_info->mfn_list = new_mfn_list; + } else + goto skip; + } + /* At this stage, cleanup_highmap has already cleaned __ka space + * from _brk_limit way up to the max_pfn_mapped (which is the end of + * the ramdisk). We continue on, erasing PMD entries that point to page + * tables - do note that they are accessible at this stage via __va. + * For good measure we also round up to the PMD - which means that if + * anybody is using __ka address to the initial boot-stack - and try + * to use it - they are going to crash. The xen_start_info has been + * taken care of already in xen_setup_kernel_pagetable. */ + addr = xen_start_info->pt_base; + size = roundup(xen_start_info->nr_pt_frames * PAGE_SIZE, PMD_SIZE); + + xen_cleanhighmap(addr, addr + size); + xen_start_info->pt_base = (unsigned long)__va(__pa(xen_start_info->pt_base)); +#ifdef DEBUG + /* This is superflous and is not neccessary, but you know what + * lets do it. The MODULES_VADDR -> MODULES_END should be clear of + * anything at this stage. */ + xen_cleanhighmap(MODULES_VADDR, roundup(MODULES_VADDR, PUD_SIZE) - 1); +#endif +skip: +#endif + xen_post_allocator_init(); +} static void xen_write_cr2(unsigned long cr2) { this_cpu_read(xen_vcpu)->arch.cr2 = cr2; @@ -1652,7 +1727,7 @@ static void set_page_prot(void *addr, pgprot_t prot) if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, 0)) BUG(); } - +#ifdef CONFIG_X86_32 static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn) { unsigned pmdidx, pteidx; @@ -1703,7 +1778,7 @@ static void __init xen_map_identity_early(pmd_t *pmd, unsigned long max_pfn) set_page_prot(pmd, PAGE_KERNEL_RO); } - +#endif void __init xen_setup_machphys_mapping(void) { struct xen_machphys_mapping mapping; @@ -1731,7 +1806,20 @@ static void convert_pfn_mfn(void *v) for (i = 0; i < PTRS_PER_PTE; i++) pte[i] = xen_make_pte(pte[i].pte); } - +static void __init check_pt_base(unsigned long *pt_base, unsigned long *pt_end, + unsigned long addr) +{ + if (*pt_base == PFN_DOWN(__pa(addr))) { + set_page_prot((void *)addr, PAGE_KERNEL); + clear_page((void *)addr); + (*pt_base)++; + } + if (*pt_end == PFN_DOWN(__pa(addr))) { + set_page_prot((void *)addr, PAGE_KERNEL); + clear_page((void *)addr); + (*pt_end)--; + } +} /* * Set up the initial kernel pagetable. * @@ -1743,11 +1831,13 @@ static void convert_pfn_mfn(void *v) * of the physical mapping once some sort of allocator has been set * up. */ -pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, - unsigned long max_pfn) +void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) { pud_t *l3; pmd_t *l2; + unsigned long addr[3]; + unsigned long pt_base, pt_end; + unsigned i; /* max_pfn_mapped is the last pfn mapped in the initial memory * mappings. Considering that on Xen after the kernel mappings we @@ -1755,32 +1845,53 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, * set max_pfn_mapped to the last real pfn mapped. */ max_pfn_mapped = PFN_DOWN(__pa(xen_start_info->mfn_list)); + pt_base = PFN_DOWN(__pa(xen_start_info->pt_base)); + pt_end = pt_base + xen_start_info->nr_pt_frames; + /* Zap identity mapping */ init_level4_pgt[0] = __pgd(0); /* Pre-constructed entries are in pfn, so convert to mfn */ + /* L4[272] -> level3_ident_pgt + * L4[511] -> level3_kernel_pgt */ convert_pfn_mfn(init_level4_pgt); + + /* L3_i[0] -> level2_ident_pgt */ convert_pfn_mfn(level3_ident_pgt); + /* L3_k[510] -> level2_kernel_pgt + * L3_i[511] -> level2_fixmap_pgt */ convert_pfn_mfn(level3_kernel_pgt); + /* We get [511][511] and have Xen's version of level2_kernel_pgt */ l3 = m2v(pgd[pgd_index(__START_KERNEL_map)].pgd); l2 = m2v(l3[pud_index(__START_KERNEL_map)].pud); - memcpy(level2_ident_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD); - memcpy(level2_kernel_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD); - + addr[0] = (unsigned long)pgd; + addr[1] = (unsigned long)l3; + addr[2] = (unsigned long)l2; + /* Graft it onto L4[272][0]. Note that we creating an aliasing problem: + * Both L4[272][0] and L4[511][511] have entries that point to the same + * L2 (PMD) tables. Meaning that if you modify it in __va space + * it will be also modified in the __ka space! (But if you just + * modify the PMD table to point to other PTE's or none, then you + * are OK - which is what cleanup_highmap does) */ + copy_page(level2_ident_pgt, l2); + /* Graft it onto L4[511][511] */ + copy_page(level2_kernel_pgt, l2); + + /* Get [511][510] and graft that in level2_fixmap_pgt */ l3 = m2v(pgd[pgd_index(__START_KERNEL_map + PMD_SIZE)].pgd); l2 = m2v(l3[pud_index(__START_KERNEL_map + PMD_SIZE)].pud); - memcpy(level2_fixmap_pgt, l2, sizeof(pmd_t) * PTRS_PER_PMD); - - /* Set up identity map */ - xen_map_identity_early(level2_ident_pgt, max_pfn); + copy_page(level2_fixmap_pgt, l2); + /* Note that we don't do anything with level1_fixmap_pgt which + * we don't need. */ /* Make pagetable pieces RO */ set_page_prot(init_level4_pgt, PAGE_KERNEL_RO); set_page_prot(level3_ident_pgt, PAGE_KERNEL_RO); set_page_prot(level3_kernel_pgt, PAGE_KERNEL_RO); set_page_prot(level3_user_vsyscall, PAGE_KERNEL_RO); + set_page_prot(level2_ident_pgt, PAGE_KERNEL_RO); set_page_prot(level2_kernel_pgt, PAGE_KERNEL_RO); set_page_prot(level2_fixmap_pgt, PAGE_KERNEL_RO); @@ -1791,22 +1902,28 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, /* Unpin Xen-provided one */ pin_pagetable_pfn(MMUEXT_UNPIN_TABLE, PFN_DOWN(__pa(pgd))); - /* Switch over */ - pgd = init_level4_pgt; - /* * At this stage there can be no user pgd, and no page * structure to attach it to, so make sure we just set kernel * pgd. */ xen_mc_batch(); - __xen_write_cr3(true, __pa(pgd)); + __xen_write_cr3(true, __pa(init_level4_pgt)); xen_mc_issue(PARAVIRT_LAZY_CPU); - memblock_reserve(__pa(xen_start_info->pt_base), - xen_start_info->nr_pt_frames * PAGE_SIZE); + /* We can't that easily rip out L3 and L2, as the Xen pagetables are + * set out this way: [L4], [L1], [L2], [L3], [L1], [L1] ... for + * the initial domain. For guests using the toolstack, they are in: + * [L4], [L3], [L2], [L1], [L1], order .. So for dom0 we can only + * rip out the [L4] (pgd), but for guests we shave off three pages. + */ + for (i = 0; i < ARRAY_SIZE(addr); i++) + check_pt_base(&pt_base, &pt_end, addr[i]); - return pgd; + /* Our (by three pages) smaller Xen pagetable that we are using */ + memblock_reserve(PFN_PHYS(pt_base), (pt_end - pt_base) * PAGE_SIZE); + /* Revector the xen_start_info */ + xen_start_info = (struct start_info *)__va(__pa(xen_start_info)); } #else /* !CONFIG_X86_64 */ static RESERVE_BRK_ARRAY(pmd_t, initial_kernel_pmd, PTRS_PER_PMD); @@ -1831,8 +1948,7 @@ static void __init xen_write_cr3_init(unsigned long cr3) */ swapper_kernel_pmd = extend_brk(sizeof(pmd_t) * PTRS_PER_PMD, PAGE_SIZE); - memcpy(swapper_kernel_pmd, initial_kernel_pmd, - sizeof(pmd_t) * PTRS_PER_PMD); + copy_page(swapper_kernel_pmd, initial_kernel_pmd); swapper_pg_dir[KERNEL_PGD_BOUNDARY] = __pgd(__pa(swapper_kernel_pmd) | _PAGE_PRESENT); set_page_prot(swapper_kernel_pmd, PAGE_KERNEL_RO); @@ -1849,8 +1965,7 @@ static void __init xen_write_cr3_init(unsigned long cr3) pv_mmu_ops.write_cr3 = &xen_write_cr3; } -pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, - unsigned long max_pfn) +void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn) { pmd_t *kernel_pmd; @@ -1862,11 +1977,11 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, 512*1024); kernel_pmd = m2v(pgd[KERNEL_PGD_BOUNDARY].pgd); - memcpy(initial_kernel_pmd, kernel_pmd, sizeof(pmd_t) * PTRS_PER_PMD); + copy_page(initial_kernel_pmd, kernel_pmd); xen_map_identity_early(initial_kernel_pmd, max_pfn); - memcpy(initial_page_table, pgd, sizeof(pgd_t) * PTRS_PER_PGD); + copy_page(initial_page_table, pgd); initial_page_table[KERNEL_PGD_BOUNDARY] = __pgd(__pa(initial_kernel_pmd) | _PAGE_PRESENT); @@ -1882,8 +1997,6 @@ pgd_t * __init xen_setup_kernel_pagetable(pgd_t *pgd, memblock_reserve(__pa(xen_start_info->pt_base), xen_start_info->nr_pt_frames * PAGE_SIZE); - - return initial_page_table; } #endif /* CONFIG_X86_64 */ @@ -2333,6 +2446,9 @@ int xen_remap_domain_mfn_range(struct vm_area_struct *vma, unsigned long range; int err = 0; + if (xen_feature(XENFEAT_auto_translated_physmap)) + return -EINVAL; + prot = __pgprot(pgprot_val(prot) | _PAGE_IOMAP); BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_RESERVED | VM_IO)) == @@ -2351,8 +2467,8 @@ int xen_remap_domain_mfn_range(struct vm_area_struct *vma, if (err) goto out; - err = -EFAULT; - if (HYPERVISOR_mmu_update(mmu_update, batch, NULL, domid) < 0) + err = HYPERVISOR_mmu_update(mmu_update, batch, NULL, domid); + if (err < 0) goto out; nr -= batch; diff --git a/arch/x86/xen/p2m.c b/arch/x86/xen/p2m.c index 72213da..95fb2aa 100644 --- a/arch/x86/xen/p2m.c +++ b/arch/x86/xen/p2m.c @@ -22,7 +22,7 @@ * * P2M_PER_PAGE depends on the architecture, as a mfn is always * unsigned long (8 bytes on 64-bit, 4 bytes on 32), leading to - * 512 and 1024 entries respectively. + * 512 and 1024 entries respectively. * * In short, these structures contain the Machine Frame Number (MFN) of the PFN. * @@ -139,11 +139,11 @@ * / | ~0, ~0, .... | * | \---------------/ * | - * p2m_missing p2m_missing - * /------------------\ /------------\ - * | [p2m_mid_missing]+---->| ~0, ~0, ~0 | - * | [p2m_mid_missing]+---->| ..., ~0 | - * \------------------/ \------------/ + * p2m_mid_missing p2m_missing + * /-----------------\ /------------\ + * | [p2m_missing] +---->| ~0, ~0, ~0 | + * | [p2m_missing] +---->| ..., ~0 | + * \-----------------/ \------------/ * * where ~0 is INVALID_P2M_ENTRY. IDENTITY is (PFN | IDENTITY_BIT) */ @@ -396,7 +396,85 @@ void __init xen_build_dynamic_phys_to_machine(void) m2p_override_init(); } +#ifdef CONFIG_X86_64 +#include <linux/bootmem.h> +unsigned long __init xen_revector_p2m_tree(void) +{ + unsigned long va_start; + unsigned long va_end; + unsigned long pfn; + unsigned long pfn_free = 0; + unsigned long *mfn_list = NULL; + unsigned long size; + + va_start = xen_start_info->mfn_list; + /*We copy in increments of P2M_PER_PAGE * sizeof(unsigned long), + * so make sure it is rounded up to that */ + size = PAGE_ALIGN(xen_start_info->nr_pages * sizeof(unsigned long)); + va_end = va_start + size; + + /* If we were revectored already, don't do it again. */ + if (va_start <= __START_KERNEL_map && va_start >= __PAGE_OFFSET) + return 0; + + mfn_list = alloc_bootmem_align(size, PAGE_SIZE); + if (!mfn_list) { + pr_warn("Could not allocate space for a new P2M tree!\n"); + return xen_start_info->mfn_list; + } + /* Fill it out with INVALID_P2M_ENTRY value */ + memset(mfn_list, 0xFF, size); + + for (pfn = 0; pfn < ALIGN(MAX_DOMAIN_PAGES, P2M_PER_PAGE); pfn += P2M_PER_PAGE) { + unsigned topidx = p2m_top_index(pfn); + unsigned mididx; + unsigned long *mid_p; + + if (!p2m_top[topidx]) + continue; + + if (p2m_top[topidx] == p2m_mid_missing) + continue; + + mididx = p2m_mid_index(pfn); + mid_p = p2m_top[topidx][mididx]; + if (!mid_p) + continue; + if ((mid_p == p2m_missing) || (mid_p == p2m_identity)) + continue; + + if ((unsigned long)mid_p == INVALID_P2M_ENTRY) + continue; + + /* The old va. Rebase it on mfn_list */ + if (mid_p >= (unsigned long *)va_start && mid_p <= (unsigned long *)va_end) { + unsigned long *new; + + if (pfn_free > (size / sizeof(unsigned long))) { + WARN(1, "Only allocated for %ld pages, but we want %ld!\n", + size / sizeof(unsigned long), pfn_free); + return 0; + } + new = &mfn_list[pfn_free]; + + copy_page(new, mid_p); + p2m_top[topidx][mididx] = &mfn_list[pfn_free]; + p2m_top_mfn_p[topidx][mididx] = virt_to_mfn(&mfn_list[pfn_free]); + + pfn_free += P2M_PER_PAGE; + } + /* This should be the leafs allocated for identity from _brk. */ + } + return (unsigned long)mfn_list; + +} +#else +unsigned long __init xen_revector_p2m_tree(void) +{ + return 0; +} +#endif unsigned long get_phys_to_machine(unsigned long pfn) { unsigned topidx, mididx, idx; @@ -430,7 +508,7 @@ static void free_p2m_page(void *p) free_page((unsigned long)p); } -/* +/* * Fully allocate the p2m structure for a given pfn. We need to check * that both the top and mid levels are allocated, and make sure the * parallel mfn tree is kept in sync. We may race with other cpus, so diff --git a/arch/x86/xen/pci-swiotlb-xen.c b/arch/x86/xen/pci-swiotlb-xen.c index 967633a..9695704 100644 --- a/arch/x86/xen/pci-swiotlb-xen.c +++ b/arch/x86/xen/pci-swiotlb-xen.c @@ -8,6 +8,14 @@ #include <xen/xen.h> #include <asm/iommu_table.h> + +#include <asm/xen/swiotlb-xen.h> +#ifdef CONFIG_X86_64 +#include <asm/iommu.h> +#include <asm/dma.h> +#endif +#include <linux/export.h> + int xen_swiotlb __read_mostly; static struct dma_map_ops xen_swiotlb_dma_ops = { @@ -34,34 +42,64 @@ static struct dma_map_ops xen_swiotlb_dma_ops = { int __init pci_xen_swiotlb_detect(void) { + if (!xen_pv_domain()) + return 0; + /* If running as PV guest, either iommu=soft, or swiotlb=force will * activate this IOMMU. If running as PV privileged, activate it * irregardless. */ - if ((xen_initial_domain() || swiotlb || swiotlb_force) && - (xen_pv_domain())) + if ((xen_initial_domain() || swiotlb || swiotlb_force)) xen_swiotlb = 1; /* If we are running under Xen, we MUST disable the native SWIOTLB. * Don't worry about swiotlb_force flag activating the native, as * the 'swiotlb' flag is the only one turning it on. */ - if (xen_pv_domain()) - swiotlb = 0; + swiotlb = 0; +#ifdef CONFIG_X86_64 + /* pci_swiotlb_detect_4gb turns on native SWIOTLB if no_iommu == 0 + * (so no iommu=X command line over-writes). + * Considering that PV guests do not want the *native SWIOTLB* but + * only Xen SWIOTLB it is not useful to us so set no_iommu=1 here. + */ + if (max_pfn > MAX_DMA32_PFN) + no_iommu = 1; +#endif return xen_swiotlb; } void __init pci_xen_swiotlb_init(void) { if (xen_swiotlb) { - xen_swiotlb_init(1); + xen_swiotlb_init(1, true /* early */); dma_ops = &xen_swiotlb_dma_ops; /* Make sure ACS will be enabled */ pci_request_acs(); } } + +int pci_xen_swiotlb_init_late(void) +{ + int rc; + + if (xen_swiotlb) + return 0; + + rc = xen_swiotlb_init(1, false /* late */); + if (rc) + return rc; + + dma_ops = &xen_swiotlb_dma_ops; + /* Make sure ACS will be enabled */ + pci_request_acs(); + + return 0; +} +EXPORT_SYMBOL_GPL(pci_xen_swiotlb_init_late); + IOMMU_INIT_FINISH(pci_xen_swiotlb_detect, - 0, + NULL, pci_xen_swiotlb_init, - 0); + NULL); diff --git a/arch/x86/xen/platform-pci-unplug.c b/arch/x86/xen/platform-pci-unplug.c index ffcf261..0a78524 100644 --- a/arch/x86/xen/platform-pci-unplug.c +++ b/arch/x86/xen/platform-pci-unplug.c @@ -24,6 +24,7 @@ #include <linux/module.h> #include <xen/platform_pci.h> +#include "xen-ops.h" #define XEN_PLATFORM_ERR_MAGIC -1 #define XEN_PLATFORM_ERR_PROTOCOL -2 diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index e2d62d6..8971a26 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -432,6 +432,24 @@ char * __init xen_memory_setup(void) * - mfn_list * - xen_start_info * See comment above "struct start_info" in <xen/interface/xen.h> + * We tried to make the the memblock_reserve more selective so + * that it would be clear what region is reserved. Sadly we ran + * in the problem wherein on a 64-bit hypervisor with a 32-bit + * initial domain, the pt_base has the cr3 value which is not + * neccessarily where the pagetable starts! As Jan put it: " + * Actually, the adjustment turns out to be correct: The page + * tables for a 32-on-64 dom0 get allocated in the order "first L1", + * "first L2", "first L3", so the offset to the page table base is + * indeed 2. When reading xen/include/public/xen.h's comment + * very strictly, this is not a violation (since there nothing is said + * that the first thing in the page table space is pointed to by + * pt_base; I admit that this seems to be implied though, namely + * do I think that it is implied that the page table space is the + * range [pt_base, pt_base + nt_pt_frames), whereas that + * range here indeed is [pt_base - 2, pt_base - 2 + nt_pt_frames), + * which - without a priori knowledge - the kernel would have + * difficulty to figure out)." - so lets just fall back to the + * easy way and reserve the whole region. */ memblock_reserve(__pa(xen_start_info->mfn_list), xen_start_info->pt_base - xen_start_info->mfn_list); diff --git a/arch/x86/xen/vga.c b/arch/x86/xen/vga.c index 1cd7f4d..6722e37 100644 --- a/arch/x86/xen/vga.c +++ b/arch/x86/xen/vga.c @@ -35,6 +35,7 @@ void __init xen_init_vga(const struct dom0_vga_console_info *info, size_t size) info->u.text_mode_3.font_height; break; + case XEN_VGATYPE_EFI_LFB: case XEN_VGATYPE_VESA_LFB: if (size < offsetof(struct dom0_vga_console_info, u.vesa_lfb.gbl_caps)) @@ -54,6 +55,12 @@ void __init xen_init_vga(const struct dom0_vga_console_info *info, size_t size) screen_info->blue_pos = info->u.vesa_lfb.blue_pos; screen_info->rsvd_size = info->u.vesa_lfb.rsvd_size; screen_info->rsvd_pos = info->u.vesa_lfb.rsvd_pos; + + if (info->video_type == XEN_VGATYPE_EFI_LFB) { + screen_info->orig_video_isVGA = VIDEO_TYPE_EFI; + break; + } + if (size >= offsetof(struct dom0_vga_console_info, u.vesa_lfb.gbl_caps) + sizeof(info->u.vesa_lfb.gbl_caps)) diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S index aaa7291..7faed58 100644 --- a/arch/x86/xen/xen-head.S +++ b/arch/x86/xen/xen-head.S @@ -28,9 +28,61 @@ ENTRY(startup_xen) __FINIT .pushsection .text - .align PAGE_SIZE + .balign PAGE_SIZE ENTRY(hypercall_page) - .skip PAGE_SIZE +#define NEXT_HYPERCALL(x) \ + ENTRY(xen_hypercall_##x) \ + .skip 32 + +NEXT_HYPERCALL(set_trap_table) +NEXT_HYPERCALL(mmu_update) +NEXT_HYPERCALL(set_gdt) +NEXT_HYPERCALL(stack_switch) +NEXT_HYPERCALL(set_callbacks) +NEXT_HYPERCALL(fpu_taskswitch) +NEXT_HYPERCALL(sched_op_compat) +NEXT_HYPERCALL(platform_op) +NEXT_HYPERCALL(set_debugreg) +NEXT_HYPERCALL(get_debugreg) +NEXT_HYPERCALL(update_descriptor) +NEXT_HYPERCALL(ni) +NEXT_HYPERCALL(memory_op) +NEXT_HYPERCALL(multicall) +NEXT_HYPERCALL(update_va_mapping) +NEXT_HYPERCALL(set_timer_op) +NEXT_HYPERCALL(event_channel_op_compat) +NEXT_HYPERCALL(xen_version) +NEXT_HYPERCALL(console_io) +NEXT_HYPERCALL(physdev_op_compat) +NEXT_HYPERCALL(grant_table_op) +NEXT_HYPERCALL(vm_assist) +NEXT_HYPERCALL(update_va_mapping_otherdomain) +NEXT_HYPERCALL(iret) +NEXT_HYPERCALL(vcpu_op) +NEXT_HYPERCALL(set_segment_base) +NEXT_HYPERCALL(mmuext_op) +NEXT_HYPERCALL(xsm_op) +NEXT_HYPERCALL(nmi_op) +NEXT_HYPERCALL(sched_op) +NEXT_HYPERCALL(callback_op) +NEXT_HYPERCALL(xenoprof_op) +NEXT_HYPERCALL(event_channel_op) +NEXT_HYPERCALL(physdev_op) +NEXT_HYPERCALL(hvm_op) +NEXT_HYPERCALL(sysctl) +NEXT_HYPERCALL(domctl) +NEXT_HYPERCALL(kexec_op) +NEXT_HYPERCALL(tmem_op) /* 38 */ +ENTRY(xen_hypercall_rsvr) + .skip 320 +NEXT_HYPERCALL(mca) /* 48 */ +NEXT_HYPERCALL(arch_1) +NEXT_HYPERCALL(arch_2) +NEXT_HYPERCALL(arch_3) +NEXT_HYPERCALL(arch_4) +NEXT_HYPERCALL(arch_5) +NEXT_HYPERCALL(arch_6) + .balign PAGE_SIZE .popsection ELFNOTE(Xen, XEN_ELFNOTE_GUEST_OS, .asciz "linux") diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h index 202d4c1..bb5a810 100644 --- a/arch/x86/xen/xen-ops.h +++ b/arch/x86/xen/xen-ops.h @@ -27,7 +27,7 @@ void xen_setup_mfn_list_list(void); void xen_setup_shared_info(void); void xen_build_mfn_list_list(void); void xen_setup_machphys_mapping(void); -pgd_t *xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn); +void xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn); void xen_reserve_top(void); extern unsigned long xen_max_p2m_pfn; @@ -45,6 +45,7 @@ void xen_hvm_init_shared_info(void); void xen_unplug_emulated_devices(void); void __init xen_build_dynamic_phys_to_machine(void); +unsigned long __init xen_revector_p2m_tree(void); void xen_init_irq_ops(void); void xen_setup_timer(int cpu); diff --git a/drivers/acpi/processor_driver.c b/drivers/acpi/processor_driver.c index bfc31cb..e78c2a5 100644 --- a/drivers/acpi/processor_driver.c +++ b/drivers/acpi/processor_driver.c @@ -475,7 +475,7 @@ static __ref int acpi_processor_start(struct acpi_processor *pr) acpi_processor_get_limit_info(pr); if (!cpuidle_get_driver() || cpuidle_get_driver() == &acpi_idle_driver) - acpi_processor_power_init(pr, device); + acpi_processor_power_init(pr); pr->cdev = thermal_cooling_device_register("Processor", device, &processor_cooling_ops); @@ -509,7 +509,7 @@ err_remove_sysfs_thermal: err_thermal_unregister: thermal_cooling_device_unregister(pr->cdev); err_power_exit: - acpi_processor_power_exit(pr, device); + acpi_processor_power_exit(pr); return result; } @@ -620,7 +620,7 @@ static int acpi_processor_remove(struct acpi_device *device, int type) return -EINVAL; } - acpi_processor_power_exit(pr, device); + acpi_processor_power_exit(pr); sysfs_remove_link(&device->dev.kobj, "sysdev"); @@ -905,8 +905,6 @@ static int __init acpi_processor_init(void) if (acpi_disabled) return 0; - memset(&errata, 0, sizeof(errata)); - result = acpi_bus_register_driver(&acpi_processor_driver); if (result < 0) return result; diff --git a/drivers/acpi/processor_idle.c b/drivers/acpi/processor_idle.c index ad3730b..3655ab9 100644 --- a/drivers/acpi/processor_idle.c +++ b/drivers/acpi/processor_idle.c @@ -79,6 +79,8 @@ module_param(bm_check_disable, uint, 0000); static unsigned int latency_factor __read_mostly = 2; module_param(latency_factor, uint, 0644); +static DEFINE_PER_CPU(struct cpuidle_device *, acpi_cpuidle_device); + static int disabled_by_idle_boot_param(void) { return boot_option_idle_override == IDLE_POLL || @@ -483,8 +485,6 @@ static int acpi_processor_get_power_info_cst(struct acpi_processor *pr) if (obj->type != ACPI_TYPE_INTEGER) continue; - cx.power = obj->integer.value; - current_count++; memcpy(&(pr->power.states[current_count]), &cx, sizeof(cx)); @@ -1000,7 +1000,7 @@ static int acpi_processor_setup_cpuidle_cx(struct acpi_processor *pr) int i, count = CPUIDLE_DRIVER_STATE_START; struct acpi_processor_cx *cx; struct cpuidle_state_usage *state_usage; - struct cpuidle_device *dev = &pr->power.dev; + struct cpuidle_device *dev = per_cpu(acpi_cpuidle_device, pr->id); if (!pr->flags.power_setup_done) return -EINVAL; @@ -1132,6 +1132,7 @@ static int acpi_processor_setup_cpuidle_states(struct acpi_processor *pr) int acpi_processor_hotplug(struct acpi_processor *pr) { int ret = 0; + struct cpuidle_device *dev = per_cpu(acpi_cpuidle_device, pr->id); if (disabled_by_idle_boot_param()) return 0; @@ -1147,11 +1148,11 @@ int acpi_processor_hotplug(struct acpi_processor *pr) return -ENODEV; cpuidle_pause_and_lock(); - cpuidle_disable_device(&pr->power.dev); + cpuidle_disable_device(dev); acpi_processor_get_power_info(pr); if (pr->flags.power) { acpi_processor_setup_cpuidle_cx(pr); - ret = cpuidle_enable_device(&pr->power.dev); + ret = cpuidle_enable_device(dev); } cpuidle_resume_and_unlock(); @@ -1162,6 +1163,7 @@ int acpi_processor_cst_has_changed(struct acpi_processor *pr) { int cpu; struct acpi_processor *_pr; + struct cpuidle_device *dev; if (disabled_by_idle_boot_param()) return 0; @@ -1192,7 +1194,8 @@ int acpi_processor_cst_has_changed(struct acpi_processor *pr) _pr = per_cpu(processors, cpu); if (!_pr || !_pr->flags.power_setup_done) continue; - cpuidle_disable_device(&_pr->power.dev); + dev = per_cpu(acpi_cpuidle_device, cpu); + cpuidle_disable_device(dev); } /* Populate Updated C-state information */ @@ -1206,7 +1209,8 @@ int acpi_processor_cst_has_changed(struct acpi_processor *pr) acpi_processor_get_power_info(_pr); if (_pr->flags.power) { acpi_processor_setup_cpuidle_cx(_pr); - cpuidle_enable_device(&_pr->power.dev); + dev = per_cpu(acpi_cpuidle_device, cpu); + cpuidle_enable_device(dev); } } put_online_cpus(); @@ -1218,11 +1222,11 @@ int acpi_processor_cst_has_changed(struct acpi_processor *pr) static int acpi_processor_registered; -int __cpuinit acpi_processor_power_init(struct acpi_processor *pr, - struct acpi_device *device) +int __cpuinit acpi_processor_power_init(struct acpi_processor *pr) { acpi_status status = 0; int retval; + struct cpuidle_device *dev; static int first_run; if (disabled_by_idle_boot_param()) @@ -1268,11 +1272,18 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr, printk(KERN_DEBUG "ACPI: %s registered with cpuidle\n", acpi_idle_driver.name); } + + dev = kzalloc(sizeof(*dev), GFP_KERNEL); + if (!dev) + return -ENOMEM; + per_cpu(acpi_cpuidle_device, pr->id) = dev; + + acpi_processor_setup_cpuidle_cx(pr); + /* Register per-cpu cpuidle_device. Cpuidle driver * must already be registered before registering device */ - acpi_processor_setup_cpuidle_cx(pr); - retval = cpuidle_register_device(&pr->power.dev); + retval = cpuidle_register_device(dev); if (retval) { if (acpi_processor_registered == 0) cpuidle_unregister_driver(&acpi_idle_driver); @@ -1283,14 +1294,15 @@ int __cpuinit acpi_processor_power_init(struct acpi_processor *pr, return 0; } -int acpi_processor_power_exit(struct acpi_processor *pr, - struct acpi_device *device) +int acpi_processor_power_exit(struct acpi_processor *pr) { + struct cpuidle_device *dev = per_cpu(acpi_cpuidle_device, pr->id); + if (disabled_by_idle_boot_param()) return 0; if (pr->flags.power) { - cpuidle_unregister_device(&pr->power.dev); + cpuidle_unregister_device(dev); acpi_processor_registered--; if (acpi_processor_registered == 0) cpuidle_unregister_driver(&acpi_idle_driver); diff --git a/drivers/acpi/processor_perflib.c b/drivers/acpi/processor_perflib.c index a093dc1..836bfe0 100644 --- a/drivers/acpi/processor_perflib.c +++ b/drivers/acpi/processor_perflib.c @@ -324,6 +324,34 @@ static int acpi_processor_get_performance_control(struct acpi_processor *pr) return result; } +#ifdef CONFIG_X86 +/* + * Some AMDs have 50MHz frequency multiples, but only provide 100MHz rounding + * in their ACPI data. Calculate the real values and fix up the _PSS data. + */ +static void amd_fixup_frequency(struct acpi_processor_px *px, int i) +{ + u32 hi, lo, fid, did; + int index = px->control & 0x00000007; + + if (boot_cpu_data.x86_vendor != X86_VENDOR_AMD) + return; + + if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10) + || boot_cpu_data.x86 == 0x11) { + rdmsr(MSR_AMD_PSTATE_DEF_BASE + index, lo, hi); + fid = lo & 0x3f; + did = (lo >> 6) & 7; + if (boot_cpu_data.x86 == 0x10) + px->core_frequency = (100 * (fid + 0x10)) >> did; + else + px->core_frequency = (100 * (fid + 8)) >> did; + } +} +#else +static void amd_fixup_frequency(struct acpi_processor_px *px, int i) {}; +#endif + static int acpi_processor_get_performance_states(struct acpi_processor *pr) { int result = 0; @@ -379,6 +407,8 @@ static int acpi_processor_get_performance_states(struct acpi_processor *pr) goto end; } + amd_fixup_frequency(px, i); + ACPI_DEBUG_PRINT((ACPI_DB_INFO, "State [%d]: core_frequency[%d] power[%d] transition_latency[%d] bus_master_latency[%d] control[0x%x] status[0x%x]\n", i, diff --git a/drivers/ata/Kconfig b/drivers/ata/Kconfig index 27cecd3..e08d322 100644 --- a/drivers/ata/Kconfig +++ b/drivers/ata/Kconfig @@ -214,6 +214,14 @@ config SATA_DWC_VDEBUG help This option enables the taskfile dumping and NCQ debugging. +config SATA_HIGHBANK + tristate "Calxeda Highbank SATA support" + help + This option enables support for the Calxeda Highbank SoC's + onboard SATA. + + If unsure, say N. + config SATA_MV tristate "Marvell SATA support" help diff --git a/drivers/ata/Makefile b/drivers/ata/Makefile index a454a13..9329daf 100644 --- a/drivers/ata/Makefile +++ b/drivers/ata/Makefile @@ -9,6 +9,7 @@ obj-$(CONFIG_SATA_FSL) += sata_fsl.o obj-$(CONFIG_SATA_INIC162X) += sata_inic162x.o obj-$(CONFIG_SATA_SIL24) += sata_sil24.o obj-$(CONFIG_SATA_DWC) += sata_dwc_460ex.o +obj-$(CONFIG_SATA_HIGHBANK) += sata_highbank.o libahci.o # SFF w/ custom DMA obj-$(CONFIG_PDC_ADMA) += pdc_adma.o diff --git a/drivers/ata/ahci.h b/drivers/ata/ahci.h index 57eb1c2..9be4712 100644 --- a/drivers/ata/ahci.h +++ b/drivers/ata/ahci.h @@ -35,6 +35,7 @@ #ifndef _AHCI_H #define _AHCI_H +#include <linux/clk.h> #include <linux/libata.h> /* Enclosure Management Control */ @@ -115,6 +116,9 @@ enum { HOST_CAP2_BOH = (1 << 0), /* BIOS/OS handoff supported */ HOST_CAP2_NVMHCI = (1 << 1), /* NVMHCI supported */ HOST_CAP2_APST = (1 << 2), /* Automatic partial to slumber */ + HOST_CAP2_SDS = (1 << 3), /* Support device sleep */ + HOST_CAP2_SADM = (1 << 4), /* Support aggressive DevSlp */ + HOST_CAP2_DESO = (1 << 5), /* DevSlp from slumber only */ /* registers for each SATA port */ PORT_LST_ADDR = 0x00, /* command list DMA addr */ @@ -133,6 +137,7 @@ enum { PORT_SCR_ACT = 0x34, /* SATA phy register: SActive */ PORT_SCR_NTF = 0x3c, /* SATA phy register: SNotification */ PORT_FBS = 0x40, /* FIS-based Switching */ + PORT_DEVSLP = 0x44, /* device sleep */ /* PORT_IRQ_{STAT,MASK} bits */ PORT_IRQ_COLD_PRES = (1 << 31), /* cold presence detect */ @@ -186,6 +191,7 @@ enum { PORT_CMD_ICC_PARTIAL = (0x2 << 28), /* Put i/f in partial state */ PORT_CMD_ICC_SLUMBER = (0x6 << 28), /* Put i/f in slumber state */ + /* PORT_FBS bits */ PORT_FBS_DWE_OFFSET = 16, /* FBS device with error offset */ PORT_FBS_ADO_OFFSET = 12, /* FBS active dev optimization offset */ PORT_FBS_DEV_OFFSET = 8, /* FBS device to issue offset */ @@ -194,6 +200,15 @@ enum { PORT_FBS_DEC = (1 << 1), /* FBS device error clear */ PORT_FBS_EN = (1 << 0), /* Enable FBS */ + /* PORT_DEVSLP bits */ + PORT_DEVSLP_DM_OFFSET = 25, /* DITO multiplier offset */ + PORT_DEVSLP_DM_MASK = (0xf << 25), /* DITO multiplier mask */ + PORT_DEVSLP_DITO_OFFSET = 15, /* DITO offset */ + PORT_DEVSLP_MDAT_OFFSET = 10, /* Minimum assertion time */ + PORT_DEVSLP_DETO_OFFSET = 2, /* DevSlp exit timeout */ + PORT_DEVSLP_DSP = (1 << 1), /* DevSlp present */ + PORT_DEVSLP_ADSE = (1 << 0), /* Aggressive DevSlp enable */ + /* hpriv->flags bits */ #define AHCI_HFLAGS(flags) .private_data = (void *)(flags) @@ -302,6 +317,7 @@ struct ahci_host_priv { u32 em_loc; /* enclosure management location */ u32 em_buf_sz; /* EM buffer size in byte */ u32 em_msg_type; /* EM message type */ + struct clk *clk; /* Only for platforms supporting clk */ }; extern int ahci_ignore_sss; diff --git a/drivers/ata/ahci_platform.c b/drivers/ata/ahci_platform.c index 09728e0..b1ae480 100644 --- a/drivers/ata/ahci_platform.c +++ b/drivers/ata/ahci_platform.c @@ -12,6 +12,7 @@ * any later version. */ +#include <linux/clk.h> #include <linux/kernel.h> #include <linux/gfp.h> #include <linux/module.h> @@ -118,6 +119,17 @@ static int __init ahci_probe(struct platform_device *pdev) return -ENOMEM; } + hpriv->clk = clk_get(dev, NULL); + if (IS_ERR(hpriv->clk)) { + dev_err(dev, "can't get clock\n"); + } else { + rc = clk_prepare_enable(hpriv->clk); + if (rc) { + dev_err(dev, "clock prepare enable failed"); + goto free_clk; + } + } + /* * Some platforms might need to prepare for mmio region access, * which could be done in the following init call. So, the mmio @@ -127,7 +139,7 @@ static int __init ahci_probe(struct platform_device *pdev) if (pdata && pdata->init) { rc = pdata->init(dev, hpriv->mmio); if (rc) - return rc; + goto disable_unprepare_clk; } ahci_save_initial_config(dev, hpriv, @@ -153,7 +165,7 @@ static int __init ahci_probe(struct platform_device *pdev) host = ata_host_alloc_pinfo(dev, ppi, n_ports); if (!host) { rc = -ENOMEM; - goto err0; + goto pdata_exit; } host->private_data = hpriv; @@ -183,7 +195,7 @@ static int __init ahci_probe(struct platform_device *pdev) rc = ahci_reset_controller(host); if (rc) - goto err0; + goto pdata_exit; ahci_init_controller(host); ahci_print_info(host, "platform"); @@ -191,12 +203,18 @@ static int __init ahci_probe(struct platform_device *pdev) rc = ata_host_activate(host, irq, ahci_interrupt, IRQF_SHARED, &ahci_platform_sht); if (rc) - goto err0; + goto pdata_exit; return 0; -err0: +pdata_exit: if (pdata && pdata->exit) pdata->exit(dev); +disable_unprepare_clk: + if (!IS_ERR(hpriv->clk)) + clk_disable_unprepare(hpriv->clk); +free_clk: + if (!IS_ERR(hpriv->clk)) + clk_put(hpriv->clk); return rc; } @@ -205,12 +223,18 @@ static int __devexit ahci_remove(struct platform_device *pdev) struct device *dev = &pdev->dev; struct ahci_platform_data *pdata = dev_get_platdata(dev); struct ata_host *host = dev_get_drvdata(dev); + struct ahci_host_priv *hpriv = host->private_data; ata_host_detach(host); if (pdata && pdata->exit) pdata->exit(dev); + if (!IS_ERR(hpriv->clk)) { + clk_disable_unprepare(hpriv->clk); + clk_put(hpriv->clk); + } + return 0; } @@ -245,6 +269,10 @@ static int ahci_suspend(struct device *dev) if (pdata && pdata->suspend) return pdata->suspend(dev); + + if (!IS_ERR(hpriv->clk)) + clk_disable_unprepare(hpriv->clk); + return 0; } @@ -252,18 +280,27 @@ static int ahci_resume(struct device *dev) { struct ahci_platform_data *pdata = dev_get_platdata(dev); struct ata_host *host = dev_get_drvdata(dev); + struct ahci_host_priv *hpriv = host->private_data; int rc; + if (!IS_ERR(hpriv->clk)) { + rc = clk_prepare_enable(hpriv->clk); + if (rc) { + dev_err(dev, "clock prepare enable failed"); + return rc; + } + } + if (pdata && pdata->resume) { rc = pdata->resume(dev); if (rc) - return rc; + goto disable_unprepare_clk; } if (dev->power.power_state.event == PM_EVENT_SUSPEND) { rc = ahci_reset_controller(host); if (rc) - return rc; + goto disable_unprepare_clk; ahci_init_controller(host); } @@ -271,13 +308,18 @@ static int ahci_resume(struct device *dev) ata_host_resume(host); return 0; + +disable_unprepare_clk: + if (!IS_ERR(hpriv->clk)) + clk_disable_unprepare(hpriv->clk); + + return rc; } #endif SIMPLE_DEV_PM_OPS(ahci_pm_ops, ahci_suspend, ahci_resume); static const struct of_device_id ahci_of_match[] = { - { .compatible = "calxeda,hb-ahci", }, { .compatible = "snps,spear-ahci", }, {}, }; diff --git a/drivers/ata/libahci.c b/drivers/ata/libahci.c index 555c07a..4201e53 100644 --- a/drivers/ata/libahci.c +++ b/drivers/ata/libahci.c @@ -45,6 +45,7 @@ #include <scsi/scsi_cmnd.h> #include <linux/libata.h> #include "ahci.h" +#include "libata.h" static int ahci_skip_host_reset; int ahci_ignore_sss; @@ -76,6 +77,7 @@ static void ahci_qc_prep(struct ata_queued_cmd *qc); static int ahci_pmp_qc_defer(struct ata_queued_cmd *qc); static void ahci_freeze(struct ata_port *ap); static void ahci_thaw(struct ata_port *ap); +static void ahci_set_aggressive_devslp(struct ata_port *ap, bool sleep); static void ahci_enable_fbs(struct ata_port *ap); static void ahci_disable_fbs(struct ata_port *ap); static void ahci_pmp_attach(struct ata_port *ap); @@ -193,6 +195,10 @@ module_param(ahci_em_messages, int, 0444); MODULE_PARM_DESC(ahci_em_messages, "AHCI Enclosure Management Message control (0 = off, 1 = on)"); +int devslp_idle_timeout = 1000; /* device sleep idle timeout in ms */ +module_param(devslp_idle_timeout, int, 0644); +MODULE_PARM_DESC(devslp_idle_timeout, "device sleep idle timeout"); + static void ahci_enable_ahci(void __iomem *mmio) { int i; @@ -702,6 +708,16 @@ static int ahci_set_lpm(struct ata_link *link, enum ata_lpm_policy policy, } } + /* set aggressive device sleep */ + if ((hpriv->cap2 & HOST_CAP2_SDS) && + (hpriv->cap2 & HOST_CAP2_SADM) && + (link->device->flags & ATA_DFLAG_DEVSLP)) { + if (policy == ATA_LPM_MIN_POWER) + ahci_set_aggressive_devslp(ap, true); + else + ahci_set_aggressive_devslp(ap, false); + } + if (policy == ATA_LPM_MAX_POWER) { sata_link_scr_lpm(link, policy, false); @@ -1890,6 +1906,81 @@ static void ahci_post_internal_cmd(struct ata_queued_cmd *qc) ahci_kick_engine(ap); } +static void ahci_set_aggressive_devslp(struct ata_port *ap, bool sleep) +{ + void __iomem *port_mmio = ahci_port_base(ap); + struct ata_device *dev = ap->link.device; + u32 devslp, dm, dito, mdat, deto; + int rc; + unsigned int err_mask; + + devslp = readl(port_mmio + PORT_DEVSLP); + if (!(devslp & PORT_DEVSLP_DSP)) { + dev_err(ap->host->dev, "port does not support device sleep\n"); + return; + } + + /* disable device sleep */ + if (!sleep) { + if (devslp & PORT_DEVSLP_ADSE) { + writel(devslp & ~PORT_DEVSLP_ADSE, + port_mmio + PORT_DEVSLP); + err_mask = ata_dev_set_feature(dev, + SETFEATURES_SATA_DISABLE, + SATA_DEVSLP); + if (err_mask && err_mask != AC_ERR_DEV) + ata_dev_warn(dev, "failed to disable DEVSLP\n"); + } + return; + } + + /* device sleep was already enabled */ + if (devslp & PORT_DEVSLP_ADSE) + return; + + /* set DITO, MDAT, DETO and enable DevSlp, need to stop engine first */ + rc = ahci_stop_engine(ap); + if (rc) + return; + + dm = (devslp & PORT_DEVSLP_DM_MASK) >> PORT_DEVSLP_DM_OFFSET; + dito = devslp_idle_timeout / (dm + 1); + if (dito > 0x3ff) + dito = 0x3ff; + + /* Use the nominal value 10 ms if the read MDAT is zero, + * the nominal value of DETO is 20 ms. + */ + if (dev->sata_settings[ATA_LOG_DEVSLP_VALID] & + ATA_LOG_DEVSLP_VALID_MASK) { + mdat = dev->sata_settings[ATA_LOG_DEVSLP_MDAT] & + ATA_LOG_DEVSLP_MDAT_MASK; + if (!mdat) + mdat = 10; + deto = dev->sata_settings[ATA_LOG_DEVSLP_DETO]; + if (!deto) + deto = 20; + } else { + mdat = 10; + deto = 20; + } + + devslp |= ((dito << PORT_DEVSLP_DITO_OFFSET) | + (mdat << PORT_DEVSLP_MDAT_OFFSET) | + (deto << PORT_DEVSLP_DETO_OFFSET) | + PORT_DEVSLP_ADSE); + writel(devslp, port_mmio + PORT_DEVSLP); + + ahci_start_engine(ap); + + /* enable device sleep feature for the drive */ + err_mask = ata_dev_set_feature(dev, + SETFEATURES_SATA_ENABLE, + SATA_DEVSLP); + if (err_mask && err_mask != AC_ERR_DEV) + ata_dev_warn(dev, "failed to enable DEVSLP\n"); +} + static void ahci_enable_fbs(struct ata_port *ap) { struct ahci_port_priv *pp = ap->private_data; @@ -2164,7 +2255,8 @@ void ahci_print_info(struct ata_host *host, const char *scc_s) "flags: " "%s%s%s%s%s%s%s" "%s%s%s%s%s%s%s" - "%s%s%s%s%s%s\n" + "%s%s%s%s%s%s%s" + "%s%s\n" , cap & HOST_CAP_64 ? "64bit " : "", @@ -2184,6 +2276,9 @@ void ahci_print_info(struct ata_host *host, const char *scc_s) cap & HOST_CAP_CCC ? "ccc " : "", cap & HOST_CAP_EMS ? "ems " : "", cap & HOST_CAP_SXS ? "sxs " : "", + cap2 & HOST_CAP2_DESO ? "deso " : "", + cap2 & HOST_CAP2_SADM ? "sadm " : "", + cap2 & HOST_CAP2_SDS ? "sds " : "", cap2 & HOST_CAP2_APST ? "apst " : "", cap2 & HOST_CAP2_NVMHCI ? "nvmp " : "", cap2 & HOST_CAP2_BOH ? "boh " : "" diff --git a/drivers/ata/libata-core.c b/drivers/ata/libata-core.c index 8e1039c..3cc7096 100644 --- a/drivers/ata/libata-core.c +++ b/drivers/ata/libata-core.c @@ -774,7 +774,7 @@ int ata_build_rw_tf(struct ata_taskfile *tf, struct ata_device *dev, tf->lbam = (block >> 8) & 0xff; tf->lbal = block & 0xff; - tf->device = 1 << 6; + tf->device = ATA_LBA; if (tf->flags & ATA_TFLAG_FUA) tf->device |= 1 << 7; } else if (dev->flags & ATA_DFLAG_LBA) { @@ -2155,6 +2155,7 @@ int ata_dev_configure(struct ata_device *dev) int print_info = ehc->i.flags & ATA_EHI_PRINTINFO; const u16 *id = dev->id; unsigned long xfer_mask; + unsigned int err_mask; char revbuf[7]; /* XYZ-99\0 */ char fwrevbuf[ATA_ID_FW_REV_LEN+1]; char modelbuf[ATA_ID_PROD_LEN+1]; @@ -2323,6 +2324,26 @@ int ata_dev_configure(struct ata_device *dev) } } + /* check and mark DevSlp capability */ + if (ata_id_has_devslp(dev->id)) + dev->flags |= ATA_DFLAG_DEVSLP; + + /* Obtain SATA Settings page from Identify Device Data Log, + * which contains DevSlp timing variables etc. + * Exclude old devices with ata_id_has_ncq() + */ + if (ata_id_has_ncq(dev->id)) { + err_mask = ata_read_log_page(dev, + ATA_LOG_SATA_ID_DEV_DATA, + ATA_LOG_SATA_SETTINGS, + dev->sata_settings, + 1); + if (err_mask) + ata_dev_dbg(dev, + "failed to get Identify Device Data, Emask 0x%x\n", + err_mask); + } + dev->cdb_len = 16; } @@ -2351,8 +2372,6 @@ int ata_dev_configure(struct ata_device *dev) (ap->flags & ATA_FLAG_AN) && ata_id_has_atapi_AN(id) && (!sata_pmp_attached(ap) || sata_scr_read(&ap->link, SCR_NOTIFICATION, &sntf) == 0)) { - unsigned int err_mask; - /* issue SET feature command to turn this on */ err_mask = ata_dev_set_feature(dev, SETFEATURES_SATA_ENABLE, SATA_AN); @@ -3598,7 +3617,7 @@ int sata_link_scr_lpm(struct ata_link *link, enum ata_lpm_policy policy, switch (policy) { case ATA_LPM_MAX_POWER: /* disable all LPM transitions */ - scontrol |= (0x3 << 8); + scontrol |= (0x7 << 8); /* initiate transition to active state */ if (spm_wakeup) { scontrol |= (0x4 << 12); @@ -3608,12 +3627,12 @@ int sata_link_scr_lpm(struct ata_link *link, enum ata_lpm_policy policy, case ATA_LPM_MED_POWER: /* allow LPM to PARTIAL */ scontrol &= ~(0x1 << 8); - scontrol |= (0x2 << 8); + scontrol |= (0x6 << 8); break; case ATA_LPM_MIN_POWER: if (ata_link_nr_enabled(link) > 0) /* no restrictions on LPM transitions */ - scontrol &= ~(0x3 << 8); + scontrol &= ~(0x7 << 8); else { /* empty port, power off */ scontrol &= ~0xf; @@ -4472,6 +4491,7 @@ unsigned int ata_dev_set_feature(struct ata_device *dev, u8 enable, u8 feature) DPRINTK("EXIT, err_mask=%x\n", err_mask); return err_mask; } +EXPORT_SYMBOL_GPL(ata_dev_set_feature); /** * ata_dev_init_params - Issue INIT DEV PARAMS command @@ -5253,16 +5273,20 @@ bool ata_link_offline(struct ata_link *link) #ifdef CONFIG_PM static int ata_port_request_pm(struct ata_port *ap, pm_message_t mesg, unsigned int action, unsigned int ehi_flags, - int wait) + int *async) { struct ata_link *link; unsigned long flags; - int rc; + int rc = 0; /* Previous resume operation might still be in * progress. Wait for PM_PENDING to clear. */ if (ap->pflags & ATA_PFLAG_PM_PENDING) { + if (async) { + *async = -EAGAIN; + return 0; + } ata_port_wait_eh(ap); WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING); } @@ -5271,10 +5295,10 @@ static int ata_port_request_pm(struct ata_port *ap, pm_message_t mesg, spin_lock_irqsave(ap->lock, flags); ap->pm_mesg = mesg; - if (wait) { - rc = 0; + if (async) + ap->pm_result = async; + else ap->pm_result = &rc; - } ap->pflags |= ATA_PFLAG_PM_PENDING; ata_for_each_link(link, ap, HOST_FIRST) { @@ -5287,7 +5311,7 @@ static int ata_port_request_pm(struct ata_port *ap, pm_message_t mesg, spin_unlock_irqrestore(ap->lock, flags); /* wait and check result */ - if (wait) { + if (!async) { ata_port_wait_eh(ap); WARN_ON(ap->pflags & ATA_PFLAG_PM_PENDING); } @@ -5295,9 +5319,8 @@ static int ata_port_request_pm(struct ata_port *ap, pm_message_t mesg, return rc; } -static int ata_port_suspend_common(struct device *dev, pm_message_t mesg) +static int __ata_port_suspend_common(struct ata_port *ap, pm_message_t mesg, int *async) { - struct ata_port *ap = to_ata_port(dev); unsigned int ehi_flags = ATA_EHI_QUIET; int rc; @@ -5312,10 +5335,17 @@ static int ata_port_suspend_common(struct device *dev, pm_message_t mesg) if (mesg.event == PM_EVENT_SUSPEND) ehi_flags |= ATA_EHI_NO_AUTOPSY | ATA_EHI_NO_RECOVERY; - rc = ata_port_request_pm(ap, mesg, 0, ehi_flags, 1); + rc = ata_port_request_pm(ap, mesg, 0, ehi_flags, async); return rc; } +static int ata_port_suspend_common(struct device *dev, pm_message_t mesg) +{ + struct ata_port *ap = to_ata_port(dev); + + return __ata_port_suspend_common(ap, mesg, NULL); +} + static int ata_port_suspend(struct device *dev) { if (pm_runtime_suspended(dev)) @@ -5340,16 +5370,22 @@ static int ata_port_poweroff(struct device *dev) return ata_port_suspend_common(dev, PMSG_HIBERNATE); } -static int ata_port_resume_common(struct device *dev) +static int __ata_port_resume_common(struct ata_port *ap, int *async) { - struct ata_port *ap = to_ata_port(dev); int rc; rc = ata_port_request_pm(ap, PMSG_ON, ATA_EH_RESET, - ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, 1); + ATA_EHI_NO_AUTOPSY | ATA_EHI_QUIET, async); return rc; } +static int ata_port_resume_common(struct device *dev) +{ + struct ata_port *ap = to_ata_port(dev); + + return __ata_port_resume_common(ap, NULL); +} + static int ata_port_resume(struct device *dev) { int rc; @@ -5382,6 +5418,24 @@ static const struct dev_pm_ops ata_port_pm_ops = { .runtime_idle = ata_port_runtime_idle, }; +/* sas ports don't participate in pm runtime management of ata_ports, + * and need to resume ata devices at the domain level, not the per-port + * level. sas suspend/resume is async to allow parallel port recovery + * since sas has multiple ata_port instances per Scsi_Host. + */ +int ata_sas_port_async_suspend(struct ata_port *ap, int *async) +{ + return __ata_port_suspend_common(ap, PMSG_SUSPEND, async); +} +EXPORT_SYMBOL_GPL(ata_sas_port_async_suspend); + +int ata_sas_port_async_resume(struct ata_port *ap, int *async) +{ + return __ata_port_resume_common(ap, async); +} +EXPORT_SYMBOL_GPL(ata_sas_port_async_resume); + + /** * ata_host_suspend - suspend host * @host: host to suspend @@ -5927,24 +5981,18 @@ int ata_host_start(struct ata_host *host) } /** - * ata_sas_host_init - Initialize a host struct + * ata_sas_host_init - Initialize a host struct for sas (ipr, libsas) * @host: host to initialize * @dev: device host is attached to - * @flags: host flags * @ops: port_ops * - * LOCKING: - * PCI/etc. bus probe sem. - * */ -/* KILLME - the only user left is ipr */ void ata_host_init(struct ata_host *host, struct device *dev, - unsigned long flags, struct ata_port_operations *ops) + struct ata_port_operations *ops) { spin_lock_init(&host->lock); mutex_init(&host->eh_mutex); host->dev = dev; - host->flags = flags; host->ops = ops; } @@ -6388,6 +6436,7 @@ static int __init ata_parse_force_one(char **cur, { "nohrst", .lflags = ATA_LFLAG_NO_HRST }, { "nosrst", .lflags = ATA_LFLAG_NO_SRST }, { "norst", .lflags = ATA_LFLAG_NO_HRST | ATA_LFLAG_NO_SRST }, + { "rstonce", .lflags = ATA_LFLAG_RST_ONCE }, }; char *start = *cur, *p = *cur; char *id, *val, *endp; diff --git a/drivers/ata/libata-eh.c b/drivers/ata/libata-eh.c index 7d4535e..e60437c 100644 --- a/drivers/ata/libata-eh.c +++ b/drivers/ata/libata-eh.c @@ -1487,6 +1487,7 @@ static const char *ata_err_string(unsigned int err_mask) /** * ata_read_log_page - read a specific log page * @dev: target device + * @log: log to read * @page: page to read * @buf: buffer to store read page * @sectors: number of sectors to read @@ -1499,17 +1500,18 @@ static const char *ata_err_string(unsigned int err_mask) * RETURNS: * 0 on success, AC_ERR_* mask otherwise. */ -static unsigned int ata_read_log_page(struct ata_device *dev, - u8 page, void *buf, unsigned int sectors) +unsigned int ata_read_log_page(struct ata_device *dev, u8 log, + u8 page, void *buf, unsigned int sectors) { struct ata_taskfile tf; unsigned int err_mask; - DPRINTK("read log page - page %d\n", page); + DPRINTK("read log page - log 0x%x, page 0x%x\n", log, page); ata_tf_init(dev, &tf); tf.command = ATA_CMD_READ_LOG_EXT; - tf.lbal = page; + tf.lbal = log; + tf.lbam = page; tf.nsect = sectors; tf.hob_nsect = sectors >> 8; tf.flags |= ATA_TFLAG_ISADDR | ATA_TFLAG_LBA48 | ATA_TFLAG_DEVICE; @@ -1545,7 +1547,7 @@ static int ata_eh_read_log_10h(struct ata_device *dev, u8 csum; int i; - err_mask = ata_read_log_page(dev, ATA_LOG_SATA_NCQ, buf, 1); + err_mask = ata_read_log_page(dev, ATA_LOG_SATA_NCQ, 0, buf, 1); if (err_mask) return -EIO; @@ -2623,6 +2625,8 @@ int ata_eh_reset(struct ata_link *link, int classify, */ while (ata_eh_reset_timeouts[max_tries] != ULONG_MAX) max_tries++; + if (link->flags & ATA_LFLAG_RST_ONCE) + max_tries = 1; if (link->flags & ATA_LFLAG_NO_HRST) hardreset = NULL; if (link->flags & ATA_LFLAG_NO_SRST) diff --git a/drivers/ata/libata-scsi.c b/drivers/ata/libata-scsi.c index 8ec81ca..e3bda074 100644 --- a/drivers/ata/libata-scsi.c +++ b/drivers/ata/libata-scsi.c @@ -1655,7 +1655,7 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc) if (unlikely(scmd->cmd_len < 10)) goto invalid_fld; scsi_10_lba_len(cdb, &block, &n_block); - if (unlikely(cdb[1] & (1 << 3))) + if (cdb[1] & (1 << 3)) tf_flags |= ATA_TFLAG_FUA; break; case READ_6: @@ -1675,7 +1675,7 @@ static unsigned int ata_scsi_rw_xlat(struct ata_queued_cmd *qc) if (unlikely(scmd->cmd_len < 16)) goto invalid_fld; scsi_16_lba_len(cdb, &block, &n_block); - if (unlikely(cdb[1] & (1 << 3))) + if (cdb[1] & (1 << 3)) tf_flags |= ATA_TFLAG_FUA; break; default: @@ -2205,9 +2205,33 @@ static unsigned int ata_scsiop_noop(struct ata_scsi_args *args, u8 *rbuf) } /** + * modecpy - Prepare response for MODE SENSE + * @dest: output buffer + * @src: data being copied + * @n: length of mode page + * @changeable: whether changeable parameters are requested + * + * Generate a generic MODE SENSE page for either current or changeable + * parameters. + * + * LOCKING: + * None. + */ +static void modecpy(u8 *dest, const u8 *src, int n, bool changeable) +{ + if (changeable) { + memcpy(dest, src, 2); + memset(dest + 2, 0, n - 2); + } else { + memcpy(dest, src, n); + } +} + +/** * ata_msense_caching - Simulate MODE SENSE caching info page * @id: device IDENTIFY data * @buf: output buffer + * @changeable: whether changeable parameters are requested * * Generate a caching info page, which conditionally indicates * write caching to the SCSI layer, depending on device @@ -2216,12 +2240,12 @@ static unsigned int ata_scsiop_noop(struct ata_scsi_args *args, u8 *rbuf) * LOCKING: * None. */ -static unsigned int ata_msense_caching(u16 *id, u8 *buf) +static unsigned int ata_msense_caching(u16 *id, u8 *buf, bool changeable) { - memcpy(buf, def_cache_mpage, sizeof(def_cache_mpage)); - if (ata_id_wcache_enabled(id)) + modecpy(buf, def_cache_mpage, sizeof(def_cache_mpage), changeable); + if (changeable || ata_id_wcache_enabled(id)) buf[2] |= (1 << 2); /* write cache enable */ - if (!ata_id_rahead_enabled(id)) + if (!changeable && !ata_id_rahead_enabled(id)) buf[12] |= (1 << 5); /* disable read ahead */ return sizeof(def_cache_mpage); } @@ -2229,30 +2253,33 @@ static unsigned int ata_msense_caching(u16 *id, u8 *buf) /** * ata_msense_ctl_mode - Simulate MODE SENSE control mode page * @buf: output buffer + * @changeable: whether changeable parameters are requested * * Generate a generic MODE SENSE control mode page. * * LOCKING: * None. */ -static unsigned int ata_msense_ctl_mode(u8 *buf) +static unsigned int ata_msense_ctl_mode(u8 *buf, bool changeable) { - memcpy(buf, def_control_mpage, sizeof(def_control_mpage)); + modecpy(buf, def_control_mpage, sizeof(def_control_mpage), changeable); return sizeof(def_control_mpage); } /** * ata_msense_rw_recovery - Simulate MODE SENSE r/w error recovery page * @buf: output buffer + * @changeable: whether changeable parameters are requested * * Generate a generic MODE SENSE r/w error recovery page. * * LOCKING: * None. */ -static unsigned int ata_msense_rw_recovery(u8 *buf) +static unsigned int ata_msense_rw_recovery(u8 *buf, bool changeable) { - memcpy(buf, def_rw_recovery_mpage, sizeof(def_rw_recovery_mpage)); + modecpy(buf, def_rw_recovery_mpage, sizeof(def_rw_recovery_mpage), + changeable); return sizeof(def_rw_recovery_mpage); } @@ -2316,11 +2343,11 @@ static unsigned int ata_scsiop_mode_sense(struct ata_scsi_args *args, u8 *rbuf) page_control = scsicmd[2] >> 6; switch (page_control) { case 0: /* current */ + case 1: /* changeable */ + case 2: /* defaults */ break; /* supported */ case 3: /* saved */ goto saving_not_supp; - case 1: /* changeable */ - case 2: /* defaults */ default: goto invalid_fld; } @@ -2341,21 +2368,21 @@ static unsigned int ata_scsiop_mode_sense(struct ata_scsi_args *args, u8 *rbuf) switch(pg) { case RW_RECOVERY_MPAGE: - p += ata_msense_rw_recovery(p); + p += ata_msense_rw_recovery(p, page_control == 1); break; case CACHE_MPAGE: - p += ata_msense_caching(args->id, p); + p += ata_msense_caching(args->id, p, page_control == 1); break; case CONTROL_MPAGE: - p += ata_msense_ctl_mode(p); + p += ata_msense_ctl_mode(p, page_control == 1); break; case ALL_MPAGES: - p += ata_msense_rw_recovery(p); - p += ata_msense_caching(args->id, p); - p += ata_msense_ctl_mode(p); + p += ata_msense_rw_recovery(p, page_control == 1); + p += ata_msense_caching(args->id, p, page_control == 1); + p += ata_msense_ctl_mode(p, page_control == 1); break; default: /* invalid page code */ @@ -3080,6 +3107,188 @@ static unsigned int ata_scsi_write_same_xlat(struct ata_queued_cmd *qc) } /** + * ata_mselect_caching - Simulate MODE SELECT for caching info page + * @qc: Storage for translated ATA taskfile + * @buf: input buffer + * @len: number of valid bytes in the input buffer + * + * Prepare a taskfile to modify caching information for the device. + * + * LOCKING: + * None. + */ +static int ata_mselect_caching(struct ata_queued_cmd *qc, + const u8 *buf, int len) +{ + struct ata_taskfile *tf = &qc->tf; + struct ata_device *dev = qc->dev; + char mpage[CACHE_MPAGE_LEN]; + u8 wce; + + /* + * The first two bytes of def_cache_mpage are a header, so offsets + * in mpage are off by 2 compared to buf. Same for len. + */ + + if (len != CACHE_MPAGE_LEN - 2) + return -EINVAL; + + wce = buf[0] & (1 << 2); + + /* + * Check that read-only bits are not modified. + */ + ata_msense_caching(dev->id, mpage, false); + mpage[2] &= ~(1 << 2); + mpage[2] |= wce; + if (memcmp(mpage + 2, buf, CACHE_MPAGE_LEN - 2) != 0) + return -EINVAL; + + tf->flags |= ATA_TFLAG_DEVICE | ATA_TFLAG_ISADDR; + tf->protocol = ATA_PROT_NODATA; + tf->nsect = 0; + tf->command = ATA_CMD_SET_FEATURES; + tf->feature = wce ? SETFEATURES_WC_ON : SETFEATURES_WC_OFF; + return 0; +} + +/** + * ata_scsiop_mode_select - Simulate MODE SELECT 6, 10 commands + * @qc: Storage for translated ATA taskfile + * + * Converts a MODE SELECT command to an ATA SET FEATURES taskfile. + * Assume this is invoked for direct access devices (e.g. disks) only. + * There should be no block descriptor for other device types. + * + * LOCKING: + * spin_lock_irqsave(host lock) + */ +static unsigned int ata_scsi_mode_select_xlat(struct ata_queued_cmd *qc) +{ + struct scsi_cmnd *scmd = qc->scsicmd; + const u8 *cdb = scmd->cmnd; + const u8 *p; + u8 pg, spg; + unsigned six_byte, pg_len, hdr_len, bd_len; + int len; + + VPRINTK("ENTER\n"); + + six_byte = (cdb[0] == MODE_SELECT); + if (six_byte) { + if (scmd->cmd_len < 5) + goto invalid_fld; + + len = cdb[4]; + hdr_len = 4; + } else { + if (scmd->cmd_len < 9) + goto invalid_fld; + + len = (cdb[7] << 8) + cdb[8]; + hdr_len = 8; + } + + /* We only support PF=1, SP=0. */ + if ((cdb[1] & 0x11) != 0x10) + goto invalid_fld; + + /* Test early for possible overrun. */ + if (!scsi_sg_count(scmd) || scsi_sglist(scmd)->length < len) + goto invalid_param_len; + + p = page_address(sg_page(scsi_sglist(scmd))); + + /* Move past header and block descriptors. */ + if (len < hdr_len) + goto invalid_param_len; + + if (six_byte) + bd_len = p[3]; + else + bd_len = (p[6] << 8) + p[7]; + + len -= hdr_len; + p += hdr_len; + if (len < bd_len) + goto invalid_param_len; + if (bd_len != 0 && bd_len != 8) + goto invalid_param; + + len -= bd_len; + p += bd_len; + if (len == 0) + goto skip; + + /* Parse both possible formats for the mode page headers. */ + pg = p[0] & 0x3f; + if (p[0] & 0x40) { + if (len < 4) + goto invalid_param_len; + + spg = p[1]; + pg_len = (p[2] << 8) | p[3]; + p += 4; + len -= 4; + } else { + if (len < 2) + goto invalid_param_len; + + spg = 0; + pg_len = p[1]; + p += 2; + len -= 2; + } + + /* + * No mode subpages supported (yet) but asking for _all_ + * subpages may be valid + */ + if (spg && (spg != ALL_SUB_MPAGES)) + goto invalid_param; + if (pg_len > len) + goto invalid_param_len; + + switch (pg) { + case CACHE_MPAGE: + if (ata_mselect_caching(qc, p, pg_len) < 0) + goto invalid_param; + break; + + default: /* invalid page code */ + goto invalid_param; + } + + /* + * Only one page has changeable data, so we only support setting one + * page at a time. + */ + if (len > pg_len) + goto invalid_param; + + return 0; + + invalid_fld: + /* "Invalid field in CDB" */ + ata_scsi_set_sense(scmd, ILLEGAL_REQUEST, 0x24, 0x0); + return 1; + + invalid_param: + /* "Invalid field in parameter list" */ + ata_scsi_set_sense(scmd, ILLEGAL_REQUEST, 0x26, 0x0); + return 1; + + invalid_param_len: + /* "Parameter list length error" */ + ata_scsi_set_sense(scmd, ILLEGAL_REQUEST, 0x1a, 0x0); + return 1; + + skip: + scmd->result = SAM_STAT_GOOD; + return 1; +} + +/** * ata_get_xlat_func - check if SCSI to ATA translation is possible * @dev: ATA device * @cmd: SCSI command opcode to consider @@ -3119,6 +3328,11 @@ static inline ata_xlat_func_t ata_get_xlat_func(struct ata_device *dev, u8 cmd) case ATA_16: return ata_scsi_pass_thru; + case MODE_SELECT: + case MODE_SELECT_10: + return ata_scsi_mode_select_xlat; + break; + case START_STOP: return ata_scsi_start_stop_xlat; } @@ -3311,11 +3525,6 @@ void ata_scsi_simulate(struct ata_device *dev, struct scsi_cmnd *cmd) ata_scsi_rbuf_fill(&args, ata_scsiop_mode_sense); break; - case MODE_SELECT: /* unconditionally return */ - case MODE_SELECT_10: /* bad-field-in-cdb */ - ata_scsi_invalid_field(cmd); - break; - case READ_CAPACITY: ata_scsi_rbuf_fill(&args, ata_scsiop_read_cap); break; diff --git a/drivers/ata/libata.h b/drivers/ata/libata.h index 50e4dff..7148a58 100644 --- a/drivers/ata/libata.h +++ b/drivers/ata/libata.h @@ -165,6 +165,8 @@ extern void ata_eh_about_to_do(struct ata_link *link, struct ata_device *dev, unsigned int action); extern void ata_eh_done(struct ata_link *link, struct ata_device *dev, unsigned int action); +extern unsigned int ata_read_log_page(struct ata_device *dev, u8 log, + u8 page, void *buf, unsigned int sectors); extern void ata_eh_autopsy(struct ata_port *ap); const char *ata_get_cmd_descript(u8 command); extern void ata_eh_report(struct ata_port *ap); diff --git a/drivers/ata/pata_arasan_cf.c b/drivers/ata/pata_arasan_cf.c index bfaa5cb..26201eb 100644 --- a/drivers/ata/pata_arasan_cf.c +++ b/drivers/ata/pata_arasan_cf.c @@ -31,6 +31,7 @@ #include <linux/kernel.h> #include <linux/libata.h> #include <linux/module.h> +#include <linux/of.h> #include <linux/pata_arasan_cf_data.h> #include <linux/platform_device.h> #include <linux/pm.h> @@ -310,7 +311,7 @@ static int cf_init(struct arasan_cf_dev *acdev) unsigned long flags; int ret = 0; - ret = clk_enable(acdev->clk); + ret = clk_prepare_enable(acdev->clk); if (ret) { dev_dbg(acdev->host->dev, "clock enable failed"); return ret; @@ -340,7 +341,7 @@ static void cf_exit(struct arasan_cf_dev *acdev) writel(readl(acdev->vbase + OP_MODE) & ~CFHOST_ENB, acdev->vbase + OP_MODE); spin_unlock_irqrestore(&acdev->host->lock, flags); - clk_disable(acdev->clk); + clk_disable_unprepare(acdev->clk); } static void dma_callback(void *dev) @@ -935,6 +936,14 @@ static int arasan_cf_resume(struct device *dev) static SIMPLE_DEV_PM_OPS(arasan_cf_pm_ops, arasan_cf_suspend, arasan_cf_resume); +#ifdef CONFIG_OF +static const struct of_device_id arasan_cf_id_table[] = { + { .compatible = "arasan,cf-spear1340" }, + {} +}; +MODULE_DEVICE_TABLE(of, arasan_cf_id_table); +#endif + static struct platform_driver arasan_cf_driver = { .probe = arasan_cf_probe, .remove = __devexit_p(arasan_cf_remove), @@ -942,6 +951,7 @@ static struct platform_driver arasan_cf_driver = { .name = DRIVER_NAME, .owner = THIS_MODULE, .pm = &arasan_cf_pm_ops, + .of_match_table = of_match_ptr(arasan_cf_id_table), }, }; diff --git a/drivers/ata/sata_fsl.c b/drivers/ata/sata_fsl.c index d6577b9..124b2c1 100644 --- a/drivers/ata/sata_fsl.c +++ b/drivers/ata/sata_fsl.c @@ -123,6 +123,7 @@ enum { ONLINE = (1 << 31), GOING_OFFLINE = (1 << 30), BIST_ERR = (1 << 29), + CLEAR_ERROR = (1 << 27), FATAL_ERR_HC_MASTER_ERR = (1 << 18), FATAL_ERR_PARITY_ERR_TX = (1 << 17), @@ -143,6 +144,7 @@ enum { FATAL_ERR_CRC_ERR_RX | FATAL_ERR_FIFO_OVRFL_TX | FATAL_ERR_FIFO_OVRFL_RX, + INT_ON_DATA_LENGTH_MISMATCH = (1 << 12), INT_ON_FATAL_ERR = (1 << 5), INT_ON_PHYRDY_CHG = (1 << 4), @@ -1181,25 +1183,54 @@ static void sata_fsl_host_intr(struct ata_port *ap) u32 hstatus, done_mask = 0; struct ata_queued_cmd *qc; u32 SError; + u32 tag; + u32 status_mask = INT_ON_ERROR; hstatus = ioread32(hcr_base + HSTATUS); sata_fsl_scr_read(&ap->link, SCR_ERROR, &SError); + /* Read command completed register */ + done_mask = ioread32(hcr_base + CC); + + /* Workaround for data length mismatch errata */ + if (unlikely(hstatus & INT_ON_DATA_LENGTH_MISMATCH)) { + for (tag = 0; tag < ATA_MAX_QUEUE; tag++) { + qc = ata_qc_from_tag(ap, tag); + if (qc && ata_is_atapi(qc->tf.protocol)) { + u32 hcontrol; + /* Set HControl[27] to clear error registers */ + hcontrol = ioread32(hcr_base + HCONTROL); + iowrite32(hcontrol | CLEAR_ERROR, + hcr_base + HCONTROL); + + /* Clear HControl[27] */ + iowrite32(hcontrol & ~CLEAR_ERROR, + hcr_base + HCONTROL); + + /* Clear SError[E] bit */ + sata_fsl_scr_write(&ap->link, SCR_ERROR, + SError); + + /* Ignore fatal error and device error */ + status_mask &= ~(INT_ON_SINGL_DEVICE_ERR + | INT_ON_FATAL_ERR); + break; + } + } + } + if (unlikely(SError & 0xFFFF0000)) { DPRINTK("serror @host_intr : 0x%x\n", SError); sata_fsl_error_intr(ap); } - if (unlikely(hstatus & INT_ON_ERROR)) { + if (unlikely(hstatus & status_mask)) { DPRINTK("error interrupt!!\n"); sata_fsl_error_intr(ap); return; } - /* Read command completed register */ - done_mask = ioread32(hcr_base + CC); - VPRINTK("Status of all queues :\n"); VPRINTK("done_mask/CC = 0x%x, CA = 0x%x, CE=0x%x,CQ=0x%x,apqa=0x%x\n", done_mask, diff --git a/drivers/ata/sata_highbank.c b/drivers/ata/sata_highbank.c new file mode 100644 index 0000000..0d7c4c2 --- /dev/null +++ b/drivers/ata/sata_highbank.c @@ -0,0 +1,450 @@ +/* + * Calxeda Highbank AHCI SATA platform driver + * Copyright 2012 Calxeda, Inc. + * + * based on the AHCI SATA platform driver by Jeff Garzik and Anton Vorontsov + * + * This program is free software; you can redistribute it and/or modify it + * under the terms and conditions of the GNU General Public License, + * version 2, as published by the Free Software Foundation. + * + * This program is distributed in the hope it will be useful, but WITHOUT + * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or + * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for + * more details. + * + * You should have received a copy of the GNU General Public License along with + * this program. If not, see <http://www.gnu.org/licenses/>. + */ +#include <linux/kernel.h> +#include <linux/gfp.h> +#include <linux/module.h> +#include <linux/init.h> +#include <linux/types.h> +#include <linux/err.h> +#include <linux/io.h> +#include <linux/spinlock.h> +#include <linux/device.h> +#include <linux/of_device.h> +#include <linux/of_address.h> +#include <linux/platform_device.h> +#include <linux/libata.h> +#include <linux/ahci_platform.h> +#include <linux/interrupt.h> +#include <linux/delay.h> +#include <linux/export.h> +#include "ahci.h" + +#define CPHY_MAP(dev, addr) ((((dev) & 0x1f) << 7) | (((addr) >> 9) & 0x7f)) +#define CPHY_ADDR(addr) (((addr) & 0x1ff) << 2) +#define SERDES_CR_CTL 0x80a0 +#define SERDES_CR_ADDR 0x80a1 +#define SERDES_CR_DATA 0x80a2 +#define CR_BUSY 0x0001 +#define CR_START 0x0001 +#define CR_WR_RDN 0x0002 +#define CPHY_RX_INPUT_STS 0x2002 +#define CPHY_SATA_OVERRIDE 0x4000 +#define CPHY_OVERRIDE 0x2005 +#define SPHY_LANE 0x100 +#define SPHY_HALF_RATE 0x0001 +#define CPHY_SATA_DPLL_MODE 0x0700 +#define CPHY_SATA_DPLL_SHIFT 8 +#define CPHY_SATA_DPLL_RESET (1 << 11) +#define CPHY_PHY_COUNT 6 +#define CPHY_LANE_COUNT 4 +#define CPHY_PORT_COUNT (CPHY_PHY_COUNT * CPHY_LANE_COUNT) + +static DEFINE_SPINLOCK(cphy_lock); +/* Each of the 6 phys can have up to 4 sata ports attached to i. Map 0-based + * sata ports to their phys and then to their lanes within the phys + */ +struct phy_lane_info { + void __iomem *phy_base; + u8 lane_mapping; + u8 phy_devs; +}; +static struct phy_lane_info port_data[CPHY_PORT_COUNT]; + +static u32 __combo_phy_reg_read(u8 sata_port, u32 addr) +{ + u32 data; + u8 dev = port_data[sata_port].phy_devs; + spin_lock(&cphy_lock); + writel(CPHY_MAP(dev, addr), port_data[sata_port].phy_base + 0x800); + data = readl(port_data[sata_port].phy_base + CPHY_ADDR(addr)); + spin_unlock(&cphy_lock); + return data; +} + +static void __combo_phy_reg_write(u8 sata_port, u32 addr, u32 data) +{ + u8 dev = port_data[sata_port].phy_devs; + spin_lock(&cphy_lock); + writel(CPHY_MAP(dev, addr), port_data[sata_port].phy_base + 0x800); + writel(data, port_data[sata_port].phy_base + CPHY_ADDR(addr)); + spin_unlock(&cphy_lock); +} + +static void combo_phy_wait_for_ready(u8 sata_port) +{ + while (__combo_phy_reg_read(sata_port, SERDES_CR_CTL) & CR_BUSY) + udelay(5); +} + +static u32 combo_phy_read(u8 sata_port, u32 addr) +{ + combo_phy_wait_for_ready(sata_port); + __combo_phy_reg_write(sata_port, SERDES_CR_ADDR, addr); + __combo_phy_reg_write(sata_port, SERDES_CR_CTL, CR_START); + combo_phy_wait_for_ready(sata_port); + return __combo_phy_reg_read(sata_port, SERDES_CR_DATA); +} + +static void combo_phy_write(u8 sata_port, u32 addr, u32 data) +{ + combo_phy_wait_for_ready(sata_port); + __combo_phy_reg_write(sata_port, SERDES_CR_ADDR, addr); + __combo_phy_reg_write(sata_port, SERDES_CR_DATA, data); + __combo_phy_reg_write(sata_port, SERDES_CR_CTL, CR_WR_RDN | CR_START); +} + +static void highbank_cphy_disable_overrides(u8 sata_port) +{ + u8 lane = port_data[sata_port].lane_mapping; + u32 tmp; + if (unlikely(port_data[sata_port].phy_base == NULL)) + return; + tmp = combo_phy_read(sata_port, CPHY_RX_INPUT_STS + lane * SPHY_LANE); + tmp &= ~CPHY_SATA_OVERRIDE; + combo_phy_write(sata_port, CPHY_OVERRIDE + lane * SPHY_LANE, tmp); +} + +static void cphy_override_rx_mode(u8 sata_port, u32 val) +{ + u8 lane = port_data[sata_port].lane_mapping; + u32 tmp; + tmp = combo_phy_read(sata_port, CPHY_RX_INPUT_STS + lane * SPHY_LANE); + tmp &= ~CPHY_SATA_OVERRIDE; + combo_phy_write(sata_port, CPHY_OVERRIDE + lane * SPHY_LANE, tmp); + + tmp |= CPHY_SATA_OVERRIDE; + combo_phy_write(sata_port, CPHY_OVERRIDE + lane * SPHY_LANE, tmp); + + tmp &= ~CPHY_SATA_DPLL_MODE; + tmp |= val << CPHY_SATA_DPLL_SHIFT; + combo_phy_write(sata_port, CPHY_OVERRIDE + lane * SPHY_LANE, tmp); + + tmp |= CPHY_SATA_DPLL_RESET; + combo_phy_write(sata_port, CPHY_OVERRIDE + lane * SPHY_LANE, tmp); + + tmp &= ~CPHY_SATA_DPLL_RESET; + combo_phy_write(sata_port, CPHY_OVERRIDE + lane * SPHY_LANE, tmp); + + msleep(15); +} + +static void highbank_cphy_override_lane(u8 sata_port) +{ + u8 lane = port_data[sata_port].lane_mapping; + u32 tmp, k = 0; + + if (unlikely(port_data[sata_port].phy_base == NULL)) + return; + do { + tmp = combo_phy_read(sata_port, CPHY_RX_INPUT_STS + + lane * SPHY_LANE); + } while ((tmp & SPHY_HALF_RATE) && (k++ < 1000)); + cphy_override_rx_mode(sata_port, 3); +} + +static int highbank_initialize_phys(struct device *dev, void __iomem *addr) +{ + struct device_node *sata_node = dev->of_node; + int phy_count = 0, phy, port = 0; + void __iomem *cphy_base[CPHY_PHY_COUNT]; + struct device_node *phy_nodes[CPHY_PHY_COUNT]; + memset(port_data, 0, sizeof(struct phy_lane_info) * CPHY_PORT_COUNT); + memset(phy_nodes, 0, sizeof(struct device_node*) * CPHY_PHY_COUNT); + + do { + u32 tmp; + struct of_phandle_args phy_data; + if (of_parse_phandle_with_args(sata_node, + "calxeda,port-phys", "#phy-cells", + port, &phy_data)) + break; + for (phy = 0; phy < phy_count; phy++) { + if (phy_nodes[phy] == phy_data.np) + break; + } + if (phy_nodes[phy] == NULL) { + phy_nodes[phy] = phy_data.np; + cphy_base[phy] = of_iomap(phy_nodes[phy], 0); + if (cphy_base[phy] == NULL) { + return 0; + } + phy_count += 1; + } + port_data[port].lane_mapping = phy_data.args[0]; + of_property_read_u32(phy_nodes[phy], "phydev", &tmp); + port_data[port].phy_devs = tmp; + port_data[port].phy_base = cphy_base[phy]; + of_node_put(phy_data.np); + port += 1; + } while (port < CPHY_PORT_COUNT); + return 0; +} + +static int ahci_highbank_hardreset(struct ata_link *link, unsigned int *class, + unsigned long deadline) +{ + const unsigned long *timing = sata_ehc_deb_timing(&link->eh_context); + struct ata_port *ap = link->ap; + struct ahci_port_priv *pp = ap->private_data; + u8 *d2h_fis = pp->rx_fis + RX_FIS_D2H_REG; + struct ata_taskfile tf; + bool online; + u32 sstatus; + int rc; + int retry = 10; + + ahci_stop_engine(ap); + + /* clear D2H reception area to properly wait for D2H FIS */ + ata_tf_init(link->device, &tf); + tf.command = 0x80; + ata_tf_to_fis(&tf, 0, 0, d2h_fis); + + do { + highbank_cphy_disable_overrides(link->ap->port_no); + rc = sata_link_hardreset(link, timing, deadline, &online, NULL); + highbank_cphy_override_lane(link->ap->port_no); + + /* If the status is 1, we are connected, but the link did not + * come up. So retry resetting the link again. + */ + if (sata_scr_read(link, SCR_STATUS, &sstatus)) + break; + if (!(sstatus & 0x3)) + break; + } while (!online && retry--); + + ahci_start_engine(ap); + + if (online) + *class = ahci_dev_classify(ap); + + return rc; +} + +static struct ata_port_operations ahci_highbank_ops = { + .inherits = &ahci_ops, + .hardreset = ahci_highbank_hardreset, +}; + +static const struct ata_port_info ahci_highbank_port_info = { + .flags = AHCI_FLAG_COMMON, + .pio_mask = ATA_PIO4, + .udma_mask = ATA_UDMA6, + .port_ops = &ahci_highbank_ops, +}; + +static struct scsi_host_template ahci_highbank_platform_sht = { + AHCI_SHT("highbank-ahci"), +}; + +static const struct of_device_id ahci_of_match[] = { + { .compatible = "calxeda,hb-ahci" }, + {}, +}; +MODULE_DEVICE_TABLE(of, ahci_of_match); + +static int __init ahci_highbank_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct ahci_host_priv *hpriv; + struct ata_host *host; + struct resource *mem; + int irq; + int n_ports; + int i; + int rc; + struct ata_port_info pi = ahci_highbank_port_info; + const struct ata_port_info *ppi[] = { &pi, NULL }; + + mem = platform_get_resource(pdev, IORESOURCE_MEM, 0); + if (!mem) { + dev_err(dev, "no mmio space\n"); + return -EINVAL; + } + + irq = platform_get_irq(pdev, 0); + if (irq <= 0) { + dev_err(dev, "no irq\n"); + return -EINVAL; + } + + hpriv = devm_kzalloc(dev, sizeof(*hpriv), GFP_KERNEL); + if (!hpriv) { + dev_err(dev, "can't alloc ahci_host_priv\n"); + return -ENOMEM; + } + + hpriv->flags |= (unsigned long)pi.private_data; + + hpriv->mmio = devm_ioremap(dev, mem->start, resource_size(mem)); + if (!hpriv->mmio) { + dev_err(dev, "can't map %pR\n", mem); + return -ENOMEM; + } + + rc = highbank_initialize_phys(dev, hpriv->mmio); + if (rc) + return rc; + + + ahci_save_initial_config(dev, hpriv, 0, 0); + + /* prepare host */ + if (hpriv->cap & HOST_CAP_NCQ) + pi.flags |= ATA_FLAG_NCQ; + + if (hpriv->cap & HOST_CAP_PMP) + pi.flags |= ATA_FLAG_PMP; + + ahci_set_em_messages(hpriv, &pi); + + /* CAP.NP sometimes indicate the index of the last enabled + * port, at other times, that of the last possible port, so + * determining the maximum port number requires looking at + * both CAP.NP and port_map. + */ + n_ports = max(ahci_nr_ports(hpriv->cap), fls(hpriv->port_map)); + + host = ata_host_alloc_pinfo(dev, ppi, n_ports); + if (!host) { + rc = -ENOMEM; + goto err0; + } + + host->private_data = hpriv; + + if (!(hpriv->cap & HOST_CAP_SSS) || ahci_ignore_sss) + host->flags |= ATA_HOST_PARALLEL_SCAN; + + if (pi.flags & ATA_FLAG_EM) + ahci_reset_em(host); + + for (i = 0; i < host->n_ports; i++) { + struct ata_port *ap = host->ports[i]; + + ata_port_desc(ap, "mmio %pR", mem); + ata_port_desc(ap, "port 0x%x", 0x100 + ap->port_no * 0x80); + + /* set enclosure management message type */ + if (ap->flags & ATA_FLAG_EM) + ap->em_message_type = hpriv->em_msg_type; + + /* disabled/not-implemented port */ + if (!(hpriv->port_map & (1 << i))) + ap->ops = &ata_dummy_port_ops; + } + + rc = ahci_reset_controller(host); + if (rc) + goto err0; + + ahci_init_controller(host); + ahci_print_info(host, "platform"); + + rc = ata_host_activate(host, irq, ahci_interrupt, 0, + &ahci_highbank_platform_sht); + if (rc) + goto err0; + + return 0; +err0: + return rc; +} + +static int __devexit ahci_highbank_remove(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct ata_host *host = dev_get_drvdata(dev); + + ata_host_detach(host); + + return 0; +} + +#ifdef CONFIG_PM +static int ahci_highbank_suspend(struct device *dev) +{ + struct ata_host *host = dev_get_drvdata(dev); + struct ahci_host_priv *hpriv = host->private_data; + void __iomem *mmio = hpriv->mmio; + u32 ctl; + int rc; + + if (hpriv->flags & AHCI_HFLAG_NO_SUSPEND) { + dev_err(dev, "firmware update required for suspend/resume\n"); + return -EIO; + } + + /* + * AHCI spec rev1.1 section 8.3.3: + * Software must disable interrupts prior to requesting a + * transition of the HBA to D3 state. + */ + ctl = readl(mmio + HOST_CTL); + ctl &= ~HOST_IRQ_EN; + writel(ctl, mmio + HOST_CTL); + readl(mmio + HOST_CTL); /* flush */ + + rc = ata_host_suspend(host, PMSG_SUSPEND); + if (rc) + return rc; + + return 0; +} + +static int ahci_highbank_resume(struct device *dev) +{ + struct ata_host *host = dev_get_drvdata(dev); + int rc; + + if (dev->power.power_state.event == PM_EVENT_SUSPEND) { + rc = ahci_reset_controller(host); + if (rc) + return rc; + + ahci_init_controller(host); + } + + ata_host_resume(host); + + return 0; +} +#endif + +SIMPLE_DEV_PM_OPS(ahci_highbank_pm_ops, + ahci_highbank_suspend, ahci_highbank_resume); + +static struct platform_driver ahci_highbank_driver = { + .remove = __devexit_p(ahci_highbank_remove), + .driver = { + .name = "highbank-ahci", + .owner = THIS_MODULE, + .of_match_table = ahci_of_match, + .pm = &ahci_highbank_pm_ops, + }, + .probe = ahci_highbank_probe, +}; + +module_platform_driver(ahci_highbank_driver); + +MODULE_DESCRIPTION("Calxeda Highbank AHCI SATA platform driver"); +MODULE_AUTHOR("Mark Langsdorf <mark.langsdorf@calxeda.com>"); +MODULE_LICENSE("GPL"); +MODULE_ALIAS("sata:highbank"); diff --git a/drivers/ata/sata_mv.c b/drivers/ata/sata_mv.c index 311be18..68f4fb5 100644 --- a/drivers/ata/sata_mv.c +++ b/drivers/ata/sata_mv.c @@ -79,8 +79,8 @@ * module options */ -static int msi; #ifdef CONFIG_PCI +static int msi; module_param(msi, int, S_IRUGO); MODULE_PARM_DESC(msi, "Enable use of PCI MSI (0=off, 1=on)"); #endif @@ -652,12 +652,13 @@ static u8 mv_sff_check_status(struct ata_port *ap); * because we have to allow room for worst case splitting of * PRDs for 64K boundaries in mv_fill_sg(). */ +#ifdef CONFIG_PCI static struct scsi_host_template mv5_sht = { ATA_BASE_SHT(DRV_NAME), .sg_tablesize = MV_MAX_SG_CT / 2, .dma_boundary = MV_DMA_BOUNDARY, }; - +#endif static struct scsi_host_template mv6_sht = { ATA_NCQ_SHT(DRV_NAME), .can_queue = MV_MAX_Q_DEPTH - 1, @@ -1252,7 +1253,7 @@ static void mv_dump_mem(void __iomem *start, unsigned bytes) } } #endif - +#if defined(ATA_DEBUG) || defined(CONFIG_PCI) static void mv_dump_pci_cfg(struct pci_dev *pdev, unsigned bytes) { #ifdef ATA_DEBUG @@ -1269,6 +1270,7 @@ static void mv_dump_pci_cfg(struct pci_dev *pdev, unsigned bytes) } #endif } +#endif static void mv_dump_all_regs(void __iomem *mmio_base, int port, struct pci_dev *pdev) { diff --git a/drivers/base/dma-buf.c b/drivers/base/dma-buf.c index c30f3e1..460e22d 100644 --- a/drivers/base/dma-buf.c +++ b/drivers/base/dma-buf.c @@ -460,8 +460,7 @@ int dma_buf_mmap(struct dma_buf *dmabuf, struct vm_area_struct *vma, if (vma->vm_file) fput(vma->vm_file); - vma->vm_file = dmabuf->file; - get_file(vma->vm_file); + vma->vm_file = get_file(dmabuf->file); vma->vm_pgoff = pgoff; diff --git a/drivers/base/dma-contiguous.c b/drivers/base/dma-contiguous.c index 34d94c7..9a14694 100644 --- a/drivers/base/dma-contiguous.c +++ b/drivers/base/dma-contiguous.c @@ -315,6 +315,7 @@ struct page *dma_alloc_from_contiguous(struct device *dev, int count, { unsigned long mask, pfn, pageno, start = 0; struct cma *cma = dev_get_cma_area(dev); + struct page *page = NULL; int ret; if (!cma || !cma->count) @@ -336,18 +337,17 @@ struct page *dma_alloc_from_contiguous(struct device *dev, int count, for (;;) { pageno = bitmap_find_next_zero_area(cma->bitmap, cma->count, start, count, mask); - if (pageno >= cma->count) { - ret = -ENOMEM; - goto error; - } + if (pageno >= cma->count) + break; pfn = cma->base_pfn + pageno; ret = alloc_contig_range(pfn, pfn + count, MIGRATE_CMA); if (ret == 0) { bitmap_set(cma->bitmap, pageno, count); + page = pfn_to_page(pfn); break; } else if (ret != -EBUSY) { - goto error; + break; } pr_debug("%s(): memory range at %p is busy, retrying\n", __func__, pfn_to_page(pfn)); @@ -356,12 +356,8 @@ struct page *dma_alloc_from_contiguous(struct device *dev, int count, } mutex_unlock(&cma_mutex); - - pr_debug("%s(): returned %p\n", __func__, pfn_to_page(pfn)); - return pfn_to_page(pfn); -error: - mutex_unlock(&cma_mutex); - return NULL; + pr_debug("%s(): returned %p\n", __func__, page); + return page; } /** diff --git a/drivers/base/platform.c b/drivers/base/platform.c index ddeca14..8727e9c 100644 --- a/drivers/base/platform.c +++ b/drivers/base/platform.c @@ -23,6 +23,7 @@ #include <linux/idr.h> #include "base.h" +#include "power/power.h" /* For automatically allocated device IDs */ static DEFINE_IDA(platform_devid_ida); @@ -983,6 +984,7 @@ void __init early_platform_add_devices(struct platform_device **devs, int num) dev = &devs[i]->dev; if (!dev->devres_head.next) { + pm_runtime_early_init(dev); INIT_LIST_HEAD(&dev->devres_head); list_add_tail(&dev->devres_head, &early_platform_device_list); diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index ba3487c..c22b869 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -53,6 +53,24 @@ static LIST_HEAD(gpd_list); static DEFINE_MUTEX(gpd_list_lock); +static struct generic_pm_domain *pm_genpd_lookup_name(const char *domain_name) +{ + struct generic_pm_domain *genpd = NULL, *gpd; + + if (IS_ERR_OR_NULL(domain_name)) + return NULL; + + mutex_lock(&gpd_list_lock); + list_for_each_entry(gpd, &gpd_list, gpd_list_node) { + if (!strcmp(gpd->name, domain_name)) { + genpd = gpd; + break; + } + } + mutex_unlock(&gpd_list_lock); + return genpd; +} + #ifdef CONFIG_PM struct generic_pm_domain *dev_to_genpd(struct device *dev) @@ -256,10 +274,28 @@ int pm_genpd_poweron(struct generic_pm_domain *genpd) return ret; } +/** + * pm_genpd_name_poweron - Restore power to a given PM domain and its masters. + * @domain_name: Name of the PM domain to power up. + */ +int pm_genpd_name_poweron(const char *domain_name) +{ + struct generic_pm_domain *genpd; + + genpd = pm_genpd_lookup_name(domain_name); + return genpd ? pm_genpd_poweron(genpd) : -EINVAL; +} + #endif /* CONFIG_PM */ #ifdef CONFIG_PM_RUNTIME +static int genpd_start_dev_no_timing(struct generic_pm_domain *genpd, + struct device *dev) +{ + return GENPD_DEV_CALLBACK(genpd, int, start, dev); +} + static int genpd_save_dev(struct generic_pm_domain *genpd, struct device *dev) { return GENPD_DEV_TIMED_CALLBACK(genpd, int, save_state, dev, @@ -436,7 +472,7 @@ static int pm_genpd_poweroff(struct generic_pm_domain *genpd) not_suspended = 0; list_for_each_entry(pdd, &genpd->dev_list, list_node) if (pdd->dev->driver && (!pm_runtime_suspended(pdd->dev) - || pdd->dev->power.irq_safe || to_gpd_data(pdd)->always_on)) + || pdd->dev->power.irq_safe)) not_suspended++; if (not_suspended > genpd->in_progress) @@ -578,9 +614,6 @@ static int pm_genpd_runtime_suspend(struct device *dev) might_sleep_if(!genpd->dev_irq_safe); - if (dev_gpd_data(dev)->always_on) - return -EBUSY; - stop_ok = genpd->gov ? genpd->gov->stop_ok : NULL; if (stop_ok && !stop_ok(dev)) return -EBUSY; @@ -629,7 +662,7 @@ static int pm_genpd_runtime_resume(struct device *dev) /* If power.irq_safe, the PM domain is never powered off. */ if (dev->power.irq_safe) - return genpd_start_dev(genpd, dev); + return genpd_start_dev_no_timing(genpd, dev); mutex_lock(&genpd->lock); ret = __pm_genpd_poweron(genpd); @@ -697,6 +730,24 @@ static inline void genpd_power_off_work_fn(struct work_struct *work) {} #ifdef CONFIG_PM_SLEEP +/** + * pm_genpd_present - Check if the given PM domain has been initialized. + * @genpd: PM domain to check. + */ +static bool pm_genpd_present(struct generic_pm_domain *genpd) +{ + struct generic_pm_domain *gpd; + + if (IS_ERR_OR_NULL(genpd)) + return false; + + list_for_each_entry(gpd, &gpd_list, gpd_list_node) + if (gpd == genpd) + return true; + + return false; +} + static bool genpd_dev_active_wakeup(struct generic_pm_domain *genpd, struct device *dev) { @@ -750,9 +801,10 @@ static int genpd_thaw_dev(struct generic_pm_domain *genpd, struct device *dev) * Check if the given PM domain can be powered off (during system suspend or * hibernation) and do that if so. Also, in that case propagate to its masters. * - * This function is only called in "noirq" stages of system power transitions, - * so it need not acquire locks (all of the "noirq" callbacks are executed - * sequentially, so it is guaranteed that it will never run twice in parallel). + * This function is only called in "noirq" and "syscore" stages of system power + * transitions, so it need not acquire locks (all of the "noirq" callbacks are + * executed sequentially, so it is guaranteed that it will never run twice in + * parallel). */ static void pm_genpd_sync_poweroff(struct generic_pm_domain *genpd) { @@ -777,6 +829,33 @@ static void pm_genpd_sync_poweroff(struct generic_pm_domain *genpd) } /** + * pm_genpd_sync_poweron - Synchronously power on a PM domain and its masters. + * @genpd: PM domain to power on. + * + * This function is only called in "noirq" and "syscore" stages of system power + * transitions, so it need not acquire locks (all of the "noirq" callbacks are + * executed sequentially, so it is guaranteed that it will never run twice in + * parallel). + */ +static void pm_genpd_sync_poweron(struct generic_pm_domain *genpd) +{ + struct gpd_link *link; + + if (genpd->status != GPD_STATE_POWER_OFF) + return; + + list_for_each_entry(link, &genpd->slave_links, slave_node) { + pm_genpd_sync_poweron(link->master); + genpd_sd_counter_inc(link->master); + } + + if (genpd->power_on) + genpd->power_on(genpd); + + genpd->status = GPD_STATE_ACTIVE; +} + +/** * resume_needed - Check whether to resume a device before system suspend. * @dev: Device to check. * @genpd: PM domain the device belongs to. @@ -937,7 +1016,7 @@ static int pm_genpd_suspend_noirq(struct device *dev) if (IS_ERR(genpd)) return -EINVAL; - if (genpd->suspend_power_off || dev_gpd_data(dev)->always_on + if (genpd->suspend_power_off || (dev->power.wakeup_path && genpd_dev_active_wakeup(genpd, dev))) return 0; @@ -970,7 +1049,7 @@ static int pm_genpd_resume_noirq(struct device *dev) if (IS_ERR(genpd)) return -EINVAL; - if (genpd->suspend_power_off || dev_gpd_data(dev)->always_on + if (genpd->suspend_power_off || (dev->power.wakeup_path && genpd_dev_active_wakeup(genpd, dev))) return 0; @@ -979,7 +1058,7 @@ static int pm_genpd_resume_noirq(struct device *dev) * guaranteed that this function will never run twice in parallel for * the same PM domain, so it is not necessary to use locking here. */ - pm_genpd_poweron(genpd); + pm_genpd_sync_poweron(genpd); genpd->suspended_count--; return genpd_start_dev(genpd, dev); @@ -1090,8 +1169,7 @@ static int pm_genpd_freeze_noirq(struct device *dev) if (IS_ERR(genpd)) return -EINVAL; - return genpd->suspend_power_off || dev_gpd_data(dev)->always_on ? - 0 : genpd_stop_dev(genpd, dev); + return genpd->suspend_power_off ? 0 : genpd_stop_dev(genpd, dev); } /** @@ -1111,8 +1189,7 @@ static int pm_genpd_thaw_noirq(struct device *dev) if (IS_ERR(genpd)) return -EINVAL; - return genpd->suspend_power_off || dev_gpd_data(dev)->always_on ? - 0 : genpd_start_dev(genpd, dev); + return genpd->suspend_power_off ? 0 : genpd_start_dev(genpd, dev); } /** @@ -1186,8 +1263,8 @@ static int pm_genpd_restore_noirq(struct device *dev) if (genpd->suspended_count++ == 0) { /* * The boot kernel might put the domain into arbitrary state, - * so make it appear as powered off to pm_genpd_poweron(), so - * that it tries to power it on in case it was really off. + * so make it appear as powered off to pm_genpd_sync_poweron(), + * so that it tries to power it on in case it was really off. */ genpd->status = GPD_STATE_POWER_OFF; if (genpd->suspend_power_off) { @@ -1205,9 +1282,9 @@ static int pm_genpd_restore_noirq(struct device *dev) if (genpd->suspend_power_off) return 0; - pm_genpd_poweron(genpd); + pm_genpd_sync_poweron(genpd); - return dev_gpd_data(dev)->always_on ? 0 : genpd_start_dev(genpd, dev); + return genpd_start_dev(genpd, dev); } /** @@ -1246,6 +1323,31 @@ static void pm_genpd_complete(struct device *dev) } } +/** + * pm_genpd_syscore_switch - Switch power during system core suspend or resume. + * @dev: Device that normally is marked as "always on" to switch power for. + * + * This routine may only be called during the system core (syscore) suspend or + * resume phase for devices whose "always on" flags are set. + */ +void pm_genpd_syscore_switch(struct device *dev, bool suspend) +{ + struct generic_pm_domain *genpd; + + genpd = dev_to_genpd(dev); + if (!pm_genpd_present(genpd)) + return; + + if (suspend) { + genpd->suspended_count++; + pm_genpd_sync_poweroff(genpd); + } else { + pm_genpd_sync_poweron(genpd); + genpd->suspended_count--; + } +} +EXPORT_SYMBOL_GPL(pm_genpd_syscore_switch); + #else #define pm_genpd_prepare NULL @@ -1393,6 +1495,19 @@ int __pm_genpd_of_add_device(struct device_node *genpd_node, struct device *dev, return __pm_genpd_add_device(genpd, dev, td); } + +/** + * __pm_genpd_name_add_device - Find I/O PM domain and add a device to it. + * @domain_name: Name of the PM domain to add the device to. + * @dev: Device to be added. + * @td: Set of PM QoS timing parameters to attach to the device. + */ +int __pm_genpd_name_add_device(const char *domain_name, struct device *dev, + struct gpd_timing_data *td) +{ + return __pm_genpd_add_device(pm_genpd_lookup_name(domain_name), dev, td); +} + /** * pm_genpd_remove_device - Remove a device from an I/O PM domain. * @genpd: PM domain to remove the device from. @@ -1455,26 +1570,6 @@ int pm_genpd_remove_device(struct generic_pm_domain *genpd, } /** - * pm_genpd_dev_always_on - Set/unset the "always on" flag for a given device. - * @dev: Device to set/unset the flag for. - * @val: The new value of the device's "always on" flag. - */ -void pm_genpd_dev_always_on(struct device *dev, bool val) -{ - struct pm_subsys_data *psd; - unsigned long flags; - - spin_lock_irqsave(&dev->power.lock, flags); - - psd = dev_to_psd(dev); - if (psd && psd->domain_data) - to_gpd_data(psd->domain_data)->always_on = val; - - spin_unlock_irqrestore(&dev->power.lock, flags); -} -EXPORT_SYMBOL_GPL(pm_genpd_dev_always_on); - -/** * pm_genpd_dev_need_restore - Set/unset the device's "need restore" flag. * @dev: Device to set/unset the flag for. * @val: The new value of the device's "need restore" flag. @@ -1505,7 +1600,8 @@ int pm_genpd_add_subdomain(struct generic_pm_domain *genpd, struct gpd_link *link; int ret = 0; - if (IS_ERR_OR_NULL(genpd) || IS_ERR_OR_NULL(subdomain)) + if (IS_ERR_OR_NULL(genpd) || IS_ERR_OR_NULL(subdomain) + || genpd == subdomain) return -EINVAL; start: @@ -1552,6 +1648,35 @@ int pm_genpd_add_subdomain(struct generic_pm_domain *genpd, } /** + * pm_genpd_add_subdomain_names - Add a subdomain to an I/O PM domain. + * @master_name: Name of the master PM domain to add the subdomain to. + * @subdomain_name: Name of the subdomain to be added. + */ +int pm_genpd_add_subdomain_names(const char *master_name, + const char *subdomain_name) +{ + struct generic_pm_domain *master = NULL, *subdomain = NULL, *gpd; + + if (IS_ERR_OR_NULL(master_name) || IS_ERR_OR_NULL(subdomain_name)) + return -EINVAL; + + mutex_lock(&gpd_list_lock); + list_for_each_entry(gpd, &gpd_list, gpd_list_node) { + if (!master && !strcmp(gpd->name, master_name)) + master = gpd; + + if (!subdomain && !strcmp(gpd->name, subdomain_name)) + subdomain = gpd; + + if (master && subdomain) + break; + } + mutex_unlock(&gpd_list_lock); + + return pm_genpd_add_subdomain(master, subdomain); +} + +/** * pm_genpd_remove_subdomain - Remove a subdomain from an I/O PM domain. * @genpd: Master PM domain to remove the subdomain from. * @subdomain: Subdomain to be removed. @@ -1704,7 +1829,16 @@ int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) } EXPORT_SYMBOL_GPL(__pm_genpd_remove_callbacks); -int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state) +/** + * pm_genpd_attach_cpuidle - Connect the given PM domain with cpuidle. + * @genpd: PM domain to be connected with cpuidle. + * @state: cpuidle state this domain can disable/enable. + * + * Make a PM domain behave as though it contained a CPU core, that is, instead + * of calling its power down routine it will enable the given cpuidle state so + * that the cpuidle subsystem can power it down (if possible and desirable). + */ +int pm_genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state) { struct cpuidle_driver *cpuidle_drv; struct gpd_cpu_data *cpu_data; @@ -1753,7 +1887,24 @@ int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state) goto out; } -int genpd_detach_cpuidle(struct generic_pm_domain *genpd) +/** + * pm_genpd_name_attach_cpuidle - Find PM domain and connect cpuidle to it. + * @name: Name of the domain to connect to cpuidle. + * @state: cpuidle state this domain can manipulate. + */ +int pm_genpd_name_attach_cpuidle(const char *name, int state) +{ + return pm_genpd_attach_cpuidle(pm_genpd_lookup_name(name), state); +} + +/** + * pm_genpd_detach_cpuidle - Remove the cpuidle connection from a PM domain. + * @genpd: PM domain to remove the cpuidle connection from. + * + * Remove the cpuidle connection set up by pm_genpd_attach_cpuidle() from the + * given PM domain. + */ +int pm_genpd_detach_cpuidle(struct generic_pm_domain *genpd) { struct gpd_cpu_data *cpu_data; struct cpuidle_state *idle_state; @@ -1784,6 +1935,15 @@ int genpd_detach_cpuidle(struct generic_pm_domain *genpd) return ret; } +/** + * pm_genpd_name_detach_cpuidle - Find PM domain and disconnect cpuidle from it. + * @name: Name of the domain to disconnect cpuidle from. + */ +int pm_genpd_name_detach_cpuidle(const char *name) +{ + return pm_genpd_detach_cpuidle(pm_genpd_lookup_name(name)); +} + /* Default device callbacks for generic PM domains. */ /** diff --git a/drivers/base/power/main.c b/drivers/base/power/main.c index b0b072a..a3c1404 100644 --- a/drivers/base/power/main.c +++ b/drivers/base/power/main.c @@ -57,20 +57,17 @@ static pm_message_t pm_transition; static int async_error; /** - * device_pm_init - Initialize the PM-related part of a device object. + * device_pm_sleep_init - Initialize system suspend-related device fields. * @dev: Device object being initialized. */ -void device_pm_init(struct device *dev) +void device_pm_sleep_init(struct device *dev) { dev->power.is_prepared = false; dev->power.is_suspended = false; init_completion(&dev->power.completion); complete_all(&dev->power.completion); dev->power.wakeup = NULL; - spin_lock_init(&dev->power.lock); - pm_runtime_init(dev); INIT_LIST_HEAD(&dev->power.entry); - dev->power.power_state = PMSG_INVALID; } /** @@ -408,6 +405,9 @@ static int device_resume_noirq(struct device *dev, pm_message_t state) TRACE_DEVICE(dev); TRACE_RESUME(0); + if (dev->power.syscore) + goto Out; + if (dev->pm_domain) { info = "noirq power domain "; callback = pm_noirq_op(&dev->pm_domain->ops, state); @@ -429,6 +429,7 @@ static int device_resume_noirq(struct device *dev, pm_message_t state) error = dpm_run_callback(callback, dev, state, info); + Out: TRACE_RESUME(error); return error; } @@ -486,6 +487,9 @@ static int device_resume_early(struct device *dev, pm_message_t state) TRACE_DEVICE(dev); TRACE_RESUME(0); + if (dev->power.syscore) + goto Out; + if (dev->pm_domain) { info = "early power domain "; callback = pm_late_early_op(&dev->pm_domain->ops, state); @@ -507,6 +511,7 @@ static int device_resume_early(struct device *dev, pm_message_t state) error = dpm_run_callback(callback, dev, state, info); + Out: TRACE_RESUME(error); return error; } @@ -565,11 +570,13 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) pm_callback_t callback = NULL; char *info = NULL; int error = 0; - bool put = false; TRACE_DEVICE(dev); TRACE_RESUME(0); + if (dev->power.syscore) + goto Complete; + dpm_wait(dev->parent, async); device_lock(dev); @@ -583,7 +590,6 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) goto Unlock; pm_runtime_enable(dev); - put = true; if (dev->pm_domain) { info = "power domain "; @@ -632,13 +638,12 @@ static int device_resume(struct device *dev, pm_message_t state, bool async) Unlock: device_unlock(dev); + + Complete: complete_all(&dev->power.completion); TRACE_RESUME(error); - if (put) - pm_runtime_put_sync(dev); - return error; } @@ -722,6 +727,9 @@ static void device_complete(struct device *dev, pm_message_t state) void (*callback)(struct device *) = NULL; char *info = NULL; + if (dev->power.syscore) + return; + device_lock(dev); if (dev->pm_domain) { @@ -749,6 +757,8 @@ static void device_complete(struct device *dev, pm_message_t state) } device_unlock(dev); + + pm_runtime_put_sync(dev); } /** @@ -834,6 +844,9 @@ static int device_suspend_noirq(struct device *dev, pm_message_t state) pm_callback_t callback = NULL; char *info = NULL; + if (dev->power.syscore) + return 0; + if (dev->pm_domain) { info = "noirq power domain "; callback = pm_noirq_op(&dev->pm_domain->ops, state); @@ -917,6 +930,9 @@ static int device_suspend_late(struct device *dev, pm_message_t state) pm_callback_t callback = NULL; char *info = NULL; + if (dev->power.syscore) + return 0; + if (dev->pm_domain) { info = "late power domain "; callback = pm_late_early_op(&dev->pm_domain->ops, state); @@ -996,7 +1012,7 @@ int dpm_suspend_end(pm_message_t state) error = dpm_suspend_noirq(state); if (error) { - dpm_resume_early(state); + dpm_resume_early(resume_event(state)); return error; } @@ -1043,16 +1059,23 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) if (async_error) goto Complete; - pm_runtime_get_noresume(dev); + /* + * If a device configured to wake up the system from sleep states + * has been suspended at run time and there's a resume request pending + * for it, this is equivalent to the device signaling wakeup, so the + * system suspend operation should be aborted. + */ if (pm_runtime_barrier(dev) && device_may_wakeup(dev)) pm_wakeup_event(dev, 0); if (pm_wakeup_pending()) { - pm_runtime_put_sync(dev); async_error = -EBUSY; goto Complete; } + if (dev->power.syscore) + goto Complete; + device_lock(dev); if (dev->pm_domain) { @@ -1111,12 +1134,10 @@ static int __device_suspend(struct device *dev, pm_message_t state, bool async) Complete: complete_all(&dev->power.completion); - if (error) { - pm_runtime_put_sync(dev); + if (error) async_error = error; - } else if (dev->power.is_suspended) { + else if (dev->power.is_suspended) __pm_runtime_disable(dev, false); - } return error; } @@ -1209,6 +1230,17 @@ static int device_prepare(struct device *dev, pm_message_t state) char *info = NULL; int error = 0; + if (dev->power.syscore) + return 0; + + /* + * If a device's parent goes into runtime suspend at the wrong time, + * it won't be possible to resume the device. To prevent this we + * block runtime suspend here, during the prepare phase, and allow + * it again during the complete phase. + */ + pm_runtime_get_noresume(dev); + device_lock(dev); dev->power.wakeup_path = device_may_wakeup(dev); diff --git a/drivers/base/power/opp.c b/drivers/base/power/opp.c index ac993ea..d946864 100644 --- a/drivers/base/power/opp.c +++ b/drivers/base/power/opp.c @@ -22,6 +22,7 @@ #include <linux/rculist.h> #include <linux/rcupdate.h> #include <linux/opp.h> +#include <linux/of.h> /* * Internal data structure organization with the OPP layer library is as @@ -674,3 +675,49 @@ struct srcu_notifier_head *opp_get_notifier(struct device *dev) return &dev_opp->head; } + +#ifdef CONFIG_OF +/** + * of_init_opp_table() - Initialize opp table from device tree + * @dev: device pointer used to lookup device OPPs. + * + * Register the initial OPP table with the OPP library for given device. + */ +int of_init_opp_table(struct device *dev) +{ + const struct property *prop; + const __be32 *val; + int nr; + + prop = of_find_property(dev->of_node, "operating-points", NULL); + if (!prop) + return -ENODEV; + if (!prop->value) + return -ENODATA; + + /* + * Each OPP is a set of tuples consisting of frequency and + * voltage like <freq-kHz vol-uV>. + */ + nr = prop->length / sizeof(u32); + if (nr % 2) { + dev_err(dev, "%s: Invalid OPP list\n", __func__); + return -EINVAL; + } + + val = prop->value; + while (nr) { + unsigned long freq = be32_to_cpup(val++) * 1000; + unsigned long volt = be32_to_cpup(val++); + + if (opp_add(dev, freq, volt)) { + dev_warn(dev, "%s: Failed to add OPP %ld\n", + __func__, freq); + continue; + } + nr -= 2; + } + + return 0; +} +#endif diff --git a/drivers/base/power/power.h b/drivers/base/power/power.h index eeb4bff..0dbfdf4 100644 --- a/drivers/base/power/power.h +++ b/drivers/base/power/power.h @@ -1,12 +1,32 @@ #include <linux/pm_qos.h> +static inline void device_pm_init_common(struct device *dev) +{ + if (!dev->power.early_init) { + spin_lock_init(&dev->power.lock); + dev->power.power_state = PMSG_INVALID; + dev->power.early_init = true; + } +} + #ifdef CONFIG_PM_RUNTIME +static inline void pm_runtime_early_init(struct device *dev) +{ + dev->power.disable_depth = 1; + device_pm_init_common(dev); +} + extern void pm_runtime_init(struct device *dev); extern void pm_runtime_remove(struct device *dev); #else /* !CONFIG_PM_RUNTIME */ +static inline void pm_runtime_early_init(struct device *dev) +{ + device_pm_init_common(dev); +} + static inline void pm_runtime_init(struct device *dev) {} static inline void pm_runtime_remove(struct device *dev) {} @@ -25,7 +45,7 @@ static inline struct device *to_device(struct list_head *entry) return container_of(entry, struct device, power.entry); } -extern void device_pm_init(struct device *dev); +extern void device_pm_sleep_init(struct device *dev); extern void device_pm_add(struct device *); extern void device_pm_remove(struct device *); extern void device_pm_move_before(struct device *, struct device *); @@ -34,12 +54,7 @@ extern void device_pm_move_last(struct device *); #else /* !CONFIG_PM_SLEEP */ -static inline void device_pm_init(struct device *dev) -{ - spin_lock_init(&dev->power.lock); - dev->power.power_state = PMSG_INVALID; - pm_runtime_init(dev); -} +static inline void device_pm_sleep_init(struct device *dev) {} static inline void device_pm_add(struct device *dev) { @@ -60,6 +75,13 @@ static inline void device_pm_move_last(struct device *dev) {} #endif /* !CONFIG_PM_SLEEP */ +static inline void device_pm_init(struct device *dev) +{ + device_pm_init_common(dev); + device_pm_sleep_init(dev); + pm_runtime_init(dev); +} + #ifdef CONFIG_PM /* diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 7d9c1cb..3148b10 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -509,6 +509,9 @@ static int rpm_resume(struct device *dev, int rpmflags) repeat: if (dev->power.runtime_error) retval = -EINVAL; + else if (dev->power.disable_depth == 1 && dev->power.is_suspended + && dev->power.runtime_status == RPM_ACTIVE) + retval = 1; else if (dev->power.disable_depth > 0) retval = -EACCES; if (retval) diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c index cbb463b..e6ee5e8 100644 --- a/drivers/base/power/wakeup.c +++ b/drivers/base/power/wakeup.c @@ -127,6 +127,8 @@ EXPORT_SYMBOL_GPL(wakeup_source_destroy); */ void wakeup_source_add(struct wakeup_source *ws) { + unsigned long flags; + if (WARN_ON(!ws)) return; @@ -135,9 +137,9 @@ void wakeup_source_add(struct wakeup_source *ws) ws->active = false; ws->last_time = ktime_get(); - spin_lock_irq(&events_lock); + spin_lock_irqsave(&events_lock, flags); list_add_rcu(&ws->entry, &wakeup_sources); - spin_unlock_irq(&events_lock); + spin_unlock_irqrestore(&events_lock, flags); } EXPORT_SYMBOL_GPL(wakeup_source_add); @@ -147,12 +149,14 @@ EXPORT_SYMBOL_GPL(wakeup_source_add); */ void wakeup_source_remove(struct wakeup_source *ws) { + unsigned long flags; + if (WARN_ON(!ws)) return; - spin_lock_irq(&events_lock); + spin_lock_irqsave(&events_lock, flags); list_del_rcu(&ws->entry); - spin_unlock_irq(&events_lock); + spin_unlock_irqrestore(&events_lock, flags); synchronize_rcu(); } EXPORT_SYMBOL_GPL(wakeup_source_remove); @@ -649,6 +653,31 @@ void pm_wakeup_event(struct device *dev, unsigned int msec) } EXPORT_SYMBOL_GPL(pm_wakeup_event); +static void print_active_wakeup_sources(void) +{ + struct wakeup_source *ws; + int active = 0; + struct wakeup_source *last_activity_ws = NULL; + + rcu_read_lock(); + list_for_each_entry_rcu(ws, &wakeup_sources, entry) { + if (ws->active) { + pr_info("active wakeup source: %s\n", ws->name); + active = 1; + } else if (!active && + (!last_activity_ws || + ktime_to_ns(ws->last_time) > + ktime_to_ns(last_activity_ws->last_time))) { + last_activity_ws = ws; + } + } + + if (!active && last_activity_ws) + pr_info("last active wakeup source: %s\n", + last_activity_ws->name); + rcu_read_unlock(); +} + /** * pm_wakeup_pending - Check if power transition in progress should be aborted. * @@ -671,6 +700,10 @@ bool pm_wakeup_pending(void) events_check_enabled = !ret; } spin_unlock_irqrestore(&events_lock, flags); + + if (ret) + print_active_wakeup_sources(); + return ret; } @@ -723,15 +756,16 @@ bool pm_get_wakeup_count(unsigned int *count, bool block) bool pm_save_wakeup_count(unsigned int count) { unsigned int cnt, inpr; + unsigned long flags; events_check_enabled = false; - spin_lock_irq(&events_lock); + spin_lock_irqsave(&events_lock, flags); split_counters(&cnt, &inpr); if (cnt == count && inpr == 0) { saved_count = count; events_check_enabled = true; } - spin_unlock_irq(&events_lock); + spin_unlock_irqrestore(&events_lock, flags); return events_check_enabled; } diff --git a/drivers/char/hw_random/Kconfig b/drivers/char/hw_random/Kconfig index 7c0d3919..fbd9b2b 100644 --- a/drivers/char/hw_random/Kconfig +++ b/drivers/char/hw_random/Kconfig @@ -289,3 +289,16 @@ config HW_RANDOM_EXYNOS module will be called exynos-rng. If unsure, say Y. + +config HW_RANDOM_TPM + tristate "TPM HW Random Number Generator support" + depends on HW_RANDOM && TCG_TPM + default HW_RANDOM + ---help--- + This driver provides kernel-side support for the Random Number + Generator in the Trusted Platform Module + + To compile this driver as a module, choose M here: the + module will be called tpm-rng. + + If unsure, say Y. diff --git a/drivers/char/hw_random/Makefile b/drivers/char/hw_random/Makefile index 39a757c..1fd7eec 100644 --- a/drivers/char/hw_random/Makefile +++ b/drivers/char/hw_random/Makefile @@ -25,3 +25,4 @@ obj-$(CONFIG_HW_RANDOM_PICOXCELL) += picoxcell-rng.o obj-$(CONFIG_HW_RANDOM_PPC4XX) += ppc4xx-rng.o obj-$(CONFIG_HW_RANDOM_PSERIES) += pseries-rng.o obj-$(CONFIG_HW_RANDOM_EXYNOS) += exynos-rng.o +obj-$(CONFIG_HW_RANDOM_TPM) += tpm-rng.o diff --git a/drivers/char/hw_random/tpm-rng.c b/drivers/char/hw_random/tpm-rng.c new file mode 100644 index 0000000..d6d4482 --- /dev/null +++ b/drivers/char/hw_random/tpm-rng.c @@ -0,0 +1,50 @@ +/* + * Copyright (C) 2012 Kent Yoder IBM Corporation + * + * HWRNG interfaces to pull RNG data from a TPM + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + * + * This program is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the + * GNU General Public License for more details. + * + * You should have received a copy of the GNU General Public License + * along with this program; if not, write to the Free Software + * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA + */ + +#include <linux/module.h> +#include <linux/hw_random.h> +#include <linux/tpm.h> + +#define MODULE_NAME "tpm-rng" + +static int tpm_rng_read(struct hwrng *rng, void *data, size_t max, bool wait) +{ + return tpm_get_random(TPM_ANY_NUM, data, max); +} + +static struct hwrng tpm_rng = { + .name = MODULE_NAME, + .read = tpm_rng_read, +}; + +static int __init rng_init(void) +{ + return hwrng_register(&tpm_rng); +} +module_init(rng_init); + +static void __exit rng_exit(void) +{ + hwrng_unregister(&tpm_rng); +} +module_exit(rng_exit); + +MODULE_LICENSE("GPL v2"); +MODULE_AUTHOR("Kent Yoder <key@linux.vnet.ibm.com>"); +MODULE_DESCRIPTION("RNG driver for TPM devices"); diff --git a/drivers/char/tpm/Kconfig b/drivers/char/tpm/Kconfig index a048199..915875e 100644 --- a/drivers/char/tpm/Kconfig +++ b/drivers/char/tpm/Kconfig @@ -33,6 +33,17 @@ config TCG_TIS from within Linux. To compile this driver as a module, choose M here; the module will be called tpm_tis. +config TCG_TIS_I2C_INFINEON + tristate "TPM Interface Specification 1.2 Interface (I2C - Infineon)" + depends on I2C + ---help--- + If you have a TPM security chip that is compliant with the + TCG TIS 1.2 TPM specification and Infineon's I2C Protocol Stack + Specification 0.20 say Yes and it will be accessible from within + Linux. + To compile this driver as a module, choose M here; the module + will be called tpm_tis_i2c_infineon. + config TCG_NSC tristate "National Semiconductor TPM Interface" depends on X86 @@ -62,4 +73,12 @@ config TCG_INFINEON Further information on this driver and the supported hardware can be found at http://www.trust.rub.de/projects/linux-device-driver-infineon-tpm/ +config TCG_IBMVTPM + tristate "IBM VTPM Interface" + depends on PPC64 + ---help--- + If you have IBM virtual TPM (VTPM) support say Yes and it + will be accessible from within Linux. To compile this driver + as a module, choose M here; the module will be called tpm_ibmvtpm. + endif # TCG_TPM diff --git a/drivers/char/tpm/Makefile b/drivers/char/tpm/Makefile index ea3a1e0..5b3fc8b 100644 --- a/drivers/char/tpm/Makefile +++ b/drivers/char/tpm/Makefile @@ -4,8 +4,16 @@ obj-$(CONFIG_TCG_TPM) += tpm.o ifdef CONFIG_ACPI obj-$(CONFIG_TCG_TPM) += tpm_bios.o + tpm_bios-objs += tpm_eventlog.o tpm_acpi.o tpm_ppi.o +else +ifdef CONFIG_TCG_IBMVTPM + obj-$(CONFIG_TCG_TPM) += tpm_bios.o + tpm_bios-objs += tpm_eventlog.o tpm_of.o +endif endif obj-$(CONFIG_TCG_TIS) += tpm_tis.o +obj-$(CONFIG_TCG_TIS_I2C_INFINEON) += tpm_i2c_infineon.o obj-$(CONFIG_TCG_NSC) += tpm_nsc.o obj-$(CONFIG_TCG_ATMEL) += tpm_atmel.o obj-$(CONFIG_TCG_INFINEON) += tpm_infineon.o +obj-$(CONFIG_TCG_IBMVTPM) += tpm_ibmvtpm.o diff --git a/drivers/char/tpm/tpm.c b/drivers/char/tpm/tpm.c index 3af9f4d..f26afdb 100644 --- a/drivers/char/tpm/tpm.c +++ b/drivers/char/tpm/tpm.c @@ -30,12 +30,7 @@ #include <linux/freezer.h> #include "tpm.h" - -enum tpm_const { - TPM_MINOR = 224, /* officially assigned */ - TPM_BUFSIZE = 4096, - TPM_NUM_DEVICES = 256, -}; +#include "tpm_eventlog.h" enum tpm_duration { TPM_SHORT = 0, @@ -482,6 +477,7 @@ static ssize_t transmit_cmd(struct tpm_chip *chip, struct tpm_cmd_t *cmd, #define TPM_INTERNAL_RESULT_SIZE 200 #define TPM_TAG_RQU_COMMAND cpu_to_be16(193) #define TPM_ORD_GET_CAP cpu_to_be32(101) +#define TPM_ORD_GET_RANDOM cpu_to_be32(70) static const struct tpm_input_header tpm_getcap_header = { .tag = TPM_TAG_RQU_COMMAND, @@ -919,7 +915,7 @@ EXPORT_SYMBOL_GPL(tpm_show_pcrs); #define READ_PUBEK_RESULT_SIZE 314 #define TPM_ORD_READPUBEK cpu_to_be32(124) -struct tpm_input_header tpm_readpubek_header = { +static struct tpm_input_header tpm_readpubek_header = { .tag = TPM_TAG_RQU_COMMAND, .length = cpu_to_be32(30), .ordinal = TPM_ORD_READPUBEK @@ -1175,7 +1171,7 @@ int tpm_release(struct inode *inode, struct file *file) flush_work(&chip->work); file->private_data = NULL; atomic_set(&chip->data_pending, 0); - kfree(chip->data_buffer); + kzfree(chip->data_buffer); clear_bit(0, &chip->is_open); put_device(chip->dev); return 0; @@ -1227,7 +1223,6 @@ ssize_t tpm_read(struct file *file, char __user *buf, del_singleshot_timer_sync(&chip->user_read_timer); flush_work(&chip->work); ret_size = atomic_read(&chip->data_pending); - atomic_set(&chip->data_pending, 0); if (ret_size > 0) { /* relay data */ ssize_t orig_ret_size = ret_size; if (size < ret_size) @@ -1242,6 +1237,8 @@ ssize_t tpm_read(struct file *file, char __user *buf, mutex_unlock(&chip->buffer_mutex); } + atomic_set(&chip->data_pending, 0); + return ret_size; } EXPORT_SYMBOL_GPL(tpm_read); @@ -1326,6 +1323,58 @@ int tpm_pm_resume(struct device *dev) } EXPORT_SYMBOL_GPL(tpm_pm_resume); +#define TPM_GETRANDOM_RESULT_SIZE 18 +static struct tpm_input_header tpm_getrandom_header = { + .tag = TPM_TAG_RQU_COMMAND, + .length = cpu_to_be32(14), + .ordinal = TPM_ORD_GET_RANDOM +}; + +/** + * tpm_get_random() - Get random bytes from the tpm's RNG + * @chip_num: A specific chip number for the request or TPM_ANY_NUM + * @out: destination buffer for the random bytes + * @max: the max number of bytes to write to @out + * + * Returns < 0 on error and the number of bytes read on success + */ +int tpm_get_random(u32 chip_num, u8 *out, size_t max) +{ + struct tpm_chip *chip; + struct tpm_cmd_t tpm_cmd; + u32 recd, num_bytes = min_t(u32, max, TPM_MAX_RNG_DATA); + int err, total = 0, retries = 5; + u8 *dest = out; + + chip = tpm_chip_find_get(chip_num); + if (chip == NULL) + return -ENODEV; + + if (!out || !num_bytes || max > TPM_MAX_RNG_DATA) + return -EINVAL; + + do { + tpm_cmd.header.in = tpm_getrandom_header; + tpm_cmd.params.getrandom_in.num_bytes = cpu_to_be32(num_bytes); + + err = transmit_cmd(chip, &tpm_cmd, + TPM_GETRANDOM_RESULT_SIZE + num_bytes, + "attempting get random"); + if (err) + break; + + recd = be32_to_cpu(tpm_cmd.params.getrandom_out.rng_data_len); + memcpy(dest, tpm_cmd.params.getrandom_out.rng_data, recd); + + dest += recd; + total += recd; + num_bytes -= recd; + } while (retries-- && total < max); + + return total ? total : -EIO; +} +EXPORT_SYMBOL_GPL(tpm_get_random); + /* In case vendor provided release function, call it too.*/ void tpm_dev_vendor_release(struct tpm_chip *chip) @@ -1346,7 +1395,7 @@ EXPORT_SYMBOL_GPL(tpm_dev_vendor_release); * Once all references to platform device are down to 0, * release all allocated structures. */ -void tpm_dev_release(struct device *dev) +static void tpm_dev_release(struct device *dev) { struct tpm_chip *chip = dev_get_drvdata(dev); @@ -1427,6 +1476,11 @@ struct tpm_chip *tpm_register_hardware(struct device *dev, goto put_device; } + if (sys_add_ppi(&dev->kobj)) { + misc_deregister(&chip->vendor.miscdev); + goto put_device; + } + chip->bios_dir = tpm_bios_log_setup(devname); /* Make chip available */ diff --git a/drivers/char/tpm/tpm.h b/drivers/char/tpm/tpm.h index 917f727..02c266a 100644 --- a/drivers/char/tpm/tpm.h +++ b/drivers/char/tpm/tpm.h @@ -28,6 +28,12 @@ #include <linux/io.h> #include <linux/tpm.h> +enum tpm_const { + TPM_MINOR = 224, /* officially assigned */ + TPM_BUFSIZE = 4096, + TPM_NUM_DEVICES = 256, +}; + enum tpm_timeout { TPM_TIMEOUT = 5, /* msecs */ }; @@ -94,6 +100,7 @@ struct tpm_vendor_specific { bool timeout_adjusted; unsigned long duration[3]; /* jiffies */ bool duration_adjusted; + void *data; wait_queue_head_t read_queue; wait_queue_head_t int_queue; @@ -269,6 +276,21 @@ struct tpm_pcrextend_in { u8 hash[TPM_DIGEST_SIZE]; }__attribute__((packed)); +/* 128 bytes is an arbitrary cap. This could be as large as TPM_BUFSIZE - 18 + * bytes, but 128 is still a relatively large number of random bytes and + * anything much bigger causes users of struct tpm_cmd_t to start getting + * compiler warnings about stack frame size. */ +#define TPM_MAX_RNG_DATA 128 + +struct tpm_getrandom_out { + __be32 rng_data_len; + u8 rng_data[TPM_MAX_RNG_DATA]; +}__attribute__((packed)); + +struct tpm_getrandom_in { + __be32 num_bytes; +}__attribute__((packed)); + typedef union { struct tpm_getcap_params_out getcap_out; struct tpm_readpubek_params_out readpubek_out; @@ -277,6 +299,8 @@ typedef union { struct tpm_pcrread_in pcrread_in; struct tpm_pcrread_out pcrread_out; struct tpm_pcrextend_in pcrextend_in; + struct tpm_getrandom_in getrandom_in; + struct tpm_getrandom_out getrandom_out; } tpm_cmd_params; struct tpm_cmd_t { @@ -303,15 +327,12 @@ extern int tpm_pm_suspend(struct device *); extern int tpm_pm_resume(struct device *); extern int wait_for_tpm_stat(struct tpm_chip *, u8, unsigned long, wait_queue_head_t *); + #ifdef CONFIG_ACPI -extern struct dentry ** tpm_bios_log_setup(char *); -extern void tpm_bios_log_teardown(struct dentry **); +extern ssize_t sys_add_ppi(struct kobject *parent); #else -static inline struct dentry ** tpm_bios_log_setup(char *name) -{ - return NULL; -} -static inline void tpm_bios_log_teardown(struct dentry **dir) +static inline ssize_t sys_add_ppi(struct kobject *parent) { + return 0; } #endif diff --git a/drivers/char/tpm/tpm_acpi.c b/drivers/char/tpm/tpm_acpi.c new file mode 100644 index 0000000..56051d0 --- /dev/null +++ b/drivers/char/tpm/tpm_acpi.c @@ -0,0 +1,109 @@ +/* + * Copyright (C) 2005 IBM Corporation + * + * Authors: + * Seiji Munetoh <munetoh@jp.ibm.com> + * Stefan Berger <stefanb@us.ibm.com> + * Reiner Sailer <sailer@watson.ibm.com> + * Kylene Hall <kjhall@us.ibm.com> + * + * Maintained by: <tpmdd-devel@lists.sourceforge.net> + * + * Access to the eventlog extended by the TCG BIOS of PC platform + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + */ + +#include <linux/seq_file.h> +#include <linux/fs.h> +#include <linux/security.h> +#include <linux/module.h> +#include <linux/slab.h> +#include <acpi/acpi.h> + +#include "tpm.h" +#include "tpm_eventlog.h" + +struct acpi_tcpa { + struct acpi_table_header hdr; + u16 platform_class; + union { + struct client_hdr { + u32 log_max_len __attribute__ ((packed)); + u64 log_start_addr __attribute__ ((packed)); + } client; + struct server_hdr { + u16 reserved; + u64 log_max_len __attribute__ ((packed)); + u64 log_start_addr __attribute__ ((packed)); + } server; + }; +}; + +/* read binary bios log */ +int read_log(struct tpm_bios_log *log) +{ + struct acpi_tcpa *buff; + acpi_status status; + void __iomem *virt; + u64 len, start; + + if (log->bios_event_log != NULL) { + printk(KERN_ERR + "%s: ERROR - Eventlog already initialized\n", + __func__); + return -EFAULT; + } + + /* Find TCPA entry in RSDT (ACPI_LOGICAL_ADDRESSING) */ + status = acpi_get_table(ACPI_SIG_TCPA, 1, + (struct acpi_table_header **)&buff); + + if (ACPI_FAILURE(status)) { + printk(KERN_ERR "%s: ERROR - Could not get TCPA table\n", + __func__); + return -EIO; + } + + switch(buff->platform_class) { + case BIOS_SERVER: + len = buff->server.log_max_len; + start = buff->server.log_start_addr; + break; + case BIOS_CLIENT: + default: + len = buff->client.log_max_len; + start = buff->client.log_start_addr; + break; + } + if (!len) { + printk(KERN_ERR "%s: ERROR - TCPA log area empty\n", __func__); + return -EIO; + } + + /* malloc EventLog space */ + log->bios_event_log = kmalloc(len, GFP_KERNEL); + if (!log->bios_event_log) { + printk("%s: ERROR - Not enough Memory for BIOS measurements\n", + __func__); + return -ENOMEM; + } + + log->bios_event_log_end = log->bios_event_log + len; + + virt = acpi_os_map_memory(start, len); + if (!virt) { + kfree(log->bios_event_log); + printk("%s: ERROR - Unable to map memory\n", __func__); + return -EIO; + } + + memcpy_fromio(log->bios_event_log, virt, len); + + acpi_os_unmap_memory(virt, len); + return 0; +} diff --git a/drivers/char/tpm/tpm_bios.c b/drivers/char/tpm/tpm_eventlog.c index 0636520..84ddc55 100644 --- a/drivers/char/tpm/tpm_bios.c +++ b/drivers/char/tpm/tpm_eventlog.c @@ -1,7 +1,8 @@ /* - * Copyright (C) 2005 IBM Corporation + * Copyright (C) 2005, 2012 IBM Corporation * * Authors: + * Kent Yoder <key@linux.vnet.ibm.com> * Seiji Munetoh <munetoh@jp.ibm.com> * Stefan Berger <stefanb@us.ibm.com> * Reiner Sailer <sailer@watson.ibm.com> @@ -9,7 +10,7 @@ * * Maintained by: <tpmdd-devel@lists.sourceforge.net> * - * Access to the eventlog extended by the TCG BIOS of PC platform + * Access to the eventlog created by a system's firmware / BIOS * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public License @@ -23,67 +24,10 @@ #include <linux/security.h> #include <linux/module.h> #include <linux/slab.h> -#include <acpi/acpi.h> -#include "tpm.h" - -#define TCG_EVENT_NAME_LEN_MAX 255 -#define MAX_TEXT_EVENT 1000 /* Max event string length */ -#define ACPI_TCPA_SIG "TCPA" /* 0x41504354 /'TCPA' */ - -enum bios_platform_class { - BIOS_CLIENT = 0x00, - BIOS_SERVER = 0x01, -}; - -struct tpm_bios_log { - void *bios_event_log; - void *bios_event_log_end; -}; - -struct acpi_tcpa { - struct acpi_table_header hdr; - u16 platform_class; - union { - struct client_hdr { - u32 log_max_len __attribute__ ((packed)); - u64 log_start_addr __attribute__ ((packed)); - } client; - struct server_hdr { - u16 reserved; - u64 log_max_len __attribute__ ((packed)); - u64 log_start_addr __attribute__ ((packed)); - } server; - }; -}; -struct tcpa_event { - u32 pcr_index; - u32 event_type; - u8 pcr_value[20]; /* SHA1 */ - u32 event_size; - u8 event_data[0]; -}; +#include "tpm.h" +#include "tpm_eventlog.h" -enum tcpa_event_types { - PREBOOT = 0, - POST_CODE, - UNUSED, - NO_ACTION, - SEPARATOR, - ACTION, - EVENT_TAG, - SCRTM_CONTENTS, - SCRTM_VERSION, - CPU_MICROCODE, - PLATFORM_CONFIG_FLAGS, - TABLE_OF_DEVICES, - COMPACT_HASH, - IPL, - IPL_PARTITION_DATA, - NONHOST_CODE, - NONHOST_CONFIG, - NONHOST_INFO, -}; static const char* tcpa_event_type_strings[] = { "PREBOOT", @@ -106,28 +50,6 @@ static const char* tcpa_event_type_strings[] = { "Non-Host Info" }; -struct tcpa_pc_event { - u32 event_id; - u32 event_size; - u8 event_data[0]; -}; - -enum tcpa_pc_event_ids { - SMBIOS = 1, - BIS_CERT, - POST_BIOS_ROM, - ESCD, - CMOS, - NVRAM, - OPTION_ROM_EXEC, - OPTION_ROM_CONFIG, - OPTION_ROM_MICROCODE = 10, - S_CRTM_VERSION, - S_CRTM_CONTENTS, - POST_CONTENTS, - HOST_TABLE_OF_DEVICES, -}; - static const char* tcpa_pc_event_id_strings[] = { "", "SMBIOS", @@ -358,65 +280,6 @@ static const struct seq_operations tpm_binary_b_measurments_seqops = { .show = tpm_binary_bios_measurements_show, }; -/* read binary bios log */ -static int read_log(struct tpm_bios_log *log) -{ - struct acpi_tcpa *buff; - acpi_status status; - struct acpi_table_header *virt; - u64 len, start; - - if (log->bios_event_log != NULL) { - printk(KERN_ERR - "%s: ERROR - Eventlog already initialized\n", - __func__); - return -EFAULT; - } - - /* Find TCPA entry in RSDT (ACPI_LOGICAL_ADDRESSING) */ - status = acpi_get_table(ACPI_SIG_TCPA, 1, - (struct acpi_table_header **)&buff); - - if (ACPI_FAILURE(status)) { - printk(KERN_ERR "%s: ERROR - Could not get TCPA table\n", - __func__); - return -EIO; - } - - switch(buff->platform_class) { - case BIOS_SERVER: - len = buff->server.log_max_len; - start = buff->server.log_start_addr; - break; - case BIOS_CLIENT: - default: - len = buff->client.log_max_len; - start = buff->client.log_start_addr; - break; - } - if (!len) { - printk(KERN_ERR "%s: ERROR - TCPA log area empty\n", __func__); - return -EIO; - } - - /* malloc EventLog space */ - log->bios_event_log = kmalloc(len, GFP_KERNEL); - if (!log->bios_event_log) { - printk("%s: ERROR - Not enough Memory for BIOS measurements\n", - __func__); - return -ENOMEM; - } - - log->bios_event_log_end = log->bios_event_log + len; - - virt = acpi_os_map_memory(start, len); - - memcpy(log->bios_event_log, virt, len); - - acpi_os_unmap_memory(virt, len); - return 0; -} - static int tpm_ascii_bios_measurements_open(struct inode *inode, struct file *file) { diff --git a/drivers/char/tpm/tpm_eventlog.h b/drivers/char/tpm/tpm_eventlog.h new file mode 100644 index 0000000..e7da086 --- /dev/null +++ b/drivers/char/tpm/tpm_eventlog.h @@ -0,0 +1,86 @@ + +#ifndef __TPM_EVENTLOG_H__ +#define __TPM_EVENTLOG_H__ + +#define TCG_EVENT_NAME_LEN_MAX 255 +#define MAX_TEXT_EVENT 1000 /* Max event string length */ +#define ACPI_TCPA_SIG "TCPA" /* 0x41504354 /'TCPA' */ + +enum bios_platform_class { + BIOS_CLIENT = 0x00, + BIOS_SERVER = 0x01, +}; + +struct tpm_bios_log { + void *bios_event_log; + void *bios_event_log_end; +}; + +struct tcpa_event { + u32 pcr_index; + u32 event_type; + u8 pcr_value[20]; /* SHA1 */ + u32 event_size; + u8 event_data[0]; +}; + +enum tcpa_event_types { + PREBOOT = 0, + POST_CODE, + UNUSED, + NO_ACTION, + SEPARATOR, + ACTION, + EVENT_TAG, + SCRTM_CONTENTS, + SCRTM_VERSION, + CPU_MICROCODE, + PLATFORM_CONFIG_FLAGS, + TABLE_OF_DEVICES, + COMPACT_HASH, + IPL, + IPL_PARTITION_DATA, + NONHOST_CODE, + NONHOST_CONFIG, + NONHOST_INFO, +}; + +struct tcpa_pc_event { + u32 event_id; + u32 event_size; + u8 event_data[0]; +}; + +enum tcpa_pc_event_ids { + SMBIOS = 1, + BIS_CERT, + POST_BIOS_ROM, + ESCD, + CMOS, + NVRAM, + OPTION_ROM_EXEC, + OPTION_ROM_CONFIG, + OPTION_ROM_MICROCODE = 10, + S_CRTM_VERSION, + S_CRTM_CONTENTS, + POST_CONTENTS, + HOST_TABLE_OF_DEVICES, +}; + +int read_log(struct tpm_bios_log *log); + +#if defined(CONFIG_TCG_IBMVTPM) || defined(CONFIG_TCG_IBMVTPM_MODULE) || \ + defined(CONFIG_ACPI) +extern struct dentry **tpm_bios_log_setup(char *); +extern void tpm_bios_log_teardown(struct dentry **); +#else +static inline struct dentry **tpm_bios_log_setup(char *name) +{ + return NULL; +} +static inline void tpm_bios_log_teardown(struct dentry **dir) +{ +} +#endif + +#endif diff --git a/drivers/char/tpm/tpm_i2c_infineon.c b/drivers/char/tpm/tpm_i2c_infineon.c new file mode 100644 index 0000000..5a831ae --- /dev/null +++ b/drivers/char/tpm/tpm_i2c_infineon.c @@ -0,0 +1,695 @@ +/* + * Copyright (C) 2012 Infineon Technologies + * + * Authors: + * Peter Huewe <peter.huewe@infineon.com> + * + * Device driver for TCG/TCPA TPM (trusted platform module). + * Specifications at www.trustedcomputinggroup.org + * + * This device driver implements the TPM interface as defined in + * the TCG TPM Interface Spec version 1.2, revision 1.0 and the + * Infineon I2C Protocol Stack Specification v0.20. + * + * It is based on the original tpm_tis device driver from Leendert van + * Dorn and Kyleen Hall. + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation, version 2 of the + * License. + * + * + */ +#include <linux/init.h> +#include <linux/i2c.h> +#include <linux/module.h> +#include <linux/moduleparam.h> +#include <linux/wait.h> +#include "tpm.h" + +/* max. buffer size supported by our TPM */ +#define TPM_BUFSIZE 1260 + +/* max. number of iterations after I2C NAK */ +#define MAX_COUNT 3 + +#define SLEEP_DURATION_LOW 55 +#define SLEEP_DURATION_HI 65 + +/* max. number of iterations after I2C NAK for 'long' commands + * we need this especially for sending TPM_READY, since the cleanup after the + * transtion to the ready state may take some time, but it is unpredictable + * how long it will take. + */ +#define MAX_COUNT_LONG 50 + +#define SLEEP_DURATION_LONG_LOW 200 +#define SLEEP_DURATION_LONG_HI 220 + +/* After sending TPM_READY to 'reset' the TPM we have to sleep even longer */ +#define SLEEP_DURATION_RESET_LOW 2400 +#define SLEEP_DURATION_RESET_HI 2600 + +/* we want to use usleep_range instead of msleep for the 5ms TPM_TIMEOUT */ +#define TPM_TIMEOUT_US_LOW (TPM_TIMEOUT * 1000) +#define TPM_TIMEOUT_US_HI (TPM_TIMEOUT_US_LOW + 2000) + +/* expected value for DIDVID register */ +#define TPM_TIS_I2C_DID_VID 0x000b15d1L + +/* Structure to store I2C TPM specific stuff */ +struct tpm_inf_dev { + struct i2c_client *client; + u8 buf[TPM_BUFSIZE + sizeof(u8)]; /* max. buffer size + addr */ + struct tpm_chip *chip; +}; + +static struct tpm_inf_dev tpm_dev; +static struct i2c_driver tpm_tis_i2c_driver; + +/* + * iic_tpm_read() - read from TPM register + * @addr: register address to read from + * @buffer: provided by caller + * @len: number of bytes to read + * + * Read len bytes from TPM register and put them into + * buffer (little-endian format, i.e. first byte is put into buffer[0]). + * + * NOTE: TPM is big-endian for multi-byte values. Multi-byte + * values have to be swapped. + * + * NOTE: We can't unfortunately use the combined read/write functions + * provided by the i2c core as the TPM currently does not support the + * repeated start condition and due to it's special requirements. + * The i2c_smbus* functions do not work for this chip. + * + * Return -EIO on error, 0 on success. + */ +static int iic_tpm_read(u8 addr, u8 *buffer, size_t len) +{ + + struct i2c_msg msg1 = { tpm_dev.client->addr, 0, 1, &addr }; + struct i2c_msg msg2 = { tpm_dev.client->addr, I2C_M_RD, len, buffer }; + + int rc; + int count; + + /* Lock the adapter for the duration of the whole sequence. */ + if (!tpm_dev.client->adapter->algo->master_xfer) + return -EOPNOTSUPP; + i2c_lock_adapter(tpm_dev.client->adapter); + + for (count = 0; count < MAX_COUNT; count++) { + rc = __i2c_transfer(tpm_dev.client->adapter, &msg1, 1); + if (rc > 0) + break; /* break here to skip sleep */ + + usleep_range(SLEEP_DURATION_LOW, SLEEP_DURATION_HI); + } + + if (rc <= 0) + goto out; + + /* After the TPM has successfully received the register address it needs + * some time, thus we're sleeping here again, before retrieving the data + */ + for (count = 0; count < MAX_COUNT; count++) { + usleep_range(SLEEP_DURATION_LOW, SLEEP_DURATION_HI); + rc = __i2c_transfer(tpm_dev.client->adapter, &msg2, 1); + if (rc > 0) + break; + + } + +out: + i2c_unlock_adapter(tpm_dev.client->adapter); + if (rc <= 0) + return -EIO; + + return 0; +} + +static int iic_tpm_write_generic(u8 addr, u8 *buffer, size_t len, + unsigned int sleep_low, + unsigned int sleep_hi, u8 max_count) +{ + int rc = -EIO; + int count; + + struct i2c_msg msg1 = { tpm_dev.client->addr, 0, len + 1, tpm_dev.buf }; + + if (len > TPM_BUFSIZE) + return -EINVAL; + + if (!tpm_dev.client->adapter->algo->master_xfer) + return -EOPNOTSUPP; + i2c_lock_adapter(tpm_dev.client->adapter); + + /* prepend the 'register address' to the buffer */ + tpm_dev.buf[0] = addr; + memcpy(&(tpm_dev.buf[1]), buffer, len); + + /* + * NOTE: We have to use these special mechanisms here and unfortunately + * cannot rely on the standard behavior of i2c_transfer. + */ + for (count = 0; count < max_count; count++) { + rc = __i2c_transfer(tpm_dev.client->adapter, &msg1, 1); + if (rc > 0) + break; + + usleep_range(sleep_low, sleep_hi); + } + + i2c_unlock_adapter(tpm_dev.client->adapter); + if (rc <= 0) + return -EIO; + + return 0; +} + +/* + * iic_tpm_write() - write to TPM register + * @addr: register address to write to + * @buffer: containing data to be written + * @len: number of bytes to write + * + * Write len bytes from provided buffer to TPM register (little + * endian format, i.e. buffer[0] is written as first byte). + * + * NOTE: TPM is big-endian for multi-byte values. Multi-byte + * values have to be swapped. + * + * NOTE: use this function instead of the iic_tpm_write_generic function. + * + * Return -EIO on error, 0 on success + */ +static int iic_tpm_write(u8 addr, u8 *buffer, size_t len) +{ + return iic_tpm_write_generic(addr, buffer, len, SLEEP_DURATION_LOW, + SLEEP_DURATION_HI, MAX_COUNT); +} + +/* + * This function is needed especially for the cleanup situation after + * sending TPM_READY + * */ +static int iic_tpm_write_long(u8 addr, u8 *buffer, size_t len) +{ + return iic_tpm_write_generic(addr, buffer, len, SLEEP_DURATION_LONG_LOW, + SLEEP_DURATION_LONG_HI, MAX_COUNT_LONG); +} + +enum tis_access { + TPM_ACCESS_VALID = 0x80, + TPM_ACCESS_ACTIVE_LOCALITY = 0x20, + TPM_ACCESS_REQUEST_PENDING = 0x04, + TPM_ACCESS_REQUEST_USE = 0x02, +}; + +enum tis_status { + TPM_STS_VALID = 0x80, + TPM_STS_COMMAND_READY = 0x40, + TPM_STS_GO = 0x20, + TPM_STS_DATA_AVAIL = 0x10, + TPM_STS_DATA_EXPECT = 0x08, +}; + +enum tis_defaults { + TIS_SHORT_TIMEOUT = 750, /* ms */ + TIS_LONG_TIMEOUT = 2000, /* 2 sec */ +}; + +#define TPM_ACCESS(l) (0x0000 | ((l) << 4)) +#define TPM_STS(l) (0x0001 | ((l) << 4)) +#define TPM_DATA_FIFO(l) (0x0005 | ((l) << 4)) +#define TPM_DID_VID(l) (0x0006 | ((l) << 4)) + +static int check_locality(struct tpm_chip *chip, int loc) +{ + u8 buf; + int rc; + + rc = iic_tpm_read(TPM_ACCESS(loc), &buf, 1); + if (rc < 0) + return rc; + + if ((buf & (TPM_ACCESS_ACTIVE_LOCALITY | TPM_ACCESS_VALID)) == + (TPM_ACCESS_ACTIVE_LOCALITY | TPM_ACCESS_VALID)) { + chip->vendor.locality = loc; + return loc; + } + + return -EIO; +} + +/* implementation similar to tpm_tis */ +static void release_locality(struct tpm_chip *chip, int loc, int force) +{ + u8 buf; + if (iic_tpm_read(TPM_ACCESS(loc), &buf, 1) < 0) + return; + + if (force || (buf & (TPM_ACCESS_REQUEST_PENDING | TPM_ACCESS_VALID)) == + (TPM_ACCESS_REQUEST_PENDING | TPM_ACCESS_VALID)) { + buf = TPM_ACCESS_ACTIVE_LOCALITY; + iic_tpm_write(TPM_ACCESS(loc), &buf, 1); + } +} + +static int request_locality(struct tpm_chip *chip, int loc) +{ + unsigned long stop; + u8 buf = TPM_ACCESS_REQUEST_USE; + + if (check_locality(chip, loc) >= 0) + return loc; + + iic_tpm_write(TPM_ACCESS(loc), &buf, 1); + + /* wait for burstcount */ + stop = jiffies + chip->vendor.timeout_a; + do { + if (check_locality(chip, loc) >= 0) + return loc; + usleep_range(TPM_TIMEOUT_US_LOW, TPM_TIMEOUT_US_HI); + } while (time_before(jiffies, stop)); + + return -ETIME; +} + +static u8 tpm_tis_i2c_status(struct tpm_chip *chip) +{ + /* NOTE: since I2C read may fail, return 0 in this case --> time-out */ + u8 buf; + if (iic_tpm_read(TPM_STS(chip->vendor.locality), &buf, 1) < 0) + return 0; + else + return buf; +} + +static void tpm_tis_i2c_ready(struct tpm_chip *chip) +{ + /* this causes the current command to be aborted */ + u8 buf = TPM_STS_COMMAND_READY; + iic_tpm_write_long(TPM_STS(chip->vendor.locality), &buf, 1); +} + +static ssize_t get_burstcount(struct tpm_chip *chip) +{ + unsigned long stop; + ssize_t burstcnt; + u8 buf[3]; + + /* wait for burstcount */ + /* which timeout value, spec has 2 answers (c & d) */ + stop = jiffies + chip->vendor.timeout_d; + do { + /* Note: STS is little endian */ + if (iic_tpm_read(TPM_STS(chip->vendor.locality)+1, buf, 3) < 0) + burstcnt = 0; + else + burstcnt = (buf[2] << 16) + (buf[1] << 8) + buf[0]; + + if (burstcnt) + return burstcnt; + + usleep_range(TPM_TIMEOUT_US_LOW, TPM_TIMEOUT_US_HI); + } while (time_before(jiffies, stop)); + return -EBUSY; +} + +static int wait_for_stat(struct tpm_chip *chip, u8 mask, unsigned long timeout, + int *status) +{ + unsigned long stop; + + /* check current status */ + *status = tpm_tis_i2c_status(chip); + if ((*status & mask) == mask) + return 0; + + stop = jiffies + timeout; + do { + /* since we just checked the status, give the TPM some time */ + usleep_range(TPM_TIMEOUT_US_LOW, TPM_TIMEOUT_US_HI); + *status = tpm_tis_i2c_status(chip); + if ((*status & mask) == mask) + return 0; + + } while (time_before(jiffies, stop)); + + return -ETIME; +} + +static int recv_data(struct tpm_chip *chip, u8 *buf, size_t count) +{ + size_t size = 0; + ssize_t burstcnt; + u8 retries = 0; + int rc; + + while (size < count) { + burstcnt = get_burstcount(chip); + + /* burstcnt < 0 = TPM is busy */ + if (burstcnt < 0) + return burstcnt; + + /* limit received data to max. left */ + if (burstcnt > (count - size)) + burstcnt = count - size; + + rc = iic_tpm_read(TPM_DATA_FIFO(chip->vendor.locality), + &(buf[size]), burstcnt); + if (rc == 0) + size += burstcnt; + else if (rc < 0) + retries++; + + /* avoid endless loop in case of broken HW */ + if (retries > MAX_COUNT_LONG) + return -EIO; + + } + return size; +} + +static int tpm_tis_i2c_recv(struct tpm_chip *chip, u8 *buf, size_t count) +{ + int size = 0; + int expected, status; + + if (count < TPM_HEADER_SIZE) { + size = -EIO; + goto out; + } + + /* read first 10 bytes, including tag, paramsize, and result */ + size = recv_data(chip, buf, TPM_HEADER_SIZE); + if (size < TPM_HEADER_SIZE) { + dev_err(chip->dev, "Unable to read header\n"); + goto out; + } + + expected = be32_to_cpu(*(__be32 *)(buf + 2)); + if ((size_t) expected > count) { + size = -EIO; + goto out; + } + + size += recv_data(chip, &buf[TPM_HEADER_SIZE], + expected - TPM_HEADER_SIZE); + if (size < expected) { + dev_err(chip->dev, "Unable to read remainder of result\n"); + size = -ETIME; + goto out; + } + + wait_for_stat(chip, TPM_STS_VALID, chip->vendor.timeout_c, &status); + if (status & TPM_STS_DATA_AVAIL) { /* retry? */ + dev_err(chip->dev, "Error left over data\n"); + size = -EIO; + goto out; + } + +out: + tpm_tis_i2c_ready(chip); + /* The TPM needs some time to clean up here, + * so we sleep rather than keeping the bus busy + */ + usleep_range(SLEEP_DURATION_RESET_LOW, SLEEP_DURATION_RESET_HI); + release_locality(chip, chip->vendor.locality, 0); + return size; +} + +static int tpm_tis_i2c_send(struct tpm_chip *chip, u8 *buf, size_t len) +{ + int rc, status; + ssize_t burstcnt; + size_t count = 0; + u8 retries = 0; + u8 sts = TPM_STS_GO; + + if (len > TPM_BUFSIZE) + return -E2BIG; /* command is too long for our tpm, sorry */ + + if (request_locality(chip, 0) < 0) + return -EBUSY; + + status = tpm_tis_i2c_status(chip); + if ((status & TPM_STS_COMMAND_READY) == 0) { + tpm_tis_i2c_ready(chip); + if (wait_for_stat + (chip, TPM_STS_COMMAND_READY, + chip->vendor.timeout_b, &status) < 0) { + rc = -ETIME; + goto out_err; + } + } + + while (count < len - 1) { + burstcnt = get_burstcount(chip); + + /* burstcnt < 0 = TPM is busy */ + if (burstcnt < 0) + return burstcnt; + + if (burstcnt > (len - 1 - count)) + burstcnt = len - 1 - count; + + rc = iic_tpm_write(TPM_DATA_FIFO(chip->vendor.locality), + &(buf[count]), burstcnt); + if (rc == 0) + count += burstcnt; + else if (rc < 0) + retries++; + + /* avoid endless loop in case of broken HW */ + if (retries > MAX_COUNT_LONG) { + rc = -EIO; + goto out_err; + } + + wait_for_stat(chip, TPM_STS_VALID, + chip->vendor.timeout_c, &status); + + if ((status & TPM_STS_DATA_EXPECT) == 0) { + rc = -EIO; + goto out_err; + } + + } + + /* write last byte */ + iic_tpm_write(TPM_DATA_FIFO(chip->vendor.locality), &(buf[count]), 1); + wait_for_stat(chip, TPM_STS_VALID, chip->vendor.timeout_c, &status); + if ((status & TPM_STS_DATA_EXPECT) != 0) { + rc = -EIO; + goto out_err; + } + + /* go and do it */ + iic_tpm_write(TPM_STS(chip->vendor.locality), &sts, 1); + + return len; +out_err: + tpm_tis_i2c_ready(chip); + /* The TPM needs some time to clean up here, + * so we sleep rather than keeping the bus busy + */ + usleep_range(SLEEP_DURATION_RESET_LOW, SLEEP_DURATION_RESET_HI); + release_locality(chip, chip->vendor.locality, 0); + return rc; +} + +static const struct file_operations tis_ops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .open = tpm_open, + .read = tpm_read, + .write = tpm_write, + .release = tpm_release, +}; + +static DEVICE_ATTR(pubek, S_IRUGO, tpm_show_pubek, NULL); +static DEVICE_ATTR(pcrs, S_IRUGO, tpm_show_pcrs, NULL); +static DEVICE_ATTR(enabled, S_IRUGO, tpm_show_enabled, NULL); +static DEVICE_ATTR(active, S_IRUGO, tpm_show_active, NULL); +static DEVICE_ATTR(owned, S_IRUGO, tpm_show_owned, NULL); +static DEVICE_ATTR(temp_deactivated, S_IRUGO, tpm_show_temp_deactivated, NULL); +static DEVICE_ATTR(caps, S_IRUGO, tpm_show_caps_1_2, NULL); +static DEVICE_ATTR(cancel, S_IWUSR | S_IWGRP, NULL, tpm_store_cancel); +static DEVICE_ATTR(durations, S_IRUGO, tpm_show_durations, NULL); +static DEVICE_ATTR(timeouts, S_IRUGO, tpm_show_timeouts, NULL); + +static struct attribute *tis_attrs[] = { + &dev_attr_pubek.attr, + &dev_attr_pcrs.attr, + &dev_attr_enabled.attr, + &dev_attr_active.attr, + &dev_attr_owned.attr, + &dev_attr_temp_deactivated.attr, + &dev_attr_caps.attr, + &dev_attr_cancel.attr, + &dev_attr_durations.attr, + &dev_attr_timeouts.attr, + NULL, +}; + +static struct attribute_group tis_attr_grp = { + .attrs = tis_attrs +}; + +static struct tpm_vendor_specific tpm_tis_i2c = { + .status = tpm_tis_i2c_status, + .recv = tpm_tis_i2c_recv, + .send = tpm_tis_i2c_send, + .cancel = tpm_tis_i2c_ready, + .req_complete_mask = TPM_STS_DATA_AVAIL | TPM_STS_VALID, + .req_complete_val = TPM_STS_DATA_AVAIL | TPM_STS_VALID, + .req_canceled = TPM_STS_COMMAND_READY, + .attr_group = &tis_attr_grp, + .miscdev.fops = &tis_ops, +}; + +static int __devinit tpm_tis_i2c_init(struct device *dev) +{ + u32 vendor; + int rc = 0; + struct tpm_chip *chip; + + chip = tpm_register_hardware(dev, &tpm_tis_i2c); + if (!chip) { + rc = -ENODEV; + goto out_err; + } + + /* Disable interrupts */ + chip->vendor.irq = 0; + + /* Default timeouts */ + chip->vendor.timeout_a = msecs_to_jiffies(TIS_SHORT_TIMEOUT); + chip->vendor.timeout_b = msecs_to_jiffies(TIS_LONG_TIMEOUT); + chip->vendor.timeout_c = msecs_to_jiffies(TIS_SHORT_TIMEOUT); + chip->vendor.timeout_d = msecs_to_jiffies(TIS_SHORT_TIMEOUT); + + if (request_locality(chip, 0) != 0) { + rc = -ENODEV; + goto out_vendor; + } + + /* read four bytes from DID_VID register */ + if (iic_tpm_read(TPM_DID_VID(0), (u8 *)&vendor, 4) < 0) { + rc = -EIO; + goto out_release; + } + + /* create DID_VID register value, after swapping to little-endian */ + vendor = be32_to_cpu((__be32) vendor); + + if (vendor != TPM_TIS_I2C_DID_VID) { + rc = -ENODEV; + goto out_release; + } + + dev_info(dev, "1.2 TPM (device-id 0x%X)\n", vendor >> 16); + + INIT_LIST_HEAD(&chip->vendor.list); + tpm_dev.chip = chip; + + tpm_get_timeouts(chip); + tpm_do_selftest(chip); + + return 0; + +out_release: + release_locality(chip, chip->vendor.locality, 1); + +out_vendor: + /* close file handles */ + tpm_dev_vendor_release(chip); + + /* remove hardware */ + tpm_remove_hardware(chip->dev); + + /* reset these pointers, otherwise we oops */ + chip->dev->release = NULL; + chip->release = NULL; + tpm_dev.client = NULL; + dev_set_drvdata(chip->dev, chip); +out_err: + return rc; +} + +static const struct i2c_device_id tpm_tis_i2c_table[] = { + {"tpm_i2c_infineon", 0}, + {}, +}; + +MODULE_DEVICE_TABLE(i2c, tpm_tis_i2c_table); +static SIMPLE_DEV_PM_OPS(tpm_tis_i2c_ops, tpm_pm_suspend, tpm_pm_resume); + +static int __devinit tpm_tis_i2c_probe(struct i2c_client *client, + const struct i2c_device_id *id) +{ + int rc; + if (tpm_dev.client != NULL) + return -EBUSY; /* We only support one client */ + + if (!i2c_check_functionality(client->adapter, I2C_FUNC_I2C)) { + dev_err(&client->dev, + "no algorithms associated to the i2c bus\n"); + return -ENODEV; + } + + client->driver = &tpm_tis_i2c_driver; + tpm_dev.client = client; + rc = tpm_tis_i2c_init(&client->dev); + if (rc != 0) { + client->driver = NULL; + tpm_dev.client = NULL; + rc = -ENODEV; + } + return rc; +} + +static int __devexit tpm_tis_i2c_remove(struct i2c_client *client) +{ + struct tpm_chip *chip = tpm_dev.chip; + release_locality(chip, chip->vendor.locality, 1); + + /* close file handles */ + tpm_dev_vendor_release(chip); + + /* remove hardware */ + tpm_remove_hardware(chip->dev); + + /* reset these pointers, otherwise we oops */ + chip->dev->release = NULL; + chip->release = NULL; + tpm_dev.client = NULL; + dev_set_drvdata(chip->dev, chip); + + return 0; +} + +static struct i2c_driver tpm_tis_i2c_driver = { + + .id_table = tpm_tis_i2c_table, + .probe = tpm_tis_i2c_probe, + .remove = tpm_tis_i2c_remove, + .driver = { + .name = "tpm_i2c_infineon", + .owner = THIS_MODULE, + .pm = &tpm_tis_i2c_ops, + }, +}; + +module_i2c_driver(tpm_tis_i2c_driver); +MODULE_AUTHOR("Peter Huewe <peter.huewe@infineon.com>"); +MODULE_DESCRIPTION("TPM TIS I2C Infineon Driver"); +MODULE_VERSION("2.1.5"); +MODULE_LICENSE("GPL"); diff --git a/drivers/char/tpm/tpm_ibmvtpm.c b/drivers/char/tpm/tpm_ibmvtpm.c new file mode 100644 index 0000000..efc4ab3 --- /dev/null +++ b/drivers/char/tpm/tpm_ibmvtpm.c @@ -0,0 +1,749 @@ +/* + * Copyright (C) 2012 IBM Corporation + * + * Author: Ashley Lai <adlai@us.ibm.com> + * + * Maintained by: <tpmdd-devel@lists.sourceforge.net> + * + * Device driver for TCG/TCPA TPM (trusted platform module). + * Specifications at www.trustedcomputinggroup.org + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation, version 2 of the + * License. + * + */ + +#include <linux/dma-mapping.h> +#include <linux/dmapool.h> +#include <linux/slab.h> +#include <asm/vio.h> +#include <asm/irq.h> +#include <linux/types.h> +#include <linux/list.h> +#include <linux/spinlock.h> +#include <linux/interrupt.h> +#include <linux/wait.h> +#include <asm/prom.h> + +#include "tpm.h" +#include "tpm_ibmvtpm.h" + +static const char tpm_ibmvtpm_driver_name[] = "tpm_ibmvtpm"; + +static struct vio_device_id tpm_ibmvtpm_device_table[] __devinitdata = { + { "IBM,vtpm", "IBM,vtpm"}, + { "", "" } +}; +MODULE_DEVICE_TABLE(vio, tpm_ibmvtpm_device_table); + +DECLARE_WAIT_QUEUE_HEAD(wq); + +/** + * ibmvtpm_send_crq - Send a CRQ request + * @vdev: vio device struct + * @w1: first word + * @w2: second word + * + * Return value: + * 0 -Sucess + * Non-zero - Failure + */ +static int ibmvtpm_send_crq(struct vio_dev *vdev, u64 w1, u64 w2) +{ + return plpar_hcall_norets(H_SEND_CRQ, vdev->unit_address, w1, w2); +} + +/** + * ibmvtpm_get_data - Retrieve ibm vtpm data + * @dev: device struct + * + * Return value: + * vtpm device struct + */ +static struct ibmvtpm_dev *ibmvtpm_get_data(const struct device *dev) +{ + struct tpm_chip *chip = dev_get_drvdata(dev); + if (chip) + return (struct ibmvtpm_dev *)chip->vendor.data; + return NULL; +} + +/** + * tpm_ibmvtpm_recv - Receive data after send + * @chip: tpm chip struct + * @buf: buffer to read + * count: size of buffer + * + * Return value: + * Number of bytes read + */ +static int tpm_ibmvtpm_recv(struct tpm_chip *chip, u8 *buf, size_t count) +{ + struct ibmvtpm_dev *ibmvtpm; + u16 len; + + ibmvtpm = (struct ibmvtpm_dev *)chip->vendor.data; + + if (!ibmvtpm->rtce_buf) { + dev_err(ibmvtpm->dev, "ibmvtpm device is not ready\n"); + return 0; + } + + wait_event_interruptible(wq, ibmvtpm->crq_res.len != 0); + + if (count < ibmvtpm->crq_res.len) { + dev_err(ibmvtpm->dev, + "Invalid size in recv: count=%ld, crq_size=%d\n", + count, ibmvtpm->crq_res.len); + return -EIO; + } + + spin_lock(&ibmvtpm->rtce_lock); + memcpy((void *)buf, (void *)ibmvtpm->rtce_buf, ibmvtpm->crq_res.len); + memset(ibmvtpm->rtce_buf, 0, ibmvtpm->crq_res.len); + ibmvtpm->crq_res.valid = 0; + ibmvtpm->crq_res.msg = 0; + len = ibmvtpm->crq_res.len; + ibmvtpm->crq_res.len = 0; + spin_unlock(&ibmvtpm->rtce_lock); + return len; +} + +/** + * tpm_ibmvtpm_send - Send tpm request + * @chip: tpm chip struct + * @buf: buffer contains data to send + * count: size of buffer + * + * Return value: + * Number of bytes sent + */ +static int tpm_ibmvtpm_send(struct tpm_chip *chip, u8 *buf, size_t count) +{ + struct ibmvtpm_dev *ibmvtpm; + struct ibmvtpm_crq crq; + u64 *word = (u64 *) &crq; + int rc; + + ibmvtpm = (struct ibmvtpm_dev *)chip->vendor.data; + + if (!ibmvtpm->rtce_buf) { + dev_err(ibmvtpm->dev, "ibmvtpm device is not ready\n"); + return 0; + } + + if (count > ibmvtpm->rtce_size) { + dev_err(ibmvtpm->dev, + "Invalid size in send: count=%ld, rtce_size=%d\n", + count, ibmvtpm->rtce_size); + return -EIO; + } + + spin_lock(&ibmvtpm->rtce_lock); + memcpy((void *)ibmvtpm->rtce_buf, (void *)buf, count); + crq.valid = (u8)IBMVTPM_VALID_CMD; + crq.msg = (u8)VTPM_TPM_COMMAND; + crq.len = (u16)count; + crq.data = ibmvtpm->rtce_dma_handle; + + rc = ibmvtpm_send_crq(ibmvtpm->vdev, word[0], word[1]); + if (rc != H_SUCCESS) { + dev_err(ibmvtpm->dev, "tpm_ibmvtpm_send failed rc=%d\n", rc); + rc = 0; + } else + rc = count; + + spin_unlock(&ibmvtpm->rtce_lock); + return rc; +} + +static void tpm_ibmvtpm_cancel(struct tpm_chip *chip) +{ + return; +} + +static u8 tpm_ibmvtpm_status(struct tpm_chip *chip) +{ + return 0; +} + +/** + * ibmvtpm_crq_get_rtce_size - Send a CRQ request to get rtce size + * @ibmvtpm: vtpm device struct + * + * Return value: + * 0 - Success + * Non-zero - Failure + */ +static int ibmvtpm_crq_get_rtce_size(struct ibmvtpm_dev *ibmvtpm) +{ + struct ibmvtpm_crq crq; + u64 *buf = (u64 *) &crq; + int rc; + + crq.valid = (u8)IBMVTPM_VALID_CMD; + crq.msg = (u8)VTPM_GET_RTCE_BUFFER_SIZE; + + rc = ibmvtpm_send_crq(ibmvtpm->vdev, buf[0], buf[1]); + if (rc != H_SUCCESS) + dev_err(ibmvtpm->dev, + "ibmvtpm_crq_get_rtce_size failed rc=%d\n", rc); + + return rc; +} + +/** + * ibmvtpm_crq_get_version - Send a CRQ request to get vtpm version + * - Note that this is vtpm version and not tpm version + * @ibmvtpm: vtpm device struct + * + * Return value: + * 0 - Success + * Non-zero - Failure + */ +static int ibmvtpm_crq_get_version(struct ibmvtpm_dev *ibmvtpm) +{ + struct ibmvtpm_crq crq; + u64 *buf = (u64 *) &crq; + int rc; + + crq.valid = (u8)IBMVTPM_VALID_CMD; + crq.msg = (u8)VTPM_GET_VERSION; + + rc = ibmvtpm_send_crq(ibmvtpm->vdev, buf[0], buf[1]); + if (rc != H_SUCCESS) + dev_err(ibmvtpm->dev, + "ibmvtpm_crq_get_version failed rc=%d\n", rc); + + return rc; +} + +/** + * ibmvtpm_crq_send_init_complete - Send a CRQ initialize complete message + * @ibmvtpm: vtpm device struct + * + * Return value: + * 0 - Success + * Non-zero - Failure + */ +static int ibmvtpm_crq_send_init_complete(struct ibmvtpm_dev *ibmvtpm) +{ + int rc; + + rc = ibmvtpm_send_crq(ibmvtpm->vdev, INIT_CRQ_COMP_CMD, 0); + if (rc != H_SUCCESS) + dev_err(ibmvtpm->dev, + "ibmvtpm_crq_send_init_complete failed rc=%d\n", rc); + + return rc; +} + +/** + * ibmvtpm_crq_send_init - Send a CRQ initialize message + * @ibmvtpm: vtpm device struct + * + * Return value: + * 0 - Success + * Non-zero - Failure + */ +static int ibmvtpm_crq_send_init(struct ibmvtpm_dev *ibmvtpm) +{ + int rc; + + rc = ibmvtpm_send_crq(ibmvtpm->vdev, INIT_CRQ_CMD, 0); + if (rc != H_SUCCESS) + dev_err(ibmvtpm->dev, + "ibmvtpm_crq_send_init failed rc=%d\n", rc); + + return rc; +} + +/** + * tpm_ibmvtpm_remove - ibm vtpm remove entry point + * @vdev: vio device struct + * + * Return value: + * 0 + */ +static int __devexit tpm_ibmvtpm_remove(struct vio_dev *vdev) +{ + struct ibmvtpm_dev *ibmvtpm = ibmvtpm_get_data(&vdev->dev); + int rc = 0; + + free_irq(vdev->irq, ibmvtpm); + tasklet_kill(&ibmvtpm->tasklet); + + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); + } while (rc == H_BUSY || H_IS_LONG_BUSY(rc)); + + dma_unmap_single(ibmvtpm->dev, ibmvtpm->crq_dma_handle, + CRQ_RES_BUF_SIZE, DMA_BIDIRECTIONAL); + free_page((unsigned long)ibmvtpm->crq_queue.crq_addr); + + if (ibmvtpm->rtce_buf) { + dma_unmap_single(ibmvtpm->dev, ibmvtpm->rtce_dma_handle, + ibmvtpm->rtce_size, DMA_BIDIRECTIONAL); + kfree(ibmvtpm->rtce_buf); + } + + tpm_remove_hardware(ibmvtpm->dev); + + kfree(ibmvtpm); + + return 0; +} + +/** + * tpm_ibmvtpm_get_desired_dma - Get DMA size needed by this driver + * @vdev: vio device struct + * + * Return value: + * Number of bytes the driver needs to DMA map + */ +static unsigned long tpm_ibmvtpm_get_desired_dma(struct vio_dev *vdev) +{ + struct ibmvtpm_dev *ibmvtpm = ibmvtpm_get_data(&vdev->dev); + return CRQ_RES_BUF_SIZE + ibmvtpm->rtce_size; +} + +/** + * tpm_ibmvtpm_suspend - Suspend + * @dev: device struct + * + * Return value: + * 0 + */ +static int tpm_ibmvtpm_suspend(struct device *dev) +{ + struct ibmvtpm_dev *ibmvtpm = ibmvtpm_get_data(dev); + struct ibmvtpm_crq crq; + u64 *buf = (u64 *) &crq; + int rc = 0; + + crq.valid = (u8)IBMVTPM_VALID_CMD; + crq.msg = (u8)VTPM_PREPARE_TO_SUSPEND; + + rc = ibmvtpm_send_crq(ibmvtpm->vdev, buf[0], buf[1]); + if (rc != H_SUCCESS) + dev_err(ibmvtpm->dev, + "tpm_ibmvtpm_suspend failed rc=%d\n", rc); + + return rc; +} + +/** + * ibmvtpm_reset_crq - Reset CRQ + * @ibmvtpm: ibm vtpm struct + * + * Return value: + * 0 - Success + * Non-zero - Failure + */ +static int ibmvtpm_reset_crq(struct ibmvtpm_dev *ibmvtpm) +{ + int rc = 0; + + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_FREE_CRQ, + ibmvtpm->vdev->unit_address); + } while (rc == H_BUSY || H_IS_LONG_BUSY(rc)); + + memset(ibmvtpm->crq_queue.crq_addr, 0, CRQ_RES_BUF_SIZE); + ibmvtpm->crq_queue.index = 0; + + return plpar_hcall_norets(H_REG_CRQ, ibmvtpm->vdev->unit_address, + ibmvtpm->crq_dma_handle, CRQ_RES_BUF_SIZE); +} + +/** + * tpm_ibmvtpm_resume - Resume from suspend + * @dev: device struct + * + * Return value: + * 0 + */ +static int tpm_ibmvtpm_resume(struct device *dev) +{ + struct ibmvtpm_dev *ibmvtpm = ibmvtpm_get_data(dev); + unsigned long flags; + int rc = 0; + + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_ENABLE_CRQ, + ibmvtpm->vdev->unit_address); + } while (rc == H_IN_PROGRESS || rc == H_BUSY || H_IS_LONG_BUSY(rc)); + + if (rc) { + dev_err(dev, "Error enabling ibmvtpm rc=%d\n", rc); + return rc; + } + + spin_lock_irqsave(&ibmvtpm->lock, flags); + vio_disable_interrupts(ibmvtpm->vdev); + tasklet_schedule(&ibmvtpm->tasklet); + spin_unlock_irqrestore(&ibmvtpm->lock, flags); + + rc = ibmvtpm_crq_send_init(ibmvtpm); + if (rc) + dev_err(dev, "Error send_init rc=%d\n", rc); + + return rc; +} + +static const struct file_operations ibmvtpm_ops = { + .owner = THIS_MODULE, + .llseek = no_llseek, + .open = tpm_open, + .read = tpm_read, + .write = tpm_write, + .release = tpm_release, +}; + +static DEVICE_ATTR(pubek, S_IRUGO, tpm_show_pubek, NULL); +static DEVICE_ATTR(pcrs, S_IRUGO, tpm_show_pcrs, NULL); +static DEVICE_ATTR(enabled, S_IRUGO, tpm_show_enabled, NULL); +static DEVICE_ATTR(active, S_IRUGO, tpm_show_active, NULL); +static DEVICE_ATTR(owned, S_IRUGO, tpm_show_owned, NULL); +static DEVICE_ATTR(temp_deactivated, S_IRUGO, tpm_show_temp_deactivated, + NULL); +static DEVICE_ATTR(caps, S_IRUGO, tpm_show_caps_1_2, NULL); +static DEVICE_ATTR(cancel, S_IWUSR | S_IWGRP, NULL, tpm_store_cancel); +static DEVICE_ATTR(durations, S_IRUGO, tpm_show_durations, NULL); +static DEVICE_ATTR(timeouts, S_IRUGO, tpm_show_timeouts, NULL); + +static struct attribute *ibmvtpm_attrs[] = { + &dev_attr_pubek.attr, + &dev_attr_pcrs.attr, + &dev_attr_enabled.attr, + &dev_attr_active.attr, + &dev_attr_owned.attr, + &dev_attr_temp_deactivated.attr, + &dev_attr_caps.attr, + &dev_attr_cancel.attr, + &dev_attr_durations.attr, + &dev_attr_timeouts.attr, NULL, +}; + +static struct attribute_group ibmvtpm_attr_grp = { .attrs = ibmvtpm_attrs }; + +static const struct tpm_vendor_specific tpm_ibmvtpm = { + .recv = tpm_ibmvtpm_recv, + .send = tpm_ibmvtpm_send, + .cancel = tpm_ibmvtpm_cancel, + .status = tpm_ibmvtpm_status, + .req_complete_mask = 0, + .req_complete_val = 0, + .req_canceled = 0, + .attr_group = &ibmvtpm_attr_grp, + .miscdev = { .fops = &ibmvtpm_ops, }, +}; + +static const struct dev_pm_ops tpm_ibmvtpm_pm_ops = { + .suspend = tpm_ibmvtpm_suspend, + .resume = tpm_ibmvtpm_resume, +}; + +/** + * ibmvtpm_crq_get_next - Get next responded crq + * @ibmvtpm vtpm device struct + * + * Return value: + * vtpm crq pointer + */ +static struct ibmvtpm_crq *ibmvtpm_crq_get_next(struct ibmvtpm_dev *ibmvtpm) +{ + struct ibmvtpm_crq_queue *crq_q = &ibmvtpm->crq_queue; + struct ibmvtpm_crq *crq = &crq_q->crq_addr[crq_q->index]; + + if (crq->valid & VTPM_MSG_RES) { + if (++crq_q->index == crq_q->num_entry) + crq_q->index = 0; + rmb(); + } else + crq = NULL; + return crq; +} + +/** + * ibmvtpm_crq_process - Process responded crq + * @crq crq to be processed + * @ibmvtpm vtpm device struct + * + * Return value: + * Nothing + */ +static void ibmvtpm_crq_process(struct ibmvtpm_crq *crq, + struct ibmvtpm_dev *ibmvtpm) +{ + int rc = 0; + + switch (crq->valid) { + case VALID_INIT_CRQ: + switch (crq->msg) { + case INIT_CRQ_RES: + dev_info(ibmvtpm->dev, "CRQ initialized\n"); + rc = ibmvtpm_crq_send_init_complete(ibmvtpm); + if (rc) + dev_err(ibmvtpm->dev, "Unable to send CRQ init complete rc=%d\n", rc); + return; + case INIT_CRQ_COMP_RES: + dev_info(ibmvtpm->dev, + "CRQ initialization completed\n"); + return; + default: + dev_err(ibmvtpm->dev, "Unknown crq message type: %d\n", crq->msg); + return; + } + return; + case IBMVTPM_VALID_CMD: + switch (crq->msg) { + case VTPM_GET_RTCE_BUFFER_SIZE_RES: + if (crq->len <= 0) { + dev_err(ibmvtpm->dev, "Invalid rtce size\n"); + return; + } + ibmvtpm->rtce_size = crq->len; + ibmvtpm->rtce_buf = kmalloc(ibmvtpm->rtce_size, + GFP_KERNEL); + if (!ibmvtpm->rtce_buf) { + dev_err(ibmvtpm->dev, "Failed to allocate memory for rtce buffer\n"); + return; + } + + ibmvtpm->rtce_dma_handle = dma_map_single(ibmvtpm->dev, + ibmvtpm->rtce_buf, ibmvtpm->rtce_size, + DMA_BIDIRECTIONAL); + + if (dma_mapping_error(ibmvtpm->dev, + ibmvtpm->rtce_dma_handle)) { + kfree(ibmvtpm->rtce_buf); + ibmvtpm->rtce_buf = NULL; + dev_err(ibmvtpm->dev, "Failed to dma map rtce buffer\n"); + } + + return; + case VTPM_GET_VERSION_RES: + ibmvtpm->vtpm_version = crq->data; + return; + case VTPM_TPM_COMMAND_RES: + ibmvtpm->crq_res.valid = crq->valid; + ibmvtpm->crq_res.msg = crq->msg; + ibmvtpm->crq_res.len = crq->len; + ibmvtpm->crq_res.data = crq->data; + wake_up_interruptible(&wq); + return; + default: + return; + } + } + return; +} + +/** + * ibmvtpm_interrupt - Interrupt handler + * @irq: irq number to handle + * @vtpm_instance: vtpm that received interrupt + * + * Returns: + * IRQ_HANDLED + **/ +static irqreturn_t ibmvtpm_interrupt(int irq, void *vtpm_instance) +{ + struct ibmvtpm_dev *ibmvtpm = (struct ibmvtpm_dev *) vtpm_instance; + unsigned long flags; + + spin_lock_irqsave(&ibmvtpm->lock, flags); + vio_disable_interrupts(ibmvtpm->vdev); + tasklet_schedule(&ibmvtpm->tasklet); + spin_unlock_irqrestore(&ibmvtpm->lock, flags); + + return IRQ_HANDLED; +} + +/** + * ibmvtpm_tasklet - Interrupt handler tasklet + * @data: ibm vtpm device struct + * + * Returns: + * Nothing + **/ +static void ibmvtpm_tasklet(void *data) +{ + struct ibmvtpm_dev *ibmvtpm = data; + struct ibmvtpm_crq *crq; + unsigned long flags; + + spin_lock_irqsave(&ibmvtpm->lock, flags); + while ((crq = ibmvtpm_crq_get_next(ibmvtpm)) != NULL) { + ibmvtpm_crq_process(crq, ibmvtpm); + crq->valid = 0; + wmb(); + } + + vio_enable_interrupts(ibmvtpm->vdev); + spin_unlock_irqrestore(&ibmvtpm->lock, flags); +} + +/** + * tpm_ibmvtpm_probe - ibm vtpm initialize entry point + * @vio_dev: vio device struct + * @id: vio device id struct + * + * Return value: + * 0 - Success + * Non-zero - Failure + */ +static int __devinit tpm_ibmvtpm_probe(struct vio_dev *vio_dev, + const struct vio_device_id *id) +{ + struct ibmvtpm_dev *ibmvtpm; + struct device *dev = &vio_dev->dev; + struct ibmvtpm_crq_queue *crq_q; + struct tpm_chip *chip; + int rc = -ENOMEM, rc1; + + chip = tpm_register_hardware(dev, &tpm_ibmvtpm); + if (!chip) { + dev_err(dev, "tpm_register_hardware failed\n"); + return -ENODEV; + } + + ibmvtpm = kzalloc(sizeof(struct ibmvtpm_dev), GFP_KERNEL); + if (!ibmvtpm) { + dev_err(dev, "kzalloc for ibmvtpm failed\n"); + goto cleanup; + } + + crq_q = &ibmvtpm->crq_queue; + crq_q->crq_addr = (struct ibmvtpm_crq *)get_zeroed_page(GFP_KERNEL); + if (!crq_q->crq_addr) { + dev_err(dev, "Unable to allocate memory for crq_addr\n"); + goto cleanup; + } + + crq_q->num_entry = CRQ_RES_BUF_SIZE / sizeof(*crq_q->crq_addr); + ibmvtpm->crq_dma_handle = dma_map_single(dev, crq_q->crq_addr, + CRQ_RES_BUF_SIZE, + DMA_BIDIRECTIONAL); + + if (dma_mapping_error(dev, ibmvtpm->crq_dma_handle)) { + dev_err(dev, "dma mapping failed\n"); + goto cleanup; + } + + rc = plpar_hcall_norets(H_REG_CRQ, vio_dev->unit_address, + ibmvtpm->crq_dma_handle, CRQ_RES_BUF_SIZE); + if (rc == H_RESOURCE) + rc = ibmvtpm_reset_crq(ibmvtpm); + + if (rc) { + dev_err(dev, "Unable to register CRQ rc=%d\n", rc); + goto reg_crq_cleanup; + } + + tasklet_init(&ibmvtpm->tasklet, (void *)ibmvtpm_tasklet, + (unsigned long)ibmvtpm); + + rc = request_irq(vio_dev->irq, ibmvtpm_interrupt, 0, + tpm_ibmvtpm_driver_name, ibmvtpm); + if (rc) { + dev_err(dev, "Error %d register irq 0x%x\n", rc, vio_dev->irq); + goto init_irq_cleanup; + } + + rc = vio_enable_interrupts(vio_dev); + if (rc) { + dev_err(dev, "Error %d enabling interrupts\n", rc); + goto init_irq_cleanup; + } + + crq_q->index = 0; + + ibmvtpm->dev = dev; + ibmvtpm->vdev = vio_dev; + chip->vendor.data = (void *)ibmvtpm; + + spin_lock_init(&ibmvtpm->lock); + spin_lock_init(&ibmvtpm->rtce_lock); + + rc = ibmvtpm_crq_send_init(ibmvtpm); + if (rc) + goto init_irq_cleanup; + + rc = ibmvtpm_crq_get_version(ibmvtpm); + if (rc) + goto init_irq_cleanup; + + rc = ibmvtpm_crq_get_rtce_size(ibmvtpm); + if (rc) + goto init_irq_cleanup; + + return rc; +init_irq_cleanup: + tasklet_kill(&ibmvtpm->tasklet); + do { + rc1 = plpar_hcall_norets(H_FREE_CRQ, vio_dev->unit_address); + } while (rc1 == H_BUSY || H_IS_LONG_BUSY(rc1)); +reg_crq_cleanup: + dma_unmap_single(dev, ibmvtpm->crq_dma_handle, CRQ_RES_BUF_SIZE, + DMA_BIDIRECTIONAL); +cleanup: + if (ibmvtpm) { + if (crq_q->crq_addr) + free_page((unsigned long)crq_q->crq_addr); + kfree(ibmvtpm); + } + + tpm_remove_hardware(dev); + + return rc; +} + +static struct vio_driver ibmvtpm_driver = { + .id_table = tpm_ibmvtpm_device_table, + .probe = tpm_ibmvtpm_probe, + .remove = tpm_ibmvtpm_remove, + .get_desired_dma = tpm_ibmvtpm_get_desired_dma, + .name = tpm_ibmvtpm_driver_name, + .pm = &tpm_ibmvtpm_pm_ops, +}; + +/** + * ibmvtpm_module_init - Initialize ibm vtpm module + * + * Return value: + * 0 -Success + * Non-zero - Failure + */ +static int __init ibmvtpm_module_init(void) +{ + return vio_register_driver(&ibmvtpm_driver); +} + +/** + * ibmvtpm_module_exit - Teardown ibm vtpm module + * + * Return value: + * Nothing + */ +static void __exit ibmvtpm_module_exit(void) +{ + vio_unregister_driver(&ibmvtpm_driver); +} + +module_init(ibmvtpm_module_init); +module_exit(ibmvtpm_module_exit); + +MODULE_AUTHOR("adlai@us.ibm.com"); +MODULE_DESCRIPTION("IBM vTPM Driver"); +MODULE_VERSION("1.0"); +MODULE_LICENSE("GPL"); diff --git a/drivers/char/tpm/tpm_ibmvtpm.h b/drivers/char/tpm/tpm_ibmvtpm.h new file mode 100644 index 0000000..4296eb4 --- /dev/null +++ b/drivers/char/tpm/tpm_ibmvtpm.h @@ -0,0 +1,77 @@ +/* + * Copyright (C) 2012 IBM Corporation + * + * Author: Ashley Lai <adlai@us.ibm.com> + * + * Maintained by: <tpmdd-devel@lists.sourceforge.net> + * + * Device driver for TCG/TCPA TPM (trusted platform module). + * Specifications at www.trustedcomputinggroup.org + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License as + * published by the Free Software Foundation, version 2 of the + * License. + * + */ + +#ifndef __TPM_IBMVTPM_H__ +#define __TPM_IBMVTPM_H__ + +/* vTPM Message Format 1 */ +struct ibmvtpm_crq { + u8 valid; + u8 msg; + u16 len; + u32 data; + u64 reserved; +} __attribute__((packed, aligned(8))); + +struct ibmvtpm_crq_queue { + struct ibmvtpm_crq *crq_addr; + u32 index; + u32 num_entry; +}; + +struct ibmvtpm_dev { + struct device *dev; + struct vio_dev *vdev; + struct ibmvtpm_crq_queue crq_queue; + dma_addr_t crq_dma_handle; + spinlock_t lock; + struct tasklet_struct tasklet; + u32 rtce_size; + void __iomem *rtce_buf; + dma_addr_t rtce_dma_handle; + spinlock_t rtce_lock; + struct ibmvtpm_crq crq_res; + u32 vtpm_version; +}; + +#define CRQ_RES_BUF_SIZE PAGE_SIZE + +/* Initialize CRQ */ +#define INIT_CRQ_CMD 0xC001000000000000LL /* Init cmd */ +#define INIT_CRQ_COMP_CMD 0xC002000000000000LL /* Init complete cmd */ +#define INIT_CRQ_RES 0x01 /* Init respond */ +#define INIT_CRQ_COMP_RES 0x02 /* Init complete respond */ +#define VALID_INIT_CRQ 0xC0 /* Valid command for init crq */ + +/* vTPM CRQ response is the message type | 0x80 */ +#define VTPM_MSG_RES 0x80 +#define IBMVTPM_VALID_CMD 0x80 + +/* vTPM CRQ message types */ +#define VTPM_GET_VERSION 0x01 +#define VTPM_GET_VERSION_RES (0x01 | VTPM_MSG_RES) + +#define VTPM_TPM_COMMAND 0x02 +#define VTPM_TPM_COMMAND_RES (0x02 | VTPM_MSG_RES) + +#define VTPM_GET_RTCE_BUFFER_SIZE 0x03 +#define VTPM_GET_RTCE_BUFFER_SIZE_RES (0x03 | VTPM_MSG_RES) + +#define VTPM_PREPARE_TO_SUSPEND 0x04 +#define VTPM_PREPARE_TO_SUSPEND_RES (0x04 | VTPM_MSG_RES) + +#endif diff --git a/drivers/char/tpm/tpm_of.c b/drivers/char/tpm/tpm_of.c new file mode 100644 index 0000000..98ba2bd --- /dev/null +++ b/drivers/char/tpm/tpm_of.c @@ -0,0 +1,73 @@ +/* + * Copyright 2012 IBM Corporation + * + * Author: Ashley Lai <adlai@us.ibm.com> + * + * Maintained by: <tpmdd-devel@lists.sourceforge.net> + * + * Read the event log created by the firmware on PPC64 + * + * This program is free software; you can redistribute it and/or + * modify it under the terms of the GNU General Public License + * as published by the Free Software Foundation; either version + * 2 of the License, or (at your option) any later version. + * + */ + +#include <linux/slab.h> +#include <linux/of.h> + +#include "tpm.h" +#include "tpm_eventlog.h" + +int read_log(struct tpm_bios_log *log) +{ + struct device_node *np; + const u32 *sizep; + const __be64 *basep; + + if (log->bios_event_log != NULL) { + pr_err("%s: ERROR - Eventlog already initialized\n", __func__); + return -EFAULT; + } + + np = of_find_node_by_name(NULL, "ibm,vtpm"); + if (!np) { + pr_err("%s: ERROR - IBMVTPM not supported\n", __func__); + return -ENODEV; + } + + sizep = of_get_property(np, "linux,sml-size", NULL); + if (sizep == NULL) { + pr_err("%s: ERROR - SML size not found\n", __func__); + goto cleanup_eio; + } + if (*sizep == 0) { + pr_err("%s: ERROR - event log area empty\n", __func__); + goto cleanup_eio; + } + + basep = of_get_property(np, "linux,sml-base", NULL); + if (basep == NULL) { + pr_err(KERN_ERR "%s: ERROR - SML not found\n", __func__); + goto cleanup_eio; + } + + of_node_put(np); + log->bios_event_log = kmalloc(*sizep, GFP_KERNEL); + if (!log->bios_event_log) { + pr_err("%s: ERROR - Not enough memory for BIOS measurements\n", + __func__); + return -ENOMEM; + } + + log->bios_event_log_end = log->bios_event_log + *sizep; + + memcpy(log->bios_event_log, __va(be64_to_cpup(basep)), *sizep); + + return 0; + +cleanup_eio: + of_node_put(np); + return -EIO; +} diff --git a/drivers/char/tpm/tpm_ppi.c b/drivers/char/tpm/tpm_ppi.c new file mode 100644 index 0000000..f27b58c --- /dev/null +++ b/drivers/char/tpm/tpm_ppi.c @@ -0,0 +1,461 @@ +#include <linux/acpi.h> +#include <acpi/acpi_drivers.h> +#include "tpm.h" + +static const u8 tpm_ppi_uuid[] = { + 0xA6, 0xFA, 0xDD, 0x3D, + 0x1B, 0x36, + 0xB4, 0x4E, + 0xA4, 0x24, + 0x8D, 0x10, 0x08, 0x9D, 0x16, 0x53 +}; +static char *tpm_device_name = "TPM"; + +#define TPM_PPI_REVISION_ID 1 +#define TPM_PPI_FN_VERSION 1 +#define TPM_PPI_FN_SUBREQ 2 +#define TPM_PPI_FN_GETREQ 3 +#define TPM_PPI_FN_GETACT 4 +#define TPM_PPI_FN_GETRSP 5 +#define TPM_PPI_FN_SUBREQ2 7 +#define TPM_PPI_FN_GETOPR 8 +#define PPI_TPM_REQ_MAX 22 +#define PPI_VS_REQ_START 128 +#define PPI_VS_REQ_END 255 +#define PPI_VERSION_LEN 3 + +static acpi_status ppi_callback(acpi_handle handle, u32 level, void *context, + void **return_value) +{ + acpi_status status; + struct acpi_buffer buffer = { ACPI_ALLOCATE_BUFFER, NULL }; + status = acpi_get_name(handle, ACPI_FULL_PATHNAME, &buffer); + if (strstr(buffer.pointer, context) != NULL) { + *return_value = handle; + kfree(buffer.pointer); + return AE_CTRL_TERMINATE; + } + return AE_OK; +} + +static inline void ppi_assign_params(union acpi_object params[4], + u64 function_num) +{ + params[0].type = ACPI_TYPE_BUFFER; + params[0].buffer.length = sizeof(tpm_ppi_uuid); + params[0].buffer.pointer = (char *)tpm_ppi_uuid; + params[1].type = ACPI_TYPE_INTEGER; + params[1].integer.value = TPM_PPI_REVISION_ID; + params[2].type = ACPI_TYPE_INTEGER; + params[2].integer.value = function_num; + params[3].type = ACPI_TYPE_PACKAGE; + params[3].package.count = 0; + params[3].package.elements = NULL; +} + +static ssize_t tpm_show_ppi_version(struct device *dev, + struct device_attribute *attr, char *buf) +{ + acpi_handle handle; + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object params[4]; + union acpi_object *obj; + + input.count = 4; + ppi_assign_params(params, TPM_PPI_FN_VERSION); + input.pointer = params; + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT, + ACPI_UINT32_MAX, ppi_callback, NULL, + tpm_device_name, &handle); + if (ACPI_FAILURE(status)) + return -ENXIO; + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_STRING); + if (ACPI_FAILURE(status)) + return -ENOMEM; + obj = (union acpi_object *)output.pointer; + status = scnprintf(buf, PAGE_SIZE, "%s\n", obj->string.pointer); + kfree(output.pointer); + return status; +} + +static ssize_t tpm_show_ppi_request(struct device *dev, + struct device_attribute *attr, char *buf) +{ + acpi_handle handle; + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object params[4]; + union acpi_object *ret_obj; + + input.count = 4; + ppi_assign_params(params, TPM_PPI_FN_GETREQ); + input.pointer = params; + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT, + ACPI_UINT32_MAX, ppi_callback, NULL, + tpm_device_name, &handle); + if (ACPI_FAILURE(status)) + return -ENXIO; + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_PACKAGE); + if (ACPI_FAILURE(status)) + return -ENOMEM; + /* + * output.pointer should be of package type, including two integers. + * The first is function return code, 0 means success and 1 means + * error. The second is pending TPM operation requested by the OS, 0 + * means none and >0 means operation value. + */ + ret_obj = ((union acpi_object *)output.pointer)->package.elements; + if (ret_obj->type == ACPI_TYPE_INTEGER) { + if (ret_obj->integer.value) { + status = -EFAULT; + goto cleanup; + } + ret_obj++; + if (ret_obj->type == ACPI_TYPE_INTEGER) + status = scnprintf(buf, PAGE_SIZE, "%llu\n", + ret_obj->integer.value); + else + status = -EINVAL; + } else { + status = -EINVAL; + } +cleanup: + kfree(output.pointer); + return status; +} + +static ssize_t tpm_store_ppi_request(struct device *dev, + struct device_attribute *attr, + const char *buf, size_t count) +{ + char version[PPI_VERSION_LEN + 1]; + acpi_handle handle; + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object params[4]; + union acpi_object obj; + u32 req; + u64 ret; + + input.count = 4; + ppi_assign_params(params, TPM_PPI_FN_VERSION); + input.pointer = params; + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT, + ACPI_UINT32_MAX, ppi_callback, NULL, + tpm_device_name, &handle); + if (ACPI_FAILURE(status)) + return -ENXIO; + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_STRING); + if (ACPI_FAILURE(status)) + return -ENOMEM; + strncpy(version, + ((union acpi_object *)output.pointer)->string.pointer, + PPI_VERSION_LEN); + kfree(output.pointer); + output.length = ACPI_ALLOCATE_BUFFER; + output.pointer = NULL; + /* + * the function to submit TPM operation request to pre-os environment + * is updated with function index from SUBREQ to SUBREQ2 since PPI + * version 1.1 + */ + if (strcmp(version, "1.1") == -1) + params[2].integer.value = TPM_PPI_FN_SUBREQ; + else + params[2].integer.value = TPM_PPI_FN_SUBREQ2; + /* + * PPI spec defines params[3].type as ACPI_TYPE_PACKAGE. Some BIOS + * accept buffer/string/integer type, but some BIOS accept buffer/ + * string/package type. For PPI version 1.0 and 1.1, use buffer type + * for compatibility, and use package type since 1.2 according to spec. + */ + if (strcmp(version, "1.2") == -1) { + params[3].type = ACPI_TYPE_BUFFER; + params[3].buffer.length = sizeof(req); + sscanf(buf, "%d", &req); + params[3].buffer.pointer = (char *)&req; + } else { + params[3].package.count = 1; + obj.type = ACPI_TYPE_INTEGER; + sscanf(buf, "%llu", &obj.integer.value); + params[3].package.elements = &obj; + } + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_INTEGER); + if (ACPI_FAILURE(status)) + return -ENOMEM; + ret = ((union acpi_object *)output.pointer)->integer.value; + if (ret == 0) + status = (acpi_status)count; + else if (ret == 1) + status = -EPERM; + else + status = -EFAULT; + kfree(output.pointer); + return status; +} + +static ssize_t tpm_show_ppi_transition_action(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + char version[PPI_VERSION_LEN + 1]; + acpi_handle handle; + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object params[4]; + u32 ret; + char *info[] = { + "None", + "Shutdown", + "Reboot", + "OS Vendor-specific", + "Error", + }; + input.count = 4; + ppi_assign_params(params, TPM_PPI_FN_VERSION); + input.pointer = params; + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT, + ACPI_UINT32_MAX, ppi_callback, NULL, + tpm_device_name, &handle); + if (ACPI_FAILURE(status)) + return -ENXIO; + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_STRING); + if (ACPI_FAILURE(status)) + return -ENOMEM; + strncpy(version, + ((union acpi_object *)output.pointer)->string.pointer, + PPI_VERSION_LEN); + /* + * PPI spec defines params[3].type as empty package, but some platforms + * (e.g. Capella with PPI 1.0) need integer/string/buffer type, so for + * compatibility, define params[3].type as buffer, if PPI version < 1.2 + */ + if (strcmp(version, "1.2") == -1) { + params[3].type = ACPI_TYPE_BUFFER; + params[3].buffer.length = 0; + params[3].buffer.pointer = NULL; + } + params[2].integer.value = TPM_PPI_FN_GETACT; + kfree(output.pointer); + output.length = ACPI_ALLOCATE_BUFFER; + output.pointer = NULL; + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_INTEGER); + if (ACPI_FAILURE(status)) + return -ENOMEM; + ret = ((union acpi_object *)output.pointer)->integer.value; + if (ret < ARRAY_SIZE(info) - 1) + status = scnprintf(buf, PAGE_SIZE, "%d: %s\n", ret, info[ret]); + else + status = scnprintf(buf, PAGE_SIZE, "%d: %s\n", ret, + info[ARRAY_SIZE(info)-1]); + kfree(output.pointer); + return status; +} + +static ssize_t tpm_show_ppi_response(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + acpi_handle handle; + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object params[4]; + union acpi_object *ret_obj; + u64 req; + + input.count = 4; + ppi_assign_params(params, TPM_PPI_FN_GETRSP); + input.pointer = params; + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT, + ACPI_UINT32_MAX, ppi_callback, NULL, + tpm_device_name, &handle); + if (ACPI_FAILURE(status)) + return -ENXIO; + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_PACKAGE); + if (ACPI_FAILURE(status)) + return -ENOMEM; + /* + * parameter output.pointer should be of package type, including + * 3 integers. The first means function return code, the second means + * most recent TPM operation request, and the last means response to + * the most recent TPM operation request. Only if the first is 0, and + * the second integer is not 0, the response makes sense. + */ + ret_obj = ((union acpi_object *)output.pointer)->package.elements; + if (ret_obj->type != ACPI_TYPE_INTEGER) { + status = -EINVAL; + goto cleanup; + } + if (ret_obj->integer.value) { + status = -EFAULT; + goto cleanup; + } + ret_obj++; + if (ret_obj->type != ACPI_TYPE_INTEGER) { + status = -EINVAL; + goto cleanup; + } + if (ret_obj->integer.value) { + req = ret_obj->integer.value; + ret_obj++; + if (ret_obj->type != ACPI_TYPE_INTEGER) { + status = -EINVAL; + goto cleanup; + } + if (ret_obj->integer.value == 0) + status = scnprintf(buf, PAGE_SIZE, "%llu %s\n", req, + "0: Success"); + else if (ret_obj->integer.value == 0xFFFFFFF0) + status = scnprintf(buf, PAGE_SIZE, "%llu %s\n", req, + "0xFFFFFFF0: User Abort"); + else if (ret_obj->integer.value == 0xFFFFFFF1) + status = scnprintf(buf, PAGE_SIZE, "%llu %s\n", req, + "0xFFFFFFF1: BIOS Failure"); + else if (ret_obj->integer.value >= 1 && + ret_obj->integer.value <= 0x00000FFF) + status = scnprintf(buf, PAGE_SIZE, "%llu %llu: %s\n", + req, ret_obj->integer.value, + "Corresponding TPM error"); + else + status = scnprintf(buf, PAGE_SIZE, "%llu %llu: %s\n", + req, ret_obj->integer.value, + "Error"); + } else { + status = scnprintf(buf, PAGE_SIZE, "%llu: %s\n", + ret_obj->integer.value, "No Recent Request"); + } +cleanup: + kfree(output.pointer); + return status; +} + +static ssize_t show_ppi_operations(char *buf, u32 start, u32 end) +{ + char *str = buf; + char version[PPI_VERSION_LEN]; + acpi_handle handle; + acpi_status status; + struct acpi_object_list input; + struct acpi_buffer output = { ACPI_ALLOCATE_BUFFER, NULL }; + union acpi_object params[4]; + union acpi_object obj; + int i; + u32 ret; + char *info[] = { + "Not implemented", + "BIOS only", + "Blocked for OS by BIOS", + "User required", + "User not required", + }; + input.count = 4; + ppi_assign_params(params, TPM_PPI_FN_VERSION); + input.pointer = params; + status = acpi_walk_namespace(ACPI_TYPE_DEVICE, ACPI_ROOT_OBJECT, + ACPI_UINT32_MAX, ppi_callback, NULL, + tpm_device_name, &handle); + if (ACPI_FAILURE(status)) + return -ENXIO; + + status = acpi_evaluate_object_typed(handle, "_DSM", &input, &output, + ACPI_TYPE_STRING); + if (ACPI_FAILURE(status)) + return -ENOMEM; + + strncpy(version, + ((union acpi_object *)output.pointer)->string.pointer, + PPI_VERSION_LEN); + kfree(output.pointer); + output.length = ACPI_ALLOCATE_BUFFER; + output.pointer = NULL; + if (strcmp(version, "1.2") == -1) + return -EPERM; + + params[2].integer.value = TPM_PPI_FN_GETOPR; + params[3].package.count = 1; + obj.type = ACPI_TYPE_INTEGER; + params[3].package.elements = &obj; + for (i = start; i <= end; i++) { + obj.integer.value = i; + status = acpi_evaluate_object_typed(handle, "_DSM", + &input, &output, ACPI_TYPE_INTEGER); + if (ACPI_FAILURE(status)) + return -ENOMEM; + + ret = ((union acpi_object *)output.pointer)->integer.value; + if (ret > 0 && ret < ARRAY_SIZE(info)) + str += scnprintf(str, PAGE_SIZE, "%d %d: %s\n", + i, ret, info[ret]); + kfree(output.pointer); + output.length = ACPI_ALLOCATE_BUFFER; + output.pointer = NULL; + } + return str - buf; +} + +static ssize_t tpm_show_ppi_tcg_operations(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return show_ppi_operations(buf, 0, PPI_TPM_REQ_MAX); +} + +static ssize_t tpm_show_ppi_vs_operations(struct device *dev, + struct device_attribute *attr, + char *buf) +{ + return show_ppi_operations(buf, PPI_VS_REQ_START, PPI_VS_REQ_END); +} + +static DEVICE_ATTR(version, S_IRUGO, tpm_show_ppi_version, NULL); +static DEVICE_ATTR(request, S_IRUGO | S_IWUSR | S_IWGRP, + tpm_show_ppi_request, tpm_store_ppi_request); +static DEVICE_ATTR(transition_action, S_IRUGO, + tpm_show_ppi_transition_action, NULL); +static DEVICE_ATTR(response, S_IRUGO, tpm_show_ppi_response, NULL); +static DEVICE_ATTR(tcg_operations, S_IRUGO, tpm_show_ppi_tcg_operations, NULL); +static DEVICE_ATTR(vs_operations, S_IRUGO, tpm_show_ppi_vs_operations, NULL); + +static struct attribute *ppi_attrs[] = { + &dev_attr_version.attr, + &dev_attr_request.attr, + &dev_attr_transition_action.attr, + &dev_attr_response.attr, + &dev_attr_tcg_operations.attr, + &dev_attr_vs_operations.attr, NULL, +}; +static struct attribute_group ppi_attr_grp = { + .attrs = ppi_attrs +}; + +ssize_t sys_add_ppi(struct kobject *parent) +{ + struct kobject *ppi; + ppi = kobject_create_and_add("ppi", parent); + if (sysfs_create_group(ppi, &ppi_attr_grp)) + return -EFAULT; + else + return 0; +} +EXPORT_SYMBOL_GPL(sys_add_ppi); + +MODULE_LICENSE("GPL"); diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c index c4be351..6bdf267 100644 --- a/drivers/char/tpm/tpm_tis.c +++ b/drivers/char/tpm/tpm_tis.c @@ -705,6 +705,7 @@ out_err: return rc; } +#if defined(CONFIG_PNP) || defined(CONFIG_PM_SLEEP) static void tpm_tis_reenable_interrupts(struct tpm_chip *chip) { u32 intmask; @@ -725,7 +726,7 @@ static void tpm_tis_reenable_interrupts(struct tpm_chip *chip) iowrite32(intmask, chip->vendor.iobase + TPM_INT_ENABLE(chip->vendor.locality)); } - +#endif #ifdef CONFIG_PNP static int __devinit tpm_tis_pnp_init(struct pnp_dev *pnp_dev, diff --git a/drivers/clocksource/sh_cmt.c b/drivers/clocksource/sh_cmt.c index 98b06ba..a5f7829 100644 --- a/drivers/clocksource/sh_cmt.c +++ b/drivers/clocksource/sh_cmt.c @@ -33,6 +33,7 @@ #include <linux/slab.h> #include <linux/module.h> #include <linux/pm_domain.h> +#include <linux/pm_runtime.h> struct sh_cmt_priv { void __iomem *mapbase; @@ -52,6 +53,7 @@ struct sh_cmt_priv { struct clock_event_device ced; struct clocksource cs; unsigned long total_cycles; + bool cs_enabled; }; static DEFINE_RAW_SPINLOCK(sh_cmt_lock); @@ -155,6 +157,9 @@ static int sh_cmt_enable(struct sh_cmt_priv *p, unsigned long *rate) { int k, ret; + pm_runtime_get_sync(&p->pdev->dev); + dev_pm_syscore_device(&p->pdev->dev, true); + /* enable clock */ ret = clk_enable(p->clk); if (ret) { @@ -221,6 +226,9 @@ static void sh_cmt_disable(struct sh_cmt_priv *p) /* stop clock */ clk_disable(p->clk); + + dev_pm_syscore_device(&p->pdev->dev, false); + pm_runtime_put(&p->pdev->dev); } /* private flags */ @@ -451,22 +459,42 @@ static int sh_cmt_clocksource_enable(struct clocksource *cs) int ret; struct sh_cmt_priv *p = cs_to_sh_cmt(cs); + WARN_ON(p->cs_enabled); + p->total_cycles = 0; ret = sh_cmt_start(p, FLAG_CLOCKSOURCE); - if (!ret) + if (!ret) { __clocksource_updatefreq_hz(cs, p->rate); + p->cs_enabled = true; + } return ret; } static void sh_cmt_clocksource_disable(struct clocksource *cs) { - sh_cmt_stop(cs_to_sh_cmt(cs), FLAG_CLOCKSOURCE); + struct sh_cmt_priv *p = cs_to_sh_cmt(cs); + + WARN_ON(!p->cs_enabled); + + sh_cmt_stop(p, FLAG_CLOCKSOURCE); + p->cs_enabled = false; +} + +static void sh_cmt_clocksource_suspend(struct clocksource *cs) +{ + struct sh_cmt_priv *p = cs_to_sh_cmt(cs); + + sh_cmt_stop(p, FLAG_CLOCKSOURCE); + pm_genpd_syscore_poweroff(&p->pdev->dev); } static void sh_cmt_clocksource_resume(struct clocksource *cs) { - sh_cmt_start(cs_to_sh_cmt(cs), FLAG_CLOCKSOURCE); + struct sh_cmt_priv *p = cs_to_sh_cmt(cs); + + pm_genpd_syscore_poweron(&p->pdev->dev); + sh_cmt_start(p, FLAG_CLOCKSOURCE); } static int sh_cmt_register_clocksource(struct sh_cmt_priv *p, @@ -479,7 +507,7 @@ static int sh_cmt_register_clocksource(struct sh_cmt_priv *p, cs->read = sh_cmt_clocksource_read; cs->enable = sh_cmt_clocksource_enable; cs->disable = sh_cmt_clocksource_disable; - cs->suspend = sh_cmt_clocksource_disable; + cs->suspend = sh_cmt_clocksource_suspend; cs->resume = sh_cmt_clocksource_resume; cs->mask = CLOCKSOURCE_MASK(sizeof(unsigned long) * 8); cs->flags = CLOCK_SOURCE_IS_CONTINUOUS; @@ -562,6 +590,16 @@ static int sh_cmt_clock_event_next(unsigned long delta, return 0; } +static void sh_cmt_clock_event_suspend(struct clock_event_device *ced) +{ + pm_genpd_syscore_poweroff(&ced_to_sh_cmt(ced)->pdev->dev); +} + +static void sh_cmt_clock_event_resume(struct clock_event_device *ced) +{ + pm_genpd_syscore_poweron(&ced_to_sh_cmt(ced)->pdev->dev); +} + static void sh_cmt_register_clockevent(struct sh_cmt_priv *p, char *name, unsigned long rating) { @@ -576,6 +614,8 @@ static void sh_cmt_register_clockevent(struct sh_cmt_priv *p, ced->cpumask = cpumask_of(0); ced->set_next_event = sh_cmt_clock_event_next; ced->set_mode = sh_cmt_clock_event_mode; + ced->suspend = sh_cmt_clock_event_suspend; + ced->resume = sh_cmt_clock_event_resume; dev_info(&p->pdev->dev, "used for clock events\n"); clockevents_register_device(ced); @@ -670,6 +710,7 @@ static int sh_cmt_setup(struct sh_cmt_priv *p, struct platform_device *pdev) dev_err(&p->pdev->dev, "registration failed\n"); goto err1; } + p->cs_enabled = false; ret = setup_irq(irq, &p->irqaction); if (ret) { @@ -688,14 +729,17 @@ err0: static int __devinit sh_cmt_probe(struct platform_device *pdev) { struct sh_cmt_priv *p = platform_get_drvdata(pdev); + struct sh_timer_config *cfg = pdev->dev.platform_data; int ret; - if (!is_early_platform_device(pdev)) - pm_genpd_dev_always_on(&pdev->dev, true); + if (!is_early_platform_device(pdev)) { + pm_runtime_set_active(&pdev->dev); + pm_runtime_enable(&pdev->dev); + } if (p) { dev_info(&pdev->dev, "kept as earlytimer\n"); - return 0; + goto out; } p = kmalloc(sizeof(*p), GFP_KERNEL); @@ -708,8 +752,19 @@ static int __devinit sh_cmt_probe(struct platform_device *pdev) if (ret) { kfree(p); platform_set_drvdata(pdev, NULL); + pm_runtime_idle(&pdev->dev); + return ret; } - return ret; + if (is_early_platform_device(pdev)) + return 0; + + out: + if (cfg->clockevent_rating || cfg->clocksource_rating) + pm_runtime_irq_safe(&pdev->dev); + else + pm_runtime_idle(&pdev->dev); + + return 0; } static int __devexit sh_cmt_remove(struct platform_device *pdev) diff --git a/drivers/clocksource/sh_mtu2.c b/drivers/clocksource/sh_mtu2.c index d9b76ca..c5eea85 100644 --- a/drivers/clocksource/sh_mtu2.c +++ b/drivers/clocksource/sh_mtu2.c @@ -32,6 +32,7 @@ #include <linux/slab.h> #include <linux/module.h> #include <linux/pm_domain.h> +#include <linux/pm_runtime.h> struct sh_mtu2_priv { void __iomem *mapbase; @@ -123,6 +124,9 @@ static int sh_mtu2_enable(struct sh_mtu2_priv *p) { int ret; + pm_runtime_get_sync(&p->pdev->dev); + dev_pm_syscore_device(&p->pdev->dev, true); + /* enable clock */ ret = clk_enable(p->clk); if (ret) { @@ -157,6 +161,9 @@ static void sh_mtu2_disable(struct sh_mtu2_priv *p) /* stop clock */ clk_disable(p->clk); + + dev_pm_syscore_device(&p->pdev->dev, false); + pm_runtime_put(&p->pdev->dev); } static irqreturn_t sh_mtu2_interrupt(int irq, void *dev_id) @@ -208,6 +215,16 @@ static void sh_mtu2_clock_event_mode(enum clock_event_mode mode, } } +static void sh_mtu2_clock_event_suspend(struct clock_event_device *ced) +{ + pm_genpd_syscore_poweroff(&ced_to_sh_mtu2(ced)->pdev->dev); +} + +static void sh_mtu2_clock_event_resume(struct clock_event_device *ced) +{ + pm_genpd_syscore_poweron(&ced_to_sh_mtu2(ced)->pdev->dev); +} + static void sh_mtu2_register_clockevent(struct sh_mtu2_priv *p, char *name, unsigned long rating) { @@ -221,6 +238,8 @@ static void sh_mtu2_register_clockevent(struct sh_mtu2_priv *p, ced->rating = rating; ced->cpumask = cpumask_of(0); ced->set_mode = sh_mtu2_clock_event_mode; + ced->suspend = sh_mtu2_clock_event_suspend; + ced->resume = sh_mtu2_clock_event_resume; dev_info(&p->pdev->dev, "used for clock events\n"); clockevents_register_device(ced); @@ -305,14 +324,17 @@ static int sh_mtu2_setup(struct sh_mtu2_priv *p, struct platform_device *pdev) static int __devinit sh_mtu2_probe(struct platform_device *pdev) { struct sh_mtu2_priv *p = platform_get_drvdata(pdev); + struct sh_timer_config *cfg = pdev->dev.platform_data; int ret; - if (!is_early_platform_device(pdev)) - pm_genpd_dev_always_on(&pdev->dev, true); + if (!is_early_platform_device(pdev)) { + pm_runtime_set_active(&pdev->dev); + pm_runtime_enable(&pdev->dev); + } if (p) { dev_info(&pdev->dev, "kept as earlytimer\n"); - return 0; + goto out; } p = kmalloc(sizeof(*p), GFP_KERNEL); @@ -325,8 +347,19 @@ static int __devinit sh_mtu2_probe(struct platform_device *pdev) if (ret) { kfree(p); platform_set_drvdata(pdev, NULL); + pm_runtime_idle(&pdev->dev); + return ret; } - return ret; + if (is_early_platform_device(pdev)) + return 0; + + out: + if (cfg->clockevent_rating) + pm_runtime_irq_safe(&pdev->dev); + else + pm_runtime_idle(&pdev->dev); + + return 0; } static int __devexit sh_mtu2_remove(struct platform_device *pdev) diff --git a/drivers/clocksource/sh_tmu.c b/drivers/clocksource/sh_tmu.c index c1b51d4..0cc4add 100644 --- a/drivers/clocksource/sh_tmu.c +++ b/drivers/clocksource/sh_tmu.c @@ -33,6 +33,7 @@ #include <linux/slab.h> #include <linux/module.h> #include <linux/pm_domain.h> +#include <linux/pm_runtime.h> struct sh_tmu_priv { void __iomem *mapbase; @@ -43,6 +44,8 @@ struct sh_tmu_priv { unsigned long periodic; struct clock_event_device ced; struct clocksource cs; + bool cs_enabled; + unsigned int enable_count; }; static DEFINE_RAW_SPINLOCK(sh_tmu_lock); @@ -107,7 +110,7 @@ static void sh_tmu_start_stop_ch(struct sh_tmu_priv *p, int start) raw_spin_unlock_irqrestore(&sh_tmu_lock, flags); } -static int sh_tmu_enable(struct sh_tmu_priv *p) +static int __sh_tmu_enable(struct sh_tmu_priv *p) { int ret; @@ -135,7 +138,18 @@ static int sh_tmu_enable(struct sh_tmu_priv *p) return 0; } -static void sh_tmu_disable(struct sh_tmu_priv *p) +static int sh_tmu_enable(struct sh_tmu_priv *p) +{ + if (p->enable_count++ > 0) + return 0; + + pm_runtime_get_sync(&p->pdev->dev); + dev_pm_syscore_device(&p->pdev->dev, true); + + return __sh_tmu_enable(p); +} + +static void __sh_tmu_disable(struct sh_tmu_priv *p) { /* disable channel */ sh_tmu_start_stop_ch(p, 0); @@ -147,6 +161,20 @@ static void sh_tmu_disable(struct sh_tmu_priv *p) clk_disable(p->clk); } +static void sh_tmu_disable(struct sh_tmu_priv *p) +{ + if (WARN_ON(p->enable_count == 0)) + return; + + if (--p->enable_count > 0) + return; + + __sh_tmu_disable(p); + + dev_pm_syscore_device(&p->pdev->dev, false); + pm_runtime_put(&p->pdev->dev); +} + static void sh_tmu_set_next(struct sh_tmu_priv *p, unsigned long delta, int periodic) { @@ -203,15 +231,53 @@ static int sh_tmu_clocksource_enable(struct clocksource *cs) struct sh_tmu_priv *p = cs_to_sh_tmu(cs); int ret; + if (WARN_ON(p->cs_enabled)) + return 0; + ret = sh_tmu_enable(p); - if (!ret) + if (!ret) { __clocksource_updatefreq_hz(cs, p->rate); + p->cs_enabled = true; + } + return ret; } static void sh_tmu_clocksource_disable(struct clocksource *cs) { - sh_tmu_disable(cs_to_sh_tmu(cs)); + struct sh_tmu_priv *p = cs_to_sh_tmu(cs); + + if (WARN_ON(!p->cs_enabled)) + return; + + sh_tmu_disable(p); + p->cs_enabled = false; +} + +static void sh_tmu_clocksource_suspend(struct clocksource *cs) +{ + struct sh_tmu_priv *p = cs_to_sh_tmu(cs); + + if (!p->cs_enabled) + return; + + if (--p->enable_count == 0) { + __sh_tmu_disable(p); + pm_genpd_syscore_poweroff(&p->pdev->dev); + } +} + +static void sh_tmu_clocksource_resume(struct clocksource *cs) +{ + struct sh_tmu_priv *p = cs_to_sh_tmu(cs); + + if (!p->cs_enabled) + return; + + if (p->enable_count++ == 0) { + pm_genpd_syscore_poweron(&p->pdev->dev); + __sh_tmu_enable(p); + } } static int sh_tmu_register_clocksource(struct sh_tmu_priv *p, @@ -224,6 +290,8 @@ static int sh_tmu_register_clocksource(struct sh_tmu_priv *p, cs->read = sh_tmu_clocksource_read; cs->enable = sh_tmu_clocksource_enable; cs->disable = sh_tmu_clocksource_disable; + cs->suspend = sh_tmu_clocksource_suspend; + cs->resume = sh_tmu_clocksource_resume; cs->mask = CLOCKSOURCE_MASK(32); cs->flags = CLOCK_SOURCE_IS_CONTINUOUS; @@ -301,6 +369,16 @@ static int sh_tmu_clock_event_next(unsigned long delta, return 0; } +static void sh_tmu_clock_event_suspend(struct clock_event_device *ced) +{ + pm_genpd_syscore_poweroff(&ced_to_sh_tmu(ced)->pdev->dev); +} + +static void sh_tmu_clock_event_resume(struct clock_event_device *ced) +{ + pm_genpd_syscore_poweron(&ced_to_sh_tmu(ced)->pdev->dev); +} + static void sh_tmu_register_clockevent(struct sh_tmu_priv *p, char *name, unsigned long rating) { @@ -316,6 +394,8 @@ static void sh_tmu_register_clockevent(struct sh_tmu_priv *p, ced->cpumask = cpumask_of(0); ced->set_next_event = sh_tmu_clock_event_next; ced->set_mode = sh_tmu_clock_event_mode; + ced->suspend = sh_tmu_clock_event_suspend; + ced->resume = sh_tmu_clock_event_resume; dev_info(&p->pdev->dev, "used for clock events\n"); @@ -392,6 +472,8 @@ static int sh_tmu_setup(struct sh_tmu_priv *p, struct platform_device *pdev) ret = PTR_ERR(p->clk); goto err1; } + p->cs_enabled = false; + p->enable_count = 0; return sh_tmu_register(p, (char *)dev_name(&p->pdev->dev), cfg->clockevent_rating, @@ -405,14 +487,17 @@ static int sh_tmu_setup(struct sh_tmu_priv *p, struct platform_device *pdev) static int __devinit sh_tmu_probe(struct platform_device *pdev) { struct sh_tmu_priv *p = platform_get_drvdata(pdev); + struct sh_timer_config *cfg = pdev->dev.platform_data; int ret; - if (!is_early_platform_device(pdev)) - pm_genpd_dev_always_on(&pdev->dev, true); + if (!is_early_platform_device(pdev)) { + pm_runtime_set_active(&pdev->dev); + pm_runtime_enable(&pdev->dev); + } if (p) { dev_info(&pdev->dev, "kept as earlytimer\n"); - return 0; + goto out; } p = kmalloc(sizeof(*p), GFP_KERNEL); @@ -425,8 +510,19 @@ static int __devinit sh_tmu_probe(struct platform_device *pdev) if (ret) { kfree(p); platform_set_drvdata(pdev, NULL); + pm_runtime_idle(&pdev->dev); + return ret; } - return ret; + if (is_early_platform_device(pdev)) + return 0; + + out: + if (cfg->clockevent_rating || cfg->clocksource_rating) + pm_runtime_irq_safe(&pdev->dev); + else + pm_runtime_idle(&pdev->dev); + + return 0; } static int __devexit sh_tmu_remove(struct platform_device *pdev) diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index e24a2a1..ea512f4 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -179,6 +179,17 @@ config CPU_FREQ_GOV_CONSERVATIVE If in doubt, say N. +config GENERIC_CPUFREQ_CPU0 + bool "Generic CPU0 cpufreq driver" + depends on HAVE_CLK && REGULATOR && PM_OPP && OF + select CPU_FREQ_TABLE + help + This adds a generic cpufreq driver for CPU0 frequency management. + It supports both uniprocessor (UP) and symmetric multiprocessor (SMP) + systems which share clock and voltage across all CPUs. + + If in doubt, say N. + menu "x86 CPU frequency scaling drivers" depends on X86 source "drivers/cpufreq/Kconfig.x86" diff --git a/drivers/cpufreq/Kconfig.x86 b/drivers/cpufreq/Kconfig.x86 index 78ff7ee..934854a 100644 --- a/drivers/cpufreq/Kconfig.x86 +++ b/drivers/cpufreq/Kconfig.x86 @@ -23,7 +23,8 @@ config X86_ACPI_CPUFREQ help This driver adds a CPUFreq driver which utilizes the ACPI Processor Performance States. - This driver also supports Intel Enhanced Speedstep. + This driver also supports Intel Enhanced Speedstep and newer + AMD CPUs. To compile this driver as a module, choose M here: the module will be called acpi-cpufreq. @@ -32,6 +33,18 @@ config X86_ACPI_CPUFREQ If in doubt, say N. +config X86_ACPI_CPUFREQ_CPB + default y + bool "Legacy cpb sysfs knob support for AMD CPUs" + depends on X86_ACPI_CPUFREQ && CPU_SUP_AMD + help + The powernow-k8 driver used to provide a sysfs knob called "cpb" + to disable the Core Performance Boosting feature of AMD CPUs. This + file has now been superseeded by the more generic "boost" entry. + + By enabling this option the acpi_cpufreq driver provides the old + entry in addition to the new boost ones, for compatibility reasons. + config ELAN_CPUFREQ tristate "AMD Elan SC400 and SC410" select CPU_FREQ_TABLE @@ -95,7 +108,8 @@ config X86_POWERNOW_K8 select CPU_FREQ_TABLE depends on ACPI && ACPI_PROCESSOR help - This adds the CPUFreq driver for K8/K10 Opteron/Athlon64 processors. + This adds the CPUFreq driver for K8/early Opteron/Athlon64 processors. + Support for K10 and newer processors is now in acpi-cpufreq. To compile this driver as a module, choose M here: the module will be called powernow-k8. diff --git a/drivers/cpufreq/Makefile b/drivers/cpufreq/Makefile index 9531fc2..1bc90e1 100644 --- a/drivers/cpufreq/Makefile +++ b/drivers/cpufreq/Makefile @@ -13,13 +13,15 @@ obj-$(CONFIG_CPU_FREQ_GOV_CONSERVATIVE) += cpufreq_conservative.o # CPUfreq cross-arch helpers obj-$(CONFIG_CPU_FREQ_TABLE) += freq_table.o +obj-$(CONFIG_GENERIC_CPUFREQ_CPU0) += cpufreq-cpu0.o + ################################################################################## # x86 drivers. # Link order matters. K8 is preferred to ACPI because of firmware bugs in early # K8 systems. ACPI is preferred to all other hardware-specific drivers. # speedstep-* is preferred over p4-clockmod. -obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o mperf.o +obj-$(CONFIG_X86_POWERNOW_K8) += powernow-k8.o obj-$(CONFIG_X86_ACPI_CPUFREQ) += acpi-cpufreq.o mperf.o obj-$(CONFIG_X86_PCC_CPUFREQ) += pcc-cpufreq.o obj-$(CONFIG_X86_POWERNOW_K6) += powernow-k6.o diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index 56c6c6b..0d048f6 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -51,13 +51,19 @@ MODULE_AUTHOR("Paul Diefenbaugh, Dominik Brodowski"); MODULE_DESCRIPTION("ACPI Processor P-States Driver"); MODULE_LICENSE("GPL"); +#define PFX "acpi-cpufreq: " + enum { UNDEFINED_CAPABLE = 0, SYSTEM_INTEL_MSR_CAPABLE, + SYSTEM_AMD_MSR_CAPABLE, SYSTEM_IO_CAPABLE, }; #define INTEL_MSR_RANGE (0xffff) +#define AMD_MSR_RANGE (0x7) + +#define MSR_K7_HWCR_CPB_DIS (1ULL << 25) struct acpi_cpufreq_data { struct acpi_processor_performance *acpi_data; @@ -74,6 +80,116 @@ static struct acpi_processor_performance __percpu *acpi_perf_data; static struct cpufreq_driver acpi_cpufreq_driver; static unsigned int acpi_pstate_strict; +static bool boost_enabled, boost_supported; +static struct msr __percpu *msrs; + +static bool boost_state(unsigned int cpu) +{ + u32 lo, hi; + u64 msr; + + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + rdmsr_on_cpu(cpu, MSR_IA32_MISC_ENABLE, &lo, &hi); + msr = lo | ((u64)hi << 32); + return !(msr & MSR_IA32_MISC_ENABLE_TURBO_DISABLE); + case X86_VENDOR_AMD: + rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); + msr = lo | ((u64)hi << 32); + return !(msr & MSR_K7_HWCR_CPB_DIS); + } + return false; +} + +static void boost_set_msrs(bool enable, const struct cpumask *cpumask) +{ + u32 cpu; + u32 msr_addr; + u64 msr_mask; + + switch (boot_cpu_data.x86_vendor) { + case X86_VENDOR_INTEL: + msr_addr = MSR_IA32_MISC_ENABLE; + msr_mask = MSR_IA32_MISC_ENABLE_TURBO_DISABLE; + break; + case X86_VENDOR_AMD: + msr_addr = MSR_K7_HWCR; + msr_mask = MSR_K7_HWCR_CPB_DIS; + break; + default: + return; + } + + rdmsr_on_cpus(cpumask, msr_addr, msrs); + + for_each_cpu(cpu, cpumask) { + struct msr *reg = per_cpu_ptr(msrs, cpu); + if (enable) + reg->q &= ~msr_mask; + else + reg->q |= msr_mask; + } + + wrmsr_on_cpus(cpumask, msr_addr, msrs); +} + +static ssize_t _store_boost(const char *buf, size_t count) +{ + int ret; + unsigned long val = 0; + + if (!boost_supported) + return -EINVAL; + + ret = kstrtoul(buf, 10, &val); + if (ret || (val > 1)) + return -EINVAL; + + if ((val && boost_enabled) || (!val && !boost_enabled)) + return count; + + get_online_cpus(); + + boost_set_msrs(val, cpu_online_mask); + + put_online_cpus(); + + boost_enabled = val; + pr_debug("Core Boosting %sabled.\n", val ? "en" : "dis"); + + return count; +} + +static ssize_t store_global_boost(struct kobject *kobj, struct attribute *attr, + const char *buf, size_t count) +{ + return _store_boost(buf, count); +} + +static ssize_t show_global_boost(struct kobject *kobj, + struct attribute *attr, char *buf) +{ + return sprintf(buf, "%u\n", boost_enabled); +} + +static struct global_attr global_boost = __ATTR(boost, 0644, + show_global_boost, + store_global_boost); + +#ifdef CONFIG_X86_ACPI_CPUFREQ_CPB +static ssize_t store_cpb(struct cpufreq_policy *policy, const char *buf, + size_t count) +{ + return _store_boost(buf, count); +} + +static ssize_t show_cpb(struct cpufreq_policy *policy, char *buf) +{ + return sprintf(buf, "%u\n", boost_enabled); +} + +static struct freq_attr cpb = __ATTR(cpb, 0644, show_cpb, store_cpb); +#endif static int check_est_cpu(unsigned int cpuid) { @@ -82,6 +198,13 @@ static int check_est_cpu(unsigned int cpuid) return cpu_has(cpu, X86_FEATURE_EST); } +static int check_amd_hwpstate_cpu(unsigned int cpuid) +{ + struct cpuinfo_x86 *cpu = &cpu_data(cpuid); + + return cpu_has(cpu, X86_FEATURE_HW_PSTATE); +} + static unsigned extract_io(u32 value, struct acpi_cpufreq_data *data) { struct acpi_processor_performance *perf; @@ -101,7 +224,11 @@ static unsigned extract_msr(u32 msr, struct acpi_cpufreq_data *data) int i; struct acpi_processor_performance *perf; - msr &= INTEL_MSR_RANGE; + if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) + msr &= AMD_MSR_RANGE; + else + msr &= INTEL_MSR_RANGE; + perf = data->acpi_data; for (i = 0; data->freq_table[i].frequency != CPUFREQ_TABLE_END; i++) { @@ -115,6 +242,7 @@ static unsigned extract_freq(u32 val, struct acpi_cpufreq_data *data) { switch (data->cpu_feature) { case SYSTEM_INTEL_MSR_CAPABLE: + case SYSTEM_AMD_MSR_CAPABLE: return extract_msr(val, data); case SYSTEM_IO_CAPABLE: return extract_io(val, data); @@ -150,6 +278,7 @@ static void do_drv_read(void *_cmd) switch (cmd->type) { case SYSTEM_INTEL_MSR_CAPABLE: + case SYSTEM_AMD_MSR_CAPABLE: rdmsr(cmd->addr.msr.reg, cmd->val, h); break; case SYSTEM_IO_CAPABLE: @@ -174,6 +303,9 @@ static void do_drv_write(void *_cmd) lo = (lo & ~INTEL_MSR_RANGE) | (cmd->val & INTEL_MSR_RANGE); wrmsr(cmd->addr.msr.reg, lo, hi); break; + case SYSTEM_AMD_MSR_CAPABLE: + wrmsr(cmd->addr.msr.reg, cmd->val, 0); + break; case SYSTEM_IO_CAPABLE: acpi_os_write_port((acpi_io_address)cmd->addr.io.port, cmd->val, @@ -217,6 +349,10 @@ static u32 get_cur_val(const struct cpumask *mask) cmd.type = SYSTEM_INTEL_MSR_CAPABLE; cmd.addr.msr.reg = MSR_IA32_PERF_STATUS; break; + case SYSTEM_AMD_MSR_CAPABLE: + cmd.type = SYSTEM_AMD_MSR_CAPABLE; + cmd.addr.msr.reg = MSR_AMD_PERF_STATUS; + break; case SYSTEM_IO_CAPABLE: cmd.type = SYSTEM_IO_CAPABLE; perf = per_cpu(acfreq_data, cpumask_first(mask))->acpi_data; @@ -326,6 +462,11 @@ static int acpi_cpufreq_target(struct cpufreq_policy *policy, cmd.addr.msr.reg = MSR_IA32_PERF_CTL; cmd.val = (u32) perf->states[next_perf_state].control; break; + case SYSTEM_AMD_MSR_CAPABLE: + cmd.type = SYSTEM_AMD_MSR_CAPABLE; + cmd.addr.msr.reg = MSR_AMD_PERF_CTL; + cmd.val = (u32) perf->states[next_perf_state].control; + break; case SYSTEM_IO_CAPABLE: cmd.type = SYSTEM_IO_CAPABLE; cmd.addr.io.port = perf->control_register.address; @@ -419,6 +560,44 @@ static void free_acpi_perf_data(void) free_percpu(acpi_perf_data); } +static int boost_notify(struct notifier_block *nb, unsigned long action, + void *hcpu) +{ + unsigned cpu = (long)hcpu; + const struct cpumask *cpumask; + + cpumask = get_cpu_mask(cpu); + + /* + * Clear the boost-disable bit on the CPU_DOWN path so that + * this cpu cannot block the remaining ones from boosting. On + * the CPU_UP path we simply keep the boost-disable flag in + * sync with the current global state. + */ + + switch (action) { + case CPU_UP_PREPARE: + case CPU_UP_PREPARE_FROZEN: + boost_set_msrs(boost_enabled, cpumask); + break; + + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + boost_set_msrs(1, cpumask); + break; + + default: + break; + } + + return NOTIFY_OK; +} + + +static struct notifier_block boost_nb = { + .notifier_call = boost_notify, +}; + /* * acpi_cpufreq_early_init - initialize ACPI P-States library * @@ -559,6 +738,14 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) policy->shared_type = CPUFREQ_SHARED_TYPE_ALL; cpumask_copy(policy->cpus, cpu_core_mask(cpu)); } + + if (check_amd_hwpstate_cpu(cpu) && !acpi_pstate_strict) { + cpumask_clear(policy->cpus); + cpumask_set_cpu(cpu, policy->cpus); + cpumask_copy(policy->related_cpus, cpu_sibling_mask(cpu)); + policy->shared_type = CPUFREQ_SHARED_TYPE_HW; + pr_info_once(PFX "overriding BIOS provided _PSD data\n"); + } #endif /* capability check */ @@ -580,12 +767,16 @@ static int acpi_cpufreq_cpu_init(struct cpufreq_policy *policy) break; case ACPI_ADR_SPACE_FIXED_HARDWARE: pr_debug("HARDWARE addr space\n"); - if (!check_est_cpu(cpu)) { - result = -ENODEV; - goto err_unreg; + if (check_est_cpu(cpu)) { + data->cpu_feature = SYSTEM_INTEL_MSR_CAPABLE; + break; } - data->cpu_feature = SYSTEM_INTEL_MSR_CAPABLE; - break; + if (check_amd_hwpstate_cpu(cpu)) { + data->cpu_feature = SYSTEM_AMD_MSR_CAPABLE; + break; + } + result = -ENODEV; + goto err_unreg; default: pr_debug("Unknown addr space %d\n", (u32) (perf->control_register.space_id)); @@ -718,6 +909,7 @@ static int acpi_cpufreq_resume(struct cpufreq_policy *policy) static struct freq_attr *acpi_cpufreq_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, + NULL, /* this is a placeholder for cpb, do not remove */ NULL, }; @@ -733,6 +925,49 @@ static struct cpufreq_driver acpi_cpufreq_driver = { .attr = acpi_cpufreq_attr, }; +static void __init acpi_cpufreq_boost_init(void) +{ + if (boot_cpu_has(X86_FEATURE_CPB) || boot_cpu_has(X86_FEATURE_IDA)) { + msrs = msrs_alloc(); + + if (!msrs) + return; + + boost_supported = true; + boost_enabled = boost_state(0); + + get_online_cpus(); + + /* Force all MSRs to the same value */ + boost_set_msrs(boost_enabled, cpu_online_mask); + + register_cpu_notifier(&boost_nb); + + put_online_cpus(); + } else + global_boost.attr.mode = 0444; + + /* We create the boost file in any case, though for systems without + * hardware support it will be read-only and hardwired to return 0. + */ + if (sysfs_create_file(cpufreq_global_kobject, &(global_boost.attr))) + pr_warn(PFX "could not register global boost sysfs file\n"); + else + pr_debug("registered global boost sysfs file\n"); +} + +static void __exit acpi_cpufreq_boost_exit(void) +{ + sysfs_remove_file(cpufreq_global_kobject, &(global_boost.attr)); + + if (msrs) { + unregister_cpu_notifier(&boost_nb); + + msrs_free(msrs); + msrs = NULL; + } +} + static int __init acpi_cpufreq_init(void) { int ret; @@ -746,9 +981,32 @@ static int __init acpi_cpufreq_init(void) if (ret) return ret; +#ifdef CONFIG_X86_ACPI_CPUFREQ_CPB + /* this is a sysfs file with a strange name and an even stranger + * semantic - per CPU instantiation, but system global effect. + * Lets enable it only on AMD CPUs for compatibility reasons and + * only if configured. This is considered legacy code, which + * will probably be removed at some point in the future. + */ + if (check_amd_hwpstate_cpu(0)) { + struct freq_attr **iter; + + pr_debug("adding sysfs entry for cpb\n"); + + for (iter = acpi_cpufreq_attr; *iter != NULL; iter++) + ; + + /* make sure there is a terminator behind it */ + if (iter[1] == NULL) + *iter = &cpb; + } +#endif + ret = cpufreq_register_driver(&acpi_cpufreq_driver); if (ret) free_acpi_perf_data(); + else + acpi_cpufreq_boost_init(); return ret; } @@ -757,6 +1015,8 @@ static void __exit acpi_cpufreq_exit(void) { pr_debug("acpi_cpufreq_exit\n"); + acpi_cpufreq_boost_exit(); + cpufreq_unregister_driver(&acpi_cpufreq_driver); free_acpi_perf_data(); diff --git a/drivers/cpufreq/cpufreq-cpu0.c b/drivers/cpufreq/cpufreq-cpu0.c new file mode 100644 index 0000000..e915827 --- /dev/null +++ b/drivers/cpufreq/cpufreq-cpu0.c @@ -0,0 +1,269 @@ +/* + * Copyright (C) 2012 Freescale Semiconductor, Inc. + * + * The OPP code in function cpu0_set_target() is reused from + * drivers/cpufreq/omap-cpufreq.c + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 as + * published by the Free Software Foundation. + */ + +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt + +#include <linux/clk.h> +#include <linux/cpu.h> +#include <linux/cpufreq.h> +#include <linux/err.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/opp.h> +#include <linux/regulator/consumer.h> +#include <linux/slab.h> + +static unsigned int transition_latency; +static unsigned int voltage_tolerance; /* in percentage */ + +static struct device *cpu_dev; +static struct clk *cpu_clk; +static struct regulator *cpu_reg; +static struct cpufreq_frequency_table *freq_table; + +static int cpu0_verify_speed(struct cpufreq_policy *policy) +{ + return cpufreq_frequency_table_verify(policy, freq_table); +} + +static unsigned int cpu0_get_speed(unsigned int cpu) +{ + return clk_get_rate(cpu_clk) / 1000; +} + +static int cpu0_set_target(struct cpufreq_policy *policy, + unsigned int target_freq, unsigned int relation) +{ + struct cpufreq_freqs freqs; + struct opp *opp; + unsigned long freq_Hz, volt = 0, volt_old = 0, tol = 0; + unsigned int index, cpu; + int ret; + + ret = cpufreq_frequency_table_target(policy, freq_table, target_freq, + relation, &index); + if (ret) { + pr_err("failed to match target freqency %d: %d\n", + target_freq, ret); + return ret; + } + + freq_Hz = clk_round_rate(cpu_clk, freq_table[index].frequency * 1000); + if (freq_Hz < 0) + freq_Hz = freq_table[index].frequency * 1000; + freqs.new = freq_Hz / 1000; + freqs.old = clk_get_rate(cpu_clk) / 1000; + + if (freqs.old == freqs.new) + return 0; + + for_each_online_cpu(cpu) { + freqs.cpu = cpu; + cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); + } + + if (cpu_reg) { + opp = opp_find_freq_ceil(cpu_dev, &freq_Hz); + if (IS_ERR(opp)) { + pr_err("failed to find OPP for %ld\n", freq_Hz); + return PTR_ERR(opp); + } + volt = opp_get_voltage(opp); + tol = volt * voltage_tolerance / 100; + volt_old = regulator_get_voltage(cpu_reg); + } + + pr_debug("%u MHz, %ld mV --> %u MHz, %ld mV\n", + freqs.old / 1000, volt_old ? volt_old / 1000 : -1, + freqs.new / 1000, volt ? volt / 1000 : -1); + + /* scaling up? scale voltage before frequency */ + if (cpu_reg && freqs.new > freqs.old) { + ret = regulator_set_voltage_tol(cpu_reg, volt, tol); + if (ret) { + pr_err("failed to scale voltage up: %d\n", ret); + freqs.new = freqs.old; + return ret; + } + } + + ret = clk_set_rate(cpu_clk, freqs.new * 1000); + if (ret) { + pr_err("failed to set clock rate: %d\n", ret); + if (cpu_reg) + regulator_set_voltage_tol(cpu_reg, volt_old, tol); + return ret; + } + + /* scaling down? scale voltage after frequency */ + if (cpu_reg && freqs.new < freqs.old) { + ret = regulator_set_voltage_tol(cpu_reg, volt, tol); + if (ret) { + pr_err("failed to scale voltage down: %d\n", ret); + clk_set_rate(cpu_clk, freqs.old * 1000); + freqs.new = freqs.old; + return ret; + } + } + + for_each_online_cpu(cpu) { + freqs.cpu = cpu; + cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); + } + + return 0; +} + +static int cpu0_cpufreq_init(struct cpufreq_policy *policy) +{ + int ret; + + if (policy->cpu != 0) + return -EINVAL; + + ret = cpufreq_frequency_table_cpuinfo(policy, freq_table); + if (ret) { + pr_err("invalid frequency table: %d\n", ret); + return ret; + } + + policy->cpuinfo.transition_latency = transition_latency; + policy->cur = clk_get_rate(cpu_clk) / 1000; + + /* + * The driver only supports the SMP configuartion where all processors + * share the clock and voltage and clock. Use cpufreq affected_cpus + * interface to have all CPUs scaled together. + */ + policy->shared_type = CPUFREQ_SHARED_TYPE_ANY; + cpumask_setall(policy->cpus); + + cpufreq_frequency_table_get_attr(freq_table, policy->cpu); + + return 0; +} + +static int cpu0_cpufreq_exit(struct cpufreq_policy *policy) +{ + cpufreq_frequency_table_put_attr(policy->cpu); + + return 0; +} + +static struct freq_attr *cpu0_cpufreq_attr[] = { + &cpufreq_freq_attr_scaling_available_freqs, + NULL, +}; + +static struct cpufreq_driver cpu0_cpufreq_driver = { + .flags = CPUFREQ_STICKY, + .verify = cpu0_verify_speed, + .target = cpu0_set_target, + .get = cpu0_get_speed, + .init = cpu0_cpufreq_init, + .exit = cpu0_cpufreq_exit, + .name = "generic_cpu0", + .attr = cpu0_cpufreq_attr, +}; + +static int __devinit cpu0_cpufreq_driver_init(void) +{ + struct device_node *np; + int ret; + + np = of_find_node_by_path("/cpus/cpu@0"); + if (!np) { + pr_err("failed to find cpu0 node\n"); + return -ENOENT; + } + + cpu_dev = get_cpu_device(0); + if (!cpu_dev) { + pr_err("failed to get cpu0 device\n"); + ret = -ENODEV; + goto out_put_node; + } + + cpu_dev->of_node = np; + + cpu_clk = clk_get(cpu_dev, NULL); + if (IS_ERR(cpu_clk)) { + ret = PTR_ERR(cpu_clk); + pr_err("failed to get cpu0 clock: %d\n", ret); + goto out_put_node; + } + + cpu_reg = regulator_get(cpu_dev, "cpu0"); + if (IS_ERR(cpu_reg)) { + pr_warn("failed to get cpu0 regulator\n"); + cpu_reg = NULL; + } + + ret = of_init_opp_table(cpu_dev); + if (ret) { + pr_err("failed to init OPP table: %d\n", ret); + goto out_put_node; + } + + ret = opp_init_cpufreq_table(cpu_dev, &freq_table); + if (ret) { + pr_err("failed to init cpufreq table: %d\n", ret); + goto out_put_node; + } + + of_property_read_u32(np, "voltage-tolerance", &voltage_tolerance); + + if (of_property_read_u32(np, "clock-latency", &transition_latency)) + transition_latency = CPUFREQ_ETERNAL; + + if (cpu_reg) { + struct opp *opp; + unsigned long min_uV, max_uV; + int i; + + /* + * OPP is maintained in order of increasing frequency, and + * freq_table initialised from OPP is therefore sorted in the + * same order. + */ + for (i = 0; freq_table[i].frequency != CPUFREQ_TABLE_END; i++) + ; + opp = opp_find_freq_exact(cpu_dev, + freq_table[0].frequency * 1000, true); + min_uV = opp_get_voltage(opp); + opp = opp_find_freq_exact(cpu_dev, + freq_table[i-1].frequency * 1000, true); + max_uV = opp_get_voltage(opp); + ret = regulator_set_voltage_time(cpu_reg, min_uV, max_uV); + if (ret > 0) + transition_latency += ret * 1000; + } + + ret = cpufreq_register_driver(&cpu0_cpufreq_driver); + if (ret) { + pr_err("failed register driver: %d\n", ret); + goto out_free_table; + } + + of_node_put(np); + return 0; + +out_free_table: + opp_free_cpufreq_table(cpu_dev, &freq_table); +out_put_node: + of_node_put(np); + return ret; +} +late_initcall(cpu0_cpufreq_driver_init); + +MODULE_AUTHOR("Shawn Guo <shawn.guo@linaro.org>"); +MODULE_DESCRIPTION("Generic CPU0 cpufreq driver"); +MODULE_LICENSE("GPL"); diff --git a/drivers/cpufreq/cpufreq_conservative.c b/drivers/cpufreq/cpufreq_conservative.c index 55f0354..a152af7 100644 --- a/drivers/cpufreq/cpufreq_conservative.c +++ b/drivers/cpufreq/cpufreq_conservative.c @@ -504,6 +504,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, j_dbs_info->prev_cpu_nice = kcpustat_cpu(j).cpustat[CPUTIME_NICE]; } + this_dbs_info->cpu = cpu; this_dbs_info->down_skip = 0; this_dbs_info->requested_freq = policy->cur; @@ -583,6 +584,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, __cpufreq_driver_target( this_dbs_info->cur_policy, policy->min, CPUFREQ_RELATION_L); + dbs_check_cpu(this_dbs_info); mutex_unlock(&this_dbs_info->timer_mutex); break; diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index 14c1af5..396322f 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -761,6 +761,7 @@ static int cpufreq_governor_dbs(struct cpufreq_policy *policy, else if (policy->min > this_dbs_info->cur_policy->cur) __cpufreq_driver_target(this_dbs_info->cur_policy, policy->min, CPUFREQ_RELATION_L); + dbs_check_cpu(this_dbs_info); mutex_unlock(&this_dbs_info->timer_mutex); break; } diff --git a/drivers/cpufreq/longhaul.h b/drivers/cpufreq/longhaul.h index cbf48fb..e2dc436 100644 --- a/drivers/cpufreq/longhaul.h +++ b/drivers/cpufreq/longhaul.h @@ -56,7 +56,7 @@ union msr_longhaul { /* * VIA C3 Samuel 1 & Samuel 2 (stepping 0) */ -static const int __cpuinitdata samuel1_mults[16] = { +static const int __cpuinitconst samuel1_mults[16] = { -1, /* 0000 -> RESERVED */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -75,7 +75,7 @@ static const int __cpuinitdata samuel1_mults[16] = { -1, /* 1111 -> RESERVED */ }; -static const int __cpuinitdata samuel1_eblcr[16] = { +static const int __cpuinitconst samuel1_eblcr[16] = { 50, /* 0000 -> RESERVED */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -97,7 +97,7 @@ static const int __cpuinitdata samuel1_eblcr[16] = { /* * VIA C3 Samuel2 Stepping 1->15 */ -static const int __cpuinitdata samuel2_eblcr[16] = { +static const int __cpuinitconst samuel2_eblcr[16] = { 50, /* 0000 -> 5.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -119,7 +119,7 @@ static const int __cpuinitdata samuel2_eblcr[16] = { /* * VIA C3 Ezra */ -static const int __cpuinitdata ezra_mults[16] = { +static const int __cpuinitconst ezra_mults[16] = { 100, /* 0000 -> 10.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -138,7 +138,7 @@ static const int __cpuinitdata ezra_mults[16] = { 120, /* 1111 -> 12.0x */ }; -static const int __cpuinitdata ezra_eblcr[16] = { +static const int __cpuinitconst ezra_eblcr[16] = { 50, /* 0000 -> 5.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -160,7 +160,7 @@ static const int __cpuinitdata ezra_eblcr[16] = { /* * VIA C3 (Ezra-T) [C5M]. */ -static const int __cpuinitdata ezrat_mults[32] = { +static const int __cpuinitconst ezrat_mults[32] = { 100, /* 0000 -> 10.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -196,7 +196,7 @@ static const int __cpuinitdata ezrat_mults[32] = { -1, /* 1111 -> RESERVED (12.0x) */ }; -static const int __cpuinitdata ezrat_eblcr[32] = { +static const int __cpuinitconst ezrat_eblcr[32] = { 50, /* 0000 -> 5.0x */ 30, /* 0001 -> 3.0x */ 40, /* 0010 -> 4.0x */ @@ -235,7 +235,7 @@ static const int __cpuinitdata ezrat_eblcr[32] = { /* * VIA C3 Nehemiah */ -static const int __cpuinitdata nehemiah_mults[32] = { +static const int __cpuinitconst nehemiah_mults[32] = { 100, /* 0000 -> 10.0x */ -1, /* 0001 -> 16.0x */ 40, /* 0010 -> 4.0x */ @@ -270,7 +270,7 @@ static const int __cpuinitdata nehemiah_mults[32] = { -1, /* 1111 -> 12.0x */ }; -static const int __cpuinitdata nehemiah_eblcr[32] = { +static const int __cpuinitconst nehemiah_eblcr[32] = { 50, /* 0000 -> 5.0x */ 160, /* 0001 -> 16.0x */ 40, /* 0010 -> 4.0x */ @@ -315,7 +315,7 @@ struct mV_pos { unsigned short pos; }; -static const struct mV_pos __cpuinitdata vrm85_mV[32] = { +static const struct mV_pos __cpuinitconst vrm85_mV[32] = { {1250, 8}, {1200, 6}, {1150, 4}, {1100, 2}, {1050, 0}, {1800, 30}, {1750, 28}, {1700, 26}, {1650, 24}, {1600, 22}, {1550, 20}, {1500, 18}, @@ -326,14 +326,14 @@ static const struct mV_pos __cpuinitdata vrm85_mV[32] = { {1475, 17}, {1425, 15}, {1375, 13}, {1325, 11} }; -static const unsigned char __cpuinitdata mV_vrm85[32] = { +static const unsigned char __cpuinitconst mV_vrm85[32] = { 0x04, 0x14, 0x03, 0x13, 0x02, 0x12, 0x01, 0x11, 0x00, 0x10, 0x0f, 0x1f, 0x0e, 0x1e, 0x0d, 0x1d, 0x0c, 0x1c, 0x0b, 0x1b, 0x0a, 0x1a, 0x09, 0x19, 0x08, 0x18, 0x07, 0x17, 0x06, 0x16, 0x05, 0x15 }; -static const struct mV_pos __cpuinitdata mobilevrm_mV[32] = { +static const struct mV_pos __cpuinitconst mobilevrm_mV[32] = { {1750, 31}, {1700, 30}, {1650, 29}, {1600, 28}, {1550, 27}, {1500, 26}, {1450, 25}, {1400, 24}, {1350, 23}, {1300, 22}, {1250, 21}, {1200, 20}, @@ -344,7 +344,7 @@ static const struct mV_pos __cpuinitdata mobilevrm_mV[32] = { {675, 3}, {650, 2}, {625, 1}, {600, 0} }; -static const unsigned char __cpuinitdata mV_mobilevrm[32] = { +static const unsigned char __cpuinitconst mV_mobilevrm[32] = { 0x1f, 0x1e, 0x1d, 0x1c, 0x1b, 0x1a, 0x19, 0x18, 0x17, 0x16, 0x15, 0x14, 0x13, 0x12, 0x11, 0x10, 0x0f, 0x0e, 0x0d, 0x0c, 0x0b, 0x0a, 0x09, 0x08, diff --git a/drivers/cpufreq/omap-cpufreq.c b/drivers/cpufreq/omap-cpufreq.c index b47034e..65f8e9a 100644 --- a/drivers/cpufreq/omap-cpufreq.c +++ b/drivers/cpufreq/omap-cpufreq.c @@ -40,16 +40,6 @@ /* OPP tolerance in percentage */ #define OPP_TOLERANCE 4 -#ifdef CONFIG_SMP -struct lpj_info { - unsigned long ref; - unsigned int freq; -}; - -static DEFINE_PER_CPU(struct lpj_info, lpj_ref); -static struct lpj_info global_lpj_ref; -#endif - static struct cpufreq_frequency_table *freq_table; static atomic_t freq_table_users = ATOMIC_INIT(0); static struct clk *mpu_clk; @@ -161,31 +151,6 @@ static int omap_target(struct cpufreq_policy *policy, } freqs.new = omap_getspeed(policy->cpu); -#ifdef CONFIG_SMP - /* - * Note that loops_per_jiffy is not updated on SMP systems in - * cpufreq driver. So, update the per-CPU loops_per_jiffy value - * on frequency transition. We need to update all dependent CPUs. - */ - for_each_cpu(i, policy->cpus) { - struct lpj_info *lpj = &per_cpu(lpj_ref, i); - if (!lpj->freq) { - lpj->ref = per_cpu(cpu_data, i).loops_per_jiffy; - lpj->freq = freqs.old; - } - - per_cpu(cpu_data, i).loops_per_jiffy = - cpufreq_scale(lpj->ref, lpj->freq, freqs.new); - } - - /* And don't forget to adjust the global one */ - if (!global_lpj_ref.freq) { - global_lpj_ref.ref = loops_per_jiffy; - global_lpj_ref.freq = freqs.old; - } - loops_per_jiffy = cpufreq_scale(global_lpj_ref.ref, global_lpj_ref.freq, - freqs.new); -#endif done: /* notifiers */ @@ -301,9 +266,9 @@ static int __init omap_cpufreq_init(void) } mpu_dev = omap_device_get_by_hwmod_name("mpu"); - if (!mpu_dev) { + if (IS_ERR(mpu_dev)) { pr_warning("%s: unable to get the mpu device\n", __func__); - return -EINVAL; + return PTR_ERR(mpu_dev); } mpu_reg = regulator_get(mpu_dev, "vcc"); diff --git a/drivers/cpufreq/powernow-k8.c b/drivers/cpufreq/powernow-k8.c index 1a40935..129e80b 100644 --- a/drivers/cpufreq/powernow-k8.c +++ b/drivers/cpufreq/powernow-k8.c @@ -48,22 +48,12 @@ #define PFX "powernow-k8: " #define VERSION "version 2.20.00" #include "powernow-k8.h" -#include "mperf.h" /* serialize freq changes */ static DEFINE_MUTEX(fidvid_mutex); static DEFINE_PER_CPU(struct powernow_k8_data *, powernow_data); -static int cpu_family = CPU_OPTERON; - -/* array to map SW pstate number to acpi state */ -static u32 ps_to_as[8]; - -/* core performance boost */ -static bool cpb_capable, cpb_enabled; -static struct msr __percpu *msrs; - static struct cpufreq_driver cpufreq_amd64_driver; #ifndef CONFIG_SMP @@ -85,12 +75,6 @@ static u32 find_khz_freq_from_fid(u32 fid) return 1000 * find_freq_from_fid(fid); } -static u32 find_khz_freq_from_pstate(struct cpufreq_frequency_table *data, - u32 pstate) -{ - return data[ps_to_as[pstate]].frequency; -} - /* Return the vco fid for an input fid * * Each "low" fid has corresponding "high" fid, and you can get to "low" fids @@ -113,9 +97,6 @@ static int pending_bit_stuck(void) { u32 lo, hi; - if (cpu_family == CPU_HW_PSTATE) - return 0; - rdmsr(MSR_FIDVID_STATUS, lo, hi); return lo & MSR_S_LO_CHANGE_PENDING ? 1 : 0; } @@ -129,20 +110,6 @@ static int query_current_values_with_pending_wait(struct powernow_k8_data *data) u32 lo, hi; u32 i = 0; - if (cpu_family == CPU_HW_PSTATE) { - rdmsr(MSR_PSTATE_STATUS, lo, hi); - i = lo & HW_PSTATE_MASK; - data->currpstate = i; - - /* - * a workaround for family 11h erratum 311 might cause - * an "out-of-range Pstate if the core is in Pstate-0 - */ - if ((boot_cpu_data.x86 == 0x11) && (i >= data->numps)) - data->currpstate = HW_PSTATE_0; - - return 0; - } do { if (i++ > 10000) { pr_debug("detected change pending stuck\n"); @@ -299,14 +266,6 @@ static int decrease_vid_code_by_step(struct powernow_k8_data *data, return 0; } -/* Change hardware pstate by single MSR write */ -static int transition_pstate(struct powernow_k8_data *data, u32 pstate) -{ - wrmsr(MSR_PSTATE_CTRL, pstate, 0); - data->currpstate = pstate; - return 0; -} - /* Change Opteron/Athlon64 fid and vid, by the 3 phases. */ static int transition_fid_vid(struct powernow_k8_data *data, u32 reqfid, u32 reqvid) @@ -523,8 +482,6 @@ static int core_voltage_post_transition(struct powernow_k8_data *data, static const struct x86_cpu_id powernow_k8_ids[] = { /* IO based frequency switching */ { X86_VENDOR_AMD, 0xf }, - /* MSR based frequency switching supported */ - X86_FEATURE_MATCH(X86_FEATURE_HW_PSTATE), {} }; MODULE_DEVICE_TABLE(x86cpu, powernow_k8_ids); @@ -560,15 +517,8 @@ static void check_supported_cpu(void *_rc) "Power state transitions not supported\n"); return; } - } else { /* must be a HW Pstate capable processor */ - cpuid(CPUID_FREQ_VOLT_CAPABILITIES, &eax, &ebx, &ecx, &edx); - if ((edx & USE_HW_PSTATE) == USE_HW_PSTATE) - cpu_family = CPU_HW_PSTATE; - else - return; + *rc = 0; } - - *rc = 0; } static int check_pst_table(struct powernow_k8_data *data, struct pst_s *pst, @@ -632,18 +582,11 @@ static void print_basics(struct powernow_k8_data *data) for (j = 0; j < data->numps; j++) { if (data->powernow_table[j].frequency != CPUFREQ_ENTRY_INVALID) { - if (cpu_family == CPU_HW_PSTATE) { - printk(KERN_INFO PFX - " %d : pstate %d (%d MHz)\n", j, - data->powernow_table[j].index, - data->powernow_table[j].frequency/1000); - } else { printk(KERN_INFO PFX "fid 0x%x (%d MHz), vid 0x%x\n", data->powernow_table[j].index & 0xff, data->powernow_table[j].frequency/1000, data->powernow_table[j].index >> 8); - } } } if (data->batps) @@ -651,20 +594,6 @@ static void print_basics(struct powernow_k8_data *data) data->batps); } -static u32 freq_from_fid_did(u32 fid, u32 did) -{ - u32 mhz = 0; - - if (boot_cpu_data.x86 == 0x10) - mhz = (100 * (fid + 0x10)) >> did; - else if (boot_cpu_data.x86 == 0x11) - mhz = (100 * (fid + 8)) >> did; - else - BUG(); - - return mhz * 1000; -} - static int fill_powernow_table(struct powernow_k8_data *data, struct pst_s *pst, u8 maxvid) { @@ -824,7 +753,7 @@ static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, { u64 control; - if (!data->acpi_data.state_count || (cpu_family == CPU_HW_PSTATE)) + if (!data->acpi_data.state_count) return; control = data->acpi_data.states[index].control; @@ -875,10 +804,7 @@ static int powernow_k8_cpu_init_acpi(struct powernow_k8_data *data) data->numps = data->acpi_data.state_count; powernow_k8_acpi_pst_values(data, 0); - if (cpu_family == CPU_HW_PSTATE) - ret_val = fill_powernow_table_pstate(data, powernow_table); - else - ret_val = fill_powernow_table_fidvid(data, powernow_table); + ret_val = fill_powernow_table_fidvid(data, powernow_table); if (ret_val) goto err_out_mem; @@ -915,51 +841,6 @@ err_out: return ret_val; } -static int fill_powernow_table_pstate(struct powernow_k8_data *data, - struct cpufreq_frequency_table *powernow_table) -{ - int i; - u32 hi = 0, lo = 0; - rdmsr(MSR_PSTATE_CUR_LIMIT, lo, hi); - data->max_hw_pstate = (lo & HW_PSTATE_MAX_MASK) >> HW_PSTATE_MAX_SHIFT; - - for (i = 0; i < data->acpi_data.state_count; i++) { - u32 index; - - index = data->acpi_data.states[i].control & HW_PSTATE_MASK; - if (index > data->max_hw_pstate) { - printk(KERN_ERR PFX "invalid pstate %d - " - "bad value %d.\n", i, index); - printk(KERN_ERR PFX "Please report to BIOS " - "manufacturer\n"); - invalidate_entry(powernow_table, i); - continue; - } - - ps_to_as[index] = i; - - /* Frequency may be rounded for these */ - if ((boot_cpu_data.x86 == 0x10 && boot_cpu_data.x86_model < 10) - || boot_cpu_data.x86 == 0x11) { - - rdmsr(MSR_PSTATE_DEF_BASE + index, lo, hi); - if (!(hi & HW_PSTATE_VALID_MASK)) { - pr_debug("invalid pstate %d, ignoring\n", index); - invalidate_entry(powernow_table, i); - continue; - } - - powernow_table[i].frequency = - freq_from_fid_did(lo & 0x3f, (lo >> 6) & 7); - } else - powernow_table[i].frequency = - data->acpi_data.states[i].core_frequency * 1000; - - powernow_table[i].index = index; - } - return 0; -} - static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table) { @@ -1036,15 +917,7 @@ static int get_transition_latency(struct powernow_k8_data *data) max_latency = cur_latency; } if (max_latency == 0) { - /* - * Fam 11h and later may return 0 as transition latency. This - * is intended and means "very fast". While cpufreq core and - * governors currently can handle that gracefully, better set it - * to 1 to avoid problems in the future. - */ - if (boot_cpu_data.x86 < 0x11) - printk(KERN_ERR FW_WARN PFX "Invalid zero transition " - "latency\n"); + pr_err(FW_WARN PFX "Invalid zero transition latency\n"); max_latency = 1; } /* value in usecs, needs to be in nanoseconds */ @@ -1104,40 +977,6 @@ static int transition_frequency_fidvid(struct powernow_k8_data *data, return res; } -/* Take a frequency, and issue the hardware pstate transition command */ -static int transition_frequency_pstate(struct powernow_k8_data *data, - unsigned int index) -{ - u32 pstate = 0; - int res, i; - struct cpufreq_freqs freqs; - - pr_debug("cpu %d transition to index %u\n", smp_processor_id(), index); - - /* get MSR index for hardware pstate transition */ - pstate = index & HW_PSTATE_MASK; - if (pstate > data->max_hw_pstate) - return -EINVAL; - - freqs.old = find_khz_freq_from_pstate(data->powernow_table, - data->currpstate); - freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate); - - for_each_cpu(i, data->available_cores) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_PRECHANGE); - } - - res = transition_pstate(data, pstate); - freqs.new = find_khz_freq_from_pstate(data->powernow_table, pstate); - - for_each_cpu(i, data->available_cores) { - freqs.cpu = i; - cpufreq_notify_transition(&freqs, CPUFREQ_POSTCHANGE); - } - return res; -} - struct powernowk8_target_arg { struct cpufreq_policy *pol; unsigned targfreq; @@ -1173,18 +1012,15 @@ static long powernowk8_target_fn(void *arg) if (query_current_values_with_pending_wait(data)) return -EIO; - if (cpu_family != CPU_HW_PSTATE) { - pr_debug("targ: curr fid 0x%x, vid 0x%x\n", - data->currfid, data->currvid); + pr_debug("targ: curr fid 0x%x, vid 0x%x\n", + data->currfid, data->currvid); - if ((checkvid != data->currvid) || - (checkfid != data->currfid)) { - printk(KERN_INFO PFX - "error - out of sync, fix 0x%x 0x%x, " - "vid 0x%x 0x%x\n", - checkfid, data->currfid, - checkvid, data->currvid); - } + if ((checkvid != data->currvid) || + (checkfid != data->currfid)) { + pr_info(PFX + "error - out of sync, fix 0x%x 0x%x, vid 0x%x 0x%x\n", + checkfid, data->currfid, + checkvid, data->currvid); } if (cpufreq_frequency_table_target(pol, data->powernow_table, @@ -1195,11 +1031,8 @@ static long powernowk8_target_fn(void *arg) powernow_k8_acpi_pst_values(data, newstate); - if (cpu_family == CPU_HW_PSTATE) - ret = transition_frequency_pstate(data, - data->powernow_table[newstate].index); - else - ret = transition_frequency_fidvid(data, newstate); + ret = transition_frequency_fidvid(data, newstate); + if (ret) { printk(KERN_ERR PFX "transition frequency failed\n"); mutex_unlock(&fidvid_mutex); @@ -1207,11 +1040,7 @@ static long powernowk8_target_fn(void *arg) } mutex_unlock(&fidvid_mutex); - if (cpu_family == CPU_HW_PSTATE) - pol->cur = find_khz_freq_from_pstate(data->powernow_table, - data->powernow_table[newstate].index); - else - pol->cur = find_khz_freq_from_fid(data->currfid); + pol->cur = find_khz_freq_from_fid(data->currfid); return 0; } @@ -1264,22 +1093,23 @@ static void __cpuinit powernowk8_cpu_init_on_cpu(void *_init_on_cpu) return; } - if (cpu_family == CPU_OPTERON) - fidvid_msr_init(); + fidvid_msr_init(); init_on_cpu->rc = 0; } +static const char missing_pss_msg[] = + KERN_ERR + FW_BUG PFX "No compatible ACPI _PSS objects found.\n" + FW_BUG PFX "First, make sure Cool'N'Quiet is enabled in the BIOS.\n" + FW_BUG PFX "If that doesn't help, try upgrading your BIOS.\n"; + /* per CPU init entry point to the driver */ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) { - static const char ACPI_PSS_BIOS_BUG_MSG[] = - KERN_ERR FW_BUG PFX "No compatible ACPI _PSS objects found.\n" - FW_BUG PFX "Try again with latest BIOS.\n"; struct powernow_k8_data *data; struct init_on_cpu init_on_cpu; int rc; - struct cpuinfo_x86 *c = &cpu_data(pol->cpu); if (!cpu_online(pol->cpu)) return -ENODEV; @@ -1295,7 +1125,6 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) } data->cpu = pol->cpu; - data->currpstate = HW_PSTATE_INVALID; if (powernow_k8_cpu_init_acpi(data)) { /* @@ -1303,7 +1132,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) * an UP version, and is deprecated by AMD. */ if (num_online_cpus() != 1) { - printk_once(ACPI_PSS_BIOS_BUG_MSG); + printk_once(missing_pss_msg); goto err_out; } if (pol->cpu != 0) { @@ -1332,17 +1161,10 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) if (rc != 0) goto err_out_exit_acpi; - if (cpu_family == CPU_HW_PSTATE) - cpumask_copy(pol->cpus, cpumask_of(pol->cpu)); - else - cpumask_copy(pol->cpus, cpu_core_mask(pol->cpu)); + cpumask_copy(pol->cpus, cpu_core_mask(pol->cpu)); data->available_cores = pol->cpus; - if (cpu_family == CPU_HW_PSTATE) - pol->cur = find_khz_freq_from_pstate(data->powernow_table, - data->currpstate); - else - pol->cur = find_khz_freq_from_fid(data->currfid); + pol->cur = find_khz_freq_from_fid(data->currfid); pr_debug("policy current frequency %d kHz\n", pol->cur); /* min/max the cpu is capable of */ @@ -1354,18 +1176,10 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol) return -EINVAL; } - /* Check for APERF/MPERF support in hardware */ - if (cpu_has(c, X86_FEATURE_APERFMPERF)) - cpufreq_amd64_driver.getavg = cpufreq_get_measured_perf; - cpufreq_frequency_table_get_attr(data->powernow_table, pol->cpu); - if (cpu_family == CPU_HW_PSTATE) - pr_debug("cpu_init done, current pstate 0x%x\n", - data->currpstate); - else - pr_debug("cpu_init done, current fid 0x%x, vid 0x%x\n", - data->currfid, data->currvid); + pr_debug("cpu_init done, current fid 0x%x, vid 0x%x\n", + data->currfid, data->currvid); per_cpu(powernow_data, pol->cpu) = data; @@ -1418,88 +1232,15 @@ static unsigned int powernowk8_get(unsigned int cpu) if (err) goto out; - if (cpu_family == CPU_HW_PSTATE) - khz = find_khz_freq_from_pstate(data->powernow_table, - data->currpstate); - else - khz = find_khz_freq_from_fid(data->currfid); + khz = find_khz_freq_from_fid(data->currfid); out: return khz; } -static void _cpb_toggle_msrs(bool t) -{ - int cpu; - - get_online_cpus(); - - rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); - - for_each_cpu(cpu, cpu_online_mask) { - struct msr *reg = per_cpu_ptr(msrs, cpu); - if (t) - reg->l &= ~BIT(25); - else - reg->l |= BIT(25); - } - wrmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); - - put_online_cpus(); -} - -/* - * Switch on/off core performance boosting. - * - * 0=disable - * 1=enable. - */ -static void cpb_toggle(bool t) -{ - if (!cpb_capable) - return; - - if (t && !cpb_enabled) { - cpb_enabled = true; - _cpb_toggle_msrs(t); - printk(KERN_INFO PFX "Core Boosting enabled.\n"); - } else if (!t && cpb_enabled) { - cpb_enabled = false; - _cpb_toggle_msrs(t); - printk(KERN_INFO PFX "Core Boosting disabled.\n"); - } -} - -static ssize_t store_cpb(struct cpufreq_policy *policy, const char *buf, - size_t count) -{ - int ret = -EINVAL; - unsigned long val = 0; - - ret = strict_strtoul(buf, 10, &val); - if (!ret && (val == 0 || val == 1) && cpb_capable) - cpb_toggle(val); - else - return -EINVAL; - - return count; -} - -static ssize_t show_cpb(struct cpufreq_policy *policy, char *buf) -{ - return sprintf(buf, "%u\n", cpb_enabled); -} - -#define define_one_rw(_name) \ -static struct freq_attr _name = \ -__ATTR(_name, 0644, show_##_name, store_##_name) - -define_one_rw(cpb); - static struct freq_attr *powernow_k8_attr[] = { &cpufreq_freq_attr_scaling_available_freqs, - &cpb, NULL, }; @@ -1515,53 +1256,18 @@ static struct cpufreq_driver cpufreq_amd64_driver = { .attr = powernow_k8_attr, }; -/* - * Clear the boost-disable flag on the CPU_DOWN path so that this cpu - * cannot block the remaining ones from boosting. On the CPU_UP path we - * simply keep the boost-disable flag in sync with the current global - * state. - */ -static int cpb_notify(struct notifier_block *nb, unsigned long action, - void *hcpu) -{ - unsigned cpu = (long)hcpu; - u32 lo, hi; - - switch (action) { - case CPU_UP_PREPARE: - case CPU_UP_PREPARE_FROZEN: - - if (!cpb_enabled) { - rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); - lo |= BIT(25); - wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi); - } - break; - - case CPU_DOWN_PREPARE: - case CPU_DOWN_PREPARE_FROZEN: - rdmsr_on_cpu(cpu, MSR_K7_HWCR, &lo, &hi); - lo &= ~BIT(25); - wrmsr_on_cpu(cpu, MSR_K7_HWCR, lo, hi); - break; - - default: - break; - } - - return NOTIFY_OK; -} - -static struct notifier_block cpb_nb = { - .notifier_call = cpb_notify, -}; - /* driver entry point for init */ static int __cpuinit powernowk8_init(void) { - unsigned int i, supported_cpus = 0, cpu; + unsigned int i, supported_cpus = 0; int rv; + if (static_cpu_has(X86_FEATURE_HW_PSTATE)) { + pr_warn(PFX "this CPU is not supported anymore, using acpi-cpufreq instead.\n"); + request_module("acpi-cpufreq"); + return -ENODEV; + } + if (!x86_match_cpu(powernow_k8_ids)) return -ENODEV; @@ -1575,38 +1281,13 @@ static int __cpuinit powernowk8_init(void) if (supported_cpus != num_online_cpus()) return -ENODEV; - printk(KERN_INFO PFX "Found %d %s (%d cpu cores) (" VERSION ")\n", - num_online_nodes(), boot_cpu_data.x86_model_id, supported_cpus); - - if (boot_cpu_has(X86_FEATURE_CPB)) { - - cpb_capable = true; - - msrs = msrs_alloc(); - if (!msrs) { - printk(KERN_ERR "%s: Error allocating msrs!\n", __func__); - return -ENOMEM; - } - - register_cpu_notifier(&cpb_nb); - - rdmsr_on_cpus(cpu_online_mask, MSR_K7_HWCR, msrs); + rv = cpufreq_register_driver(&cpufreq_amd64_driver); - for_each_cpu(cpu, cpu_online_mask) { - struct msr *reg = per_cpu_ptr(msrs, cpu); - cpb_enabled |= !(!!(reg->l & BIT(25))); - } + if (!rv) + pr_info(PFX "Found %d %s (%d cpu cores) (" VERSION ")\n", + num_online_nodes(), boot_cpu_data.x86_model_id, + supported_cpus); - printk(KERN_INFO PFX "Core Performance Boosting: %s.\n", - (cpb_enabled ? "on" : "off")); - } - - rv = cpufreq_register_driver(&cpufreq_amd64_driver); - if (rv < 0 && boot_cpu_has(X86_FEATURE_CPB)) { - unregister_cpu_notifier(&cpb_nb); - msrs_free(msrs); - msrs = NULL; - } return rv; } @@ -1615,13 +1296,6 @@ static void __exit powernowk8_exit(void) { pr_debug("exit\n"); - if (boot_cpu_has(X86_FEATURE_CPB)) { - msrs_free(msrs); - msrs = NULL; - - unregister_cpu_notifier(&cpb_nb); - } - cpufreq_unregister_driver(&cpufreq_amd64_driver); } diff --git a/drivers/cpufreq/powernow-k8.h b/drivers/cpufreq/powernow-k8.h index 3744d26..79329d4 100644 --- a/drivers/cpufreq/powernow-k8.h +++ b/drivers/cpufreq/powernow-k8.h @@ -5,24 +5,11 @@ * http://www.gnu.org/licenses/gpl.html */ -enum pstate { - HW_PSTATE_INVALID = 0xff, - HW_PSTATE_0 = 0, - HW_PSTATE_1 = 1, - HW_PSTATE_2 = 2, - HW_PSTATE_3 = 3, - HW_PSTATE_4 = 4, - HW_PSTATE_5 = 5, - HW_PSTATE_6 = 6, - HW_PSTATE_7 = 7, -}; - struct powernow_k8_data { unsigned int cpu; u32 numps; /* number of p-states */ u32 batps; /* number of p-states supported on battery */ - u32 max_hw_pstate; /* maximum legal hardware pstate */ /* these values are constant when the PSB is used to determine * vid/fid pairings, but are modified during the ->target() call @@ -37,7 +24,6 @@ struct powernow_k8_data { /* keep track of the current fid / vid or pstate */ u32 currvid; u32 currfid; - enum pstate currpstate; /* the powernow_table includes all frequency and vid/fid pairings: * fid are the lower 8 bits of the index, vid are the upper 8 bits. @@ -97,23 +83,6 @@ struct powernow_k8_data { #define MSR_S_HI_CURRENT_VID 0x0000003f #define MSR_C_HI_STP_GNT_BENIGN 0x00000001 - -/* Hardware Pstate _PSS and MSR definitions */ -#define USE_HW_PSTATE 0x00000080 -#define HW_PSTATE_MASK 0x00000007 -#define HW_PSTATE_VALID_MASK 0x80000000 -#define HW_PSTATE_MAX_MASK 0x000000f0 -#define HW_PSTATE_MAX_SHIFT 4 -#define MSR_PSTATE_DEF_BASE 0xc0010064 /* base of Pstate MSRs */ -#define MSR_PSTATE_STATUS 0xc0010063 /* Pstate Status MSR */ -#define MSR_PSTATE_CTRL 0xc0010062 /* Pstate control MSR */ -#define MSR_PSTATE_CUR_LIMIT 0xc0010061 /* pstate current limit MSR */ - -/* define the two driver architectures */ -#define CPU_OPTERON 0 -#define CPU_HW_PSTATE 1 - - /* * There are restrictions frequencies have to follow: * - only 1 entry in the low fid table ( <=1.4GHz ) @@ -218,5 +187,4 @@ static int core_frequency_transition(struct powernow_k8_data *data, u32 reqfid); static void powernow_k8_acpi_pst_values(struct powernow_k8_data *data, unsigned int index); -static int fill_powernow_table_pstate(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table); static int fill_powernow_table_fidvid(struct powernow_k8_data *data, struct cpufreq_frequency_table *powernow_table); diff --git a/drivers/cpuidle/driver.c b/drivers/cpuidle/driver.c index 58bf3b1..87db387 100644 --- a/drivers/cpuidle/driver.c +++ b/drivers/cpuidle/driver.c @@ -18,9 +18,10 @@ static struct cpuidle_driver *cpuidle_curr_driver; DEFINE_SPINLOCK(cpuidle_driver_lock); int cpuidle_driver_refcount; -static void __cpuidle_register_driver(struct cpuidle_driver *drv) +static void set_power_states(struct cpuidle_driver *drv) { int i; + /* * cpuidle driver should set the drv->power_specified bit * before registering if the driver provides @@ -35,13 +36,10 @@ static void __cpuidle_register_driver(struct cpuidle_driver *drv) * an power value of -1. So we use -2, -3, etc, for other * c-states. */ - if (!drv->power_specified) { - for (i = CPUIDLE_DRIVER_STATE_START; i < drv->state_count; i++) - drv->states[i].power_usage = -1 - i; - } + for (i = CPUIDLE_DRIVER_STATE_START; i < drv->state_count; i++) + drv->states[i].power_usage = -1 - i; } - /** * cpuidle_register_driver - registers a driver * @drv: the driver @@ -59,13 +57,16 @@ int cpuidle_register_driver(struct cpuidle_driver *drv) spin_unlock(&cpuidle_driver_lock); return -EBUSY; } - __cpuidle_register_driver(drv); + + if (!drv->power_specified) + set_power_states(drv); + cpuidle_curr_driver = drv; + spin_unlock(&cpuidle_driver_lock); return 0; } - EXPORT_SYMBOL_GPL(cpuidle_register_driver); /** @@ -96,7 +97,6 @@ void cpuidle_unregister_driver(struct cpuidle_driver *drv) spin_unlock(&cpuidle_driver_lock); } - EXPORT_SYMBOL_GPL(cpuidle_unregister_driver); struct cpuidle_driver *cpuidle_driver_ref(void) diff --git a/drivers/cpuidle/governors/ladder.c b/drivers/cpuidle/governors/ladder.c index b6a09ea..9b78405 100644 --- a/drivers/cpuidle/governors/ladder.c +++ b/drivers/cpuidle/governors/ladder.c @@ -88,6 +88,8 @@ static int ladder_select_state(struct cpuidle_driver *drv, /* consider promotion */ if (last_idx < drv->state_count - 1 && + !drv->states[last_idx + 1].disabled && + !dev->states_usage[last_idx + 1].disable && last_residency > last_state->threshold.promotion_time && drv->states[last_idx + 1].exit_latency <= latency_req) { last_state->stats.promotion_count++; @@ -100,7 +102,9 @@ static int ladder_select_state(struct cpuidle_driver *drv, /* consider demotion */ if (last_idx > CPUIDLE_DRIVER_STATE_START && - drv->states[last_idx].exit_latency > latency_req) { + (drv->states[last_idx].disabled || + dev->states_usage[last_idx].disable || + drv->states[last_idx].exit_latency > latency_req)) { int i; for (i = last_idx - 1; i > CPUIDLE_DRIVER_STATE_START; i--) { diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c index 7972bae2..2709ff5 100644 --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -1183,7 +1183,7 @@ static ssize_t ucma_migrate_id(struct ucma_file *new_file, struct rdma_ucm_migrate_id cmd; struct rdma_ucm_migrate_resp resp; struct ucma_context *ctx; - struct file *filp; + struct fd f; struct ucma_file *cur_file; int ret = 0; @@ -1191,12 +1191,12 @@ static ssize_t ucma_migrate_id(struct ucma_file *new_file, return -EFAULT; /* Get current fd to protect against it being closed */ - filp = fget(cmd.fd); - if (!filp) + f = fdget(cmd.fd); + if (!f.file) return -ENOENT; /* Validate current fd and prevent destruction of id. */ - ctx = ucma_get_ctx(filp->private_data, cmd.id); + ctx = ucma_get_ctx(f.file->private_data, cmd.id); if (IS_ERR(ctx)) { ret = PTR_ERR(ctx); goto file_put; @@ -1230,7 +1230,7 @@ response: ucma_put_ctx(ctx); file_put: - fput(filp); + fdput(f); return ret; } diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index f9d0d7c..0cb0007 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -705,7 +705,7 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, struct ib_udata udata; struct ib_uxrcd_object *obj; struct ib_xrcd *xrcd = NULL; - struct file *f = NULL; + struct fd f = {NULL, 0}; struct inode *inode = NULL; int ret = 0; int new_xrcd = 0; @@ -724,18 +724,13 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, if (cmd.fd != -1) { /* search for file descriptor */ - f = fget(cmd.fd); - if (!f) { - ret = -EBADF; - goto err_tree_mutex_unlock; - } - - inode = f->f_dentry->d_inode; - if (!inode) { + f = fdget(cmd.fd); + if (!f.file) { ret = -EBADF; goto err_tree_mutex_unlock; } + inode = f.file->f_path.dentry->d_inode; xrcd = find_xrcd(file->device, inode); if (!xrcd && !(cmd.oflags & O_CREAT)) { /* no file descriptor. Need CREATE flag */ @@ -800,8 +795,8 @@ ssize_t ib_uverbs_open_xrcd(struct ib_uverbs_file *file, goto err_copy; } - if (f) - fput(f); + if (f.file) + fdput(f); mutex_lock(&file->mutex); list_add_tail(&obj->uobject.list, &file->ucontext->xrcd_list); @@ -830,8 +825,8 @@ err: put_uobj_write(&obj->uobject); err_tree_mutex_unlock: - if (f) - fput(f); + if (f.file) + fdput(f); mutex_unlock(&file->device->xrcd_tree_mutex); diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 604556d..6f2ce6f 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -541,16 +541,15 @@ struct file *ib_uverbs_alloc_event_file(struct ib_uverbs_file *uverbs_file, struct ib_uverbs_event_file *ib_uverbs_lookup_comp_file(int fd) { struct ib_uverbs_event_file *ev_file = NULL; - struct file *filp; + struct fd f = fdget(fd); - filp = fget(fd); - if (!filp) + if (!f.file) return NULL; - if (filp->f_op != &uverbs_event_fops) + if (f.file->f_op != &uverbs_event_fops) goto out; - ev_file = filp->private_data; + ev_file = f.file->private_data; if (ev_file->is_async) { ev_file = NULL; goto out; @@ -559,7 +558,7 @@ struct ib_uverbs_event_file *ib_uverbs_lookup_comp_file(int fd) kref_get(&ev_file->ref); out: - fput(filp); + fdput(f); return ev_file; } diff --git a/drivers/message/fusion/mptbase.c b/drivers/message/fusion/mptbase.c index d99db56..fb69baa 100644 --- a/drivers/message/fusion/mptbase.c +++ b/drivers/message/fusion/mptbase.c @@ -1666,7 +1666,7 @@ mpt_mapresources(MPT_ADAPTER *ioc) if (pci_request_selected_regions(pdev, ioc->bars, "mpt")) { printk(MYIOC_s_ERR_FMT "pci_request_selected_regions() with " "MEM failed\n", ioc->name); - return r; + goto out_pci_disable_device; } if (sizeof(dma_addr_t) > 4) { @@ -1690,8 +1690,7 @@ mpt_mapresources(MPT_ADAPTER *ioc) } else { printk(MYIOC_s_WARN_FMT "no suitable DMA mask for %s\n", ioc->name, pci_name(pdev)); - pci_release_selected_regions(pdev, ioc->bars); - return r; + goto out_pci_release_region; } } else { if (!pci_set_dma_mask(pdev, DMA_BIT_MASK(32)) @@ -1704,8 +1703,7 @@ mpt_mapresources(MPT_ADAPTER *ioc) } else { printk(MYIOC_s_WARN_FMT "no suitable DMA mask for %s\n", ioc->name, pci_name(pdev)); - pci_release_selected_regions(pdev, ioc->bars); - return r; + goto out_pci_release_region; } } @@ -1735,8 +1733,8 @@ mpt_mapresources(MPT_ADAPTER *ioc) if (mem == NULL) { printk(MYIOC_s_ERR_FMT ": ERROR - Unable to map adapter" " memory!\n", ioc->name); - pci_release_selected_regions(pdev, ioc->bars); - return -EINVAL; + r = -EINVAL; + goto out_pci_release_region; } ioc->memmap = mem; dinitprintk(ioc, printk(MYIOC_s_INFO_FMT "mem = %p, mem_phys = %llx\n", @@ -1750,6 +1748,12 @@ mpt_mapresources(MPT_ADAPTER *ioc) ioc->pio_chip = (SYSIF_REGS __iomem *)port; return 0; + +out_pci_release_region: + pci_release_selected_regions(pdev, ioc->bars); +out_pci_disable_device: + pci_disable_device(pdev); + return r; } /*=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=*/ diff --git a/drivers/mtd/mtdpart.c b/drivers/mtd/mtdpart.c index d518e4d..3a49e6d 100644 --- a/drivers/mtd/mtdpart.c +++ b/drivers/mtd/mtdpart.c @@ -744,7 +744,7 @@ int parse_mtd_partitions(struct mtd_info *master, const char **types, return ret; } -int mtd_is_partition(struct mtd_info *mtd) +int mtd_is_partition(const struct mtd_info *mtd) { struct mtd_part *part; int ispart = 0; @@ -760,3 +760,13 @@ int mtd_is_partition(struct mtd_info *mtd) return ispart; } EXPORT_SYMBOL_GPL(mtd_is_partition); + +/* Returns the size of the entire flash chip */ +uint64_t mtd_get_device_size(const struct mtd_info *mtd) +{ + if (!mtd_is_partition(mtd)) + return mtd->size; + + return PART(mtd)->master->size; +} +EXPORT_SYMBOL_GPL(mtd_get_device_size); diff --git a/drivers/mtd/ubi/Kconfig b/drivers/mtd/ubi/Kconfig index ea4b95b..271a842 100644 --- a/drivers/mtd/ubi/Kconfig +++ b/drivers/mtd/ubi/Kconfig @@ -27,20 +27,34 @@ config MTD_UBI_WL_THRESHOLD life-cycle less than 10000, the threshold should be lessened (e.g., to 128 or 256, although it does not have to be power of 2). -config MTD_UBI_BEB_RESERVE - int "Percentage of reserved eraseblocks for bad eraseblocks handling" - default 2 - range 0 25 +config MTD_UBI_BEB_LIMIT + int "Maximum expected bad eraseblock count per 1024 eraseblocks" + default 20 + range 0 768 help - If the MTD device admits of bad eraseblocks (e.g. NAND flash), UBI - reserves some amount of physical eraseblocks to handle new bad - eraseblocks. For example, if a flash physical eraseblock becomes bad, - UBI uses these reserved physical eraseblocks to relocate the bad one. - This option specifies how many physical eraseblocks will be reserved - for bad eraseblock handling (percents of total number of good flash - eraseblocks). If the underlying flash does not admit of bad - eraseblocks (e.g. NOR flash), this value is ignored and nothing is - reserved. Leave the default value if unsure. + This option specifies the maximum bad physical eraseblocks UBI + expects on the MTD device (per 1024 eraseblocks). If the underlying + flash does not admit of bad eraseblocks (e.g. NOR flash), this value + is ignored. + + NAND datasheets often specify the minimum and maximum NVM (Number of + Valid Blocks) for the flashes' endurance lifetime. The maximum + expected bad eraseblocks per 1024 eraseblocks then can be calculated + as "1024 * (1 - MinNVB / MaxNVB)", which gives 20 for most NANDs + (MaxNVB is basically the total count of eraseblocks on the chip). + + To put it differently, if this value is 20, UBI will try to reserve + about 1.9% of physical eraseblocks for bad blocks handling. And that + will be 1.9% of eraseblocks on the entire NAND chip, not just the MTD + partition UBI attaches. This means that if you have, say, a NAND + flash chip admits maximum 40 bad eraseblocks, and it is split on two + MTD partitions of the same size, UBI will reserve 40 eraseblocks when + attaching a partition. + + This option can be overridden by the "mtd=" UBI module parameter or + by the "attach" ioctl. + + Leave the default value if unsure. config MTD_UBI_GLUEBI tristate "MTD devices emulation driver (gluebi)" diff --git a/drivers/mtd/ubi/attach.c b/drivers/mtd/ubi/attach.c index bd27cbb..f7adf53 100644 --- a/drivers/mtd/ubi/attach.c +++ b/drivers/mtd/ubi/attach.c @@ -79,7 +79,7 @@ * NAND), it is probably a PEB which was being erased when power cut * happened, so this is corruption type 1. However, this is just a guess, * which might be wrong. - * o Otherwise this it corruption type 2. + * o Otherwise this is corruption type 2. */ #include <linux/err.h> @@ -378,8 +378,8 @@ static int compare_lebs(struct ubi_device *ubi, const struct ubi_ainf_peb *aeb, if (err == UBI_IO_BITFLIPS) bitflips = 1; else { - ubi_err("VID of PEB %d header is bad, but it " - "was OK earlier, err %d", pnum, err); + ubi_err("VID of PEB %d header is bad, but it was OK earlier, err %d", + pnum, err); if (err > 0) err = -EIO; @@ -790,12 +790,12 @@ static int check_corruption(struct ubi_device *ubi, struct ubi_vid_hdr *vid_hdr, if (ubi_check_pattern(ubi->peb_buf, 0xFF, ubi->leb_size)) goto out_unlock; - ubi_err("PEB %d contains corrupted VID header, and the data does not " - "contain all 0xFF, this may be a non-UBI PEB or a severe VID " - "header corruption which requires manual inspection", pnum); + ubi_err("PEB %d contains corrupted VID header, and the data does not contain all 0xFF", + pnum); + ubi_err("this may be a non-UBI PEB or a severe VID header corruption which requires manual inspection"); ubi_dump_vid_hdr(vid_hdr); - dbg_msg("hexdump of PEB %d offset %d, length %d", - pnum, ubi->leb_start, ubi->leb_size); + pr_err("hexdump of PEB %d offset %d, length %d", + pnum, ubi->leb_start, ubi->leb_size); ubi_dbg_print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, ubi->peb_buf, ubi->leb_size, 1); err = 1; @@ -907,8 +907,8 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, ubi->image_seq = image_seq; if (ubi->image_seq && image_seq && ubi->image_seq != image_seq) { - ubi_err("bad image sequence number %d in PEB %d, " - "expected %d", image_seq, pnum, ubi->image_seq); + ubi_err("bad image sequence number %d in PEB %d, expected %d", + image_seq, pnum, ubi->image_seq); ubi_dump_ec_hdr(ech); return -EINVAL; } @@ -975,7 +975,7 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, return err; goto adjust_mean_ec; case UBI_IO_FF: - if (ec_err) + if (ec_err || bitflips) err = add_to_list(ai, pnum, UBI_UNKNOWN, UBI_UNKNOWN, ec, 1, &ai->erase); else @@ -997,8 +997,8 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, /* Unsupported internal volume */ switch (vidh->compat) { case UBI_COMPAT_DELETE: - ubi_msg("\"delete\" compatible internal volume %d:%d" - " found, will remove it", vol_id, lnum); + ubi_msg("\"delete\" compatible internal volume %d:%d found, will remove it", + vol_id, lnum); err = add_to_list(ai, pnum, vol_id, lnum, ec, 1, &ai->erase); if (err) @@ -1006,15 +1006,14 @@ static int scan_peb(struct ubi_device *ubi, struct ubi_attach_info *ai, return 0; case UBI_COMPAT_RO: - ubi_msg("read-only compatible internal volume %d:%d" - " found, switch to read-only mode", + ubi_msg("read-only compatible internal volume %d:%d found, switch to read-only mode", vol_id, lnum); ubi->ro_mode = 1; break; case UBI_COMPAT_PRESERVE: - ubi_msg("\"preserve\" compatible internal volume %d:%d" - " found", vol_id, lnum); + ubi_msg("\"preserve\" compatible internal volume %d:%d found", + vol_id, lnum); err = add_to_list(ai, pnum, vol_id, lnum, ec, 0, &ai->alien); if (err) @@ -1075,10 +1074,10 @@ static int late_analysis(struct ubi_device *ubi, struct ubi_attach_info *ai) if (ai->corr_peb_count) { ubi_err("%d PEBs are corrupted and preserved", ai->corr_peb_count); - printk(KERN_ERR "Corrupted PEBs are:"); + pr_err("Corrupted PEBs are:"); list_for_each_entry(aeb, &ai->corr, u.list) - printk(KERN_CONT " %d", aeb->pnum); - printk(KERN_CONT "\n"); + pr_cont(" %d", aeb->pnum); + pr_cont("\n"); /* * If too many PEBs are corrupted, we refuse attaching, @@ -1112,8 +1111,7 @@ static int late_analysis(struct ubi_device *ubi, struct ubi_attach_info *ai) get_random_bytes(&ubi->image_seq, sizeof(ubi->image_seq)); } else { - ubi_err("MTD device is not UBI-formatted and possibly " - "contains non-UBI data - refusing it"); + ubi_err("MTD device is not UBI-formatted and possibly contains non-UBI data - refusing it"); return -EINVAL; } @@ -1172,7 +1170,7 @@ static struct ubi_attach_info *scan_all(struct ubi_device *ubi) goto out_vidh; } - dbg_msg("scanning is finished"); + ubi_msg("scanning is finished"); /* Calculate mean erase counter */ if (ai->ec_count) @@ -1244,7 +1242,7 @@ int ubi_attach(struct ubi_device *ubi) ubi->corr_peb_count = ai->corr_peb_count; ubi->max_ec = ai->max_ec; ubi->mean_ec = ai->mean_ec; - ubi_msg("max. sequence number: %llu", ai->max_sqnum); + dbg_gen("max. sequence number: %llu", ai->max_sqnum); err = ubi_read_volume_table(ubi, ai); if (err) diff --git a/drivers/mtd/ubi/build.c b/drivers/mtd/ubi/build.c index 2c5ed5c..3497703 100644 --- a/drivers/mtd/ubi/build.c +++ b/drivers/mtd/ubi/build.c @@ -36,6 +36,7 @@ #include <linux/namei.h> #include <linux/stat.h> #include <linux/miscdevice.h> +#include <linux/mtd/partitions.h> #include <linux/log2.h> #include <linux/kthread.h> #include <linux/kernel.h> @@ -45,6 +46,12 @@ /* Maximum length of the 'mtd=' parameter */ #define MTD_PARAM_LEN_MAX 64 +/* Maximum number of comma-separated items in the 'mtd=' parameter */ +#define MTD_PARAM_MAX_COUNT 3 + +/* Maximum value for the number of bad PEBs per 1024 PEBs */ +#define MAX_MTD_UBI_BEB_LIMIT 768 + #ifdef CONFIG_MTD_UBI_MODULE #define ubi_is_module() 1 #else @@ -56,10 +63,12 @@ * @name: MTD character device node path, MTD device name, or MTD device number * string * @vid_hdr_offs: VID header offset + * @max_beb_per1024: maximum expected number of bad PEBs per 1024 PEBs */ struct mtd_dev_param { char name[MTD_PARAM_LEN_MAX]; int vid_hdr_offs; + int max_beb_per1024; }; /* Numbers of elements set in the @mtd_dev_param array */ @@ -564,9 +573,38 @@ void ubi_free_internal_volumes(struct ubi_device *ubi) } } +static int get_bad_peb_limit(const struct ubi_device *ubi, int max_beb_per1024) +{ + int limit, device_pebs; + uint64_t device_size; + + if (!max_beb_per1024) + return 0; + + /* + * Here we are using size of the entire flash chip and + * not just the MTD partition size because the maximum + * number of bad eraseblocks is a percentage of the + * whole device and bad eraseblocks are not fairly + * distributed over the flash chip. So the worst case + * is that all the bad eraseblocks of the chip are in + * the MTD partition we are attaching (ubi->mtd). + */ + device_size = mtd_get_device_size(ubi->mtd); + device_pebs = mtd_div_by_eb(device_size, ubi->mtd); + limit = mult_frac(device_pebs, max_beb_per1024, 1024); + + /* Round it up */ + if (mult_frac(limit, 1024, max_beb_per1024) < device_pebs) + limit += 1; + + return limit; +} + /** * io_init - initialize I/O sub-system for a given UBI device. * @ubi: UBI device description object + * @max_beb_per1024: maximum expected number of bad PEB per 1024 PEBs * * If @ubi->vid_hdr_offset or @ubi->leb_start is zero, default offsets are * assumed: @@ -579,8 +617,11 @@ void ubi_free_internal_volumes(struct ubi_device *ubi) * This function returns zero in case of success and a negative error code in * case of failure. */ -static int io_init(struct ubi_device *ubi) +static int io_init(struct ubi_device *ubi, int max_beb_per1024) { + dbg_gen("sizeof(struct ubi_ainf_peb) %zu", sizeof(struct ubi_ainf_peb)); + dbg_gen("sizeof(struct ubi_wl_entry) %zu", sizeof(struct ubi_wl_entry)); + if (ubi->mtd->numeraseregions != 0) { /* * Some flashes have several erase regions. Different regions @@ -607,8 +648,10 @@ static int io_init(struct ubi_device *ubi) ubi->peb_count = mtd_div_by_eb(ubi->mtd->size, ubi->mtd); ubi->flash_size = ubi->mtd->size; - if (mtd_can_have_bb(ubi->mtd)) + if (mtd_can_have_bb(ubi->mtd)) { ubi->bad_allowed = 1; + ubi->bad_peb_limit = get_bad_peb_limit(ubi, max_beb_per1024); + } if (ubi->mtd->type == MTD_NORFLASH) { ubi_assert(ubi->mtd->writesize == 1); @@ -650,11 +693,11 @@ static int io_init(struct ubi_device *ubi) ubi->ec_hdr_alsize = ALIGN(UBI_EC_HDR_SIZE, ubi->hdrs_min_io_size); ubi->vid_hdr_alsize = ALIGN(UBI_VID_HDR_SIZE, ubi->hdrs_min_io_size); - dbg_msg("min_io_size %d", ubi->min_io_size); - dbg_msg("max_write_size %d", ubi->max_write_size); - dbg_msg("hdrs_min_io_size %d", ubi->hdrs_min_io_size); - dbg_msg("ec_hdr_alsize %d", ubi->ec_hdr_alsize); - dbg_msg("vid_hdr_alsize %d", ubi->vid_hdr_alsize); + dbg_gen("min_io_size %d", ubi->min_io_size); + dbg_gen("max_write_size %d", ubi->max_write_size); + dbg_gen("hdrs_min_io_size %d", ubi->hdrs_min_io_size); + dbg_gen("ec_hdr_alsize %d", ubi->ec_hdr_alsize); + dbg_gen("vid_hdr_alsize %d", ubi->vid_hdr_alsize); if (ubi->vid_hdr_offset == 0) /* Default offset */ @@ -671,10 +714,10 @@ static int io_init(struct ubi_device *ubi) ubi->leb_start = ubi->vid_hdr_offset + UBI_VID_HDR_SIZE; ubi->leb_start = ALIGN(ubi->leb_start, ubi->min_io_size); - dbg_msg("vid_hdr_offset %d", ubi->vid_hdr_offset); - dbg_msg("vid_hdr_aloffset %d", ubi->vid_hdr_aloffset); - dbg_msg("vid_hdr_shift %d", ubi->vid_hdr_shift); - dbg_msg("leb_start %d", ubi->leb_start); + dbg_gen("vid_hdr_offset %d", ubi->vid_hdr_offset); + dbg_gen("vid_hdr_aloffset %d", ubi->vid_hdr_aloffset); + dbg_gen("vid_hdr_shift %d", ubi->vid_hdr_shift); + dbg_gen("leb_start %d", ubi->leb_start); /* The shift must be aligned to 32-bit boundary */ if (ubi->vid_hdr_shift % 4) { @@ -700,7 +743,7 @@ static int io_init(struct ubi_device *ubi) ubi->max_erroneous = ubi->peb_count / 10; if (ubi->max_erroneous < 16) ubi->max_erroneous = 16; - dbg_msg("max_erroneous %d", ubi->max_erroneous); + dbg_gen("max_erroneous %d", ubi->max_erroneous); /* * It may happen that EC and VID headers are situated in one minimal @@ -708,30 +751,18 @@ static int io_init(struct ubi_device *ubi) * read-only mode. */ if (ubi->vid_hdr_offset + UBI_VID_HDR_SIZE <= ubi->hdrs_min_io_size) { - ubi_warn("EC and VID headers are in the same minimal I/O unit, " - "switch to read-only mode"); + ubi_warn("EC and VID headers are in the same minimal I/O unit, switch to read-only mode"); ubi->ro_mode = 1; } ubi->leb_size = ubi->peb_size - ubi->leb_start; if (!(ubi->mtd->flags & MTD_WRITEABLE)) { - ubi_msg("MTD device %d is write-protected, attach in " - "read-only mode", ubi->mtd->index); + ubi_msg("MTD device %d is write-protected, attach in read-only mode", + ubi->mtd->index); ubi->ro_mode = 1; } - ubi_msg("physical eraseblock size: %d bytes (%d KiB)", - ubi->peb_size, ubi->peb_size >> 10); - ubi_msg("logical eraseblock size: %d bytes", ubi->leb_size); - ubi_msg("smallest flash I/O unit: %d", ubi->min_io_size); - if (ubi->hdrs_min_io_size != ubi->min_io_size) - ubi_msg("sub-page size: %d", - ubi->hdrs_min_io_size); - ubi_msg("VID header offset: %d (aligned %d)", - ubi->vid_hdr_offset, ubi->vid_hdr_aloffset); - ubi_msg("data offset: %d", ubi->leb_start); - /* * Note, ideally, we have to initialize @ubi->bad_peb_count here. But * unfortunately, MTD does not provide this information. We should loop @@ -759,6 +790,11 @@ static int autoresize(struct ubi_device *ubi, int vol_id) struct ubi_volume *vol = ubi->volumes[vol_id]; int err, old_reserved_pebs = vol->reserved_pebs; + if (ubi->ro_mode) { + ubi_warn("skip auto-resize because of R/O mode"); + return 0; + } + /* * Clear the auto-resize flag in the volume in-memory copy of the * volume table, and 'ubi_resize_volume()' will propagate this change @@ -800,6 +836,7 @@ static int autoresize(struct ubi_device *ubi, int vol_id) * @mtd: MTD device description object * @ubi_num: number to assign to the new UBI device * @vid_hdr_offset: VID header offset + * @max_beb_per1024: maximum expected number of bad PEB per 1024 PEBs * * This function attaches MTD device @mtd_dev to UBI and assign @ubi_num number * to the newly created UBI device, unless @ubi_num is %UBI_DEV_NUM_AUTO, in @@ -810,11 +847,18 @@ static int autoresize(struct ubi_device *ubi, int vol_id) * Note, the invocations of this function has to be serialized by the * @ubi_devices_mutex. */ -int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) +int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, + int vid_hdr_offset, int max_beb_per1024) { struct ubi_device *ubi; int i, err, ref = 0; + if (max_beb_per1024 < 0 || max_beb_per1024 > MAX_MTD_UBI_BEB_LIMIT) + return -EINVAL; + + if (!max_beb_per1024) + max_beb_per1024 = CONFIG_MTD_UBI_BEB_LIMIT; + /* * Check if we already have the same MTD device attached. * @@ -839,8 +883,8 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) * no sense to attach emulated MTD devices, so we prohibit this. */ if (mtd->type == MTD_UBIVOLUME) { - ubi_err("refuse attaching mtd%d - it is already emulated on " - "top of UBI", mtd->index); + ubi_err("refuse attaching mtd%d - it is already emulated on top of UBI", + mtd->index); return -EINVAL; } @@ -880,10 +924,8 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) spin_lock_init(&ubi->volumes_lock); ubi_msg("attaching mtd%d to ubi%d", mtd->index, ubi_num); - dbg_msg("sizeof(struct ubi_ainf_peb) %zu", sizeof(struct ubi_ainf_peb)); - dbg_msg("sizeof(struct ubi_wl_entry) %zu", sizeof(struct ubi_wl_entry)); - err = io_init(ubi); + err = io_init(ubi, max_beb_per1024); if (err) goto out_free; @@ -924,23 +966,24 @@ int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset) goto out_debugfs; } - ubi_msg("attached mtd%d to ubi%d", mtd->index, ubi_num); - ubi_msg("MTD device name: \"%s\"", mtd->name); - ubi_msg("MTD device size: %llu MiB", ubi->flash_size >> 20); - ubi_msg("number of good PEBs: %d", ubi->good_peb_count); - ubi_msg("number of bad PEBs: %d", ubi->bad_peb_count); - ubi_msg("number of corrupted PEBs: %d", ubi->corr_peb_count); - ubi_msg("max. allowed volumes: %d", ubi->vtbl_slots); - ubi_msg("wear-leveling threshold: %d", CONFIG_MTD_UBI_WL_THRESHOLD); - ubi_msg("number of internal volumes: %d", UBI_INT_VOL_COUNT); - ubi_msg("number of user volumes: %d", - ubi->vol_count - UBI_INT_VOL_COUNT); - ubi_msg("available PEBs: %d", ubi->avail_pebs); - ubi_msg("total number of reserved PEBs: %d", ubi->rsvd_pebs); - ubi_msg("number of PEBs reserved for bad PEB handling: %d", - ubi->beb_rsvd_pebs); - ubi_msg("max/mean erase counter: %d/%d", ubi->max_ec, ubi->mean_ec); - ubi_msg("image sequence number: %d", ubi->image_seq); + ubi_msg("attached mtd%d (name \"%s\", size %llu MiB) to ubi%d", + mtd->index, mtd->name, ubi->flash_size >> 20, ubi_num); + ubi_msg("PEB size: %d bytes (%d KiB), LEB size: %d bytes", + ubi->peb_size, ubi->peb_size >> 10, ubi->leb_size); + ubi_msg("min./max. I/O unit sizes: %d/%d, sub-page size %d", + ubi->min_io_size, ubi->max_write_size, ubi->hdrs_min_io_size); + ubi_msg("VID header offset: %d (aligned %d), data offset: %d", + ubi->vid_hdr_offset, ubi->vid_hdr_aloffset, ubi->leb_start); + ubi_msg("good PEBs: %d, bad PEBs: %d, corrupted PEBs: %d", + ubi->good_peb_count, ubi->bad_peb_count, ubi->corr_peb_count); + ubi_msg("user volume: %d, internal volumes: %d, max. volumes count: %d", + ubi->vol_count - UBI_INT_VOL_COUNT, UBI_INT_VOL_COUNT, + ubi->vtbl_slots); + ubi_msg("max/mean erase counter: %d/%d, WL threshold: %d, image sequence number: %u", + ubi->max_ec, ubi->mean_ec, CONFIG_MTD_UBI_WL_THRESHOLD, + ubi->image_seq); + ubi_msg("available PEBs: %d, total reserved PEBs: %d, PEBs reserved for bad PEB handling: %d", + ubi->avail_pebs, ubi->rsvd_pebs, ubi->beb_rsvd_pebs); /* * The below lock makes sure we do not race with 'ubi_thread()' which @@ -1017,7 +1060,7 @@ int ubi_detach_mtd_dev(int ubi_num, int anyway) ubi_assert(ubi_num == ubi->ubi_num); ubi_notify_all(ubi, UBI_VOLUME_REMOVED, NULL); - dbg_msg("detaching mtd%d from ubi%d", ubi->mtd->index, ubi_num); + ubi_msg("detaching mtd%d from ubi%d", ubi->mtd->index, ubi_num); /* * Before freeing anything, we have to stop the background thread to @@ -1172,7 +1215,7 @@ static int __init ubi_init(void) mutex_lock(&ubi_devices_mutex); err = ubi_attach_mtd_dev(mtd, UBI_DEV_NUM_AUTO, - p->vid_hdr_offs); + p->vid_hdr_offs, p->max_beb_per1024); mutex_unlock(&ubi_devices_mutex); if (err < 0) { ubi_err("cannot attach mtd%d", mtd->index); @@ -1218,7 +1261,7 @@ out: ubi_err("UBI error: cannot initialize UBI, error %d", err); return err; } -module_init(ubi_init); +late_initcall(ubi_init); static void __exit ubi_exit(void) { @@ -1252,8 +1295,7 @@ static int __init bytes_str_to_int(const char *str) result = simple_strtoul(str, &endp, 0); if (str == endp || result >= INT_MAX) { - printk(KERN_ERR "UBI error: incorrect bytes count: \"%s\"\n", - str); + ubi_err("UBI error: incorrect bytes count: \"%s\"\n", str); return -EINVAL; } @@ -1269,8 +1311,7 @@ static int __init bytes_str_to_int(const char *str) case '\0': break; default: - printk(KERN_ERR "UBI error: incorrect bytes count: \"%s\"\n", - str); + ubi_err("UBI error: incorrect bytes count: \"%s\"\n", str); return -EINVAL; } @@ -1291,27 +1332,26 @@ static int __init ubi_mtd_param_parse(const char *val, struct kernel_param *kp) struct mtd_dev_param *p; char buf[MTD_PARAM_LEN_MAX]; char *pbuf = &buf[0]; - char *tokens[2] = {NULL, NULL}; + char *tokens[MTD_PARAM_MAX_COUNT]; if (!val) return -EINVAL; if (mtd_devs == UBI_MAX_DEVICES) { - printk(KERN_ERR "UBI error: too many parameters, max. is %d\n", - UBI_MAX_DEVICES); + ubi_err("UBI error: too many parameters, max. is %d\n", + UBI_MAX_DEVICES); return -EINVAL; } len = strnlen(val, MTD_PARAM_LEN_MAX); if (len == MTD_PARAM_LEN_MAX) { - printk(KERN_ERR "UBI error: parameter \"%s\" is too long, " - "max. is %d\n", val, MTD_PARAM_LEN_MAX); + ubi_err("UBI error: parameter \"%s\" is too long, max. is %d\n", + val, MTD_PARAM_LEN_MAX); return -EINVAL; } if (len == 0) { - printk(KERN_WARNING "UBI warning: empty 'mtd=' parameter - " - "ignored\n"); + pr_warn("UBI warning: empty 'mtd=' parameter - ignored\n"); return 0; } @@ -1321,12 +1361,11 @@ static int __init ubi_mtd_param_parse(const char *val, struct kernel_param *kp) if (buf[len - 1] == '\n') buf[len - 1] = '\0'; - for (i = 0; i < 2; i++) + for (i = 0; i < MTD_PARAM_MAX_COUNT; i++) tokens[i] = strsep(&pbuf, ","); if (pbuf) { - printk(KERN_ERR "UBI error: too many arguments at \"%s\"\n", - val); + ubi_err("UBI error: too many arguments at \"%s\"\n", val); return -EINVAL; } @@ -1339,23 +1378,32 @@ static int __init ubi_mtd_param_parse(const char *val, struct kernel_param *kp) if (p->vid_hdr_offs < 0) return p->vid_hdr_offs; + if (tokens[2]) { + int err = kstrtoint(tokens[2], 10, &p->max_beb_per1024); + + if (err) { + ubi_err("UBI error: bad value for max_beb_per1024 parameter: %s", + tokens[2]); + return -EINVAL; + } + } + mtd_devs += 1; return 0; } module_param_call(mtd, ubi_mtd_param_parse, NULL, NULL, 000); -MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: " - "mtd=<name|num|path>[,<vid_hdr_offs>].\n" +MODULE_PARM_DESC(mtd, "MTD devices to attach. Parameter format: mtd=<name|num|path>[,<vid_hdr_offs>[,max_beb_per1024]].\n" "Multiple \"mtd\" parameters may be specified.\n" - "MTD devices may be specified by their number, name, or " - "path to the MTD character device node.\n" - "Optional \"vid_hdr_offs\" parameter specifies UBI VID " - "header position to be used by UBI.\n" - "Example 1: mtd=/dev/mtd0 - attach MTD device " - "/dev/mtd0.\n" - "Example 2: mtd=content,1984 mtd=4 - attach MTD device " - "with name \"content\" using VID header offset 1984, and " - "MTD device number 4 with default VID header offset."); + "MTD devices may be specified by their number, name, or path to the MTD character device node.\n" + "Optional \"vid_hdr_offs\" parameter specifies UBI VID header position to be used by UBI. (default value if 0)\n" + "Optional \"max_beb_per1024\" parameter specifies the maximum expected bad eraseblock per 1024 eraseblocks. (default value (" + __stringify(CONFIG_MTD_UBI_BEB_LIMIT) ") if 0)\n" + "\n" + "Example 1: mtd=/dev/mtd0 - attach MTD device /dev/mtd0.\n" + "Example 2: mtd=content,1984 mtd=4 - attach MTD device with name \"content\" using VID header offset 1984, and MTD device number 4 with default VID header offset.\n" + "Example 3: mtd=/dev/mtd1,0,25 - attach MTD device /dev/mtd1 using default VID header offset and reserve 25*nand_size_in_blocks/1024 erase blocks for bad block handling.\n" + "\t(e.g. if the NAND *chipset* has 4096 PEB, 100 will be reserved for this UBI device)."); MODULE_VERSION(__stringify(UBI_VERSION)); MODULE_DESCRIPTION("UBI - Unsorted Block Images"); diff --git a/drivers/mtd/ubi/cdev.c b/drivers/mtd/ubi/cdev.c index fb55678..dfcc65b 100644 --- a/drivers/mtd/ubi/cdev.c +++ b/drivers/mtd/ubi/cdev.c @@ -140,9 +140,9 @@ static int vol_cdev_release(struct inode *inode, struct file *file) vol->updating = 0; vfree(vol->upd_buf); } else if (vol->changing_leb) { - dbg_gen("only %lld of %lld bytes received for atomic LEB change" - " for volume %d:%d, cancel", vol->upd_received, - vol->upd_bytes, vol->ubi->ubi_num, vol->vol_id); + dbg_gen("only %lld of %lld bytes received for atomic LEB change for volume %d:%d, cancel", + vol->upd_received, vol->upd_bytes, vol->ubi->ubi_num, + vol->vol_id); vol->changing_leb = 0; vfree(vol->upd_buf); } @@ -189,7 +189,8 @@ static loff_t vol_cdev_llseek(struct file *file, loff_t offset, int origin) return new_offset; } -static int vol_cdev_fsync(struct file *file, loff_t start, loff_t end, int datasync) +static int vol_cdev_fsync(struct file *file, loff_t start, loff_t end, + int datasync) { struct ubi_volume_desc *desc = file->private_data; struct ubi_device *ubi = desc->vol->ubi; @@ -753,7 +754,7 @@ static int rename_volumes(struct ubi_device *ubi, re->new_name_len = name_len; memcpy(re->new_name, name, name_len); list_add_tail(&re->list, &rename_list); - dbg_msg("will rename volume %d from \"%s\" to \"%s\"", + dbg_gen("will rename volume %d from \"%s\" to \"%s\"", vol_id, re->desc->vol->name, name); } @@ -811,7 +812,7 @@ static int rename_volumes(struct ubi_device *ubi, re1->remove = 1; re1->desc = desc; list_add(&re1->list, &rename_list); - dbg_msg("will remove volume %d, name \"%s\"", + dbg_gen("will remove volume %d, name \"%s\"", re1->desc->vol->vol_id, re1->desc->vol->name); } @@ -942,7 +943,7 @@ static long ubi_cdev_ioctl(struct file *file, unsigned int cmd, { struct ubi_rnvol_req *req; - dbg_msg("re-name volumes"); + dbg_gen("re-name volumes"); req = kmalloc(sizeof(struct ubi_rnvol_req), GFP_KERNEL); if (!req) { err = -ENOMEM; @@ -1010,7 +1011,8 @@ static long ctrl_cdev_ioctl(struct file *file, unsigned int cmd, * 'ubi_attach_mtd_dev()'. */ mutex_lock(&ubi_devices_mutex); - err = ubi_attach_mtd_dev(mtd, req.ubi_num, req.vid_hdr_offset); + err = ubi_attach_mtd_dev(mtd, req.ubi_num, req.vid_hdr_offset, + req.max_beb_per1024); mutex_unlock(&ubi_devices_mutex); if (err < 0) put_mtd_device(mtd); diff --git a/drivers/mtd/ubi/debug.c b/drivers/mtd/ubi/debug.c index 7c13803..26908a5 100644 --- a/drivers/mtd/ubi/debug.c +++ b/drivers/mtd/ubi/debug.c @@ -43,8 +43,8 @@ void ubi_dump_flash(struct ubi_device *ubi, int pnum, int offset, int len) return; err = mtd_read(ubi->mtd, addr, len, &read, buf); if (err && err != -EUCLEAN) { - ubi_err("error %d while reading %d bytes from PEB %d:%d, " - "read %zd bytes", err, len, pnum, offset, read); + ubi_err("error %d while reading %d bytes from PEB %d:%d, read %zd bytes", + err, len, pnum, offset, read); goto out; } @@ -62,21 +62,15 @@ out: */ void ubi_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr) { - printk(KERN_DEBUG "Erase counter header dump:\n"); - printk(KERN_DEBUG "\tmagic %#08x\n", - be32_to_cpu(ec_hdr->magic)); - printk(KERN_DEBUG "\tversion %d\n", (int)ec_hdr->version); - printk(KERN_DEBUG "\tec %llu\n", - (long long)be64_to_cpu(ec_hdr->ec)); - printk(KERN_DEBUG "\tvid_hdr_offset %d\n", - be32_to_cpu(ec_hdr->vid_hdr_offset)); - printk(KERN_DEBUG "\tdata_offset %d\n", - be32_to_cpu(ec_hdr->data_offset)); - printk(KERN_DEBUG "\timage_seq %d\n", - be32_to_cpu(ec_hdr->image_seq)); - printk(KERN_DEBUG "\thdr_crc %#08x\n", - be32_to_cpu(ec_hdr->hdr_crc)); - printk(KERN_DEBUG "erase counter header hexdump:\n"); + pr_err("Erase counter header dump:\n"); + pr_err("\tmagic %#08x\n", be32_to_cpu(ec_hdr->magic)); + pr_err("\tversion %d\n", (int)ec_hdr->version); + pr_err("\tec %llu\n", (long long)be64_to_cpu(ec_hdr->ec)); + pr_err("\tvid_hdr_offset %d\n", be32_to_cpu(ec_hdr->vid_hdr_offset)); + pr_err("\tdata_offset %d\n", be32_to_cpu(ec_hdr->data_offset)); + pr_err("\timage_seq %d\n", be32_to_cpu(ec_hdr->image_seq)); + pr_err("\thdr_crc %#08x\n", be32_to_cpu(ec_hdr->hdr_crc)); + pr_err("erase counter header hexdump:\n"); print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, ec_hdr, UBI_EC_HDR_SIZE, 1); } @@ -87,21 +81,21 @@ void ubi_dump_ec_hdr(const struct ubi_ec_hdr *ec_hdr) */ void ubi_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr) { - printk(KERN_DEBUG "Volume identifier header dump:\n"); - printk(KERN_DEBUG "\tmagic %08x\n", be32_to_cpu(vid_hdr->magic)); - printk(KERN_DEBUG "\tversion %d\n", (int)vid_hdr->version); - printk(KERN_DEBUG "\tvol_type %d\n", (int)vid_hdr->vol_type); - printk(KERN_DEBUG "\tcopy_flag %d\n", (int)vid_hdr->copy_flag); - printk(KERN_DEBUG "\tcompat %d\n", (int)vid_hdr->compat); - printk(KERN_DEBUG "\tvol_id %d\n", be32_to_cpu(vid_hdr->vol_id)); - printk(KERN_DEBUG "\tlnum %d\n", be32_to_cpu(vid_hdr->lnum)); - printk(KERN_DEBUG "\tdata_size %d\n", be32_to_cpu(vid_hdr->data_size)); - printk(KERN_DEBUG "\tused_ebs %d\n", be32_to_cpu(vid_hdr->used_ebs)); - printk(KERN_DEBUG "\tdata_pad %d\n", be32_to_cpu(vid_hdr->data_pad)); - printk(KERN_DEBUG "\tsqnum %llu\n", + pr_err("Volume identifier header dump:\n"); + pr_err("\tmagic %08x\n", be32_to_cpu(vid_hdr->magic)); + pr_err("\tversion %d\n", (int)vid_hdr->version); + pr_err("\tvol_type %d\n", (int)vid_hdr->vol_type); + pr_err("\tcopy_flag %d\n", (int)vid_hdr->copy_flag); + pr_err("\tcompat %d\n", (int)vid_hdr->compat); + pr_err("\tvol_id %d\n", be32_to_cpu(vid_hdr->vol_id)); + pr_err("\tlnum %d\n", be32_to_cpu(vid_hdr->lnum)); + pr_err("\tdata_size %d\n", be32_to_cpu(vid_hdr->data_size)); + pr_err("\tused_ebs %d\n", be32_to_cpu(vid_hdr->used_ebs)); + pr_err("\tdata_pad %d\n", be32_to_cpu(vid_hdr->data_pad)); + pr_err("\tsqnum %llu\n", (unsigned long long)be64_to_cpu(vid_hdr->sqnum)); - printk(KERN_DEBUG "\thdr_crc %08x\n", be32_to_cpu(vid_hdr->hdr_crc)); - printk(KERN_DEBUG "Volume identifier header hexdump:\n"); + pr_err("\thdr_crc %08x\n", be32_to_cpu(vid_hdr->hdr_crc)); + pr_err("Volume identifier header hexdump:\n"); print_hex_dump(KERN_DEBUG, "", DUMP_PREFIX_OFFSET, 32, 1, vid_hdr, UBI_VID_HDR_SIZE, 1); } @@ -112,25 +106,25 @@ void ubi_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr) */ void ubi_dump_vol_info(const struct ubi_volume *vol) { - printk(KERN_DEBUG "Volume information dump:\n"); - printk(KERN_DEBUG "\tvol_id %d\n", vol->vol_id); - printk(KERN_DEBUG "\treserved_pebs %d\n", vol->reserved_pebs); - printk(KERN_DEBUG "\talignment %d\n", vol->alignment); - printk(KERN_DEBUG "\tdata_pad %d\n", vol->data_pad); - printk(KERN_DEBUG "\tvol_type %d\n", vol->vol_type); - printk(KERN_DEBUG "\tname_len %d\n", vol->name_len); - printk(KERN_DEBUG "\tusable_leb_size %d\n", vol->usable_leb_size); - printk(KERN_DEBUG "\tused_ebs %d\n", vol->used_ebs); - printk(KERN_DEBUG "\tused_bytes %lld\n", vol->used_bytes); - printk(KERN_DEBUG "\tlast_eb_bytes %d\n", vol->last_eb_bytes); - printk(KERN_DEBUG "\tcorrupted %d\n", vol->corrupted); - printk(KERN_DEBUG "\tupd_marker %d\n", vol->upd_marker); + pr_err("Volume information dump:\n"); + pr_err("\tvol_id %d\n", vol->vol_id); + pr_err("\treserved_pebs %d\n", vol->reserved_pebs); + pr_err("\talignment %d\n", vol->alignment); + pr_err("\tdata_pad %d\n", vol->data_pad); + pr_err("\tvol_type %d\n", vol->vol_type); + pr_err("\tname_len %d\n", vol->name_len); + pr_err("\tusable_leb_size %d\n", vol->usable_leb_size); + pr_err("\tused_ebs %d\n", vol->used_ebs); + pr_err("\tused_bytes %lld\n", vol->used_bytes); + pr_err("\tlast_eb_bytes %d\n", vol->last_eb_bytes); + pr_err("\tcorrupted %d\n", vol->corrupted); + pr_err("\tupd_marker %d\n", vol->upd_marker); if (vol->name_len <= UBI_VOL_NAME_MAX && strnlen(vol->name, vol->name_len + 1) == vol->name_len) { - printk(KERN_DEBUG "\tname %s\n", vol->name); + pr_err("\tname %s\n", vol->name); } else { - printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n", + pr_err("\t1st 5 characters of name: %c%c%c%c%c\n", vol->name[0], vol->name[1], vol->name[2], vol->name[3], vol->name[4]); } @@ -145,29 +139,28 @@ void ubi_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx) { int name_len = be16_to_cpu(r->name_len); - printk(KERN_DEBUG "Volume table record %d dump:\n", idx); - printk(KERN_DEBUG "\treserved_pebs %d\n", - be32_to_cpu(r->reserved_pebs)); - printk(KERN_DEBUG "\talignment %d\n", be32_to_cpu(r->alignment)); - printk(KERN_DEBUG "\tdata_pad %d\n", be32_to_cpu(r->data_pad)); - printk(KERN_DEBUG "\tvol_type %d\n", (int)r->vol_type); - printk(KERN_DEBUG "\tupd_marker %d\n", (int)r->upd_marker); - printk(KERN_DEBUG "\tname_len %d\n", name_len); + pr_err("Volume table record %d dump:\n", idx); + pr_err("\treserved_pebs %d\n", be32_to_cpu(r->reserved_pebs)); + pr_err("\talignment %d\n", be32_to_cpu(r->alignment)); + pr_err("\tdata_pad %d\n", be32_to_cpu(r->data_pad)); + pr_err("\tvol_type %d\n", (int)r->vol_type); + pr_err("\tupd_marker %d\n", (int)r->upd_marker); + pr_err("\tname_len %d\n", name_len); if (r->name[0] == '\0') { - printk(KERN_DEBUG "\tname NULL\n"); + pr_err("\tname NULL\n"); return; } if (name_len <= UBI_VOL_NAME_MAX && strnlen(&r->name[0], name_len + 1) == name_len) { - printk(KERN_DEBUG "\tname %s\n", &r->name[0]); + pr_err("\tname %s\n", &r->name[0]); } else { - printk(KERN_DEBUG "\t1st 5 characters of name: %c%c%c%c%c\n", + pr_err("\t1st 5 characters of name: %c%c%c%c%c\n", r->name[0], r->name[1], r->name[2], r->name[3], r->name[4]); } - printk(KERN_DEBUG "\tcrc %#08x\n", be32_to_cpu(r->crc)); + pr_err("\tcrc %#08x\n", be32_to_cpu(r->crc)); } /** @@ -176,15 +169,15 @@ void ubi_dump_vtbl_record(const struct ubi_vtbl_record *r, int idx) */ void ubi_dump_av(const struct ubi_ainf_volume *av) { - printk(KERN_DEBUG "Volume attaching information dump:\n"); - printk(KERN_DEBUG "\tvol_id %d\n", av->vol_id); - printk(KERN_DEBUG "\thighest_lnum %d\n", av->highest_lnum); - printk(KERN_DEBUG "\tleb_count %d\n", av->leb_count); - printk(KERN_DEBUG "\tcompat %d\n", av->compat); - printk(KERN_DEBUG "\tvol_type %d\n", av->vol_type); - printk(KERN_DEBUG "\tused_ebs %d\n", av->used_ebs); - printk(KERN_DEBUG "\tlast_data_size %d\n", av->last_data_size); - printk(KERN_DEBUG "\tdata_pad %d\n", av->data_pad); + pr_err("Volume attaching information dump:\n"); + pr_err("\tvol_id %d\n", av->vol_id); + pr_err("\thighest_lnum %d\n", av->highest_lnum); + pr_err("\tleb_count %d\n", av->leb_count); + pr_err("\tcompat %d\n", av->compat); + pr_err("\tvol_type %d\n", av->vol_type); + pr_err("\tused_ebs %d\n", av->used_ebs); + pr_err("\tlast_data_size %d\n", av->last_data_size); + pr_err("\tdata_pad %d\n", av->data_pad); } /** @@ -194,13 +187,13 @@ void ubi_dump_av(const struct ubi_ainf_volume *av) */ void ubi_dump_aeb(const struct ubi_ainf_peb *aeb, int type) { - printk(KERN_DEBUG "eraseblock attaching information dump:\n"); - printk(KERN_DEBUG "\tec %d\n", aeb->ec); - printk(KERN_DEBUG "\tpnum %d\n", aeb->pnum); + pr_err("eraseblock attaching information dump:\n"); + pr_err("\tec %d\n", aeb->ec); + pr_err("\tpnum %d\n", aeb->pnum); if (type == 0) { - printk(KERN_DEBUG "\tlnum %d\n", aeb->lnum); - printk(KERN_DEBUG "\tscrub %d\n", aeb->scrub); - printk(KERN_DEBUG "\tsqnum %llu\n", aeb->sqnum); + pr_err("\tlnum %d\n", aeb->lnum); + pr_err("\tscrub %d\n", aeb->scrub); + pr_err("\tsqnum %llu\n", aeb->sqnum); } } @@ -212,16 +205,16 @@ void ubi_dump_mkvol_req(const struct ubi_mkvol_req *req) { char nm[17]; - printk(KERN_DEBUG "Volume creation request dump:\n"); - printk(KERN_DEBUG "\tvol_id %d\n", req->vol_id); - printk(KERN_DEBUG "\talignment %d\n", req->alignment); - printk(KERN_DEBUG "\tbytes %lld\n", (long long)req->bytes); - printk(KERN_DEBUG "\tvol_type %d\n", req->vol_type); - printk(KERN_DEBUG "\tname_len %d\n", req->name_len); + pr_err("Volume creation request dump:\n"); + pr_err("\tvol_id %d\n", req->vol_id); + pr_err("\talignment %d\n", req->alignment); + pr_err("\tbytes %lld\n", (long long)req->bytes); + pr_err("\tvol_type %d\n", req->vol_type); + pr_err("\tname_len %d\n", req->name_len); memcpy(nm, req->name, 16); nm[16] = 0; - printk(KERN_DEBUG "\t1st 16 characters of name: %s\n", nm); + pr_err("\t1st 16 characters of name: %s\n", nm); } /** diff --git a/drivers/mtd/ubi/debug.h b/drivers/mtd/ubi/debug.h index d5d2645..3dbc877 100644 --- a/drivers/mtd/ubi/debug.h +++ b/drivers/mtd/ubi/debug.h @@ -29,22 +29,18 @@ void ubi_dump_vid_hdr(const struct ubi_vid_hdr *vid_hdr); #define ubi_assert(expr) do { \ if (unlikely(!(expr))) { \ - printk(KERN_CRIT "UBI assert failed in %s at %u (pid %d)\n", \ + pr_crit("UBI assert failed in %s at %u (pid %d)\n", \ __func__, __LINE__, current->pid); \ dump_stack(); \ } \ } while (0) -#define ubi_dbg_print_hex_dump(l, ps, pt, r, g, b, len, a) \ +#define ubi_dbg_print_hex_dump(l, ps, pt, r, g, b, len, a) \ print_hex_dump(l, ps, pt, r, g, b, len, a) #define ubi_dbg_msg(type, fmt, ...) \ - pr_debug("UBI DBG " type ": " fmt "\n", ##__VA_ARGS__) - -/* Just a debugging messages not related to any specific UBI subsystem */ -#define dbg_msg(fmt, ...) \ - printk(KERN_DEBUG "UBI DBG (pid %d): %s: " fmt "\n", \ - current->pid, __func__, ##__VA_ARGS__) + pr_debug("UBI DBG " type " (pid %d): " fmt "\n", current->pid, \ + ##__VA_ARGS__) /* General debugging messages */ #define dbg_gen(fmt, ...) ubi_dbg_msg("gen", fmt, ##__VA_ARGS__) diff --git a/drivers/mtd/ubi/eba.c b/drivers/mtd/ubi/eba.c index b703ac7..a26d7d2 100644 --- a/drivers/mtd/ubi/eba.c +++ b/drivers/mtd/ubi/eba.c @@ -420,9 +420,8 @@ retry: */ if (err == UBI_IO_BAD_HDR_EBADMSG || err == UBI_IO_BAD_HDR) { - ubi_warn("corrupted VID header at PEB " - "%d, LEB %d:%d", pnum, vol_id, - lnum); + ubi_warn("corrupted VID header at PEB %d, LEB %d:%d", + pnum, vol_id, lnum); err = -EBADMSG; } else ubi_ro_mode(ubi); @@ -660,9 +659,8 @@ retry: if (len) { err = ubi_io_write_data(ubi, buf, pnum, offset, len); if (err) { - ubi_warn("failed to write %d bytes at offset %d of " - "LEB %d:%d, PEB %d", len, offset, vol_id, - lnum, pnum); + ubi_warn("failed to write %d bytes at offset %d of LEB %d:%d, PEB %d", + len, offset, vol_id, lnum, pnum); goto write_error; } } @@ -1040,9 +1038,8 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, * cancel it. */ if (vol->eba_tbl[lnum] != from) { - dbg_wl("LEB %d:%d is no longer mapped to PEB %d, mapped to " - "PEB %d, cancel", vol_id, lnum, from, - vol->eba_tbl[lnum]); + dbg_wl("LEB %d:%d is no longer mapped to PEB %d, mapped to PEB %d, cancel", + vol_id, lnum, from, vol->eba_tbl[lnum]); err = MOVE_CANCEL_RACE; goto out_unlock_leb; } @@ -1107,8 +1104,8 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, err = ubi_io_read_vid_hdr(ubi, to, vid_hdr, 1); if (err) { if (err != UBI_IO_BITFLIPS) { - ubi_warn("error %d while reading VID header back from " - "PEB %d", err, to); + ubi_warn("error %d while reading VID header back from PEB %d", + err, to); if (is_error_sane(err)) err = MOVE_TARGET_RD_ERR; } else @@ -1134,8 +1131,8 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, err = ubi_io_read_data(ubi, ubi->peb_buf, to, 0, aldata_size); if (err) { if (err != UBI_IO_BITFLIPS) { - ubi_warn("error %d while reading data back " - "from PEB %d", err, to); + ubi_warn("error %d while reading data back from PEB %d", + err, to); if (is_error_sane(err)) err = MOVE_TARGET_RD_ERR; } else @@ -1146,8 +1143,8 @@ int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to, cond_resched(); if (crc != crc32(UBI_CRC32_INIT, ubi->peb_buf, data_size)) { - ubi_warn("read data back from PEB %d and it is " - "different", to); + ubi_warn("read data back from PEB %d and it is different", + to); err = -EINVAL; goto out_unlock_buf; } @@ -1197,11 +1194,11 @@ static void print_rsvd_warning(struct ubi_device *ubi, return; } - ubi_warn("cannot reserve enough PEBs for bad PEB handling, reserved %d," - " need %d", ubi->beb_rsvd_pebs, ubi->beb_rsvd_level); + ubi_warn("cannot reserve enough PEBs for bad PEB handling, reserved %d, need %d", + ubi->beb_rsvd_pebs, ubi->beb_rsvd_level); if (ubi->corr_peb_count) ubi_warn("%d PEBs are corrupted and not used", - ubi->corr_peb_count); + ubi->corr_peb_count); } /** diff --git a/drivers/mtd/ubi/gluebi.c b/drivers/mtd/ubi/gluebi.c index 4e44bee..4bd4db8 100644 --- a/drivers/mtd/ubi/gluebi.c +++ b/drivers/mtd/ubi/gluebi.c @@ -41,7 +41,7 @@ #include "ubi-media.h" #define err_msg(fmt, ...) \ - printk(KERN_DEBUG "gluebi (pid %d): %s: " fmt "\n", \ + pr_err("gluebi (pid %d): %s: " fmt "\n", \ current->pid, __func__, ##__VA_ARGS__) /** @@ -341,9 +341,8 @@ static int gluebi_create(struct ubi_device_info *di, mutex_lock(&devices_mutex); g = find_gluebi_nolock(vi->ubi_num, vi->vol_id); if (g) - err_msg("gluebi MTD device %d form UBI device %d volume %d " - "already exists", g->mtd.index, vi->ubi_num, - vi->vol_id); + err_msg("gluebi MTD device %d form UBI device %d volume %d already exists", + g->mtd.index, vi->ubi_num, vi->vol_id); mutex_unlock(&devices_mutex); if (mtd_device_register(mtd, NULL, 0)) { @@ -376,8 +375,8 @@ static int gluebi_remove(struct ubi_volume_info *vi) mutex_lock(&devices_mutex); gluebi = find_gluebi_nolock(vi->ubi_num, vi->vol_id); if (!gluebi) { - err_msg("got remove notification for unknown UBI device %d " - "volume %d", vi->ubi_num, vi->vol_id); + err_msg("got remove notification for unknown UBI device %d volume %d", + vi->ubi_num, vi->vol_id); err = -ENOENT; } else if (gluebi->refcnt) err = -EBUSY; @@ -390,9 +389,8 @@ static int gluebi_remove(struct ubi_volume_info *vi) mtd = &gluebi->mtd; err = mtd_device_unregister(mtd); if (err) { - err_msg("cannot remove fake MTD device %d, UBI device %d, " - "volume %d, error %d", mtd->index, gluebi->ubi_num, - gluebi->vol_id, err); + err_msg("cannot remove fake MTD device %d, UBI device %d, volume %d, error %d", + mtd->index, gluebi->ubi_num, gluebi->vol_id, err); mutex_lock(&devices_mutex); list_add_tail(&gluebi->list, &gluebi_devices); mutex_unlock(&devices_mutex); @@ -422,8 +420,8 @@ static int gluebi_updated(struct ubi_volume_info *vi) gluebi = find_gluebi_nolock(vi->ubi_num, vi->vol_id); if (!gluebi) { mutex_unlock(&devices_mutex); - err_msg("got update notification for unknown UBI device %d " - "volume %d", vi->ubi_num, vi->vol_id); + err_msg("got update notification for unknown UBI device %d volume %d", + vi->ubi_num, vi->vol_id); return -ENOENT; } @@ -449,8 +447,8 @@ static int gluebi_resized(struct ubi_volume_info *vi) gluebi = find_gluebi_nolock(vi->ubi_num, vi->vol_id); if (!gluebi) { mutex_unlock(&devices_mutex); - err_msg("got update notification for unknown UBI device %d " - "volume %d", vi->ubi_num, vi->vol_id); + err_msg("got update notification for unknown UBI device %d volume %d", + vi->ubi_num, vi->vol_id); return -ENOENT; } gluebi->mtd.size = vi->used_bytes; @@ -507,9 +505,9 @@ static void __exit ubi_gluebi_exit(void) err = mtd_device_unregister(mtd); if (err) - err_msg("error %d while removing gluebi MTD device %d, " - "UBI device %d, volume %d - ignoring", err, - mtd->index, gluebi->ubi_num, gluebi->vol_id); + err_msg("error %d while removing gluebi MTD device %d, UBI device %d, volume %d - ignoring", + err, mtd->index, gluebi->ubi_num, + gluebi->vol_id); kfree(mtd->name); kfree(gluebi); } diff --git a/drivers/mtd/ubi/io.c b/drivers/mtd/ubi/io.c index a8d5237..78a1dcb 100644 --- a/drivers/mtd/ubi/io.c +++ b/drivers/mtd/ubi/io.c @@ -177,21 +177,20 @@ retry: * enabled. A corresponding message will be printed * later, when it is has been scrubbed. */ - dbg_msg("fixable bit-flip detected at PEB %d", pnum); + ubi_msg("fixable bit-flip detected at PEB %d", pnum); ubi_assert(len == read); return UBI_IO_BITFLIPS; } if (retries++ < UBI_IO_RETRIES) { - ubi_warn("error %d%s while reading %d bytes from PEB " - "%d:%d, read only %zd bytes, retry", + ubi_warn("error %d%s while reading %d bytes from PEB %d:%d, read only %zd bytes, retry", err, errstr, len, pnum, offset, read); yield(); goto retry; } - ubi_err("error %d%s while reading %d bytes from PEB %d:%d, " - "read %zd bytes", err, errstr, len, pnum, offset, read); + ubi_err("error %d%s while reading %d bytes from PEB %d:%d, read %zd bytes", + err, errstr, len, pnum, offset, read); dump_stack(); /* @@ -274,8 +273,8 @@ int ubi_io_write(struct ubi_device *ubi, const void *buf, int pnum, int offset, } if (ubi_dbg_is_write_failure(ubi)) { - ubi_err("cannot write %d bytes to PEB %d:%d " - "(emulated)", len, pnum, offset); + ubi_err("cannot write %d bytes to PEB %d:%d (emulated)", + len, pnum, offset); dump_stack(); return -EIO; } @@ -283,8 +282,8 @@ int ubi_io_write(struct ubi_device *ubi, const void *buf, int pnum, int offset, addr = (loff_t)pnum * ubi->peb_size + offset; err = mtd_write(ubi->mtd, addr, len, &written, buf); if (err) { - ubi_err("error %d while writing %d bytes to PEB %d:%d, written " - "%zd bytes", err, len, pnum, offset, written); + ubi_err("error %d while writing %d bytes to PEB %d:%d, written %zd bytes", + err, len, pnum, offset, written); dump_stack(); ubi_dump_flash(ubi, pnum, offset, len); } else @@ -685,8 +684,7 @@ static int validate_ec_hdr(const struct ubi_device *ubi, leb_start = be32_to_cpu(ec_hdr->data_offset); if (ec_hdr->version != UBI_VERSION) { - ubi_err("node with incompatible UBI version found: " - "this UBI version is %d, image version is %d", + ubi_err("node with incompatible UBI version found: this UBI version is %d, image version is %d", UBI_VERSION, (int)ec_hdr->version); goto bad; } @@ -777,10 +775,10 @@ int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum, if (ubi_check_pattern(ec_hdr, 0xFF, UBI_EC_HDR_SIZE)) { /* The physical eraseblock is supposedly empty */ if (verbose) - ubi_warn("no EC header found at PEB %d, " - "only 0xFF bytes", pnum); - dbg_bld("no EC header found at PEB %d, " - "only 0xFF bytes", pnum); + ubi_warn("no EC header found at PEB %d, only 0xFF bytes", + pnum); + dbg_bld("no EC header found at PEB %d, only 0xFF bytes", + pnum); if (!read_err) return UBI_IO_FF; else @@ -792,12 +790,12 @@ int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum, * 0xFF bytes. Report that the header is corrupted. */ if (verbose) { - ubi_warn("bad magic number at PEB %d: %08x instead of " - "%08x", pnum, magic, UBI_EC_HDR_MAGIC); + ubi_warn("bad magic number at PEB %d: %08x instead of %08x", + pnum, magic, UBI_EC_HDR_MAGIC); ubi_dump_ec_hdr(ec_hdr); } - dbg_bld("bad magic number at PEB %d: %08x instead of " - "%08x", pnum, magic, UBI_EC_HDR_MAGIC); + dbg_bld("bad magic number at PEB %d: %08x instead of %08x", + pnum, magic, UBI_EC_HDR_MAGIC); return UBI_IO_BAD_HDR; } @@ -806,12 +804,12 @@ int ubi_io_read_ec_hdr(struct ubi_device *ubi, int pnum, if (hdr_crc != crc) { if (verbose) { - ubi_warn("bad EC header CRC at PEB %d, calculated " - "%#08x, read %#08x", pnum, crc, hdr_crc); + ubi_warn("bad EC header CRC at PEB %d, calculated %#08x, read %#08x", + pnum, crc, hdr_crc); ubi_dump_ec_hdr(ec_hdr); } - dbg_bld("bad EC header CRC at PEB %d, calculated " - "%#08x, read %#08x", pnum, crc, hdr_crc); + dbg_bld("bad EC header CRC at PEB %d, calculated %#08x, read %#08x", + pnum, crc, hdr_crc); if (!read_err) return UBI_IO_BAD_HDR; @@ -1032,10 +1030,10 @@ int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum, if (ubi_check_pattern(vid_hdr, 0xFF, UBI_VID_HDR_SIZE)) { if (verbose) - ubi_warn("no VID header found at PEB %d, " - "only 0xFF bytes", pnum); - dbg_bld("no VID header found at PEB %d, " - "only 0xFF bytes", pnum); + ubi_warn("no VID header found at PEB %d, only 0xFF bytes", + pnum); + dbg_bld("no VID header found at PEB %d, only 0xFF bytes", + pnum); if (!read_err) return UBI_IO_FF; else @@ -1043,12 +1041,12 @@ int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum, } if (verbose) { - ubi_warn("bad magic number at PEB %d: %08x instead of " - "%08x", pnum, magic, UBI_VID_HDR_MAGIC); + ubi_warn("bad magic number at PEB %d: %08x instead of %08x", + pnum, magic, UBI_VID_HDR_MAGIC); ubi_dump_vid_hdr(vid_hdr); } - dbg_bld("bad magic number at PEB %d: %08x instead of " - "%08x", pnum, magic, UBI_VID_HDR_MAGIC); + dbg_bld("bad magic number at PEB %d: %08x instead of %08x", + pnum, magic, UBI_VID_HDR_MAGIC); return UBI_IO_BAD_HDR; } @@ -1057,12 +1055,12 @@ int ubi_io_read_vid_hdr(struct ubi_device *ubi, int pnum, if (hdr_crc != crc) { if (verbose) { - ubi_warn("bad CRC at PEB %d, calculated %#08x, " - "read %#08x", pnum, crc, hdr_crc); + ubi_warn("bad CRC at PEB %d, calculated %#08x, read %#08x", + pnum, crc, hdr_crc); ubi_dump_vid_hdr(vid_hdr); } - dbg_bld("bad CRC at PEB %d, calculated %#08x, " - "read %#08x", pnum, crc, hdr_crc); + dbg_bld("bad CRC at PEB %d, calculated %#08x, read %#08x", + pnum, crc, hdr_crc); if (!read_err) return UBI_IO_BAD_HDR; else @@ -1300,8 +1298,8 @@ static int self_check_peb_vid_hdr(const struct ubi_device *ubi, int pnum) crc = crc32(UBI_CRC32_INIT, vid_hdr, UBI_EC_HDR_SIZE_CRC); hdr_crc = be32_to_cpu(vid_hdr->hdr_crc); if (hdr_crc != crc) { - ubi_err("bad VID header CRC at PEB %d, calculated %#08x, " - "read %#08x", pnum, crc, hdr_crc); + ubi_err("bad VID header CRC at PEB %d, calculated %#08x, read %#08x", + pnum, crc, hdr_crc); ubi_err("self-check failed for PEB %d", pnum); ubi_dump_vid_hdr(vid_hdr); dump_stack(); @@ -1411,15 +1409,15 @@ int ubi_self_check_all_ff(struct ubi_device *ubi, int pnum, int offset, int len) err = mtd_read(ubi->mtd, addr, len, &read, buf); if (err && !mtd_is_bitflip(err)) { - ubi_err("error %d while reading %d bytes from PEB %d:%d, " - "read %zd bytes", err, len, pnum, offset, read); + ubi_err("error %d while reading %d bytes from PEB %d:%d, read %zd bytes", + err, len, pnum, offset, read); goto error; } err = ubi_check_pattern(buf, 0xFF, len); if (err == 0) { - ubi_err("flash region at PEB %d:%d, length %d does not " - "contain all 0xFF bytes", pnum, offset, len); + ubi_err("flash region at PEB %d:%d, length %d does not contain all 0xFF bytes", + pnum, offset, len); goto fail; } diff --git a/drivers/mtd/ubi/misc.c b/drivers/mtd/ubi/misc.c index 8bbfb44..f913d70 100644 --- a/drivers/mtd/ubi/misc.c +++ b/drivers/mtd/ubi/misc.c @@ -121,10 +121,16 @@ void ubi_update_reserved(struct ubi_device *ubi) */ void ubi_calculate_reserved(struct ubi_device *ubi) { - ubi->beb_rsvd_level = ubi->good_peb_count/100; - ubi->beb_rsvd_level *= CONFIG_MTD_UBI_BEB_RESERVE; - if (ubi->beb_rsvd_level < MIN_RESEVED_PEBS) - ubi->beb_rsvd_level = MIN_RESEVED_PEBS; + /* + * Calculate the actual number of PEBs currently needed to be reserved + * for future bad eraseblock handling. + */ + ubi->beb_rsvd_level = ubi->bad_peb_limit - ubi->bad_peb_count; + if (ubi->beb_rsvd_level < 0) { + ubi->beb_rsvd_level = 0; + ubi_warn("number of bad PEBs (%d) is above the expected limit (%d), not reserving any PEBs for bad PEB handling, will use available PEBs (if any)", + ubi->bad_peb_count, ubi->bad_peb_limit); + } } /** diff --git a/drivers/mtd/ubi/ubi.h b/drivers/mtd/ubi/ubi.h index 84f66e3..383ee43 100644 --- a/drivers/mtd/ubi/ubi.h +++ b/drivers/mtd/ubi/ubi.h @@ -51,17 +51,14 @@ #define UBI_NAME_STR "ubi" /* Normal UBI messages */ -#define ubi_msg(fmt, ...) printk(KERN_NOTICE "UBI: " fmt "\n", ##__VA_ARGS__) +#define ubi_msg(fmt, ...) pr_notice("UBI: " fmt "\n", ##__VA_ARGS__) /* UBI warning messages */ -#define ubi_warn(fmt, ...) printk(KERN_WARNING "UBI warning: %s: " fmt "\n", \ - __func__, ##__VA_ARGS__) +#define ubi_warn(fmt, ...) pr_warn("UBI warning: %s: " fmt "\n", \ + __func__, ##__VA_ARGS__) /* UBI error messages */ -#define ubi_err(fmt, ...) printk(KERN_ERR "UBI error: %s: " fmt "\n", \ +#define ubi_err(fmt, ...) pr_err("UBI error: %s: " fmt "\n", \ __func__, ##__VA_ARGS__) -/* Lowest number PEBs reserved for bad PEB handling */ -#define MIN_RESEVED_PEBS 2 - /* Background thread name pattern */ #define UBI_BGT_NAME_PATTERN "ubi_bgt%dd" @@ -363,6 +360,7 @@ struct ubi_wl_entry; * @flash_size: underlying MTD device size (in bytes) * @peb_count: count of physical eraseblocks on the MTD device * @peb_size: physical eraseblock size + * @bad_peb_limit: top limit of expected bad physical eraseblocks * @bad_peb_count: count of bad physical eraseblocks * @good_peb_count: count of good physical eraseblocks * @corr_peb_count: count of corrupted physical eraseblocks (preserved and not @@ -410,6 +408,7 @@ struct ubi_device { int avail_pebs; int beb_rsvd_pebs; int beb_rsvd_level; + int bad_peb_limit; int autoresize_vol_id; int vtbl_slots; @@ -694,7 +693,8 @@ int ubi_io_write_vid_hdr(struct ubi_device *ubi, int pnum, struct ubi_vid_hdr *vid_hdr); /* build.c */ -int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, int vid_hdr_offset); +int ubi_attach_mtd_dev(struct mtd_info *mtd, int ubi_num, + int vid_hdr_offset, int max_beb_per1024); int ubi_detach_mtd_dev(int ubi_num, int anyway); struct ubi_device *ubi_get_device(int ubi_num); void ubi_put_device(struct ubi_device *ubi); diff --git a/drivers/mtd/ubi/vtbl.c b/drivers/mtd/ubi/vtbl.c index 568307c..926e3df 100644 --- a/drivers/mtd/ubi/vtbl.c +++ b/drivers/mtd/ubi/vtbl.c @@ -270,8 +270,8 @@ static int vtbl_check(const struct ubi_device *ubi, if (len1 > 0 && len1 == len2 && !strncmp(vtbl[i].name, vtbl[n].name, len1)) { - ubi_err("volumes %d and %d have the same name" - " \"%s\"", i, n, vtbl[i].name); + ubi_err("volumes %d and %d have the same name \"%s\"", + i, n, vtbl[i].name); ubi_dump_vtbl_record(&vtbl[i], i); ubi_dump_vtbl_record(&vtbl[n], n); return -EINVAL; @@ -304,7 +304,7 @@ static int create_vtbl(struct ubi_device *ubi, struct ubi_attach_info *ai, struct ubi_vid_hdr *vid_hdr; struct ubi_ainf_peb *new_aeb; - ubi_msg("create volume table (copy #%d)", copy + 1); + dbg_gen("create volume table (copy #%d)", copy + 1); vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_KERNEL); if (!vid_hdr) @@ -562,8 +562,8 @@ static int init_volumes(struct ubi_device *ubi, if (vtbl[i].flags & UBI_VTBL_AUTORESIZE_FLG) { /* Auto re-size flag may be set only for one volume */ if (ubi->autoresize_vol_id != -1) { - ubi_err("more than one auto-resize volume (%d " - "and %d)", ubi->autoresize_vol_id, i); + ubi_err("more than one auto-resize volume (%d and %d)", + ubi->autoresize_vol_id, i); kfree(vol); return -EINVAL; } diff --git a/drivers/mtd/ubi/wl.c b/drivers/mtd/ubi/wl.c index b6be644..032fc57 100644 --- a/drivers/mtd/ubi/wl.c +++ b/drivers/mtd/ubi/wl.c @@ -978,9 +978,10 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, int cancel) { struct ubi_wl_entry *e = wl_wrk->e; - int pnum = e->pnum, err, need; + int pnum = e->pnum; int vol_id = wl_wrk->vol_id; int lnum = wl_wrk->lnum; + int err, available_consumed = 0; if (cancel) { dbg_wl("cancel erasure of PEB %d EC %d", pnum, e->ec); @@ -1045,20 +1046,14 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, } spin_lock(&ubi->volumes_lock); - need = ubi->beb_rsvd_level - ubi->beb_rsvd_pebs + 1; - if (need > 0) { - need = ubi->avail_pebs >= need ? need : ubi->avail_pebs; - ubi->avail_pebs -= need; - ubi->rsvd_pebs += need; - ubi->beb_rsvd_pebs += need; - if (need > 0) - ubi_msg("reserve more %d PEBs", need); - } - if (ubi->beb_rsvd_pebs == 0) { - spin_unlock(&ubi->volumes_lock); - ubi_err("no reserved physical eraseblocks"); - goto out_ro; + if (ubi->avail_pebs == 0) { + spin_unlock(&ubi->volumes_lock); + ubi_err("no reserved/available physical eraseblocks"); + goto out_ro; + } + ubi->avail_pebs -= 1; + available_consumed = 1; } spin_unlock(&ubi->volumes_lock); @@ -1068,19 +1063,36 @@ static int erase_worker(struct ubi_device *ubi, struct ubi_work *wl_wrk, goto out_ro; spin_lock(&ubi->volumes_lock); - ubi->beb_rsvd_pebs -= 1; + if (ubi->beb_rsvd_pebs > 0) { + if (available_consumed) { + /* + * The amount of reserved PEBs increased since we last + * checked. + */ + ubi->avail_pebs += 1; + available_consumed = 0; + } + ubi->beb_rsvd_pebs -= 1; + } ubi->bad_peb_count += 1; ubi->good_peb_count -= 1; ubi_calculate_reserved(ubi); - if (ubi->beb_rsvd_pebs) + if (available_consumed) + ubi_warn("no PEBs in the reserved pool, used an available PEB"); + else if (ubi->beb_rsvd_pebs) ubi_msg("%d PEBs left in the reserve", ubi->beb_rsvd_pebs); else - ubi_warn("last PEB from the reserved pool was used"); + ubi_warn("last PEB from the reserve was used"); spin_unlock(&ubi->volumes_lock); return err; out_ro: + if (available_consumed) { + spin_lock(&ubi->volumes_lock); + ubi->avail_pebs += 1; + spin_unlock(&ubi->volumes_lock); + } ubi_ro_mode(ubi); return err; } @@ -1189,7 +1201,7 @@ int ubi_wl_scrub_peb(struct ubi_device *ubi, int pnum) { struct ubi_wl_entry *e; - dbg_msg("schedule PEB %d for scrubbing", pnum); + ubi_msg("schedule PEB %d for scrubbing", pnum); retry: spin_lock(&ubi->wl_lock); diff --git a/drivers/net/xen-netback/netback.c b/drivers/net/xen-netback/netback.c index 682633bf..05593d8 100644 --- a/drivers/net/xen-netback/netback.c +++ b/drivers/net/xen-netback/netback.c @@ -635,9 +635,7 @@ static void xen_netbk_rx_action(struct xen_netbk *netbk) return; BUG_ON(npo.copy_prod > ARRAY_SIZE(netbk->grant_copy_op)); - ret = HYPERVISOR_grant_table_op(GNTTABOP_copy, &netbk->grant_copy_op, - npo.copy_prod); - BUG_ON(ret != 0); + gnttab_batch_copy(netbk->grant_copy_op, npo.copy_prod); while ((skb = __skb_dequeue(&rxq)) != NULL) { sco = (struct skb_cb_overlay *)skb->cb; @@ -1460,18 +1458,15 @@ static void xen_netbk_tx_submit(struct xen_netbk *netbk) static void xen_netbk_tx_action(struct xen_netbk *netbk) { unsigned nr_gops; - int ret; nr_gops = xen_netbk_tx_build_gops(netbk); if (nr_gops == 0) return; - ret = HYPERVISOR_grant_table_op(GNTTABOP_copy, - netbk->tx_copy_ops, nr_gops); - BUG_ON(ret); - xen_netbk_tx_submit(netbk); + gnttab_batch_copy(netbk->tx_copy_ops, nr_gops); + xen_netbk_tx_submit(netbk); } static void xen_netbk_idx_release(struct xen_netbk *netbk, u16 pending_idx) diff --git a/drivers/pci/pci-driver.c b/drivers/pci/pci-driver.c index 9e1d295..94c6e2a 100644 --- a/drivers/pci/pci-driver.c +++ b/drivers/pci/pci-driver.c @@ -607,21 +607,6 @@ static int pci_pm_prepare(struct device *dev) int error = 0; /* - * If a PCI device configured to wake up the system from sleep states - * has been suspended at run time and there's a resume request pending - * for it, this is equivalent to the device signaling wakeup, so the - * system suspend operation should be aborted. - */ - pm_runtime_get_noresume(dev); - if (pm_runtime_barrier(dev) && device_may_wakeup(dev)) - pm_wakeup_event(dev, 0); - - if (pm_wakeup_pending()) { - pm_runtime_put_sync(dev); - return -EBUSY; - } - - /* * PCI devices suspended at run time need to be resumed at this * point, because in general it is necessary to reconfigure them for * system suspend. Namely, if the device is supposed to wake up the @@ -644,8 +629,6 @@ static void pci_pm_complete(struct device *dev) if (drv && drv->pm && drv->pm->complete) drv->pm->complete(dev); - - pm_runtime_put_sync(dev); } #else /* !CONFIG_PM_SLEEP */ diff --git a/drivers/pci/xen-pcifront.c b/drivers/pci/xen-pcifront.c index def8d0b..0aab85a 100644 --- a/drivers/pci/xen-pcifront.c +++ b/drivers/pci/xen-pcifront.c @@ -21,6 +21,7 @@ #include <linux/bitops.h> #include <linux/time.h> +#include <asm/xen/swiotlb-xen.h> #define INVALID_GRANT_REF (0) #define INVALID_EVTCHN (-1) @@ -236,7 +237,7 @@ static int pcifront_bus_write(struct pci_bus *bus, unsigned int devfn, return errno_to_pcibios_err(do_pci_op(pdev, &op)); } -struct pci_ops pcifront_bus_ops = { +static struct pci_ops pcifront_bus_ops = { .read = pcifront_bus_read, .write = pcifront_bus_write, }; @@ -668,7 +669,7 @@ static irqreturn_t pcifront_handler_aer(int irq, void *dev) schedule_pcifront_aer_op(pdev); return IRQ_HANDLED; } -static int pcifront_connect(struct pcifront_device *pdev) +static int pcifront_connect_and_init_dma(struct pcifront_device *pdev) { int err = 0; @@ -681,9 +682,13 @@ static int pcifront_connect(struct pcifront_device *pdev) dev_err(&pdev->xdev->dev, "PCI frontend already installed!\n"); err = -EEXIST; } - spin_unlock(&pcifront_dev_lock); + if (!err && !swiotlb_nr_tbl()) { + err = pci_xen_swiotlb_init_late(); + if (err) + dev_err(&pdev->xdev->dev, "Could not setup SWIOTLB!\n"); + } return err; } @@ -842,10 +847,10 @@ static int __devinit pcifront_try_connect(struct pcifront_device *pdev) XenbusStateInitialised) goto out; - err = pcifront_connect(pdev); + err = pcifront_connect_and_init_dma(pdev); if (err) { xenbus_dev_fatal(pdev->xdev, err, - "Error connecting PCI Frontend"); + "Error setting up PCI Frontend"); goto out; } diff --git a/drivers/s390/scsi/zfcp_aux.c b/drivers/s390/scsi/zfcp_aux.c index aff8621..f6adde4 100644 --- a/drivers/s390/scsi/zfcp_aux.c +++ b/drivers/s390/scsi/zfcp_aux.c @@ -519,6 +519,7 @@ struct zfcp_port *zfcp_port_enqueue(struct zfcp_adapter *adapter, u64 wwpn, rwlock_init(&port->unit_list_lock); INIT_LIST_HEAD(&port->unit_list); + atomic_set(&port->units, 0); INIT_WORK(&port->gid_pn_work, zfcp_fc_port_did_lookup); INIT_WORK(&port->test_link_work, zfcp_fc_link_test_work); diff --git a/drivers/s390/scsi/zfcp_ccw.c b/drivers/s390/scsi/zfcp_ccw.c index e37f045..f2dd3a0 100644 --- a/drivers/s390/scsi/zfcp_ccw.c +++ b/drivers/s390/scsi/zfcp_ccw.c @@ -39,19 +39,25 @@ void zfcp_ccw_adapter_put(struct zfcp_adapter *adapter) spin_unlock_irqrestore(&zfcp_ccw_adapter_ref_lock, flags); } -static int zfcp_ccw_activate(struct ccw_device *cdev) - +/** + * zfcp_ccw_activate - activate adapter and wait for it to finish + * @cdev: pointer to belonging ccw device + * @clear: Status flags to clear. + * @tag: s390dbf trace record tag + */ +static int zfcp_ccw_activate(struct ccw_device *cdev, int clear, char *tag) { struct zfcp_adapter *adapter = zfcp_ccw_adapter_by_cdev(cdev); if (!adapter) return 0; + zfcp_erp_clear_adapter_status(adapter, clear); zfcp_erp_set_adapter_status(adapter, ZFCP_STATUS_COMMON_RUNNING); zfcp_erp_adapter_reopen(adapter, ZFCP_STATUS_COMMON_ERP_FAILED, - "ccresu2"); + tag); zfcp_erp_wait(adapter); - flush_work(&adapter->scan_work); + flush_work(&adapter->scan_work); /* ok to call even if nothing queued */ zfcp_ccw_adapter_put(adapter); @@ -164,26 +170,34 @@ static int zfcp_ccw_set_online(struct ccw_device *cdev) BUG_ON(!zfcp_reqlist_isempty(adapter->req_list)); adapter->req_no = 0; - zfcp_ccw_activate(cdev); + zfcp_ccw_activate(cdev, 0, "ccsonl1"); + /* scan for remote ports + either at the end of any successful adapter recovery + or only after the adapter recovery for setting a device online */ + zfcp_fc_inverse_conditional_port_scan(adapter); + flush_work(&adapter->scan_work); /* ok to call even if nothing queued */ zfcp_ccw_adapter_put(adapter); return 0; } /** - * zfcp_ccw_set_offline - set_offline function of zfcp driver + * zfcp_ccw_offline_sync - shut down adapter and wait for it to finish * @cdev: pointer to belonging ccw device + * @set: Status flags to set. + * @tag: s390dbf trace record tag * * This function gets called by the common i/o layer and sets an adapter * into state offline. */ -static int zfcp_ccw_set_offline(struct ccw_device *cdev) +static int zfcp_ccw_offline_sync(struct ccw_device *cdev, int set, char *tag) { struct zfcp_adapter *adapter = zfcp_ccw_adapter_by_cdev(cdev); if (!adapter) return 0; - zfcp_erp_adapter_shutdown(adapter, 0, "ccsoff1"); + zfcp_erp_set_adapter_status(adapter, set); + zfcp_erp_adapter_shutdown(adapter, 0, tag); zfcp_erp_wait(adapter); zfcp_ccw_adapter_put(adapter); @@ -191,6 +205,18 @@ static int zfcp_ccw_set_offline(struct ccw_device *cdev) } /** + * zfcp_ccw_set_offline - set_offline function of zfcp driver + * @cdev: pointer to belonging ccw device + * + * This function gets called by the common i/o layer and sets an adapter + * into state offline. + */ +static int zfcp_ccw_set_offline(struct ccw_device *cdev) +{ + return zfcp_ccw_offline_sync(cdev, 0, "ccsoff1"); +} + +/** * zfcp_ccw_notify - ccw notify function * @cdev: pointer to belonging ccw device * @event: indicates if adapter was detached or attached @@ -207,6 +233,11 @@ static int zfcp_ccw_notify(struct ccw_device *cdev, int event) switch (event) { case CIO_GONE: + if (atomic_read(&adapter->status) & + ZFCP_STATUS_ADAPTER_SUSPENDED) { /* notification ignore */ + zfcp_dbf_hba_basic("ccnigo1", adapter); + break; + } dev_warn(&cdev->dev, "The FCP device has been detached\n"); zfcp_erp_adapter_shutdown(adapter, 0, "ccnoti1"); break; @@ -216,6 +247,11 @@ static int zfcp_ccw_notify(struct ccw_device *cdev, int event) zfcp_erp_adapter_shutdown(adapter, 0, "ccnoti2"); break; case CIO_OPER: + if (atomic_read(&adapter->status) & + ZFCP_STATUS_ADAPTER_SUSPENDED) { /* notification ignore */ + zfcp_dbf_hba_basic("ccniop1", adapter); + break; + } dev_info(&cdev->dev, "The FCP device is operational again\n"); zfcp_erp_set_adapter_status(adapter, ZFCP_STATUS_COMMON_RUNNING); @@ -251,6 +287,28 @@ static void zfcp_ccw_shutdown(struct ccw_device *cdev) zfcp_ccw_adapter_put(adapter); } +static int zfcp_ccw_suspend(struct ccw_device *cdev) +{ + zfcp_ccw_offline_sync(cdev, ZFCP_STATUS_ADAPTER_SUSPENDED, "ccsusp1"); + return 0; +} + +static int zfcp_ccw_thaw(struct ccw_device *cdev) +{ + /* trace records for thaw and final shutdown during suspend + can only be found in system dump until the end of suspend + but not after resume because it's based on the memory image + right after the very first suspend (freeze) callback */ + zfcp_ccw_activate(cdev, 0, "ccthaw1"); + return 0; +} + +static int zfcp_ccw_resume(struct ccw_device *cdev) +{ + zfcp_ccw_activate(cdev, ZFCP_STATUS_ADAPTER_SUSPENDED, "ccresu1"); + return 0; +} + struct ccw_driver zfcp_ccw_driver = { .driver = { .owner = THIS_MODULE, @@ -263,7 +321,7 @@ struct ccw_driver zfcp_ccw_driver = { .set_offline = zfcp_ccw_set_offline, .notify = zfcp_ccw_notify, .shutdown = zfcp_ccw_shutdown, - .freeze = zfcp_ccw_set_offline, - .thaw = zfcp_ccw_activate, - .restore = zfcp_ccw_activate, + .freeze = zfcp_ccw_suspend, + .thaw = zfcp_ccw_thaw, + .restore = zfcp_ccw_resume, }; diff --git a/drivers/s390/scsi/zfcp_cfdc.c b/drivers/s390/scsi/zfcp_cfdc.c index fbd8b4d..49b82e4 100644 --- a/drivers/s390/scsi/zfcp_cfdc.c +++ b/drivers/s390/scsi/zfcp_cfdc.c @@ -293,7 +293,7 @@ void zfcp_cfdc_adapter_access_changed(struct zfcp_adapter *adapter) } read_unlock_irqrestore(&adapter->port_list_lock, flags); - shost_for_each_device(sdev, port->adapter->scsi_host) { + shost_for_each_device(sdev, adapter->scsi_host) { zfcp_sdev = sdev_to_zfcp(sdev); status = atomic_read(&zfcp_sdev->status); if ((status & ZFCP_STATUS_COMMON_ACCESS_DENIED) || diff --git a/drivers/s390/scsi/zfcp_dbf.c b/drivers/s390/scsi/zfcp_dbf.c index 3c1d220..e1a8cc2 100644 --- a/drivers/s390/scsi/zfcp_dbf.c +++ b/drivers/s390/scsi/zfcp_dbf.c @@ -191,7 +191,7 @@ void zfcp_dbf_hba_def_err(struct zfcp_adapter *adapter, u64 req_id, u16 scount, length = min((u16)sizeof(struct qdio_buffer), (u16)ZFCP_DBF_PAY_MAX_REC); - while ((char *)pl[payload->counter] && payload->counter < scount) { + while (payload->counter < scount && (char *)pl[payload->counter]) { memcpy(payload->data, (char *)pl[payload->counter], length); debug_event(dbf->pay, 1, payload, zfcp_dbf_plen(length)); payload->counter++; @@ -200,6 +200,26 @@ void zfcp_dbf_hba_def_err(struct zfcp_adapter *adapter, u64 req_id, u16 scount, spin_unlock_irqrestore(&dbf->pay_lock, flags); } +/** + * zfcp_dbf_hba_basic - trace event for basic adapter events + * @adapter: pointer to struct zfcp_adapter + */ +void zfcp_dbf_hba_basic(char *tag, struct zfcp_adapter *adapter) +{ + struct zfcp_dbf *dbf = adapter->dbf; + struct zfcp_dbf_hba *rec = &dbf->hba_buf; + unsigned long flags; + + spin_lock_irqsave(&dbf->hba_lock, flags); + memset(rec, 0, sizeof(*rec)); + + memcpy(rec->tag, tag, ZFCP_DBF_TAG_LEN); + rec->id = ZFCP_DBF_HBA_BASIC; + + debug_event(dbf->hba, 1, rec, sizeof(*rec)); + spin_unlock_irqrestore(&dbf->hba_lock, flags); +} + static void zfcp_dbf_set_common(struct zfcp_dbf_rec *rec, struct zfcp_adapter *adapter, struct zfcp_port *port, diff --git a/drivers/s390/scsi/zfcp_dbf.h b/drivers/s390/scsi/zfcp_dbf.h index 714f087..3ac7a4b 100644 --- a/drivers/s390/scsi/zfcp_dbf.h +++ b/drivers/s390/scsi/zfcp_dbf.h @@ -154,6 +154,7 @@ enum zfcp_dbf_hba_id { ZFCP_DBF_HBA_RES = 1, ZFCP_DBF_HBA_USS = 2, ZFCP_DBF_HBA_BIT = 3, + ZFCP_DBF_HBA_BASIC = 4, }; /** diff --git a/drivers/s390/scsi/zfcp_def.h b/drivers/s390/scsi/zfcp_def.h index 2955e1a..1305955 100644 --- a/drivers/s390/scsi/zfcp_def.h +++ b/drivers/s390/scsi/zfcp_def.h @@ -77,6 +77,7 @@ struct zfcp_reqlist; #define ZFCP_STATUS_ADAPTER_SIOSL_ISSUED 0x00000004 #define ZFCP_STATUS_ADAPTER_XCONFIG_OK 0x00000008 #define ZFCP_STATUS_ADAPTER_HOST_CON_INIT 0x00000010 +#define ZFCP_STATUS_ADAPTER_SUSPENDED 0x00000040 #define ZFCP_STATUS_ADAPTER_ERP_PENDING 0x00000100 #define ZFCP_STATUS_ADAPTER_LINK_UNPLUGGED 0x00000200 #define ZFCP_STATUS_ADAPTER_DATA_DIV_ENABLED 0x00000400 @@ -204,6 +205,7 @@ struct zfcp_port { struct zfcp_adapter *adapter; /* adapter used to access port */ struct list_head unit_list; /* head of logical unit list */ rwlock_t unit_list_lock; /* unit list lock */ + atomic_t units; /* zfcp_unit count */ atomic_t status; /* status of this remote port */ u64 wwnn; /* WWNN if known */ u64 wwpn; /* WWPN */ diff --git a/drivers/s390/scsi/zfcp_erp.c b/drivers/s390/scsi/zfcp_erp.c index 92d3df6..4133ab6 100644 --- a/drivers/s390/scsi/zfcp_erp.c +++ b/drivers/s390/scsi/zfcp_erp.c @@ -1230,7 +1230,7 @@ static void zfcp_erp_action_cleanup(struct zfcp_erp_action *act, int result) case ZFCP_ERP_ACTION_REOPEN_ADAPTER: if (result == ZFCP_ERP_SUCCEEDED) { register_service_level(&adapter->service_level); - queue_work(adapter->work_queue, &adapter->scan_work); + zfcp_fc_conditional_port_scan(adapter); queue_work(adapter->work_queue, &adapter->ns_up_work); } else unregister_service_level(&adapter->service_level); diff --git a/drivers/s390/scsi/zfcp_ext.h b/drivers/s390/scsi/zfcp_ext.h index 36f4227..1d3dd3f 100644 --- a/drivers/s390/scsi/zfcp_ext.h +++ b/drivers/s390/scsi/zfcp_ext.h @@ -54,6 +54,7 @@ extern void zfcp_dbf_hba_fsf_res(char *, struct zfcp_fsf_req *); extern void zfcp_dbf_hba_bit_err(char *, struct zfcp_fsf_req *); extern void zfcp_dbf_hba_berr(struct zfcp_dbf *, struct zfcp_fsf_req *); extern void zfcp_dbf_hba_def_err(struct zfcp_adapter *, u64, u16, void **); +extern void zfcp_dbf_hba_basic(char *, struct zfcp_adapter *); extern void zfcp_dbf_san_req(char *, struct zfcp_fsf_req *, u32); extern void zfcp_dbf_san_res(char *, struct zfcp_fsf_req *); extern void zfcp_dbf_san_in_els(char *, struct zfcp_fsf_req *); @@ -98,6 +99,8 @@ extern void zfcp_fc_gs_destroy(struct zfcp_adapter *); extern int zfcp_fc_exec_bsg_job(struct fc_bsg_job *); extern int zfcp_fc_timeout_bsg_job(struct fc_bsg_job *); extern void zfcp_fc_sym_name_update(struct work_struct *); +extern void zfcp_fc_conditional_port_scan(struct zfcp_adapter *); +extern void zfcp_fc_inverse_conditional_port_scan(struct zfcp_adapter *); /* zfcp_fsf.c */ extern struct kmem_cache *zfcp_fsf_qtcb_cache; @@ -158,6 +161,7 @@ extern void zfcp_scsi_dif_sense_error(struct scsi_cmnd *, int); extern struct attribute_group zfcp_sysfs_unit_attrs; extern struct attribute_group zfcp_sysfs_adapter_attrs; extern struct attribute_group zfcp_sysfs_port_attrs; +extern struct mutex zfcp_sysfs_port_units_mutex; extern struct device_attribute *zfcp_sysfs_sdev_attrs[]; extern struct device_attribute *zfcp_sysfs_shost_attrs[]; diff --git a/drivers/s390/scsi/zfcp_fc.c b/drivers/s390/scsi/zfcp_fc.c index 88688a8..ff598cd 100644 --- a/drivers/s390/scsi/zfcp_fc.c +++ b/drivers/s390/scsi/zfcp_fc.c @@ -26,6 +26,27 @@ static u32 zfcp_fc_rscn_range_mask[] = { [ELS_ADDR_FMT_FAB] = 0x000000, }; +static bool no_auto_port_rescan; +module_param_named(no_auto_port_rescan, no_auto_port_rescan, bool, 0600); +MODULE_PARM_DESC(no_auto_port_rescan, + "no automatic port_rescan (default off)"); + +void zfcp_fc_conditional_port_scan(struct zfcp_adapter *adapter) +{ + if (no_auto_port_rescan) + return; + + queue_work(adapter->work_queue, &adapter->scan_work); +} + +void zfcp_fc_inverse_conditional_port_scan(struct zfcp_adapter *adapter) +{ + if (!no_auto_port_rescan) + return; + + queue_work(adapter->work_queue, &adapter->scan_work); +} + /** * zfcp_fc_post_event - post event to userspace via fc_transport * @work: work struct with enqueued events @@ -206,7 +227,7 @@ static void zfcp_fc_incoming_rscn(struct zfcp_fsf_req *fsf_req) zfcp_fc_enqueue_event(fsf_req->adapter, FCH_EVT_RSCN, *(u32 *)page); } - queue_work(fsf_req->adapter->work_queue, &fsf_req->adapter->scan_work); + zfcp_fc_conditional_port_scan(fsf_req->adapter); } static void zfcp_fc_incoming_wwpn(struct zfcp_fsf_req *req, u64 wwpn) diff --git a/drivers/s390/scsi/zfcp_fsf.c b/drivers/s390/scsi/zfcp_fsf.c index e1c1efc..c96320d 100644 --- a/drivers/s390/scsi/zfcp_fsf.c +++ b/drivers/s390/scsi/zfcp_fsf.c @@ -219,7 +219,7 @@ static void zfcp_fsf_status_read_handler(struct zfcp_fsf_req *req) return; } - zfcp_dbf_hba_fsf_uss("fssrh_2", req); + zfcp_dbf_hba_fsf_uss("fssrh_4", req); switch (sr_buf->status_type) { case FSF_STATUS_READ_PORT_CLOSED: @@ -257,7 +257,7 @@ static void zfcp_fsf_status_read_handler(struct zfcp_fsf_req *req) if (sr_buf->status_subtype & FSF_STATUS_READ_SUB_ACT_UPDATED) zfcp_cfdc_adapter_access_changed(adapter); if (sr_buf->status_subtype & FSF_STATUS_READ_SUB_INCOMING_ELS) - queue_work(adapter->work_queue, &adapter->scan_work); + zfcp_fc_conditional_port_scan(adapter); break; case FSF_STATUS_READ_CFDC_UPDATED: zfcp_cfdc_adapter_access_changed(adapter); @@ -437,6 +437,34 @@ void zfcp_fsf_req_dismiss_all(struct zfcp_adapter *adapter) } } +#define ZFCP_FSF_PORTSPEED_1GBIT (1 << 0) +#define ZFCP_FSF_PORTSPEED_2GBIT (1 << 1) +#define ZFCP_FSF_PORTSPEED_4GBIT (1 << 2) +#define ZFCP_FSF_PORTSPEED_10GBIT (1 << 3) +#define ZFCP_FSF_PORTSPEED_8GBIT (1 << 4) +#define ZFCP_FSF_PORTSPEED_16GBIT (1 << 5) +#define ZFCP_FSF_PORTSPEED_NOT_NEGOTIATED (1 << 15) + +static u32 zfcp_fsf_convert_portspeed(u32 fsf_speed) +{ + u32 fdmi_speed = 0; + if (fsf_speed & ZFCP_FSF_PORTSPEED_1GBIT) + fdmi_speed |= FC_PORTSPEED_1GBIT; + if (fsf_speed & ZFCP_FSF_PORTSPEED_2GBIT) + fdmi_speed |= FC_PORTSPEED_2GBIT; + if (fsf_speed & ZFCP_FSF_PORTSPEED_4GBIT) + fdmi_speed |= FC_PORTSPEED_4GBIT; + if (fsf_speed & ZFCP_FSF_PORTSPEED_10GBIT) + fdmi_speed |= FC_PORTSPEED_10GBIT; + if (fsf_speed & ZFCP_FSF_PORTSPEED_8GBIT) + fdmi_speed |= FC_PORTSPEED_8GBIT; + if (fsf_speed & ZFCP_FSF_PORTSPEED_16GBIT) + fdmi_speed |= FC_PORTSPEED_16GBIT; + if (fsf_speed & ZFCP_FSF_PORTSPEED_NOT_NEGOTIATED) + fdmi_speed |= FC_PORTSPEED_NOT_NEGOTIATED; + return fdmi_speed; +} + static int zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *req) { struct fsf_qtcb_bottom_config *bottom = &req->qtcb->bottom.config; @@ -456,7 +484,8 @@ static int zfcp_fsf_exchange_config_evaluate(struct zfcp_fsf_req *req) fc_host_port_name(shost) = nsp->fl_wwpn; fc_host_node_name(shost) = nsp->fl_wwnn; fc_host_port_id(shost) = ntoh24(bottom->s_id); - fc_host_speed(shost) = bottom->fc_link_speed; + fc_host_speed(shost) = + zfcp_fsf_convert_portspeed(bottom->fc_link_speed); fc_host_supported_classes(shost) = FC_COS_CLASS2 | FC_COS_CLASS3; adapter->hydra_version = bottom->adapter_type; @@ -580,7 +609,8 @@ static void zfcp_fsf_exchange_port_evaluate(struct zfcp_fsf_req *req) } else fc_host_permanent_port_name(shost) = fc_host_port_name(shost); fc_host_maxframe_size(shost) = bottom->maximum_frame_size; - fc_host_supported_speeds(shost) = bottom->supported_speed; + fc_host_supported_speeds(shost) = + zfcp_fsf_convert_portspeed(bottom->supported_speed); memcpy(fc_host_supported_fc4s(shost), bottom->supported_fc4_types, FC_FC4_LIST_SIZE); memcpy(fc_host_active_fc4s(shost), bottom->active_fc4_types, @@ -771,12 +801,14 @@ out: static void zfcp_fsf_abort_fcp_command_handler(struct zfcp_fsf_req *req) { struct scsi_device *sdev = req->data; - struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev); + struct zfcp_scsi_dev *zfcp_sdev; union fsf_status_qual *fsq = &req->qtcb->header.fsf_status_qual; if (req->status & ZFCP_STATUS_FSFREQ_ERROR) return; + zfcp_sdev = sdev_to_zfcp(sdev); + switch (req->qtcb->header.fsf_status) { case FSF_PORT_HANDLE_NOT_VALID: if (fsq->word[0] == fsq->word[1]) { @@ -885,7 +917,7 @@ static void zfcp_fsf_send_ct_handler(struct zfcp_fsf_req *req) switch (header->fsf_status) { case FSF_GOOD: - zfcp_dbf_san_res("fsscth1", req); + zfcp_dbf_san_res("fsscth2", req); ct->status = 0; break; case FSF_SERVICE_CLASS_NOT_SUPPORTED: @@ -1739,13 +1771,15 @@ static void zfcp_fsf_open_lun_handler(struct zfcp_fsf_req *req) { struct zfcp_adapter *adapter = req->adapter; struct scsi_device *sdev = req->data; - struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev); + struct zfcp_scsi_dev *zfcp_sdev; struct fsf_qtcb_header *header = &req->qtcb->header; struct fsf_qtcb_bottom_support *bottom = &req->qtcb->bottom.support; if (req->status & ZFCP_STATUS_FSFREQ_ERROR) return; + zfcp_sdev = sdev_to_zfcp(sdev); + atomic_clear_mask(ZFCP_STATUS_COMMON_ACCESS_DENIED | ZFCP_STATUS_COMMON_ACCESS_BOXED | ZFCP_STATUS_LUN_SHARED | @@ -1856,11 +1890,13 @@ out: static void zfcp_fsf_close_lun_handler(struct zfcp_fsf_req *req) { struct scsi_device *sdev = req->data; - struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev); + struct zfcp_scsi_dev *zfcp_sdev; if (req->status & ZFCP_STATUS_FSFREQ_ERROR) return; + zfcp_sdev = sdev_to_zfcp(sdev); + switch (req->qtcb->header.fsf_status) { case FSF_PORT_HANDLE_NOT_VALID: zfcp_erp_adapter_reopen(zfcp_sdev->port->adapter, 0, "fscuh_1"); @@ -1950,7 +1986,7 @@ static void zfcp_fsf_req_trace(struct zfcp_fsf_req *req, struct scsi_cmnd *scsi) { struct fsf_qual_latency_info *lat_in; struct latency_cont *lat = NULL; - struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(scsi->device); + struct zfcp_scsi_dev *zfcp_sdev; struct zfcp_blk_drv_data blktrc; int ticks = req->adapter->timer_ticks; @@ -1965,6 +2001,7 @@ static void zfcp_fsf_req_trace(struct zfcp_fsf_req *req, struct scsi_cmnd *scsi) if (req->adapter->adapter_features & FSF_FEATURE_MEASUREMENT_DATA && !(req->status & ZFCP_STATUS_FSFREQ_ERROR)) { + zfcp_sdev = sdev_to_zfcp(scsi->device); blktrc.flags |= ZFCP_BLK_LAT_VALID; blktrc.channel_lat = lat_in->channel_lat * ticks; blktrc.fabric_lat = lat_in->fabric_lat * ticks; @@ -2002,12 +2039,14 @@ static void zfcp_fsf_fcp_handler_common(struct zfcp_fsf_req *req) { struct scsi_cmnd *scmnd = req->data; struct scsi_device *sdev = scmnd->device; - struct zfcp_scsi_dev *zfcp_sdev = sdev_to_zfcp(sdev); + struct zfcp_scsi_dev *zfcp_sdev; struct fsf_qtcb_header *header = &req->qtcb->header; if (unlikely(req->status & ZFCP_STATUS_FSFREQ_ERROR)) return; + zfcp_sdev = sdev_to_zfcp(sdev); + switch (header->fsf_status) { case FSF_HANDLE_MISMATCH: case FSF_PORT_HANDLE_NOT_VALID: diff --git a/drivers/s390/scsi/zfcp_qdio.c b/drivers/s390/scsi/zfcp_qdio.c index b9fffc8..50b5615 100644 --- a/drivers/s390/scsi/zfcp_qdio.c +++ b/drivers/s390/scsi/zfcp_qdio.c @@ -102,18 +102,22 @@ static void zfcp_qdio_int_resp(struct ccw_device *cdev, unsigned int qdio_err, { struct zfcp_qdio *qdio = (struct zfcp_qdio *) parm; struct zfcp_adapter *adapter = qdio->adapter; - struct qdio_buffer_element *sbale; int sbal_no, sbal_idx; - void *pl[ZFCP_QDIO_MAX_SBALS_PER_REQ + 1]; - u64 req_id; - u8 scount; if (unlikely(qdio_err)) { - memset(pl, 0, ZFCP_QDIO_MAX_SBALS_PER_REQ * sizeof(void *)); if (zfcp_adapter_multi_buffer_active(adapter)) { + void *pl[ZFCP_QDIO_MAX_SBALS_PER_REQ + 1]; + struct qdio_buffer_element *sbale; + u64 req_id; + u8 scount; + + memset(pl, 0, + ZFCP_QDIO_MAX_SBALS_PER_REQ * sizeof(void *)); sbale = qdio->res_q[idx]->element; req_id = (u64) sbale->addr; - scount = sbale->scount + 1; /* incl. signaling SBAL */ + scount = min(sbale->scount + 1, + ZFCP_QDIO_MAX_SBALS_PER_REQ + 1); + /* incl. signaling SBAL */ for (sbal_no = 0; sbal_no < scount; sbal_no++) { sbal_idx = (idx + sbal_no) % diff --git a/drivers/s390/scsi/zfcp_sysfs.c b/drivers/s390/scsi/zfcp_sysfs.c index c66af27..1e0eb08 100644 --- a/drivers/s390/scsi/zfcp_sysfs.c +++ b/drivers/s390/scsi/zfcp_sysfs.c @@ -227,6 +227,8 @@ static ssize_t zfcp_sysfs_port_rescan_store(struct device *dev, static ZFCP_DEV_ATTR(adapter, port_rescan, S_IWUSR, NULL, zfcp_sysfs_port_rescan_store); +DEFINE_MUTEX(zfcp_sysfs_port_units_mutex); + static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, struct device_attribute *attr, const char *buf, size_t count) @@ -249,6 +251,16 @@ static ssize_t zfcp_sysfs_port_remove_store(struct device *dev, else retval = 0; + mutex_lock(&zfcp_sysfs_port_units_mutex); + if (atomic_read(&port->units) > 0) { + retval = -EBUSY; + mutex_unlock(&zfcp_sysfs_port_units_mutex); + goto out; + } + /* port is about to be removed, so no more unit_add */ + atomic_set(&port->units, -1); + mutex_unlock(&zfcp_sysfs_port_units_mutex); + write_lock_irq(&adapter->port_list_lock); list_del(&port->list); write_unlock_irq(&adapter->port_list_lock); @@ -289,12 +301,14 @@ static ssize_t zfcp_sysfs_unit_add_store(struct device *dev, { struct zfcp_port *port = container_of(dev, struct zfcp_port, dev); u64 fcp_lun; + int retval; if (strict_strtoull(buf, 0, (unsigned long long *) &fcp_lun)) return -EINVAL; - if (zfcp_unit_add(port, fcp_lun)) - return -EINVAL; + retval = zfcp_unit_add(port, fcp_lun); + if (retval) + return retval; return count; } diff --git a/drivers/s390/scsi/zfcp_unit.c b/drivers/s390/scsi/zfcp_unit.c index 3f2bff0..1cd2b99 100644 --- a/drivers/s390/scsi/zfcp_unit.c +++ b/drivers/s390/scsi/zfcp_unit.c @@ -104,7 +104,7 @@ static void zfcp_unit_release(struct device *dev) { struct zfcp_unit *unit = container_of(dev, struct zfcp_unit, dev); - put_device(&unit->port->dev); + atomic_dec(&unit->port->units); kfree(unit); } @@ -119,16 +119,27 @@ static void zfcp_unit_release(struct device *dev) int zfcp_unit_add(struct zfcp_port *port, u64 fcp_lun) { struct zfcp_unit *unit; + int retval = 0; + + mutex_lock(&zfcp_sysfs_port_units_mutex); + if (atomic_read(&port->units) == -1) { + /* port is already gone */ + retval = -ENODEV; + goto out; + } unit = zfcp_unit_find(port, fcp_lun); if (unit) { put_device(&unit->dev); - return -EEXIST; + retval = -EEXIST; + goto out; } unit = kzalloc(sizeof(struct zfcp_unit), GFP_KERNEL); - if (!unit) - return -ENOMEM; + if (!unit) { + retval = -ENOMEM; + goto out; + } unit->port = port; unit->fcp_lun = fcp_lun; @@ -139,28 +150,33 @@ int zfcp_unit_add(struct zfcp_port *port, u64 fcp_lun) if (dev_set_name(&unit->dev, "0x%016llx", (unsigned long long) fcp_lun)) { kfree(unit); - return -ENOMEM; + retval = -ENOMEM; + goto out; } - get_device(&port->dev); - if (device_register(&unit->dev)) { put_device(&unit->dev); - return -ENOMEM; + retval = -ENOMEM; + goto out; } if (sysfs_create_group(&unit->dev.kobj, &zfcp_sysfs_unit_attrs)) { device_unregister(&unit->dev); - return -EINVAL; + retval = -EINVAL; + goto out; } + atomic_inc(&port->units); /* under zfcp_sysfs_port_units_mutex ! */ + write_lock_irq(&port->unit_list_lock); list_add_tail(&unit->list, &port->unit_list); write_unlock_irq(&port->unit_list_lock); zfcp_unit_scsi_scan(unit); - return 0; +out: + mutex_unlock(&zfcp_sysfs_port_units_mutex); + return retval; } /** diff --git a/drivers/scsi/be2iscsi/be_cmds.c b/drivers/scsi/be2iscsi/be_cmds.c index d2e9e93..07d2cb1 100644 --- a/drivers/scsi/be2iscsi/be_cmds.c +++ b/drivers/scsi/be2iscsi/be_cmds.c @@ -48,7 +48,8 @@ int beiscsi_pci_soft_reset(struct beiscsi_hba *phba) } if (sreset & BE2_SET_RESET) { - printk(KERN_ERR "Soft Reset did not deassert\n"); + printk(KERN_ERR DRV_NAME + " Soft Reset did not deassert\n"); return -EIO; } pconline1 = BE2_MPU_IRAM_ONLINE; @@ -67,7 +68,8 @@ int beiscsi_pci_soft_reset(struct beiscsi_hba *phba) i++; } if (sreset & BE2_SET_RESET) { - printk(KERN_ERR "MPU Online Soft Reset did not deassert\n"); + printk(KERN_ERR DRV_NAME + " MPU Online Soft Reset did not deassert\n"); return -EIO; } return 0; @@ -93,8 +95,9 @@ int be_chk_reset_complete(struct beiscsi_hba *phba) } if ((status & 0x80000000) || (!num_loop)) { - printk(KERN_ERR "Failed in be_chk_reset_complete" - "status = 0x%x\n", status); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BC_%d : Failed in be_chk_reset_complete" + "status = 0x%x\n", status); return -EIO; } @@ -169,6 +172,7 @@ static int be_mcc_compl_process(struct be_ctrl_info *ctrl, struct be_mcc_compl *compl) { u16 compl_status, extd_status; + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); be_dws_le_to_cpu(compl, 4); @@ -177,9 +181,12 @@ static int be_mcc_compl_process(struct be_ctrl_info *ctrl, if (compl_status != MCC_STATUS_SUCCESS) { extd_status = (compl->status >> CQE_STATUS_EXTD_SHIFT) & CQE_STATUS_EXTD_MASK; - dev_err(&ctrl->pdev->dev, - "error in cmd completion: status(compl/extd)=%d/%d\n", - compl_status, extd_status); + + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : error in cmd completion: status(compl/extd)=%d/%d\n", + compl_status, extd_status); + return -EBUSY; } return 0; @@ -233,22 +240,29 @@ void beiscsi_async_link_state_process(struct beiscsi_hba *phba, { switch (evt->port_link_status) { case ASYNC_EVENT_LINK_DOWN: - SE_DEBUG(DBG_LVL_1, "Link Down on Physical Port %d\n", - evt->physical_port); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_INIT, + "BC_%d : Link Down on Physical Port %d\n", + evt->physical_port); + phba->state |= BE_ADAPTER_LINK_DOWN; iscsi_host_for_each_session(phba->shost, be2iscsi_fail_session); break; case ASYNC_EVENT_LINK_UP: phba->state = BE_ADAPTER_UP; - SE_DEBUG(DBG_LVL_1, "Link UP on Physical Port %d\n", - evt->physical_port); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_INIT, + "BC_%d : Link UP on Physical Port %d\n", + evt->physical_port); break; default: - SE_DEBUG(DBG_LVL_1, "Unexpected Async Notification %d on" - "Physical Port %d\n", - evt->port_link_status, - evt->physical_port); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_INIT, + "BC_%d : Unexpected Async Notification %d on" + "Physical Port %d\n", + evt->port_link_status, + evt->physical_port); } } @@ -279,9 +293,11 @@ int beiscsi_process_mcc(struct beiscsi_hba *phba) beiscsi_async_link_state_process(phba, (struct be_async_event_link_state *) compl); else - SE_DEBUG(DBG_LVL_1, - " Unsupported Async Event, flags" - " = 0x%08x\n", compl->flags); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | + BEISCSI_LOG_MBOX, + "BC_%d : Unsupported Async Event, flags" + " = 0x%08x\n", compl->flags); } else if (compl->flags & CQE_FLAGS_COMPLETED_MASK) { status = be_mcc_compl_process(ctrl, compl); @@ -312,7 +328,10 @@ static int be_mcc_wait_compl(struct beiscsi_hba *phba) udelay(100); } if (i == mcc_timeout) { - dev_err(&phba->pcidev->dev, "mccq poll timed out\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : mccq poll timed out\n"); + return -EBUSY; } return 0; @@ -338,7 +357,11 @@ static int be_mbox_db_ready_wait(struct be_ctrl_info *ctrl) break; if (cnt > 12000000) { - dev_err(&ctrl->pdev->dev, "mbox_db poll timed out\n"); + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : mbox_db poll timed out\n"); + return -EBUSY; } @@ -360,6 +383,7 @@ int be_mbox_notify(struct be_ctrl_info *ctrl) struct be_dma_mem *mbox_mem = &ctrl->mbox_mem; struct be_mcc_mailbox *mbox = mbox_mem->va; struct be_mcc_compl *compl = &mbox->compl; + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); val &= ~MPU_MAILBOX_DB_RDY_MASK; val |= MPU_MAILBOX_DB_HI_MASK; @@ -368,7 +392,10 @@ int be_mbox_notify(struct be_ctrl_info *ctrl) status = be_mbox_db_ready_wait(ctrl); if (status != 0) { - SE_DEBUG(DBG_LVL_1, " be_mbox_db_ready_wait failed\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : be_mbox_db_ready_wait failed\n"); + return status; } val = 0; @@ -379,18 +406,27 @@ int be_mbox_notify(struct be_ctrl_info *ctrl) status = be_mbox_db_ready_wait(ctrl); if (status != 0) { - SE_DEBUG(DBG_LVL_1, " be_mbox_db_ready_wait failed\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : be_mbox_db_ready_wait failed\n"); + return status; } if (be_mcc_compl_is_new(compl)) { status = be_mcc_compl_process(ctrl, &mbox->compl); be_mcc_compl_use(compl); if (status) { - SE_DEBUG(DBG_LVL_1, "After be_mcc_compl_process\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : After be_mcc_compl_process\n"); + return status; } } else { - dev_err(&ctrl->pdev->dev, "invalid mailbox completion\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : Invalid Mailbox Completion\n"); + return -EBUSY; } return 0; @@ -436,7 +472,10 @@ static int be_mbox_notify_wait(struct beiscsi_hba *phba) if (status) return status; } else { - dev_err(&phba->pcidev->dev, "invalid mailbox completion\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BC_%d : invalid mailbox completion\n"); + return -EBUSY; } return 0; @@ -528,7 +567,6 @@ int beiscsi_cmd_eq_create(struct be_ctrl_info *ctrl, struct be_dma_mem *q_mem = &eq->dma_mem; int status; - SE_DEBUG(DBG_LVL_8, "In beiscsi_cmd_eq_create\n"); spin_lock(&ctrl->mbox_lock); memset(wrb, 0, sizeof(*wrb)); @@ -563,10 +601,10 @@ int beiscsi_cmd_eq_create(struct be_ctrl_info *ctrl, int be_cmd_fw_initialize(struct be_ctrl_info *ctrl) { struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem); + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); int status; u8 *endian_check; - SE_DEBUG(DBG_LVL_8, "In be_cmd_fw_initialize\n"); spin_lock(&ctrl->mbox_lock); memset(wrb, 0, sizeof(*wrb)); @@ -583,7 +621,8 @@ int be_cmd_fw_initialize(struct be_ctrl_info *ctrl) status = be_mbox_notify(ctrl); if (status) - SE_DEBUG(DBG_LVL_1, "be_cmd_fw_initialize Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BC_%d : be_cmd_fw_initialize Failed\n"); spin_unlock(&ctrl->mbox_lock); return status; @@ -596,11 +635,11 @@ int beiscsi_cmd_cq_create(struct be_ctrl_info *ctrl, struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem); struct be_cmd_req_cq_create *req = embedded_payload(wrb); struct be_cmd_resp_cq_create *resp = embedded_payload(wrb); + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); struct be_dma_mem *q_mem = &cq->dma_mem; void *ctxt = &req->context; int status; - SE_DEBUG(DBG_LVL_8, "In beiscsi_cmd_cq_create\n"); spin_lock(&ctrl->mbox_lock); memset(wrb, 0, sizeof(*wrb)); @@ -608,8 +647,6 @@ int beiscsi_cmd_cq_create(struct be_ctrl_info *ctrl, be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_COMMON, OPCODE_COMMON_CQ_CREATE, sizeof(*req)); - if (!q_mem->va) - SE_DEBUG(DBG_LVL_1, "uninitialized q_mem->va\n"); req->num_pages = cpu_to_le16(PAGES_4K_SPANNED(q_mem->va, q_mem->size)); @@ -633,8 +670,10 @@ int beiscsi_cmd_cq_create(struct be_ctrl_info *ctrl, cq->id = le16_to_cpu(resp->cq_id); cq->created = true; } else - SE_DEBUG(DBG_LVL_1, "In be_cmd_cq_create, status=ox%08x\n", - status); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BC_%d : In be_cmd_cq_create, status=ox%08x\n", + status); + spin_unlock(&ctrl->mbox_lock); return status; @@ -700,10 +739,14 @@ int beiscsi_cmd_q_destroy(struct be_ctrl_info *ctrl, struct be_queue_info *q, { struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem); struct be_cmd_req_q_destroy *req = embedded_payload(wrb); + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); u8 subsys = 0, opcode = 0; int status; - SE_DEBUG(DBG_LVL_8, "In beiscsi_cmd_q_destroy\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BC_%d : In beiscsi_cmd_q_destroy " + "queue_type : %d\n", queue_type); + spin_lock(&ctrl->mbox_lock); memset(wrb, 0, sizeof(*wrb)); be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0); @@ -759,7 +802,6 @@ int be_cmd_create_default_pdu_queue(struct be_ctrl_info *ctrl, void *ctxt = &req->context; int status; - SE_DEBUG(DBG_LVL_8, "In be_cmd_create_default_pdu_queue\n"); spin_lock(&ctrl->mbox_lock); memset(wrb, 0, sizeof(*wrb)); @@ -830,6 +872,7 @@ int be_cmd_iscsi_post_sgl_pages(struct be_ctrl_info *ctrl, { struct be_mcc_wrb *wrb = wrb_from_mbox(&ctrl->mbox_mem); struct be_post_sgl_pages_req *req = embedded_payload(wrb); + struct beiscsi_hba *phba = pci_get_drvdata(ctrl->pdev); int status; unsigned int curr_pages; u32 internal_page_offset = 0; @@ -860,8 +903,9 @@ int be_cmd_iscsi_post_sgl_pages(struct be_ctrl_info *ctrl, status = be_mbox_notify(ctrl); if (status) { - SE_DEBUG(DBG_LVL_1, - "FW CMD to map iscsi frags failed.\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BC_%d : FW CMD to map iscsi frags failed.\n"); + goto error; } } while (num_pages > 0); @@ -890,3 +934,45 @@ int beiscsi_cmd_reset_function(struct beiscsi_hba *phba) spin_unlock(&ctrl->mbox_lock); return status; } + +/** + * be_cmd_set_vlan()- Configure VLAN paramters on the adapter + * @phba: device priv structure instance + * @vlan_tag: TAG to be set + * + * Set the VLAN_TAG for the adapter or Disable VLAN on adapter + * + * returns + * TAG for the MBX Cmd + * **/ +int be_cmd_set_vlan(struct beiscsi_hba *phba, + uint16_t vlan_tag) +{ + unsigned int tag = 0; + struct be_mcc_wrb *wrb; + struct be_cmd_set_vlan_req *req; + struct be_ctrl_info *ctrl = &phba->ctrl; + + spin_lock(&ctrl->mbox_lock); + tag = alloc_mcc_tag(phba); + if (!tag) { + spin_unlock(&ctrl->mbox_lock); + return tag; + } + + wrb = wrb_from_mccq(phba); + req = embedded_payload(wrb); + wrb->tag0 |= tag; + be_wrb_hdr_prepare(wrb, sizeof(*wrb), true, 0); + be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI, + OPCODE_COMMON_ISCSI_NTWK_SET_VLAN, + sizeof(*req)); + + req->interface_hndl = phba->interface_handle; + req->vlan_priority = vlan_tag; + + be_mcc_notify(phba); + spin_unlock(&ctrl->mbox_lock); + + return tag; +} diff --git a/drivers/scsi/be2iscsi/be_cmds.h b/drivers/scsi/be2iscsi/be_cmds.h index b0b36c6..2c8f98d 100644 --- a/drivers/scsi/be2iscsi/be_cmds.h +++ b/drivers/scsi/be2iscsi/be_cmds.h @@ -348,6 +348,23 @@ struct be_cmd_get_boot_target_resp { int boot_session_handle; }; +struct be_cmd_reopen_session_req { + struct be_cmd_req_hdr hdr; +#define BE_REOPEN_ALL_SESSIONS 0x00 +#define BE_REOPEN_BOOT_SESSIONS 0x01 +#define BE_REOPEN_A_SESSION 0x02 + u16 reopen_type; + u16 rsvd; + u32 session_handle; +} __packed; + +struct be_cmd_reopen_session_resp { + struct be_cmd_resp_hdr hdr; + u32 rsvd; + u32 session_handle; +} __packed; + + struct be_cmd_mac_query_req { struct be_cmd_req_hdr hdr; u8 type; @@ -432,6 +449,12 @@ struct be_cmd_get_def_gateway_resp { struct ip_addr_format ip_addr; } __packed; +#define BEISCSI_VLAN_DISABLE 0xFFFF +struct be_cmd_set_vlan_req { + struct be_cmd_req_hdr hdr; + u32 interface_hndl; + u32 vlan_priority; +} __packed; /******************** Create CQ ***************************/ /** * Pseudo amap definition in which each bit of the actual structure is defined @@ -671,6 +694,9 @@ int be_cmd_wrbq_create(struct be_ctrl_info *ctrl, struct be_dma_mem *q_mem, bool is_link_state_evt(u32 trailer); +/* Configuration Functions */ +int be_cmd_set_vlan(struct beiscsi_hba *phba, uint16_t vlan_tag); + struct be_default_pdu_context { u32 dw[4]; } __packed; @@ -911,6 +937,7 @@ struct be_cmd_get_all_if_id_req { #define OPCODE_ISCSI_INI_CFG_GET_HBA_NAME 6 #define OPCODE_ISCSI_INI_CFG_SET_HBA_NAME 7 #define OPCODE_ISCSI_INI_SESSION_GET_A_SESSION 14 +#define OPCODE_ISCSI_INI_DRIVER_REOPEN_ALL_SESSIONS 36 #define OPCODE_ISCSI_INI_DRIVER_OFFLOAD_SESSION 41 #define OPCODE_ISCSI_INI_DRIVER_INVALIDATE_CONNECTION 42 #define OPCODE_ISCSI_INI_BOOT_GET_BOOT_TARGET 52 diff --git a/drivers/scsi/be2iscsi/be_iscsi.c b/drivers/scsi/be2iscsi/be_iscsi.c index 43f3503..aedb0d9 100644 --- a/drivers/scsi/be2iscsi/be_iscsi.c +++ b/drivers/scsi/be2iscsi/be_iscsi.c @@ -50,21 +50,27 @@ struct iscsi_cls_session *beiscsi_session_create(struct iscsi_endpoint *ep, struct beiscsi_session *beiscsi_sess; struct beiscsi_io_task *io_task; - SE_DEBUG(DBG_LVL_8, "In beiscsi_session_create\n"); if (!ep) { - SE_DEBUG(DBG_LVL_1, "beiscsi_session_create: invalid ep\n"); + printk(KERN_ERR + "beiscsi_session_create: invalid ep\n"); return NULL; } beiscsi_ep = ep->dd_data; phba = beiscsi_ep->phba; shost = phba->shost; + + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_session_create\n"); + if (cmds_max > beiscsi_ep->phba->params.wrbs_per_cxn) { - shost_printk(KERN_ERR, shost, "Cannot handle %d cmds." - "Max cmds per session supported is %d. Using %d. " - "\n", cmds_max, - beiscsi_ep->phba->params.wrbs_per_cxn, - beiscsi_ep->phba->params.wrbs_per_cxn); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Cannot handle %d cmds." + "Max cmds per session supported is %d. Using %d." + "\n", cmds_max, + beiscsi_ep->phba->params.wrbs_per_cxn, + beiscsi_ep->phba->params.wrbs_per_cxn); + cmds_max = beiscsi_ep->phba->params.wrbs_per_cxn; } @@ -102,7 +108,7 @@ void beiscsi_session_destroy(struct iscsi_cls_session *cls_session) struct iscsi_session *sess = cls_session->dd_data; struct beiscsi_session *beiscsi_sess = sess->dd_data; - SE_DEBUG(DBG_LVL_8, "In beiscsi_session_destroy\n"); + printk(KERN_INFO "In beiscsi_session_destroy\n"); pci_pool_destroy(beiscsi_sess->bhs_pool); iscsi_session_teardown(cls_session); } @@ -123,11 +129,13 @@ beiscsi_conn_create(struct iscsi_cls_session *cls_session, u32 cid) struct iscsi_session *sess; struct beiscsi_session *beiscsi_sess; - SE_DEBUG(DBG_LVL_8, "In beiscsi_conn_create ,cid" - "from iscsi layer=%d\n", cid); shost = iscsi_session_to_shost(cls_session); phba = iscsi_host_priv(shost); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_conn_create ,cid" + "from iscsi layer=%d\n", cid); + cls_conn = iscsi_conn_setup(cls_session, sizeof(*beiscsi_conn), cid); if (!cls_conn) return NULL; @@ -154,12 +162,15 @@ static int beiscsi_bindconn_cid(struct beiscsi_hba *phba, unsigned int cid) { if (phba->conn_table[cid]) { - SE_DEBUG(DBG_LVL_1, - "Connection table already occupied. Detected clash\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Connection table already occupied. Detected clash\n"); + return -EINVAL; } else { - SE_DEBUG(DBG_LVL_8, "phba->conn_table[%d]=%p(beiscsi_conn)\n", - cid, beiscsi_conn); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : phba->conn_table[%d]=%p(beiscsi_conn)\n", + cid, beiscsi_conn); + phba->conn_table[cid] = beiscsi_conn; } return 0; @@ -184,7 +195,6 @@ int beiscsi_conn_bind(struct iscsi_cls_session *cls_session, struct beiscsi_endpoint *beiscsi_ep; struct iscsi_endpoint *ep; - SE_DEBUG(DBG_LVL_8, "In beiscsi_conn_bind\n"); ep = iscsi_lookup_endpoint(transport_fd); if (!ep) return -EINVAL; @@ -195,17 +205,21 @@ int beiscsi_conn_bind(struct iscsi_cls_session *cls_session, return -EINVAL; if (beiscsi_ep->phba != phba) { - SE_DEBUG(DBG_LVL_8, - "beiscsi_ep->hba=%p not equal to phba=%p\n", - beiscsi_ep->phba, phba); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : beiscsi_ep->hba=%p not equal to phba=%p\n", + beiscsi_ep->phba, phba); + return -EEXIST; } beiscsi_conn->beiscsi_conn_cid = beiscsi_ep->ep_cid; beiscsi_conn->ep = beiscsi_ep; beiscsi_ep->conn = beiscsi_conn; - SE_DEBUG(DBG_LVL_8, "beiscsi_conn=%p conn=%p ep_cid=%d\n", - beiscsi_conn, conn, beiscsi_ep->ep_cid); + + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : beiscsi_conn=%p conn=%p ep_cid=%d\n", + beiscsi_conn, conn, beiscsi_ep->ep_cid); + return beiscsi_bindconn_cid(phba, beiscsi_conn, beiscsi_ep->ep_cid); } @@ -219,8 +233,9 @@ static int beiscsi_create_ipv4_iface(struct beiscsi_hba *phba) ISCSI_IFACE_TYPE_IPV4, 0, 0); if (!phba->ipv4_iface) { - shost_printk(KERN_ERR, phba->shost, "Could not " - "create default IPv4 address.\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Could not " + "create default IPv4 address.\n"); return -ENODEV; } @@ -237,8 +252,9 @@ static int beiscsi_create_ipv6_iface(struct beiscsi_hba *phba) ISCSI_IFACE_TYPE_IPV6, 0, 0); if (!phba->ipv6_iface) { - shost_printk(KERN_ERR, phba->shost, "Could not " - "create default IPv6 address.\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Could not " + "create default IPv6 address.\n"); return -ENODEV; } @@ -299,12 +315,14 @@ beiscsi_set_static_ip(struct Scsi_Host *shost, iface_ip = nla_data(nla); break; default: - shost_printk(KERN_ERR, shost, "Unsupported param %d\n", - iface_param->param); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Unsupported param %d\n", + iface_param->param); } if (!iface_ip || !iface_subnet) { - shost_printk(KERN_ERR, shost, "IP and Subnet Mask required\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : IP and Subnet Mask required\n"); return -EINVAL; } @@ -314,6 +332,51 @@ beiscsi_set_static_ip(struct Scsi_Host *shost, return ret; } +/** + * beiscsi_set_vlan_tag()- Set the VLAN TAG + * @shost: Scsi Host for the driver instance + * @iface_param: Interface paramters + * + * Set the VLAN TAG for the adapter or disable + * the VLAN config + * + * returns + * Success: 0 + * Failure: Non-Zero Value + **/ +static int +beiscsi_set_vlan_tag(struct Scsi_Host *shost, + struct iscsi_iface_param_info *iface_param) +{ + struct beiscsi_hba *phba = iscsi_host_priv(shost); + int ret = 0; + + /* Get the Interface Handle */ + if (mgmt_get_all_if_id(phba)) { + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Getting Interface Handle Failed\n"); + return -EIO; + } + + switch (iface_param->param) { + case ISCSI_NET_PARAM_VLAN_ENABLED: + if (iface_param->value[0] != ISCSI_VLAN_ENABLE) + ret = mgmt_set_vlan(phba, BEISCSI_VLAN_DISABLE); + break; + case ISCSI_NET_PARAM_VLAN_TAG: + ret = mgmt_set_vlan(phba, + *((uint16_t *)iface_param->value)); + break; + default: + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BS_%d : Unkown Param Type : %d\n", + iface_param->param); + return -ENOSYS; + } + return ret; +} + + static int beiscsi_set_ipv4(struct Scsi_Host *shost, struct iscsi_iface_param_info *iface_param, @@ -335,8 +398,9 @@ beiscsi_set_ipv4(struct Scsi_Host *shost, ret = beiscsi_set_static_ip(shost, iface_param, data, dt_len); else - shost_printk(KERN_ERR, shost, "Invalid BOOTPROTO: %d\n", - iface_param->value[0]); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Invalid BOOTPROTO: %d\n", + iface_param->value[0]); break; case ISCSI_NET_PARAM_IFACE_ENABLE: if (iface_param->value[0] == ISCSI_IFACE_ENABLE) @@ -349,9 +413,14 @@ beiscsi_set_ipv4(struct Scsi_Host *shost, ret = beiscsi_set_static_ip(shost, iface_param, data, dt_len); break; + case ISCSI_NET_PARAM_VLAN_ENABLED: + case ISCSI_NET_PARAM_VLAN_TAG: + ret = beiscsi_set_vlan_tag(shost, iface_param); + break; default: - shost_printk(KERN_ERR, shost, "Param %d not supported\n", - iface_param->param); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Param %d not supported\n", + iface_param->param); } return ret; @@ -379,8 +448,9 @@ beiscsi_set_ipv6(struct Scsi_Host *shost, ISCSI_BOOTPROTO_STATIC); break; default: - shost_printk(KERN_ERR, shost, "Param %d not supported\n", - iface_param->param); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Param %d not supported\n", + iface_param->param); } return ret; @@ -390,6 +460,7 @@ int be2iscsi_iface_set_param(struct Scsi_Host *shost, void *data, uint32_t dt_len) { struct iscsi_iface_param_info *iface_param = NULL; + struct beiscsi_hba *phba = iscsi_host_priv(shost); struct nlattr *attrib; uint32_t rm_len = dt_len; int ret = 0 ; @@ -404,9 +475,11 @@ int be2iscsi_iface_set_param(struct Scsi_Host *shost, * BE2ISCSI only supports 1 interface */ if (iface_param->iface_num) { - shost_printk(KERN_ERR, shost, "Invalid iface_num %d." - "Only iface_num 0 is supported.\n", - iface_param->iface_num); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Invalid iface_num %d." + "Only iface_num 0 is supported.\n", + iface_param->iface_num); + return -EINVAL; } @@ -420,9 +493,9 @@ int be2iscsi_iface_set_param(struct Scsi_Host *shost, data, dt_len); break; default: - shost_printk(KERN_ERR, shost, - "Invalid iface type :%d passed\n", - iface_param->iface_type); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Invalid iface type :%d passed\n", + iface_param->iface_type); break; } @@ -465,6 +538,27 @@ static int be2iscsi_get_if_param(struct beiscsi_hba *phba, case ISCSI_NET_PARAM_IPV4_SUBNET: len = sprintf(buf, "%pI4\n", &if_info.ip_addr.subnet_mask); break; + case ISCSI_NET_PARAM_VLAN_ENABLED: + len = sprintf(buf, "%s\n", + (if_info.vlan_priority == BEISCSI_VLAN_DISABLE) + ? "Disabled" : "Enabled"); + break; + case ISCSI_NET_PARAM_VLAN_ID: + if (if_info.vlan_priority == BEISCSI_VLAN_DISABLE) + return -EINVAL; + else + len = sprintf(buf, "%d\n", + (if_info.vlan_priority & + ISCSI_MAX_VLAN_ID)); + break; + case ISCSI_NET_PARAM_VLAN_PRIORITY: + if (if_info.vlan_priority == BEISCSI_VLAN_DISABLE) + return -EINVAL; + else + len = sprintf(buf, "%d\n", + ((if_info.vlan_priority >> 13) & + ISCSI_MAX_VLAN_PRIORITY)); + break; default: WARN_ON(1); } @@ -486,6 +580,9 @@ int be2iscsi_iface_get_param(struct iscsi_iface *iface, case ISCSI_NET_PARAM_IPV4_SUBNET: case ISCSI_NET_PARAM_IPV4_BOOTPROTO: case ISCSI_NET_PARAM_IPV6_ADDR: + case ISCSI_NET_PARAM_VLAN_ENABLED: + case ISCSI_NET_PARAM_VLAN_ID: + case ISCSI_NET_PARAM_VLAN_PRIORITY: len = be2iscsi_get_if_param(phba, iface, param, buf); break; case ISCSI_NET_PARAM_IFACE_ENABLE: @@ -518,7 +615,10 @@ int beiscsi_ep_get_param(struct iscsi_endpoint *ep, struct beiscsi_endpoint *beiscsi_ep = ep->dd_data; int len = 0; - SE_DEBUG(DBG_LVL_8, "In beiscsi_ep_get_param, param= %d\n", param); + beiscsi_log(beiscsi_ep->phba, KERN_INFO, + BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_ep_get_param," + " param= %d\n", param); switch (param) { case ISCSI_PARAM_CONN_PORT: @@ -541,9 +641,14 @@ int beiscsi_set_param(struct iscsi_cls_conn *cls_conn, { struct iscsi_conn *conn = cls_conn->dd_data; struct iscsi_session *session = conn->session; + struct beiscsi_hba *phba = NULL; int ret; - SE_DEBUG(DBG_LVL_8, "In beiscsi_conn_set_param, param= %d\n", param); + phba = ((struct beiscsi_conn *)conn->dd_data)->phba; + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_conn_set_param," + " param= %d\n", param); + ret = iscsi_set_param(cls_conn, param, buf, buflen); if (ret) return ret; @@ -593,7 +698,9 @@ static int beiscsi_get_initname(char *buf, struct beiscsi_hba *phba) tag = be_cmd_get_initname(phba); if (!tag) { - SE_DEBUG(DBG_LVL_1, "Getting Initiator Name Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Getting Initiator Name Failed\n"); + return -EBUSY; } else wait_event_interruptible(phba->ctrl.mcc_wait[tag], @@ -604,9 +711,12 @@ static int beiscsi_get_initname(char *buf, struct beiscsi_hba *phba) status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, "MailBox Command Failed with " - "status = %d extd_status = %d\n", - status, extd_status); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BS_%d : MailBox Command Failed with " + "status = %d extd_status = %d\n", + status, extd_status); + free_mcc_tag(&phba->ctrl, tag); return -EAGAIN; } @@ -650,7 +760,9 @@ static int beiscsi_get_port_speed(struct Scsi_Host *shost) tag = be_cmd_get_port_speed(phba); if (!tag) { - SE_DEBUG(DBG_LVL_1, "Getting Port Speed Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Getting Port Speed Failed\n"); + return -EBUSY; } else wait_event_interruptible(phba->ctrl.mcc_wait[tag], @@ -661,9 +773,12 @@ static int beiscsi_get_port_speed(struct Scsi_Host *shost) status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, "MailBox Command Failed with " - "status = %d extd_status = %d\n", - status, extd_status); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BS_%d : MailBox Command Failed with " + "status = %d extd_status = %d\n", + status, extd_status); + free_mcc_tag(&phba->ctrl, tag); return -EAGAIN; } @@ -704,20 +819,24 @@ int beiscsi_get_host_param(struct Scsi_Host *shost, struct beiscsi_hba *phba = iscsi_host_priv(shost); int status = 0; - SE_DEBUG(DBG_LVL_8, "In beiscsi_get_host_param, param= %d\n", param); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_get_host_param," + " param= %d\n", param); + switch (param) { case ISCSI_HOST_PARAM_HWADDRESS: status = beiscsi_get_macaddr(buf, phba); if (status < 0) { - SE_DEBUG(DBG_LVL_1, "beiscsi_get_macaddr Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : beiscsi_get_macaddr Failed\n"); return status; } break; case ISCSI_HOST_PARAM_INITIATOR_NAME: status = beiscsi_get_initname(buf, phba); if (status < 0) { - SE_DEBUG(DBG_LVL_1, - "Retreiving Initiator Name Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Retreiving Initiator Name Failed\n"); return status; } break; @@ -728,8 +847,8 @@ int beiscsi_get_host_param(struct Scsi_Host *shost, case ISCSI_HOST_PARAM_PORT_SPEED: status = beiscsi_get_port_speed(shost); if (status) { - SE_DEBUG(DBG_LVL_1, - "Retreiving Port Speed Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Retreiving Port Speed Failed\n"); return status; } status = sprintf(buf, "%s\n", iscsi_get_port_speed_name(shost)); @@ -746,7 +865,7 @@ int beiscsi_get_macaddr(char *buf, struct beiscsi_hba *phba) int rc; if (strlen(phba->mac_address)) - return strlcpy(buf, phba->mac_address, PAGE_SIZE); + return sysfs_format_mac(buf, phba->mac_address, ETH_ALEN); memset(&resp, 0, sizeof(resp)); rc = mgmt_get_nic_conf(phba, &resp); @@ -768,8 +887,12 @@ void beiscsi_conn_get_stats(struct iscsi_cls_conn *cls_conn, struct iscsi_stats *stats) { struct iscsi_conn *conn = cls_conn->dd_data; + struct beiscsi_hba *phba = NULL; + + phba = ((struct beiscsi_conn *)conn->dd_data)->phba; + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_conn_get_stats\n"); - SE_DEBUG(DBG_LVL_8, "In beiscsi_conn_get_stats\n"); stats->txdata_octets = conn->txdata_octets; stats->rxdata_octets = conn->rxdata_octets; stats->dataout_pdus = conn->dataout_pdus_cnt; @@ -829,11 +952,16 @@ int beiscsi_conn_start(struct iscsi_cls_conn *cls_conn) struct beiscsi_endpoint *beiscsi_ep; struct beiscsi_offload_params params; - SE_DEBUG(DBG_LVL_8, "In beiscsi_conn_start\n"); + beiscsi_log(beiscsi_conn->phba, KERN_INFO, + BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_conn_start\n"); + memset(¶ms, 0, sizeof(struct beiscsi_offload_params)); beiscsi_ep = beiscsi_conn->ep; if (!beiscsi_ep) - SE_DEBUG(DBG_LVL_1, "In beiscsi_conn_start , no beiscsi_ep\n"); + beiscsi_log(beiscsi_conn->phba, KERN_ERR, + BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_conn_start , no beiscsi_ep\n"); beiscsi_conn->login_in_progress = 0; beiscsi_set_params_for_offld(beiscsi_conn, ¶ms); @@ -907,19 +1035,27 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep, unsigned int tag, wrb_num; int ret = -ENOMEM; - SE_DEBUG(DBG_LVL_8, "In beiscsi_open_conn\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_open_conn\n"); + beiscsi_ep->ep_cid = beiscsi_get_cid(phba); if (beiscsi_ep->ep_cid == 0xFFFF) { - SE_DEBUG(DBG_LVL_1, "No free cid available\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : No free cid available\n"); return ret; } - SE_DEBUG(DBG_LVL_8, "In beiscsi_open_conn, ep_cid=%d\n", - beiscsi_ep->ep_cid); + + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_open_conn, ep_cid=%d\n", + beiscsi_ep->ep_cid); + phba->ep_array[beiscsi_ep->ep_cid - phba->fw_config.iscsi_cid_start] = ep; if (beiscsi_ep->ep_cid > (phba->fw_config.iscsi_cid_start + phba->params.cxns_per_ctrl * 2)) { - SE_DEBUG(DBG_LVL_1, "Failed in allocate iscsi cid\n"); + + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Failed in allocate iscsi cid\n"); goto free_ep; } @@ -928,9 +1064,11 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep, sizeof(struct tcp_connect_and_offload_in), &nonemb_cmd.dma); if (nonemb_cmd.va == NULL) { - SE_DEBUG(DBG_LVL_1, - "Failed to allocate memory for mgmt_open_connection" - "\n"); + + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Failed to allocate memory for" + " mgmt_open_connection\n"); + beiscsi_put_cid(phba, beiscsi_ep->ep_cid); return -ENOMEM; } @@ -938,9 +1076,10 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep, memset(nonemb_cmd.va, 0, nonemb_cmd.size); tag = mgmt_open_connection(phba, dst_addr, beiscsi_ep, &nonemb_cmd); if (!tag) { - SE_DEBUG(DBG_LVL_1, - "mgmt_open_connection Failed for cid=%d\n", - beiscsi_ep->ep_cid); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : mgmt_open_connection Failed for cid=%d\n", + beiscsi_ep->ep_cid); + beiscsi_put_cid(phba, beiscsi_ep->ep_cid); pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); @@ -953,9 +1092,12 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep, extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, "mgmt_open_connection Failed" - " status = %d extd_status = %d\n", - status, extd_status); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BS_%d : mgmt_open_connection Failed" + " status = %d extd_status = %d\n", + status, extd_status); + free_mcc_tag(&phba->ctrl, tag); pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); @@ -968,7 +1110,8 @@ static int beiscsi_open_conn(struct iscsi_endpoint *ep, beiscsi_ep = ep->dd_data; beiscsi_ep->fw_handle = ptcpcnct_out->connection_handle; beiscsi_ep->cid_vld = 1; - SE_DEBUG(DBG_LVL_8, "mgmt_open_connection Success\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : mgmt_open_connection Success\n"); } pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); @@ -996,18 +1139,19 @@ beiscsi_ep_connect(struct Scsi_Host *shost, struct sockaddr *dst_addr, struct iscsi_endpoint *ep; int ret; - SE_DEBUG(DBG_LVL_8, "In beiscsi_ep_connect\n"); if (shost) phba = iscsi_host_priv(shost); else { ret = -ENXIO; - SE_DEBUG(DBG_LVL_1, "shost is NULL\n"); + printk(KERN_ERR + "beiscsi_ep_connect shost is NULL\n"); return ERR_PTR(ret); } if (phba->state != BE_ADAPTER_UP) { ret = -EBUSY; - SE_DEBUG(DBG_LVL_1, "The Adapter state is Not UP\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : The Adapter state is Not UP\n"); return ERR_PTR(ret); } @@ -1022,7 +1166,8 @@ beiscsi_ep_connect(struct Scsi_Host *shost, struct sockaddr *dst_addr, beiscsi_ep->openiscsi_ep = ep; ret = beiscsi_open_conn(ep, NULL, dst_addr, non_blocking); if (ret) { - SE_DEBUG(DBG_LVL_1, "Failed in beiscsi_open_conn\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : Failed in beiscsi_open_conn\n"); goto free_ep; } @@ -1044,7 +1189,9 @@ int beiscsi_ep_poll(struct iscsi_endpoint *ep, int timeout_ms) { struct beiscsi_endpoint *beiscsi_ep = ep->dd_data; - SE_DEBUG(DBG_LVL_8, "In beiscsi_ep_poll\n"); + beiscsi_log(beiscsi_ep->phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_ep_poll\n"); + if (beiscsi_ep->cid_vld == 1) return 1; else @@ -1064,8 +1211,10 @@ static int beiscsi_close_conn(struct beiscsi_endpoint *beiscsi_ep, int flag) tag = mgmt_upload_connection(phba, beiscsi_ep->ep_cid, flag); if (!tag) { - SE_DEBUG(DBG_LVL_8, "upload failed for cid 0x%x\n", - beiscsi_ep->ep_cid); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : upload failed for cid 0x%x\n", + beiscsi_ep->ep_cid); + ret = -EAGAIN; } else { wait_event_interruptible(phba->ctrl.mcc_wait[tag], @@ -1086,7 +1235,8 @@ static int beiscsi_unbind_conn_to_cid(struct beiscsi_hba *phba, if (phba->conn_table[cid]) phba->conn_table[cid] = NULL; else { - SE_DEBUG(DBG_LVL_8, "Connection table Not occupied.\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : Connection table Not occupied.\n"); return -EINVAL; } return 0; @@ -1104,38 +1254,40 @@ void beiscsi_ep_disconnect(struct iscsi_endpoint *ep) struct beiscsi_endpoint *beiscsi_ep; struct beiscsi_hba *phba; unsigned int tag; + uint8_t mgmt_invalidate_flag, tcp_upload_flag; unsigned short savecfg_flag = CMD_ISCSI_SESSION_SAVE_CFG_ON_FLASH; beiscsi_ep = ep->dd_data; phba = beiscsi_ep->phba; - SE_DEBUG(DBG_LVL_8, "In beiscsi_ep_disconnect for ep_cid = %d\n", - beiscsi_ep->ep_cid); - - if (!beiscsi_ep->conn) { - SE_DEBUG(DBG_LVL_8, "In beiscsi_ep_disconnect, no " - "beiscsi_ep\n"); - return; + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BS_%d : In beiscsi_ep_disconnect for ep_cid = %d\n", + beiscsi_ep->ep_cid); + + if (beiscsi_ep->conn) { + beiscsi_conn = beiscsi_ep->conn; + iscsi_suspend_queue(beiscsi_conn->conn); + mgmt_invalidate_flag = ~BEISCSI_NO_RST_ISSUE; + tcp_upload_flag = CONNECTION_UPLOAD_GRACEFUL; + } else { + mgmt_invalidate_flag = BEISCSI_NO_RST_ISSUE; + tcp_upload_flag = CONNECTION_UPLOAD_ABORT; } - beiscsi_conn = beiscsi_ep->conn; - iscsi_suspend_queue(beiscsi_conn->conn); - - SE_DEBUG(DBG_LVL_8, "In beiscsi_ep_disconnect ep_cid = %d\n", - beiscsi_ep->ep_cid); tag = mgmt_invalidate_connection(phba, beiscsi_ep, - beiscsi_ep->ep_cid, 1, - savecfg_flag); + beiscsi_ep->ep_cid, + mgmt_invalidate_flag, + savecfg_flag); if (!tag) { - SE_DEBUG(DBG_LVL_1, - "mgmt_invalidate_connection Failed for cid=%d\n", - beiscsi_ep->ep_cid); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BS_%d : mgmt_invalidate_connection Failed for cid=%d\n", + beiscsi_ep->ep_cid); } else { wait_event_interruptible(phba->ctrl.mcc_wait[tag], phba->ctrl.mcc_numtag[tag]); free_mcc_tag(&phba->ctrl, tag); } - beiscsi_close_conn(beiscsi_ep, CONNECTION_UPLOAD_GRACEFUL); + beiscsi_close_conn(beiscsi_ep, tcp_upload_flag); beiscsi_free_ep(beiscsi_ep); beiscsi_unbind_conn_to_cid(phba, beiscsi_ep->ep_cid); iscsi_destroy_endpoint(beiscsi_ep->openiscsi_ep); @@ -1152,6 +1304,9 @@ umode_t be2iscsi_attr_is_visible(int param_type, int param) case ISCSI_NET_PARAM_IPV4_BOOTPROTO: case ISCSI_NET_PARAM_IPV4_GW: case ISCSI_NET_PARAM_IPV6_ADDR: + case ISCSI_NET_PARAM_VLAN_ID: + case ISCSI_NET_PARAM_VLAN_PRIORITY: + case ISCSI_NET_PARAM_VLAN_ENABLED: return S_IRUGO; default: return 0; diff --git a/drivers/scsi/be2iscsi/be_main.c b/drivers/scsi/be2iscsi/be_main.c index 0b1d99c..ff73f95 100644 --- a/drivers/scsi/be2iscsi/be_main.c +++ b/drivers/scsi/be2iscsi/be_main.c @@ -42,6 +42,7 @@ #include "be_main.h" #include "be_iscsi.h" #include "be_mgmt.h" +#include "be_cmds.h" static unsigned int be_iopoll_budget = 10; static unsigned int be_max_phys_size = 64; @@ -57,9 +58,105 @@ MODULE_LICENSE("GPL"); module_param(be_iopoll_budget, int, 0); module_param(enable_msix, int, 0); module_param(be_max_phys_size, uint, S_IRUGO); -MODULE_PARM_DESC(be_max_phys_size, "Maximum Size (In Kilobytes) of physically" - "contiguous memory that can be allocated." - "Range is 16 - 128"); +MODULE_PARM_DESC(be_max_phys_size, + "Maximum Size (In Kilobytes) of physically contiguous " + "memory that can be allocated. Range is 16 - 128"); + +#define beiscsi_disp_param(_name)\ +ssize_t \ +beiscsi_##_name##_disp(struct device *dev,\ + struct device_attribute *attrib, char *buf) \ +{ \ + struct Scsi_Host *shost = class_to_shost(dev);\ + struct beiscsi_hba *phba = iscsi_host_priv(shost); \ + uint32_t param_val = 0; \ + param_val = phba->attr_##_name;\ + return snprintf(buf, PAGE_SIZE, "%d\n",\ + phba->attr_##_name);\ +} + +#define beiscsi_change_param(_name, _minval, _maxval, _defaval)\ +int \ +beiscsi_##_name##_change(struct beiscsi_hba *phba, uint32_t val)\ +{\ + if (val >= _minval && val <= _maxval) {\ + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT,\ + "BA_%d : beiscsi_"#_name" updated "\ + "from 0x%x ==> 0x%x\n",\ + phba->attr_##_name, val); \ + phba->attr_##_name = val;\ + return 0;\ + } \ + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, \ + "BA_%d beiscsi_"#_name" attribute "\ + "cannot be updated to 0x%x, "\ + "range allowed is ["#_minval" - "#_maxval"]\n", val);\ + return -EINVAL;\ +} + +#define beiscsi_store_param(_name) \ +ssize_t \ +beiscsi_##_name##_store(struct device *dev,\ + struct device_attribute *attr, const char *buf,\ + size_t count) \ +{ \ + struct Scsi_Host *shost = class_to_shost(dev);\ + struct beiscsi_hba *phba = iscsi_host_priv(shost);\ + uint32_t param_val = 0;\ + if (!isdigit(buf[0]))\ + return -EINVAL;\ + if (sscanf(buf, "%i", ¶m_val) != 1)\ + return -EINVAL;\ + if (beiscsi_##_name##_change(phba, param_val) == 0) \ + return strlen(buf);\ + else \ + return -EINVAL;\ +} + +#define beiscsi_init_param(_name, _minval, _maxval, _defval) \ +int \ +beiscsi_##_name##_init(struct beiscsi_hba *phba, uint32_t val) \ +{ \ + if (val >= _minval && val <= _maxval) {\ + phba->attr_##_name = val;\ + return 0;\ + } \ + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT,\ + "BA_%d beiscsi_"#_name" attribute " \ + "cannot be updated to 0x%x, "\ + "range allowed is ["#_minval" - "#_maxval"]\n", val);\ + phba->attr_##_name = _defval;\ + return -EINVAL;\ +} + +#define BEISCSI_RW_ATTR(_name, _minval, _maxval, _defval, _descp) \ +static uint beiscsi_##_name = _defval;\ +module_param(beiscsi_##_name, uint, S_IRUGO);\ +MODULE_PARM_DESC(beiscsi_##_name, _descp);\ +beiscsi_disp_param(_name)\ +beiscsi_change_param(_name, _minval, _maxval, _defval)\ +beiscsi_store_param(_name)\ +beiscsi_init_param(_name, _minval, _maxval, _defval)\ +DEVICE_ATTR(beiscsi_##_name, S_IRUGO | S_IWUSR,\ + beiscsi_##_name##_disp, beiscsi_##_name##_store) + +/* + * When new log level added update the + * the MAX allowed value for log_enable + */ +BEISCSI_RW_ATTR(log_enable, 0x00, + 0xFF, 0x00, "Enable logging Bit Mask\n" + "\t\t\t\tInitialization Events : 0x01\n" + "\t\t\t\tMailbox Events : 0x02\n" + "\t\t\t\tMiscellaneous Events : 0x04\n" + "\t\t\t\tError Handling : 0x08\n" + "\t\t\t\tIO Path Events : 0x10\n" + "\t\t\t\tConfiguration Path : 0x20\n"); + +struct device_attribute *beiscsi_attrs[] = { + &dev_attr_beiscsi_log_enable, + NULL, +}; static int beiscsi_slave_configure(struct scsi_device *sdev) { @@ -112,9 +209,9 @@ static int beiscsi_eh_abort(struct scsi_cmnd *sc) sizeof(struct invalidate_commands_params_in), &nonemb_cmd.dma); if (nonemb_cmd.va == NULL) { - SE_DEBUG(DBG_LVL_1, - "Failed to allocate memory for" - "mgmt_invalidate_icds\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_EH, + "BM_%d : Failed to allocate memory for" + "mgmt_invalidate_icds\n"); return FAILED; } nonemb_cmd.size = sizeof(struct invalidate_commands_params_in); @@ -122,9 +219,9 @@ static int beiscsi_eh_abort(struct scsi_cmnd *sc) tag = mgmt_invalidate_icds(phba, inv_tbl, num_invalidate, cid, &nonemb_cmd); if (!tag) { - shost_printk(KERN_WARNING, phba->shost, - "mgmt_invalidate_icds could not be" - " submitted\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_EH, + "BM_%d : mgmt_invalidate_icds could not be" + "submitted\n"); pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); @@ -188,9 +285,9 @@ static int beiscsi_eh_device_reset(struct scsi_cmnd *sc) sizeof(struct invalidate_commands_params_in), &nonemb_cmd.dma); if (nonemb_cmd.va == NULL) { - SE_DEBUG(DBG_LVL_1, - "Failed to allocate memory for" - "mgmt_invalidate_icds\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_EH, + "BM_%d : Failed to allocate memory for" + "mgmt_invalidate_icds\n"); return FAILED; } nonemb_cmd.size = sizeof(struct invalidate_commands_params_in); @@ -198,9 +295,9 @@ static int beiscsi_eh_device_reset(struct scsi_cmnd *sc) tag = mgmt_invalidate_icds(phba, inv_tbl, num_invalidate, cid, &nonemb_cmd); if (!tag) { - shost_printk(KERN_WARNING, phba->shost, - "mgmt_invalidate_icds could not be" - " submitted\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_EH, + "BM_%d : mgmt_invalidate_icds could not be" + " submitted\n"); pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); return FAILED; @@ -389,6 +486,7 @@ static DEFINE_PCI_DEVICE_TABLE(beiscsi_pci_id_table) = { }; MODULE_DEVICE_TABLE(pci, beiscsi_pci_id_table); + static struct scsi_host_template beiscsi_sht = { .module = THIS_MODULE, .name = "Emulex 10Gbe open-iscsi Initiator Driver", @@ -400,6 +498,7 @@ static struct scsi_host_template beiscsi_sht = { .eh_abort_handler = beiscsi_eh_abort, .eh_device_reset_handler = beiscsi_eh_device_reset, .eh_target_reset_handler = iscsi_eh_session_reset, + .shost_attrs = beiscsi_attrs, .sg_tablesize = BEISCSI_SGLIST_ELEMENTS, .can_queue = BE2_IO_DEPTH, .this_id = -1, @@ -419,8 +518,8 @@ static struct beiscsi_hba *beiscsi_hba_alloc(struct pci_dev *pcidev) shost = iscsi_host_alloc(&beiscsi_sht, sizeof(*phba), 0); if (!shost) { - dev_err(&pcidev->dev, "beiscsi_hba_alloc -" - "iscsi_host_alloc failed\n"); + dev_err(&pcidev->dev, + "beiscsi_hba_alloc - iscsi_host_alloc failed\n"); return NULL; } shost->dma_boundary = pcidev->dma_mask; @@ -510,8 +609,8 @@ static int beiscsi_enable_pci(struct pci_dev *pcidev) ret = pci_enable_device(pcidev); if (ret) { - dev_err(&pcidev->dev, "beiscsi_enable_pci - enable device " - "failed. Returning -ENODEV\n"); + dev_err(&pcidev->dev, + "beiscsi_enable_pci - enable device failed\n"); return ret; } @@ -576,8 +675,9 @@ static void beiscsi_get_params(struct beiscsi_hba *phba) + BE2_TMFS) / 512) + 1) * 512; phba->params.num_eq_entries = (phba->params.num_eq_entries < 1024) ? 1024 : phba->params.num_eq_entries; - SE_DEBUG(DBG_LVL_8, "phba->params.num_eq_entries=%d\n", - phba->params.num_eq_entries); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : phba->params.num_eq_entries=%d\n", + phba->params.num_eq_entries); phba->params.num_cq_entries = (((BE2_CMDS_PER_CXN * 2 + phba->fw_config.iscsi_cid_count * 2 + BE2_TMFS) / 512) + 1) * 512; @@ -621,8 +721,6 @@ static irqreturn_t be_isr_mcc(int irq, void *dev_id) phba = pbe_eq->phba; mcc = &phba->ctrl.mcc_obj.cq; eqe = queue_tail_node(eq); - if (!eqe) - SE_DEBUG(DBG_LVL_1, "eqe is NULL\n"); num_eq_processed = 0; @@ -667,8 +765,6 @@ static irqreturn_t be_isr_msix(int irq, void *dev_id) eq = &pbe_eq->q; cq = pbe_eq->cq; eqe = queue_tail_node(eq); - if (!eqe) - SE_DEBUG(DBG_LVL_1, "eqe is NULL\n"); phba = pbe_eq->phba; num_eq_processed = 0; @@ -743,8 +839,6 @@ static irqreturn_t be_isr(int irq, void *dev_id) mcc = &phba->ctrl.mcc_obj.cq; index = 0; eqe = queue_tail_node(eq); - if (!eqe) - SE_DEBUG(DBG_LVL_1, "eqe is NULL\n"); num_ioeq_processed = 0; num_mcceq_processed = 0; @@ -842,9 +936,10 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba) phba->msi_name[i], &phwi_context->be_eq[i]); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "beiscsi_init_irqs-Failed to" - "register msix for i = %d\n", i); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_init_irqs-Failed to" + "register msix for i = %d\n", + i); kfree(phba->msi_name[i]); goto free_msix_irqs; } @@ -860,8 +955,9 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba) ret = request_irq(msix_vec, be_isr_mcc, 0, phba->msi_name[i], &phwi_context->be_eq[i]); if (ret) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_init_irqs-" - "Failed to register beiscsi_msix_mcc\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT , + "BM_%d : beiscsi_init_irqs-" + "Failed to register beiscsi_msix_mcc\n"); kfree(phba->msi_name[i]); goto free_msix_irqs; } @@ -870,8 +966,9 @@ static int beiscsi_init_irqs(struct beiscsi_hba *phba) ret = request_irq(pcidev->irq, be_isr, IRQF_SHARED, "beiscsi", phba); if (ret) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_init_irqs-" - "Failed to register irq\\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_init_irqs-" + "Failed to register irq\\n"); return ret; } } @@ -922,7 +1019,9 @@ beiscsi_process_async_pdu(struct beiscsi_conn *beiscsi_conn, case ISCSI_OP_REJECT: WARN_ON(!pbuffer); WARN_ON(!(buf_len == 48)); - SE_DEBUG(DBG_LVL_1, "In ISCSI_OP_REJECT\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d : In ISCSI_OP_REJECT\n"); break; case ISCSI_OP_LOGIN_RSP: case ISCSI_OP_TEXT_RSP: @@ -932,11 +1031,12 @@ beiscsi_process_async_pdu(struct beiscsi_conn *beiscsi_conn, login_hdr->itt = io_task->libiscsi_itt; break; default: - shost_printk(KERN_WARNING, phba->shost, - "Unrecognized opcode 0x%x in async msg\n", - (ppdu-> + beiscsi_log(phba, KERN_WARNING, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Unrecognized opcode 0x%x in async msg\n", + (ppdu-> dw[offsetof(struct amap_pdu_base, opcode) / 32] - & PDUBASE_OPCODE_MASK)); + & PDUBASE_OPCODE_MASK)); return 1; } @@ -951,9 +1051,11 @@ static struct sgl_handle *alloc_io_sgl_handle(struct beiscsi_hba *phba) struct sgl_handle *psgl_handle; if (phba->io_sgl_hndl_avbl) { - SE_DEBUG(DBG_LVL_8, - "In alloc_io_sgl_handle,io_sgl_alloc_index=%d\n", - phba->io_sgl_alloc_index); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_IO, + "BM_%d : In alloc_io_sgl_handle," + " io_sgl_alloc_index=%d\n", + phba->io_sgl_alloc_index); + psgl_handle = phba->io_sgl_hndl_base[phba-> io_sgl_alloc_index]; phba->io_sgl_hndl_base[phba->io_sgl_alloc_index] = NULL; @@ -971,17 +1073,20 @@ static struct sgl_handle *alloc_io_sgl_handle(struct beiscsi_hba *phba) static void free_io_sgl_handle(struct beiscsi_hba *phba, struct sgl_handle *psgl_handle) { - SE_DEBUG(DBG_LVL_8, "In free_,io_sgl_free_index=%d\n", - phba->io_sgl_free_index); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_IO, + "BM_%d : In free_,io_sgl_free_index=%d\n", + phba->io_sgl_free_index); + if (phba->io_sgl_hndl_base[phba->io_sgl_free_index]) { /* * this can happen if clean_task is called on a task that * failed in xmit_task or alloc_pdu. */ - SE_DEBUG(DBG_LVL_8, - "Double Free in IO SGL io_sgl_free_index=%d," - "value there=%p\n", phba->io_sgl_free_index, - phba->io_sgl_hndl_base[phba->io_sgl_free_index]); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_IO, + "BM_%d : Double Free in IO SGL io_sgl_free_index=%d," + "value there=%p\n", phba->io_sgl_free_index, + phba->io_sgl_hndl_base + [phba->io_sgl_free_index]); return; } phba->io_sgl_hndl_base[phba->io_sgl_free_index] = psgl_handle; @@ -1043,11 +1148,12 @@ free_wrb_handle(struct beiscsi_hba *phba, struct hwi_wrb_context *pwrb_context, else pwrb_context->free_index++; - SE_DEBUG(DBG_LVL_8, - "FREE WRB: pwrb_handle=%p free_index=0x%x" - "wrb_handles_available=%d\n", - pwrb_handle, pwrb_context->free_index, - pwrb_context->wrb_handles_available); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : FREE WRB: pwrb_handle=%p free_index=0x%x" + "wrb_handles_available=%d\n", + pwrb_handle, pwrb_context->free_index, + pwrb_context->wrb_handles_available); } static struct sgl_handle *alloc_mgmt_sgl_handle(struct beiscsi_hba *phba) @@ -1057,8 +1163,11 @@ static struct sgl_handle *alloc_mgmt_sgl_handle(struct beiscsi_hba *phba) if (phba->eh_sgl_hndl_avbl) { psgl_handle = phba->eh_sgl_hndl_base[phba->eh_sgl_alloc_index]; phba->eh_sgl_hndl_base[phba->eh_sgl_alloc_index] = NULL; - SE_DEBUG(DBG_LVL_8, "mgmt_sgl_alloc_index=%d=0x%x\n", - phba->eh_sgl_alloc_index, phba->eh_sgl_alloc_index); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BM_%d : mgmt_sgl_alloc_index=%d=0x%x\n", + phba->eh_sgl_alloc_index, + phba->eh_sgl_alloc_index); + phba->eh_sgl_hndl_avbl--; if (phba->eh_sgl_alloc_index == (phba->params.icds_per_ctrl - phba->params.ios_per_ctrl - @@ -1075,16 +1184,20 @@ void free_mgmt_sgl_handle(struct beiscsi_hba *phba, struct sgl_handle *psgl_handle) { - SE_DEBUG(DBG_LVL_8, "In free_mgmt_sgl_handle,eh_sgl_free_index=%d\n", - phba->eh_sgl_free_index); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BM_%d : In free_mgmt_sgl_handle," + "eh_sgl_free_index=%d\n", + phba->eh_sgl_free_index); + if (phba->eh_sgl_hndl_base[phba->eh_sgl_free_index]) { /* * this can happen if clean_task is called on a task that * failed in xmit_task or alloc_pdu. */ - SE_DEBUG(DBG_LVL_8, - "Double Free in eh SGL ,eh_sgl_free_index=%d\n", - phba->eh_sgl_free_index); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BM_%d : Double Free in eh SGL ," + "eh_sgl_free_index=%d\n", + phba->eh_sgl_free_index); return; } phba->eh_sgl_hndl_base[phba->eh_sgl_free_index] = psgl_handle; @@ -1326,9 +1439,10 @@ static void hwi_complete_cmd(struct beiscsi_conn *beiscsi_conn, break; case HWH_TYPE_LOGIN: - SE_DEBUG(DBG_LVL_1, - "\t\t No HWH_TYPE_LOGIN Expected in hwi_complete_cmd" - "- Solicited path\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d :\t\t No HWH_TYPE_LOGIN Expected in" + " hwi_complete_cmd- Solicited path\n"); break; case HWH_TYPE_NOP: @@ -1336,13 +1450,14 @@ static void hwi_complete_cmd(struct beiscsi_conn *beiscsi_conn, break; default: - shost_printk(KERN_WARNING, phba->shost, - "In hwi_complete_cmd, unknown type = %d" - "wrb_index 0x%x CID 0x%x\n", type, - ((psol->dw[offsetof(struct amap_iscsi_wrb, - type) / 32] & SOL_WRB_INDEX_MASK) >> 16), - ((psol->dw[offsetof(struct amap_sol_cqe, - cid) / 32] & SOL_CID_MASK) >> 6)); + beiscsi_log(phba, KERN_WARNING, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d : In hwi_complete_cmd, unknown type = %d" + "wrb_index 0x%x CID 0x%x\n", type, + ((psol->dw[offsetof(struct amap_iscsi_wrb, + type) / 32] & SOL_WRB_INDEX_MASK) >> 16), + ((psol->dw[offsetof(struct amap_sol_cqe, + cid) / 32] & SOL_CID_MASK) >> 6)); break; } @@ -1397,10 +1512,11 @@ hwi_get_async_handle(struct beiscsi_hba *phba, break; default: pbusy_list = NULL; - shost_printk(KERN_WARNING, phba->shost, - "Unexpected code=%d\n", - pdpdu_cqe->dw[offsetof(struct amap_i_t_dpdu_cqe, - code) / 32] & PDUCQE_CODE_MASK); + beiscsi_log(phba, KERN_WARNING, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Unexpected code=%d\n", + pdpdu_cqe->dw[offsetof(struct amap_i_t_dpdu_cqe, + code) / 32] & PDUCQE_CODE_MASK); return NULL; } @@ -1425,8 +1541,9 @@ hwi_get_async_handle(struct beiscsi_hba *phba, } static unsigned int -hwi_update_async_writables(struct hwi_async_pdu_context *pasync_ctx, - unsigned int is_header, unsigned int cq_index) +hwi_update_async_writables(struct beiscsi_hba *phba, + struct hwi_async_pdu_context *pasync_ctx, + unsigned int is_header, unsigned int cq_index) { struct list_head *pbusy_list; struct async_pdu_handle *pasync_handle; @@ -1463,9 +1580,10 @@ hwi_update_async_writables(struct hwi_async_pdu_context *pasync_ctx, } if (!writables) { - SE_DEBUG(DBG_LVL_1, - "Duplicate notification received - index 0x%x!!\n", - cq_index); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d : Duplicate notification received - index 0x%x!!\n", + cq_index); WARN_ON(1); } @@ -1616,8 +1734,8 @@ static void hwi_flush_default_pdu_buffer(struct beiscsi_hba *phba, pdpdu_cqe, &cq_index); BUG_ON(pasync_handle->is_header != 0); if (pasync_handle->consumed == 0) - hwi_update_async_writables(pasync_ctx, pasync_handle->is_header, - cq_index); + hwi_update_async_writables(phba, pasync_ctx, + pasync_handle->is_header, cq_index); hwi_free_async_msg(phba, pasync_handle->cri); hwi_post_async_buffers(phba, pasync_handle->is_header); @@ -1745,8 +1863,9 @@ static void hwi_process_default_pdu_ring(struct beiscsi_conn *beiscsi_conn, pdpdu_cqe, &cq_index); if (pasync_handle->consumed == 0) - hwi_update_async_writables(pasync_ctx, pasync_handle->is_header, - cq_index); + hwi_update_async_writables(phba, pasync_ctx, + pasync_handle->is_header, cq_index); + hwi_gather_async_pdu(beiscsi_conn, phba, pasync_handle); hwi_post_async_buffers(phba, pasync_handle->is_header); } @@ -1774,9 +1893,10 @@ static void beiscsi_process_mcc_isr(struct beiscsi_hba *phba) beiscsi_async_link_state_process(phba, (struct be_async_event_link_state *) mcc_compl); else - SE_DEBUG(DBG_LVL_1, - " Unsupported Async Event, flags" - " = 0x%08x\n", mcc_compl->flags); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_MBOX, + "BM_%d : Unsupported Async Event, flags" + " = 0x%08x\n", + mcc_compl->flags); } else if (mcc_compl->flags & CQE_FLAGS_COMPLETED_MASK) { be_mcc_compl_process_isr(&phba->ctrl, mcc_compl); atomic_dec(&phba->ctrl.mcc_obj.q.used); @@ -1801,6 +1921,7 @@ static unsigned int beiscsi_process_cq(struct be_eq_obj *pbe_eq) struct dmsg_cqe *dmsg; unsigned int num_processed = 0; unsigned int tot_nump = 0; + unsigned short code = 0, cid = 0; struct beiscsi_conn *beiscsi_conn; struct beiscsi_endpoint *beiscsi_ep; struct iscsi_endpoint *ep; @@ -1814,10 +1935,11 @@ static unsigned int beiscsi_process_cq(struct be_eq_obj *pbe_eq) CQE_VALID_MASK) { be_dws_le_to_cpu(sol, sizeof(struct sol_cqe)); - ep = phba->ep_array[(u32) ((sol-> - dw[offsetof(struct amap_sol_cqe, cid) / 32] & - SOL_CID_MASK) >> 6) - - phba->fw_config.iscsi_cid_start]; + cid = ((sol->dw[offsetof(struct amap_sol_cqe, cid)/32] & + CQE_CID_MASK) >> 6); + code = (sol->dw[offsetof(struct amap_sol_cqe, code)/32] & + CQE_CODE_MASK); + ep = phba->ep_array[cid - phba->fw_config.iscsi_cid_start]; beiscsi_ep = ep->dd_data; beiscsi_conn = beiscsi_ep->conn; @@ -1829,32 +1951,41 @@ static unsigned int beiscsi_process_cq(struct be_eq_obj *pbe_eq) num_processed = 0; } - switch ((u32) sol->dw[offsetof(struct amap_sol_cqe, code) / - 32] & CQE_CODE_MASK) { + switch (code) { case SOL_CMD_COMPLETE: hwi_complete_cmd(beiscsi_conn, phba, sol); break; case DRIVERMSG_NOTIFY: - SE_DEBUG(DBG_LVL_8, "Received DRIVERMSG_NOTIFY\n"); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Received DRIVERMSG_NOTIFY\n"); + dmsg = (struct dmsg_cqe *)sol; hwi_complete_drvr_msgs(beiscsi_conn, phba, sol); break; case UNSOL_HDR_NOTIFY: - SE_DEBUG(DBG_LVL_8, "Received UNSOL_HDR_ NOTIFY\n"); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Received UNSOL_HDR_ NOTIFY\n"); + hwi_process_default_pdu_ring(beiscsi_conn, phba, (struct i_t_dpdu_cqe *)sol); break; case UNSOL_DATA_NOTIFY: - SE_DEBUG(DBG_LVL_8, "Received UNSOL_DATA_NOTIFY\n"); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d : Received UNSOL_DATA_NOTIFY\n"); + hwi_process_default_pdu_ring(beiscsi_conn, phba, (struct i_t_dpdu_cqe *)sol); break; case CXN_INVALIDATE_INDEX_NOTIFY: case CMD_INVALIDATED_NOTIFY: case CXN_INVALIDATE_NOTIFY: - SE_DEBUG(DBG_LVL_1, - "Ignoring CQ Error notification for cmd/cxn" - "invalidate\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Ignoring CQ Error notification for" + " cmd/cxn invalidate\n"); break; case SOL_CMD_KILLED_DATA_DIGEST_ERR: case CMD_KILLED_INVALID_STATSN_RCVD: @@ -1864,17 +1995,16 @@ static unsigned int beiscsi_process_cq(struct be_eq_obj *pbe_eq) case CMD_CXN_KILLED_ITT_INVALID: case CMD_CXN_KILLED_SEQ_OUTOFORDER: case CMD_CXN_KILLED_INVALID_DATASN_RCVD: - SE_DEBUG(DBG_LVL_1, - "CQ Error notification for cmd.. " - "code %d cid 0x%x\n", - sol->dw[offsetof(struct amap_sol_cqe, code) / - 32] & CQE_CODE_MASK, - (sol->dw[offsetof(struct amap_sol_cqe, cid) / - 32] & SOL_CID_MASK)); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d : CQ Error notification for cmd.. " + "code %d cid 0x%x\n", code, cid); break; case UNSOL_DATA_DIGEST_ERROR_NOTIFY: - SE_DEBUG(DBG_LVL_1, - "Digest error on def pdu ring, dropping..\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Digest error on def pdu ring," + " dropping..\n"); hwi_flush_default_pdu_buffer(phba, beiscsi_conn, (struct i_t_dpdu_cqe *) sol); break; @@ -1892,33 +2022,31 @@ static unsigned int beiscsi_process_cq(struct be_eq_obj *pbe_eq) case CXN_KILLED_OVER_RUN_RESIDUAL: case CXN_KILLED_UNDER_RUN_RESIDUAL: case CXN_KILLED_CMND_DATA_NOT_ON_SAME_CONN: - SE_DEBUG(DBG_LVL_1, "CQ Error %d, reset CID " - "0x%x...\n", - sol->dw[offsetof(struct amap_sol_cqe, code) / - 32] & CQE_CODE_MASK, - (sol->dw[offsetof(struct amap_sol_cqe, cid) / - 32] & CQE_CID_MASK)); - iscsi_conn_failure(beiscsi_conn->conn, - ISCSI_ERR_CONN_FAILED); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : CQ Error %d, reset CID 0x%x...\n", + code, cid); + if (beiscsi_conn) + iscsi_conn_failure(beiscsi_conn->conn, + ISCSI_ERR_CONN_FAILED); break; case CXN_KILLED_RST_SENT: case CXN_KILLED_RST_RCVD: - SE_DEBUG(DBG_LVL_1, "CQ Error %d, reset" - "received/sent on CID 0x%x...\n", - sol->dw[offsetof(struct amap_sol_cqe, code) / - 32] & CQE_CODE_MASK, - (sol->dw[offsetof(struct amap_sol_cqe, cid) / - 32] & CQE_CID_MASK)); - iscsi_conn_failure(beiscsi_conn->conn, - ISCSI_ERR_CONN_FAILED); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : CQ Error %d, reset" + "received/sent on CID 0x%x...\n", + code, cid); + if (beiscsi_conn) + iscsi_conn_failure(beiscsi_conn->conn, + ISCSI_ERR_CONN_FAILED); break; default: - SE_DEBUG(DBG_LVL_1, "CQ Error Invalid code= %d " - "received on CID 0x%x...\n", - sol->dw[offsetof(struct amap_sol_cqe, code) / - 32] & CQE_CODE_MASK, - (sol->dw[offsetof(struct amap_sol_cqe, cid) / - 32] & CQE_CID_MASK)); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : CQ Error Invalid code= %d " + "received on CID 0x%x...\n", + code, cid); break; } @@ -1977,7 +2105,10 @@ static int be_iopoll(struct blk_iopoll *iop, int budget) if (ret < budget) { phba = pbe_eq->phba; blk_iopoll_complete(iop); - SE_DEBUG(DBG_LVL_8, "rearm pbe_eq->q.id =%d\n", pbe_eq->q.id); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_IO, + "BM_%d : rearm pbe_eq->q.id =%d\n", + pbe_eq->q.id); hwi_ring_eq_db(phba, pbe_eq->q.id, 0, 0, 1, 1); } return ret; @@ -2348,16 +2479,16 @@ static int beiscsi_init_wrb_handle(struct beiscsi_hba *phba) kzalloc(sizeof(struct wrb_handle *) * phba->params.wrbs_per_cxn, GFP_KERNEL); if (!pwrb_context->pwrb_handle_base) { - shost_printk(KERN_ERR, phba->shost, - "Mem Alloc Failed. Failing to load\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Mem Alloc Failed. Failing to load\n"); goto init_wrb_hndl_failed; } pwrb_context->pwrb_handle_basestd = kzalloc(sizeof(struct wrb_handle *) * phba->params.wrbs_per_cxn, GFP_KERNEL); if (!pwrb_context->pwrb_handle_basestd) { - shost_printk(KERN_ERR, phba->shost, - "Mem Alloc Failed. Failing to load\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Mem Alloc Failed. Failing to load\n"); goto init_wrb_hndl_failed; } if (!num_cxn_wrbh) { @@ -2438,12 +2569,13 @@ static void hwi_init_async_pdu_ctx(struct beiscsi_hba *phba) mem_descr = (struct be_mem_descriptor *)phba->init_mem; mem_descr += HWI_MEM_ASYNC_HEADER_BUF; if (mem_descr->mem_array[0].virtual_address) { - SE_DEBUG(DBG_LVL_8, - "hwi_init_async_pdu_ctx HWI_MEM_ASYNC_HEADER_BUF" - "va=%p\n", mem_descr->mem_array[0].virtual_address); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_async_pdu_ctx" + " HWI_MEM_ASYNC_HEADER_BUF va=%p\n", + mem_descr->mem_array[0].virtual_address); } else - shost_printk(KERN_WARNING, phba->shost, - "No Virtual address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : No Virtual address\n"); pasync_ctx->async_header.va_base = mem_descr->mem_array[0].virtual_address; @@ -2454,24 +2586,27 @@ static void hwi_init_async_pdu_ctx(struct beiscsi_hba *phba) mem_descr = (struct be_mem_descriptor *)phba->init_mem; mem_descr += HWI_MEM_ASYNC_HEADER_RING; if (mem_descr->mem_array[0].virtual_address) { - SE_DEBUG(DBG_LVL_8, - "hwi_init_async_pdu_ctx HWI_MEM_ASYNC_HEADER_RING" - "va=%p\n", mem_descr->mem_array[0].virtual_address); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_async_pdu_ctx" + " HWI_MEM_ASYNC_HEADER_RING va=%p\n", + mem_descr->mem_array[0].virtual_address); } else - shost_printk(KERN_WARNING, phba->shost, - "No Virtual address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : No Virtual address\n"); + pasync_ctx->async_header.ring_base = mem_descr->mem_array[0].virtual_address; mem_descr = (struct be_mem_descriptor *)phba->init_mem; mem_descr += HWI_MEM_ASYNC_HEADER_HANDLE; if (mem_descr->mem_array[0].virtual_address) { - SE_DEBUG(DBG_LVL_8, - "hwi_init_async_pdu_ctx HWI_MEM_ASYNC_HEADER_HANDLE" - "va=%p\n", mem_descr->mem_array[0].virtual_address); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_async_pdu_ctx" + " HWI_MEM_ASYNC_HEADER_HANDLE va=%p\n", + mem_descr->mem_array[0].virtual_address); } else - shost_printk(KERN_WARNING, phba->shost, - "No Virtual address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : No Virtual address\n"); pasync_ctx->async_header.handle_base = mem_descr->mem_array[0].virtual_address; @@ -2482,12 +2617,13 @@ static void hwi_init_async_pdu_ctx(struct beiscsi_hba *phba) mem_descr = (struct be_mem_descriptor *)phba->init_mem; mem_descr += HWI_MEM_ASYNC_DATA_RING; if (mem_descr->mem_array[0].virtual_address) { - SE_DEBUG(DBG_LVL_8, - "hwi_init_async_pdu_ctx HWI_MEM_ASYNC_DATA_RING" - "va=%p\n", mem_descr->mem_array[0].virtual_address); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_async_pdu_ctx" + " HWI_MEM_ASYNC_DATA_RING va=%p\n", + mem_descr->mem_array[0].virtual_address); } else - shost_printk(KERN_WARNING, phba->shost, - "No Virtual address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : No Virtual address\n"); pasync_ctx->async_data.ring_base = mem_descr->mem_array[0].virtual_address; @@ -2495,8 +2631,8 @@ static void hwi_init_async_pdu_ctx(struct beiscsi_hba *phba) mem_descr = (struct be_mem_descriptor *)phba->init_mem; mem_descr += HWI_MEM_ASYNC_DATA_HANDLE; if (!mem_descr->mem_array[0].virtual_address) - shost_printk(KERN_WARNING, phba->shost, - "No Virtual address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : No Virtual address\n"); pasync_ctx->async_data.handle_base = mem_descr->mem_array[0].virtual_address; @@ -2511,12 +2647,14 @@ static void hwi_init_async_pdu_ctx(struct beiscsi_hba *phba) mem_descr = (struct be_mem_descriptor *)phba->init_mem; mem_descr += HWI_MEM_ASYNC_DATA_BUF; if (mem_descr->mem_array[0].virtual_address) { - SE_DEBUG(DBG_LVL_8, - "hwi_init_async_pdu_ctx HWI_MEM_ASYNC_DATA_BUF" - "va=%p\n", mem_descr->mem_array[0].virtual_address); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_async_pdu_ctx" + " HWI_MEM_ASYNC_DATA_BUF va=%p\n", + mem_descr->mem_array[0].virtual_address); } else - shost_printk(KERN_WARNING, phba->shost, - "No Virtual address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : No Virtual address\n"); + idx = 0; pasync_ctx->async_data.va_base = mem_descr->mem_array[idx].virtual_address; @@ -2657,7 +2795,7 @@ static int beiscsi_create_eqs(struct beiscsi_hba *phba, struct hwi_context_memory *phwi_context) { unsigned int i, num_eq_pages; - int ret, eq_for_mcc; + int ret = 0, eq_for_mcc; struct be_queue_info *eq; struct be_dma_mem *mem; void *eq_vaddress; @@ -2684,8 +2822,8 @@ static int beiscsi_create_eqs(struct beiscsi_hba *phba, ret = be_fill_queue(eq, phba->params.num_eq_entries, sizeof(struct be_eq_entry), eq_vaddress); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "be_fill_queue Failed for EQ\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : be_fill_queue Failed for EQ\n"); goto create_eq_error; } @@ -2693,12 +2831,15 @@ static int beiscsi_create_eqs(struct beiscsi_hba *phba, ret = beiscsi_cmd_eq_create(&phba->ctrl, eq, phwi_context->cur_eqd); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "beiscsi_cmd_eq_create" - "Failedfor EQ\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_cmd_eq_create" + "Failed for EQ\n"); goto create_eq_error; } - SE_DEBUG(DBG_LVL_8, "eqid = %d\n", phwi_context->be_eq[i].q.id); + + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : eqid = %d\n", + phwi_context->be_eq[i].q.id); } return 0; create_eq_error: @@ -2717,7 +2858,7 @@ static int beiscsi_create_cqs(struct beiscsi_hba *phba, struct hwi_context_memory *phwi_context) { unsigned int i, num_cq_pages; - int ret; + int ret = 0; struct be_queue_info *cq, *eq; struct be_dma_mem *mem; struct be_eq_obj *pbe_eq; @@ -2742,8 +2883,9 @@ static int beiscsi_create_cqs(struct beiscsi_hba *phba, ret = be_fill_queue(cq, phba->params.num_cq_entries, sizeof(struct sol_cqe), cq_vaddress); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "be_fill_queue Failed for ISCSI CQ\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : be_fill_queue Failed " + "for ISCSI CQ\n"); goto create_cq_error; } @@ -2751,14 +2893,14 @@ static int beiscsi_create_cqs(struct beiscsi_hba *phba, ret = beiscsi_cmd_cq_create(&phba->ctrl, cq, eq, false, false, 0); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "beiscsi_cmd_eq_create" - "Failed for ISCSI CQ\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_cmd_eq_create" + "Failed for ISCSI CQ\n"); goto create_cq_error; } - SE_DEBUG(DBG_LVL_8, "iscsi cq_id is %d for eq_id %d\n", - cq->id, eq->id); - SE_DEBUG(DBG_LVL_8, "ISCSI CQ CREATED\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : iscsi cq_id is %d for eq_id %d\n" + "iSCSI CQ CREATED\n", cq->id, eq->id); } return 0; @@ -2799,8 +2941,8 @@ beiscsi_create_def_hdr(struct beiscsi_hba *phba, sizeof(struct phys_addr), sizeof(struct phys_addr), dq_vaddress); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "be_fill_queue Failed for DEF PDU HDR\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : be_fill_queue Failed for DEF PDU HDR\n"); return ret; } mem->dma = (unsigned long)mem_descr->mem_array[idx]. @@ -2809,13 +2951,15 @@ beiscsi_create_def_hdr(struct beiscsi_hba *phba, def_pdu_ring_sz, phba->params.defpdu_hdr_sz); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "be_cmd_create_default_pdu_queue Failed DEFHDR\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : be_cmd_create_default_pdu_queue Failed DEFHDR\n"); return ret; } phwi_ctrlr->default_pdu_hdr.id = phwi_context->be_def_hdrq.id; - SE_DEBUG(DBG_LVL_8, "iscsi def pdu id is %d\n", - phwi_context->be_def_hdrq.id); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : iscsi def pdu id is %d\n", + phwi_context->be_def_hdrq.id); + hwi_post_async_buffers(phba, 1); return 0; } @@ -2844,8 +2988,8 @@ beiscsi_create_def_data(struct beiscsi_hba *phba, sizeof(struct phys_addr), sizeof(struct phys_addr), dq_vaddress); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "be_fill_queue Failed for DEF PDU DATA\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : be_fill_queue Failed for DEF PDU DATA\n"); return ret; } mem->dma = (unsigned long)mem_descr->mem_array[idx]. @@ -2854,16 +2998,20 @@ beiscsi_create_def_data(struct beiscsi_hba *phba, def_pdu_ring_sz, phba->params.defpdu_data_sz); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "be_cmd_create_default_pdu_queue Failed" - " for DEF PDU DATA\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d be_cmd_create_default_pdu_queue" + " Failed for DEF PDU DATA\n"); return ret; } phwi_ctrlr->default_pdu_data.id = phwi_context->be_def_dataq.id; - SE_DEBUG(DBG_LVL_8, "iscsi def data id is %d\n", - phwi_context->be_def_dataq.id); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : iscsi def data id is %d\n", + phwi_context->be_def_dataq.id); + hwi_post_async_buffers(phba, 0); - SE_DEBUG(DBG_LVL_8, "DEFAULT PDU DATA RING CREATED\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : DEFAULT PDU DATA RING CREATED\n"); + return 0; } @@ -2889,13 +3037,14 @@ beiscsi_post_pages(struct beiscsi_hba *phba) (pm_arr->size / PAGE_SIZE)); page_offset += pm_arr->size / PAGE_SIZE; if (status != 0) { - shost_printk(KERN_ERR, phba->shost, - "post sgl failed.\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : post sgl failed.\n"); return status; } pm_arr++; } - SE_DEBUG(DBG_LVL_8, "POSTED PAGES\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : POSTED PAGES\n"); return 0; } @@ -2945,8 +3094,8 @@ beiscsi_create_wrb_rings(struct beiscsi_hba *phba, pwrb_arr = kmalloc(sizeof(*pwrb_arr) * phba->params.cxns_per_ctrl, GFP_KERNEL); if (!pwrb_arr) { - shost_printk(KERN_ERR, phba->shost, - "Memory alloc failed in create wrb ring.\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Memory alloc failed in create wrb ring.\n"); return -ENOMEM; } wrb_vaddr = mem_descr->mem_array[idx].virtual_address; @@ -2990,8 +3139,8 @@ beiscsi_create_wrb_rings(struct beiscsi_hba *phba, status = be_cmd_wrbq_create(&phba->ctrl, &sgl, &phwi_context->be_wrbq[i]); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, - "wrbq create failed."); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : wrbq create failed."); kfree(pwrb_arr); return status; } @@ -3127,7 +3276,6 @@ static int find_num_cpus(void) if (num_cpus >= MAX_CPUS) num_cpus = MAX_CPUS - 1; - SE_DEBUG(DBG_LVL_8, "num_cpus = %d\n", num_cpus); return num_cpus; } @@ -3150,7 +3298,8 @@ static int hwi_init_port(struct beiscsi_hba *phba) status = beiscsi_create_eqs(phba, phwi_context); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, "EQ not created\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : EQ not created\n"); goto error; } @@ -3160,51 +3309,55 @@ static int hwi_init_port(struct beiscsi_hba *phba) status = mgmt_check_supported_fw(ctrl, phba); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, - "Unsupported fw version\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Unsupported fw version\n"); goto error; } status = beiscsi_create_cqs(phba, phwi_context); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, "CQ not created\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : CQ not created\n"); goto error; } status = beiscsi_create_def_hdr(phba, phwi_context, phwi_ctrlr, def_pdu_ring_sz); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, - "Default Header not created\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Default Header not created\n"); goto error; } status = beiscsi_create_def_data(phba, phwi_context, phwi_ctrlr, def_pdu_ring_sz); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, - "Default Data not created\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Default Data not created\n"); goto error; } status = beiscsi_post_pages(phba); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, "Post SGL Pages Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Post SGL Pages Failed\n"); goto error; } status = beiscsi_create_wrb_rings(phba, phwi_context, phwi_ctrlr); if (status != 0) { - shost_printk(KERN_ERR, phba->shost, - "WRB Rings not created\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : WRB Rings not created\n"); goto error; } - SE_DEBUG(DBG_LVL_8, "hwi_init_port success\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_port success\n"); return 0; error: - shost_printk(KERN_ERR, phba->shost, "hwi_init_port failed"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_port failed"); hwi_cleanup(phba); return status; } @@ -3217,12 +3370,13 @@ static int hwi_init_controller(struct beiscsi_hba *phba) if (1 == phba->init_mem[HWI_MEM_ADDN_CONTEXT].num_elements) { phwi_ctrlr->phwi_ctxt = (struct hwi_context_memory *)phba-> init_mem[HWI_MEM_ADDN_CONTEXT].mem_array[0].virtual_address; - SE_DEBUG(DBG_LVL_8, " phwi_ctrlr->phwi_ctxt=%p\n", - phwi_ctrlr->phwi_ctxt); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : phwi_ctrlr->phwi_ctxt=%p\n", + phwi_ctrlr->phwi_ctxt); } else { - shost_printk(KERN_ERR, phba->shost, - "HWI_MEM_ADDN_CONTEXT is more than one element." - "Failing to load\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : HWI_MEM_ADDN_CONTEXT is more " + "than one element.Failing to load\n"); return -ENOMEM; } @@ -3232,8 +3386,9 @@ static int hwi_init_controller(struct beiscsi_hba *phba) hwi_init_async_pdu_ctx(phba); if (hwi_init_port(phba) != 0) { - shost_printk(KERN_ERR, phba->shost, - "hwi_init_controller failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : hwi_init_controller failed\n"); + return -ENOMEM; } return 0; @@ -3268,15 +3423,18 @@ static int beiscsi_init_controller(struct beiscsi_hba *phba) ret = beiscsi_get_memory(phba); if (ret < 0) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_dev_probe -" - "Failed in beiscsi_alloc_memory\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe -" + "Failed in beiscsi_alloc_memory\n"); return ret; } ret = hwi_init_controller(phba); if (ret) goto free_init; - SE_DEBUG(DBG_LVL_8, "Return success from beiscsi_init_controller"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : Return success from beiscsi_init_controller"); + return 0; free_init: @@ -3301,8 +3459,8 @@ static int beiscsi_init_sgl_handle(struct beiscsi_hba *phba) phba->params.ios_per_ctrl, GFP_KERNEL); if (!phba->io_sgl_hndl_base) { - shost_printk(KERN_ERR, phba->shost, - "Mem Alloc Failed. Failing to load\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Mem Alloc Failed. Failing to load\n"); return -ENOMEM; } phba->eh_sgl_hndl_base = kzalloc(sizeof(struct sgl_handle *) * @@ -3311,14 +3469,14 @@ static int beiscsi_init_sgl_handle(struct beiscsi_hba *phba) GFP_KERNEL); if (!phba->eh_sgl_hndl_base) { kfree(phba->io_sgl_hndl_base); - shost_printk(KERN_ERR, phba->shost, - "Mem Alloc Failed. Failing to load\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Mem Alloc Failed. Failing to load\n"); return -ENOMEM; } } else { - shost_printk(KERN_ERR, phba->shost, - "HWI_MEM_SGLH is more than one element." - "Failing to load\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : HWI_MEM_SGLH is more than one element." + "Failing to load\n"); return -ENOMEM; } @@ -3344,15 +3502,18 @@ static int beiscsi_init_sgl_handle(struct beiscsi_hba *phba) } idx++; } - SE_DEBUG(DBG_LVL_8, - "phba->io_sgl_hndl_avbl=%d" - "phba->eh_sgl_hndl_avbl=%d\n", - phba->io_sgl_hndl_avbl, - phba->eh_sgl_hndl_avbl); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : phba->io_sgl_hndl_avbl=%d" + "phba->eh_sgl_hndl_avbl=%d\n", + phba->io_sgl_hndl_avbl, + phba->eh_sgl_hndl_avbl); + mem_descr_sg = phba->init_mem; mem_descr_sg += HWI_MEM_SGE; - SE_DEBUG(DBG_LVL_8, "\n mem_descr_sg->num_elements=%d\n", - mem_descr_sg->num_elements); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "\n BM_%d : mem_descr_sg->num_elements=%d\n", + mem_descr_sg->num_elements); + arr_index = 0; idx = 0; while (idx < mem_descr_sg->num_elements) { @@ -3390,17 +3551,17 @@ static int hba_setup_cid_tbls(struct beiscsi_hba *phba) phba->cid_array = kzalloc(sizeof(void *) * phba->params.cxns_per_ctrl, GFP_KERNEL); if (!phba->cid_array) { - shost_printk(KERN_ERR, phba->shost, - "Failed to allocate memory in " - "hba_setup_cid_tbls\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Failed to allocate memory in " + "hba_setup_cid_tbls\n"); return -ENOMEM; } phba->ep_array = kzalloc(sizeof(struct iscsi_endpoint *) * phba->params.cxns_per_ctrl * 2, GFP_KERNEL); if (!phba->ep_array) { - shost_printk(KERN_ERR, phba->shost, - "Failed to allocate memory in " - "hba_setup_cid_tbls\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Failed to allocate memory in " + "hba_setup_cid_tbls\n"); kfree(phba->cid_array); return -ENOMEM; } @@ -3433,18 +3594,22 @@ static void hwi_enable_intr(struct beiscsi_hba *phba) enabled = reg & MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK; if (!enabled) { reg |= MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK; - SE_DEBUG(DBG_LVL_8, "reg =x%08x addr=%p\n", reg, addr); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : reg =x%08x addr=%p\n", reg, addr); iowrite32(reg, addr); } if (!phba->msix_enabled) { eq = &phwi_context->be_eq[0].q; - SE_DEBUG(DBG_LVL_8, "eq->id=%d\n", eq->id); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : eq->id=%d\n", eq->id); + hwi_ring_eq_db(phba, eq->id, 0, 0, 1, 1); } else { for (i = 0; i <= phba->num_cpus; i++) { eq = &phwi_context->be_eq[i].q; - SE_DEBUG(DBG_LVL_8, "eq->id=%d\n", eq->id); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : eq->id=%d\n", eq->id); hwi_ring_eq_db(phba, eq->id, 0, 0, 1, 1); } } @@ -3462,64 +3627,60 @@ static void hwi_disable_intr(struct beiscsi_hba *phba) reg &= ~MEMBAR_CTRL_INT_CTRL_HOSTINTR_MASK; iowrite32(reg, addr); } else - shost_printk(KERN_WARNING, phba->shost, - "In hwi_disable_intr, Already Disabled\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : In hwi_disable_intr, Already Disabled\n"); } +/** + * beiscsi_get_boot_info()- Get the boot session info + * @phba: The device priv structure instance + * + * Get the boot target info and store in driver priv structure + * + * return values + * Success: 0 + * Failure: Non-Zero Value + **/ static int beiscsi_get_boot_info(struct beiscsi_hba *phba) { - struct be_cmd_get_boot_target_resp *boot_resp; struct be_cmd_get_session_resp *session_resp; struct be_mcc_wrb *wrb; struct be_dma_mem nonemb_cmd; unsigned int tag, wrb_num; unsigned short status, extd_status; + unsigned int s_handle; struct be_queue_info *mccq = &phba->ctrl.mcc_obj.q; int ret = -ENOMEM; - tag = mgmt_get_boot_target(phba); - if (!tag) { - SE_DEBUG(DBG_LVL_1, "beiscsi_get_boot_info Failed\n"); - return -EAGAIN; - } else - wait_event_interruptible(phba->ctrl.mcc_wait[tag], - phba->ctrl.mcc_numtag[tag]); - - wrb_num = (phba->ctrl.mcc_numtag[tag] & 0x00FF0000) >> 16; - extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; - status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; - if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, "beiscsi_get_boot_info Failed" - " status = %d extd_status = %d\n", - status, extd_status); - free_mcc_tag(&phba->ctrl, tag); - return -EBUSY; - } - wrb = queue_get_wrb(mccq, wrb_num); - free_mcc_tag(&phba->ctrl, tag); - boot_resp = embedded_payload(wrb); - - if (boot_resp->boot_session_handle < 0) { - shost_printk(KERN_INFO, phba->shost, "No Boot Session.\n"); - return -ENXIO; + /* Get the session handle of the boot target */ + ret = be_mgmt_get_boot_shandle(phba, &s_handle); + if (ret) { + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BM_%d : No boot session\n"); + return ret; } - nonemb_cmd.va = pci_alloc_consistent(phba->ctrl.pdev, sizeof(*session_resp), &nonemb_cmd.dma); if (nonemb_cmd.va == NULL) { - SE_DEBUG(DBG_LVL_1, - "Failed to allocate memory for" - "beiscsi_get_session_info\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BM_%d : Failed to allocate memory for" + "beiscsi_get_session_info\n"); + return -ENOMEM; } memset(nonemb_cmd.va, 0, sizeof(*session_resp)); - tag = mgmt_get_session_info(phba, boot_resp->boot_session_handle, + tag = mgmt_get_session_info(phba, s_handle, &nonemb_cmd); if (!tag) { - SE_DEBUG(DBG_LVL_1, "beiscsi_get_session_info" - " Failed\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BM_%d : beiscsi_get_session_info" + " Failed\n"); + goto boot_freemem; } else wait_event_interruptible(phba->ctrl.mcc_wait[tag], @@ -3529,9 +3690,12 @@ static int beiscsi_get_boot_info(struct beiscsi_hba *phba) extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, "beiscsi_get_session_info Failed" - " status = %d extd_status = %d\n", - status, extd_status); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BM_%d : beiscsi_get_session_info Failed" + " status = %d extd_status = %d\n", + status, extd_status); + free_mcc_tag(&phba->ctrl, tag); goto boot_freemem; } @@ -3611,22 +3775,22 @@ static int beiscsi_init_port(struct beiscsi_hba *phba) ret = beiscsi_init_controller(phba); if (ret < 0) { - shost_printk(KERN_ERR, phba->shost, - "beiscsi_dev_probe - Failed in" - "beiscsi_init_controller\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe - Failed in" + "beiscsi_init_controller\n"); return ret; } ret = beiscsi_init_sgl_handle(phba); if (ret < 0) { - shost_printk(KERN_ERR, phba->shost, - "beiscsi_dev_probe - Failed in" - "beiscsi_init_sgl_handle\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe - Failed in" + "beiscsi_init_sgl_handle\n"); goto do_cleanup_ctrlr; } if (hba_setup_cid_tbls(phba)) { - shost_printk(KERN_ERR, phba->shost, - "Failed in hba_setup_cid_tbls\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Failed in hba_setup_cid_tbls\n"); kfree(phba->io_sgl_hndl_base); kfree(phba->eh_sgl_hndl_base); goto do_cleanup_ctrlr; @@ -3678,8 +3842,8 @@ static void beiscsi_clean_port(struct beiscsi_hba *phba) mgmt_status = mgmt_epfw_cleanup(phba, CMD_CONNECTION_CHUTE_0); if (mgmt_status) - shost_printk(KERN_WARNING, phba->shost, - "mgmt_epfw_cleanup FAILED\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BM_%d : mgmt_epfw_cleanup FAILED\n"); hwi_purge_eq(phba); hwi_cleanup(phba); @@ -3960,7 +4124,9 @@ free_hndls: pci_pool_free(beiscsi_sess->bhs_pool, io_task->cmd_bhs, io_task->bhs_pa.u.a64.address); io_task->cmd_bhs = NULL; - SE_DEBUG(DBG_LVL_1, "Alloc of SGL_ICD Failed\n"); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_IO | BEISCSI_LOG_CONFIG, + "BM_%d : Alloc of SGL_ICD Failed\n"); return -ENOMEM; } @@ -3981,15 +4147,6 @@ static int beiscsi_iotask(struct iscsi_task *task, struct scatterlist *sg, io_task->bhs_len = sizeof(struct be_cmd_bhs); if (writedir) { - memset(&io_task->cmd_bhs->iscsi_data_pdu, 0, 48); - AMAP_SET_BITS(struct amap_pdu_data_out, itt, - &io_task->cmd_bhs->iscsi_data_pdu, - (unsigned int)io_task->cmd_bhs->iscsi_hdr.itt); - AMAP_SET_BITS(struct amap_pdu_data_out, opcode, - &io_task->cmd_bhs->iscsi_data_pdu, - ISCSI_OPCODE_SCSI_DATA_OUT); - AMAP_SET_BITS(struct amap_pdu_data_out, final_bit, - &io_task->cmd_bhs->iscsi_data_pdu, 1); AMAP_SET_BITS(struct amap_iscsi_wrb, type, pwrb, INI_WR_CMD); AMAP_SET_BITS(struct amap_iscsi_wrb, dsp, pwrb, 1); @@ -3998,9 +4155,6 @@ static int beiscsi_iotask(struct iscsi_task *task, struct scatterlist *sg, INI_RD_CMD); AMAP_SET_BITS(struct amap_iscsi_wrb, dsp, pwrb, 0); } - memcpy(&io_task->cmd_bhs->iscsi_data_pdu. - dw[offsetof(struct amap_pdu_data_out, lun) / 32], - &io_task->cmd_bhs->iscsi_hdr.lun, sizeof(struct scsi_lun)); AMAP_SET_BITS(struct amap_iscsi_wrb, lun, pwrb, cpu_to_be16(*(unsigned short *) @@ -4090,8 +4244,10 @@ static int beiscsi_mtask(struct iscsi_task *task) break; default: - SE_DEBUG(DBG_LVL_1, "opcode =%d Not supported\n", - task->hdr->opcode & ISCSI_OPCODE_MASK); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BM_%d : opcode =%d Not supported\n", + task->hdr->opcode & ISCSI_OPCODE_MASK); + return -EINVAL; } @@ -4123,17 +4279,22 @@ static int beiscsi_task_xmit(struct iscsi_task *task) io_task->scsi_cmnd = sc; num_sg = scsi_dma_map(sc); if (num_sg < 0) { - SE_DEBUG(DBG_LVL_1, " scsi_dma_map Failed\n") + struct iscsi_conn *conn = task->conn; + struct beiscsi_hba *phba = NULL; + + phba = ((struct beiscsi_conn *)conn->dd_data)->phba; + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_IO, + "BM_%d : scsi_dma_map Failed\n"); + return num_sg; } xferlen = scsi_bufflen(sc); sg = scsi_sglist(sc); - if (sc->sc_data_direction == DMA_TO_DEVICE) { + if (sc->sc_data_direction == DMA_TO_DEVICE) writedir = 1; - SE_DEBUG(DBG_LVL_4, "task->imm_count=0x%08x\n", - task->imm_count); - } else + else writedir = 0; + return beiscsi_iotask(task, sg, num_sg, xferlen, writedir); } @@ -4162,14 +4323,17 @@ static int beiscsi_bsg_request(struct bsg_job *job) job->request_payload.payload_len, &nonemb_cmd.dma); if (nonemb_cmd.va == NULL) { - SE_DEBUG(DBG_LVL_1, "Failed to allocate memory for " - "beiscsi_bsg_request\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BM_%d : Failed to allocate memory for " + "beiscsi_bsg_request\n"); return -EIO; } tag = mgmt_vendor_specific_fw_cmd(&phba->ctrl, phba, job, &nonemb_cmd); if (!tag) { - SE_DEBUG(DBG_LVL_1, "be_cmd_get_mac_addr Failed\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BM_%d : be_cmd_get_mac_addr Failed\n"); + pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); return -EAGAIN; @@ -4191,22 +4355,31 @@ static int beiscsi_bsg_request(struct bsg_job *job) pci_free_consistent(phba->ctrl.pdev, nonemb_cmd.size, nonemb_cmd.va, nonemb_cmd.dma); if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, "be_cmd_get_mac_addr Failed" - " status = %d extd_status = %d\n", - status, extd_status); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BM_%d : be_cmd_get_mac_addr Failed" + " status = %d extd_status = %d\n", + status, extd_status); + return -EIO; } break; default: - SE_DEBUG(DBG_LVL_1, "Unsupported bsg command: 0x%x\n", - bsg_req->msgcode); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BM_%d : Unsupported bsg command: 0x%x\n", + bsg_req->msgcode); break; } return rc; } +void beiscsi_hba_attrs_init(struct beiscsi_hba *phba) +{ + /* Set the logging parameter */ + beiscsi_log_enable_init(phba, beiscsi_log_enable); +} + static void beiscsi_quiesce(struct beiscsi_hba *phba) { struct hwi_controller *phwi_ctrlr; @@ -4316,18 +4489,21 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, ret = beiscsi_enable_pci(pcidev); if (ret < 0) { - dev_err(&pcidev->dev, "beiscsi_dev_probe-" - " Failed to enable pci device\n"); + dev_err(&pcidev->dev, + "beiscsi_dev_probe - Failed to enable pci device\n"); return ret; } phba = beiscsi_hba_alloc(pcidev); if (!phba) { - dev_err(&pcidev->dev, "beiscsi_dev_probe-" - " Failed in beiscsi_hba_alloc\n"); + dev_err(&pcidev->dev, + "beiscsi_dev_probe - Failed in beiscsi_hba_alloc\n"); goto disable_pci; } + /* Initialize Driver configuration Paramters */ + beiscsi_hba_attrs_init(phba); + switch (pcidev->device) { case BE_DEVICE_ID1: case OC_DEVICE_ID1: @@ -4347,7 +4523,9 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, else num_cpus = 1; phba->num_cpus = num_cpus; - SE_DEBUG(DBG_LVL_8, "num_cpus = %d\n", phba->num_cpus); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : num_cpus = %d\n", + phba->num_cpus); if (enable_msix) { beiscsi_msix_enable(phba); @@ -4356,8 +4534,9 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, } ret = be_ctrl_init(phba, pcidev); if (ret) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_dev_probe-" - "Failed in be_ctrl_init\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe-" + "Failed in be_ctrl_init\n"); goto hba_free; } @@ -4366,19 +4545,19 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, value = readl((void *)real_offset); if (value & 0x00010000) { gcrashmode++; - shost_printk(KERN_ERR, phba->shost, - "Loading Driver in crashdump mode\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Loading Driver in crashdump mode\n"); ret = beiscsi_cmd_reset_function(phba); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "Reset Failed. Aborting Crashdump\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Reset Failed. Aborting Crashdump\n"); goto hba_free; } ret = be_chk_reset_complete(phba); if (ret) { - shost_printk(KERN_ERR, phba->shost, - "Failed to get out of reset." - "Aborting Crashdump\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Failed to get out of reset." + "Aborting Crashdump\n"); goto hba_free; } } else { @@ -4393,8 +4572,8 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, spin_lock_init(&phba->isr_lock); ret = mgmt_get_fw_config(&phba->ctrl, phba); if (ret != 0) { - shost_printk(KERN_ERR, phba->shost, - "Error getting fw config\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Error getting fw config\n"); goto free_port; } phba->shost->max_id = phba->fw_config.iscsi_cid_count; @@ -4402,8 +4581,9 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, phba->shost->can_queue = phba->params.ios_per_ctrl; ret = beiscsi_init_port(phba); if (ret < 0) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_dev_probe-" - "Failed in beiscsi_init_port\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe-" + "Failed in beiscsi_init_port\n"); goto free_port; } @@ -4420,8 +4600,9 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, phba->shost->host_no); phba->wq = alloc_workqueue(phba->wq_name, WQ_MEM_RECLAIM, 1); if (!phba->wq) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_dev_probe-" - "Failed to allocate work queue\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe-" + "Failed to allocate work queue\n"); goto free_twq; } @@ -4439,8 +4620,9 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, } ret = beiscsi_init_irqs(phba); if (ret < 0) { - shost_printk(KERN_ERR, phba->shost, "beiscsi_dev_probe-" - "Failed to beiscsi_init_irqs\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : beiscsi_dev_probe-" + "Failed to beiscsi_init_irqs\n"); goto free_blkenbld; } hwi_enable_intr(phba); @@ -4450,11 +4632,13 @@ static int __devinit beiscsi_dev_probe(struct pci_dev *pcidev, * log error but continue, because we may not be using * iscsi boot. */ - shost_printk(KERN_ERR, phba->shost, "Could not set up " - "iSCSI boot info.\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BM_%d : Could not set up " + "iSCSI boot info.\n"); beiscsi_create_def_ifaces(phba); - SE_DEBUG(DBG_LVL_8, "\n\n\n SUCCESS - DRIVER LOADED\n\n\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "\n\n\n BM_%d : SUCCESS - DRIVER LOADED\n\n\n"); return 0; free_blkenbld: @@ -4542,19 +4726,17 @@ static int __init beiscsi_module_init(void) beiscsi_scsi_transport = iscsi_register_transport(&beiscsi_iscsi_transport); if (!beiscsi_scsi_transport) { - SE_DEBUG(DBG_LVL_1, - "beiscsi_module_init - Unable to register beiscsi" - "transport.\n"); + printk(KERN_ERR + "beiscsi_module_init - Unable to register beiscsi transport.\n"); return -ENOMEM; } - SE_DEBUG(DBG_LVL_8, "In beiscsi_module_init, tt=%p\n", - &beiscsi_iscsi_transport); + printk(KERN_INFO "In beiscsi_module_init, tt=%p\n", + &beiscsi_iscsi_transport); ret = pci_register_driver(&beiscsi_pci_driver); if (ret) { - SE_DEBUG(DBG_LVL_1, - "beiscsi_module_init - Unable to register" - "beiscsi pci driver.\n"); + printk(KERN_ERR + "beiscsi_module_init - Unable to register beiscsi pci driver.\n"); goto unregister_iscsi_transport; } return 0; diff --git a/drivers/scsi/be2iscsi/be_main.h b/drivers/scsi/be2iscsi/be_main.h index 40fea6e..b891226 100644 --- a/drivers/scsi/be2iscsi/be_main.h +++ b/drivers/scsi/be2iscsi/be_main.h @@ -24,6 +24,8 @@ #include <linux/pci.h> #include <linux/if_ether.h> #include <linux/in.h> +#include <linux/ctype.h> +#include <linux/module.h> #include <scsi/scsi.h> #include <scsi/scsi_cmnd.h> #include <scsi/scsi_device.h> @@ -34,7 +36,7 @@ #include "be.h" #define DRV_NAME "be2iscsi" -#define BUILD_STR "4.2.162.0" +#define BUILD_STR "4.4.58.0" #define BE_NAME "Emulex OneConnect" \ "Open-iSCSI Driver version" BUILD_STR #define DRV_DESC BE_NAME " " "Driver" @@ -84,23 +86,7 @@ #define MAX_CMD_SZ 65536 #define IIOC_SCSI_DATA 0x05 /* Write Operation */ -#define DBG_LVL 0x00000001 -#define DBG_LVL_1 0x00000001 -#define DBG_LVL_2 0x00000002 -#define DBG_LVL_3 0x00000004 -#define DBG_LVL_4 0x00000008 -#define DBG_LVL_5 0x00000010 -#define DBG_LVL_6 0x00000020 -#define DBG_LVL_7 0x00000040 -#define DBG_LVL_8 0x00000080 - -#define SE_DEBUG(debug_mask, fmt, args...) \ -do { \ - if (debug_mask & DBG_LVL) { \ - printk(KERN_ERR "(%s():%d):", __func__, __LINE__);\ - printk(fmt, ##args); \ - } \ -} while (0); +#define INVALID_SESS_HANDLE 0xFFFFFFFF #define BE_ADAPTER_UP 0x00000000 #define BE_ADAPTER_LINK_DOWN 0x00000001 @@ -351,6 +337,8 @@ struct beiscsi_hba { struct mgmt_session_info boot_sess; struct invalidate_command_table inv_tbl[128]; + unsigned int attr_log_enable; + }; struct beiscsi_session { @@ -860,4 +848,20 @@ struct hwi_context_memory { struct hwi_async_pdu_context *pasync_ctx; }; +/* Logging related definitions */ +#define BEISCSI_LOG_INIT 0x0001 /* Initialization events */ +#define BEISCSI_LOG_MBOX 0x0002 /* Mailbox Events */ +#define BEISCSI_LOG_MISC 0x0004 /* Miscllaneous Events */ +#define BEISCSI_LOG_EH 0x0008 /* Error Handler */ +#define BEISCSI_LOG_IO 0x0010 /* IO Code Path */ +#define BEISCSI_LOG_CONFIG 0x0020 /* CONFIG Code Path */ + +#define beiscsi_log(phba, level, mask, fmt, arg...) \ +do { \ + uint32_t log_value = phba->attr_log_enable; \ + if (((mask) & log_value) || (level[1] <= '3')) \ + shost_printk(level, phba->shost, \ + fmt, __LINE__, ##arg); \ +} while (0) + #endif diff --git a/drivers/scsi/be2iscsi/be_mgmt.c b/drivers/scsi/be2iscsi/be_mgmt.c index 2a09679..aab5dd3 100644 --- a/drivers/scsi/be2iscsi/be_mgmt.c +++ b/drivers/scsi/be2iscsi/be_mgmt.c @@ -23,6 +23,53 @@ #include "be_mgmt.h" #include "be_iscsi.h" +/** + * mgmt_reopen_session()- Reopen a session based on reopen_type + * @phba: Device priv structure instance + * @reopen_type: Type of reopen_session FW should do. + * @sess_handle: Session Handle of the session to be re-opened + * + * return + * the TAG used for MBOX Command + * + **/ +unsigned int mgmt_reopen_session(struct beiscsi_hba *phba, + unsigned int reopen_type, + unsigned int sess_handle) +{ + struct be_ctrl_info *ctrl = &phba->ctrl; + struct be_mcc_wrb *wrb; + struct be_cmd_reopen_session_req *req; + unsigned int tag = 0; + + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BG_%d : In bescsi_get_boot_target\n"); + + spin_lock(&ctrl->mbox_lock); + tag = alloc_mcc_tag(phba); + if (!tag) { + spin_unlock(&ctrl->mbox_lock); + return tag; + } + + wrb = wrb_from_mccq(phba); + req = embedded_payload(wrb); + wrb->tag0 |= tag; + be_wrb_hdr_prepare(wrb, sizeof(*req), true, 0); + be_cmd_hdr_prepare(&req->hdr, CMD_SUBSYSTEM_ISCSI_INI, + OPCODE_ISCSI_INI_DRIVER_REOPEN_ALL_SESSIONS, + sizeof(struct be_cmd_reopen_session_resp)); + + /* set the reopen_type,sess_handle */ + req->reopen_type = reopen_type; + req->session_handle = sess_handle; + + be_mcc_notify(phba); + spin_unlock(&ctrl->mbox_lock); + return tag; +} + unsigned int mgmt_get_boot_target(struct beiscsi_hba *phba) { struct be_ctrl_info *ctrl = &phba->ctrl; @@ -30,7 +77,10 @@ unsigned int mgmt_get_boot_target(struct beiscsi_hba *phba) struct be_cmd_get_boot_target_req *req; unsigned int tag = 0; - SE_DEBUG(DBG_LVL_8, "In bescsi_get_boot_target\n"); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BG_%d : In bescsi_get_boot_target\n"); + spin_lock(&ctrl->mbox_lock); tag = alloc_mcc_tag(phba); if (!tag) { @@ -62,7 +112,10 @@ unsigned int mgmt_get_session_info(struct beiscsi_hba *phba, struct be_cmd_get_session_resp *resp; struct be_sge *sge; - SE_DEBUG(DBG_LVL_8, "In beiscsi_get_session_info\n"); + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BG_%d : In beiscsi_get_session_info\n"); + spin_lock(&ctrl->mbox_lock); tag = alloc_mcc_tag(phba); if (!tag) { @@ -121,16 +174,16 @@ int mgmt_get_fw_config(struct be_ctrl_info *ctrl, phba->fw_config.iscsi_cid_count = pfw_cfg->ulp[0].sq_count; if (phba->fw_config.iscsi_cid_count > (BE2_MAX_SESSIONS / 2)) { - SE_DEBUG(DBG_LVL_8, - "FW reported MAX CXNS as %d\t" - "Max Supported = %d.\n", - phba->fw_config.iscsi_cid_count, - BE2_MAX_SESSIONS); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BG_%d : FW reported MAX CXNS as %d\t" + "Max Supported = %d.\n", + phba->fw_config.iscsi_cid_count, + BE2_MAX_SESSIONS); phba->fw_config.iscsi_cid_count = BE2_MAX_SESSIONS / 2; } } else { - shost_printk(KERN_WARNING, phba->shost, - "Failed in mgmt_get_fw_config\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BG_%d : Failed in mgmt_get_fw_config\n"); } spin_unlock(&ctrl->mbox_lock); @@ -150,9 +203,9 @@ int mgmt_check_supported_fw(struct be_ctrl_info *ctrl, sizeof(struct be_mgmt_controller_attributes), &nonemb_cmd.dma); if (nonemb_cmd.va == NULL) { - SE_DEBUG(DBG_LVL_1, - "Failed to allocate memory for mgmt_check_supported_fw" - "\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BG_%d : Failed to allocate memory for " + "mgmt_check_supported_fw\n"); return -ENOMEM; } nonemb_cmd.size = sizeof(struct be_mgmt_controller_attributes); @@ -169,18 +222,23 @@ int mgmt_check_supported_fw(struct be_ctrl_info *ctrl, status = be_mbox_notify(ctrl); if (!status) { struct be_mgmt_controller_attributes_resp *resp = nonemb_cmd.va; - SE_DEBUG(DBG_LVL_8, "Firmware version of CMD: %s\n", - resp->params.hba_attribs.flashrom_version_string); - SE_DEBUG(DBG_LVL_8, "Firmware version is : %s\n", - resp->params.hba_attribs.firmware_version_string); - SE_DEBUG(DBG_LVL_8, - "Developer Build, not performing version check...\n"); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BG_%d : Firmware Version of CMD : %s\n" + "Firmware Version is : %s\n" + "Developer Build, not performing version check...\n", + resp->params.hba_attribs + .flashrom_version_string, + resp->params.hba_attribs. + firmware_version_string); + phba->fw_config.iscsi_features = resp->params.hba_attribs.iscsi_features; - SE_DEBUG(DBG_LVL_8, " phba->fw_config.iscsi_features = %d\n", - phba->fw_config.iscsi_features); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_INIT, + "BM_%d : phba->fw_config.iscsi_features = %d\n", + phba->fw_config.iscsi_features); } else - SE_DEBUG(DBG_LVL_1, " Failed in mgmt_check_supported_fw\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_INIT, + "BG_%d : Failed in mgmt_check_supported_fw\n"); spin_unlock(&ctrl->mbox_lock); if (nonemb_cmd.va) pci_free_consistent(ctrl->pdev, nonemb_cmd.size, @@ -229,9 +287,10 @@ unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl, OPCODE_COMMON_READ_FLASH, sizeof(*req)); break; default: - shost_printk(KERN_WARNING, phba->shost, - "Unsupported cmd = 0x%x\n\n", bsg_req->rqst_data. - h_vendor.vendor_cmd[0]); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Unsupported cmd = 0x%x\n\n", + bsg_req->rqst_data.h_vendor.vendor_cmd[0]); + spin_unlock(&ctrl->mbox_lock); return -ENOSYS; } @@ -275,8 +334,8 @@ int mgmt_epfw_cleanup(struct beiscsi_hba *phba, unsigned short chute) status = be_mcc_notify_wait(phba); if (status) - shost_printk(KERN_WARNING, phba->shost, - " mgmt_epfw_cleanup , FAILED\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_INIT, + "BG_%d : mgmt_epfw_cleanup , FAILED\n"); spin_unlock(&ctrl->mbox_lock); return status; } @@ -459,8 +518,9 @@ int mgmt_open_connection(struct beiscsi_hba *phba, &daddr_in6->sin6_addr.in6_u.u6_addr8, 16); beiscsi_ep->ip_type = BE2_IPV6; } else{ - shost_printk(KERN_ERR, phba->shost, "unknown addr family %d\n", - dst_addr->sa_family); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BG_%d : unknown addr family %d\n", + dst_addr->sa_family); spin_unlock(&ctrl->mbox_lock); free_mcc_tag(&phba->ctrl, tag); return -EINVAL; @@ -471,7 +531,8 @@ int mgmt_open_connection(struct beiscsi_hba *phba, if (phba->nxt_cqid == phba->num_cpus) phba->nxt_cqid = 0; req->cq_id = phwi_context->be_cq[i].id; - SE_DEBUG(DBG_LVL_8, "i=%d cq_id=%d\n", i, req->cq_id); + beiscsi_log(phba, KERN_INFO, BEISCSI_LOG_CONFIG, + "BG_%d : i=%d cq_id=%d\n", i, req->cq_id); req->defq_id = def_hdr_id; req->hdr_ring_id = def_hdr_id; req->data_ring_id = def_data_id; @@ -506,8 +567,8 @@ unsigned int mgmt_get_all_if_id(struct beiscsi_hba *phba) if (!status) phba->interface_handle = pbe_allid->if_hndl_list[0]; else { - shost_printk(KERN_WARNING, phba->shost, - "Failed in mgmt_get_all_if_id\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Failed in mgmt_get_all_if_id\n"); } spin_unlock(&ctrl->mbox_lock); @@ -550,9 +611,10 @@ static int mgmt_exec_nonemb_cmd(struct beiscsi_hba *phba, extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; if (status || extd_status) { - SE_DEBUG(DBG_LVL_1, - "mgmt_exec_nonemb_cmd Failed status = %d" - "extd_status = %d\n", status, extd_status); + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX, + "BG_%d : mgmt_exec_nonemb_cmd Failed status = %d" + "extd_status = %d\n", status, extd_status); rc = -EIO; goto free_tag; } @@ -573,7 +635,8 @@ static int mgmt_alloc_cmd_data(struct beiscsi_hba *phba, struct be_dma_mem *cmd, { cmd->va = pci_alloc_consistent(phba->ctrl.pdev, size, &cmd->dma); if (!cmd->va) { - SE_DEBUG(DBG_LVL_1, "Failed to allocate memory for if info\n"); + beiscsi_log(phba, KERN_ERR, BEISCSI_LOG_CONFIG, + "BG_%d : Failed to allocate memory for if info\n"); return -ENOMEM; } memset(cmd->va, 0, size); @@ -629,8 +692,8 @@ mgmt_static_ip_modify(struct beiscsi_hba *phba, rc = mgmt_exec_nonemb_cmd(phba, &nonemb_cmd, NULL, 0); if (rc < 0) - shost_printk(KERN_WARNING, phba->shost, - "Failed to Modify existing IP Address\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Failed to Modify existing IP Address\n"); return rc; } @@ -684,8 +747,8 @@ int mgmt_set_ip(struct beiscsi_hba *phba, if (boot_proto == ISCSI_BOOTPROTO_DHCP) { if (if_info.dhcp_state) { - shost_printk(KERN_WARNING, phba->shost, - "DHCP Already Enabled\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : DHCP Already Enabled\n"); return 0; } /* The ip_param->len is 1 in DHCP case. Setting @@ -712,8 +775,9 @@ int mgmt_set_ip(struct beiscsi_hba *phba, rc = mgmt_exec_nonemb_cmd(phba, &nonemb_cmd, NULL, 0); if (rc < 0) { - shost_printk(KERN_WARNING, phba->shost, - "Failed to Delete existing dhcp\n"); + beiscsi_log(phba, KERN_WARNING, + BEISCSI_LOG_CONFIG, + "BG_%d : Failed to Delete existing dhcp\n"); return rc; } } @@ -732,8 +796,8 @@ int mgmt_set_ip(struct beiscsi_hba *phba, memset(>way_addr_set, 0, sizeof(gtway_addr_set)); rc = mgmt_get_gateway(phba, BE2_IPV4, >way_addr_set); if (rc) { - shost_printk(KERN_WARNING, phba->shost, - "Failed to Get Gateway Addr\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Failed to Get Gateway Addr\n"); return rc; } @@ -743,8 +807,9 @@ int mgmt_set_ip(struct beiscsi_hba *phba, IP_ACTION_DEL, IP_V4_LEN); if (rc) { - shost_printk(KERN_WARNING, phba->shost, - "Failed to clear Gateway Addr Set\n"); + beiscsi_log(phba, KERN_WARNING, + BEISCSI_LOG_CONFIG, + "BG_%d : Failed to clear Gateway Addr Set\n"); return rc; } } @@ -783,8 +848,8 @@ int mgmt_set_gateway(struct beiscsi_hba *phba, memset(>way_addr_set, 0, sizeof(gtway_addr_set)); rt_val = mgmt_get_gateway(phba, BE2_IPV4, >way_addr_set); if (rt_val) { - shost_printk(KERN_WARNING, phba->shost, - "Failed to Get Gateway Addr\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Failed to Get Gateway Addr\n"); return rt_val; } @@ -793,8 +858,8 @@ int mgmt_set_gateway(struct beiscsi_hba *phba, rt_val = mgmt_modify_gateway(phba, gtway_addr, IP_ACTION_DEL, gateway_param->len); if (rt_val) { - shost_printk(KERN_WARNING, phba->shost, - "Failed to clear Gateway Addr Set\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Failed to clear Gateway Addr Set\n"); return rt_val; } } @@ -804,8 +869,8 @@ int mgmt_set_gateway(struct beiscsi_hba *phba, gateway_param->len); if (rt_val) - shost_printk(KERN_WARNING, phba->shost, - "Failed to Set Gateway Addr\n"); + beiscsi_log(phba, KERN_WARNING, BEISCSI_LOG_CONFIG, + "BG_%d : Failed to Set Gateway Addr\n"); return rt_val; } @@ -924,3 +989,150 @@ unsigned int be_cmd_get_port_speed(struct beiscsi_hba *phba) spin_unlock(&ctrl->mbox_lock); return tag; } + +/** + * be_mgmt_get_boot_shandle()- Get the session handle + * @phba: device priv structure instance + * @s_handle: session handle returned for boot session. + * + * Get the boot target session handle. In case of + * crashdump mode driver has to issue and MBX Cmd + * for FW to login to boot target + * + * return + * Success: 0 + * Failure: Non-Zero value + * + **/ +int be_mgmt_get_boot_shandle(struct beiscsi_hba *phba, + unsigned int *s_handle) +{ + struct be_cmd_get_boot_target_resp *boot_resp; + struct be_mcc_wrb *wrb; + unsigned int tag, wrb_num; + uint8_t boot_retry = 3; + unsigned short status, extd_status; + struct be_queue_info *mccq = &phba->ctrl.mcc_obj.q; + + do { + /* Get the Boot Target Session Handle and Count*/ + tag = mgmt_get_boot_target(phba); + if (!tag) { + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_CONFIG | BEISCSI_LOG_INIT, + "BG_%d : Getting Boot Target Info Failed\n"); + return -EAGAIN; + } else + wait_event_interruptible(phba->ctrl.mcc_wait[tag], + phba->ctrl.mcc_numtag[tag]); + + wrb_num = (phba->ctrl.mcc_numtag[tag] & 0x00FF0000) >> 16; + extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; + status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; + if (status || extd_status) { + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BG_%d : mgmt_get_boot_target Failed" + " status = %d extd_status = %d\n", + status, extd_status); + free_mcc_tag(&phba->ctrl, tag); + return -EBUSY; + } + wrb = queue_get_wrb(mccq, wrb_num); + free_mcc_tag(&phba->ctrl, tag); + boot_resp = embedded_payload(wrb); + + /* Check if the there are any Boot targets configured */ + if (!boot_resp->boot_session_count) { + beiscsi_log(phba, KERN_INFO, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BG_%d ;No boot targets configured\n"); + return -ENXIO; + } + + /* FW returns the session handle of the boot session */ + if (boot_resp->boot_session_handle != INVALID_SESS_HANDLE) { + *s_handle = boot_resp->boot_session_handle; + return 0; + } + + /* Issue MBX Cmd to FW to login to the boot target */ + tag = mgmt_reopen_session(phba, BE_REOPEN_BOOT_SESSIONS, + INVALID_SESS_HANDLE); + if (!tag) { + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BG_%d : mgmt_reopen_session Failed\n"); + return -EAGAIN; + } else + wait_event_interruptible(phba->ctrl.mcc_wait[tag], + phba->ctrl.mcc_numtag[tag]); + + wrb_num = (phba->ctrl.mcc_numtag[tag] & 0x00FF0000) >> 16; + extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; + status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; + if (status || extd_status) { + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BG_%d : mgmt_reopen_session Failed" + " status = %d extd_status = %d\n", + status, extd_status); + free_mcc_tag(&phba->ctrl, tag); + return -EBUSY; + } + free_mcc_tag(&phba->ctrl, tag); + + } while (--boot_retry); + + /* Couldn't log into the boot target */ + beiscsi_log(phba, KERN_ERR, + BEISCSI_LOG_INIT | BEISCSI_LOG_CONFIG, + "BG_%d : Login to Boot Target Failed\n"); + return -ENXIO; +} + +/** + * mgmt_set_vlan()- Issue and wait for CMD completion + * @phba: device private structure instance + * @vlan_tag: VLAN tag + * + * Issue the MBX Cmd and wait for the completion of the + * command. + * + * returns + * Success: 0 + * Failure: Non-Xero Value + **/ +int mgmt_set_vlan(struct beiscsi_hba *phba, + uint16_t vlan_tag) +{ + unsigned int tag, wrb_num; + unsigned short status, extd_status; + + tag = be_cmd_set_vlan(phba, vlan_tag); + if (!tag) { + beiscsi_log(phba, KERN_ERR, + (BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX), + "BG_%d : VLAN Setting Failed\n"); + return -EBUSY; + } else + wait_event_interruptible(phba->ctrl.mcc_wait[tag], + phba->ctrl.mcc_numtag[tag]); + + wrb_num = (phba->ctrl.mcc_numtag[tag] & 0x00FF0000) >> 16; + extd_status = (phba->ctrl.mcc_numtag[tag] & 0x0000FF00) >> 8; + status = phba->ctrl.mcc_numtag[tag] & 0x000000FF; + + if (status || extd_status) { + beiscsi_log(phba, KERN_ERR, + (BEISCSI_LOG_CONFIG | BEISCSI_LOG_MBOX), + "BS_%d : status : %d extd_status : %d\n", + status, extd_status); + + free_mcc_tag(&phba->ctrl, tag); + return -EAGAIN; + } + + free_mcc_tag(&phba->ctrl, tag); + return 0; +} diff --git a/drivers/scsi/be2iscsi/be_mgmt.h b/drivers/scsi/be2iscsi/be_mgmt.h index 5c2e376..c50cef6 100644 --- a/drivers/scsi/be2iscsi/be_mgmt.h +++ b/drivers/scsi/be2iscsi/be_mgmt.h @@ -108,6 +108,7 @@ unsigned int mgmt_vendor_specific_fw_cmd(struct be_ctrl_info *ctrl, struct bsg_job *job, struct be_dma_mem *nonemb_cmd); +#define BEISCSI_NO_RST_ISSUE 0 struct iscsi_invalidate_connection_params_in { struct be_cmd_req_hdr hdr; unsigned int session_handle; @@ -274,6 +275,10 @@ int mgmt_set_ip(struct beiscsi_hba *phba, unsigned int mgmt_get_boot_target(struct beiscsi_hba *phba); +unsigned int mgmt_reopen_session(struct beiscsi_hba *phba, + unsigned int reopen_type, + unsigned sess_handle); + unsigned int mgmt_get_session_info(struct beiscsi_hba *phba, u32 boot_session_handle, struct be_dma_mem *nonemb_cmd); @@ -290,4 +295,10 @@ int mgmt_get_gateway(struct beiscsi_hba *phba, int ip_type, int mgmt_set_gateway(struct beiscsi_hba *phba, struct iscsi_iface_param_info *gateway_param); +int be_mgmt_get_boot_shandle(struct beiscsi_hba *phba, + unsigned int *s_handle); + +unsigned int mgmt_get_all_if_id(struct beiscsi_hba *phba); + +int mgmt_set_vlan(struct beiscsi_hba *phba, uint16_t vlan_tag); #endif diff --git a/drivers/scsi/bfa/bfa_core.c b/drivers/scsi/bfa/bfa_core.c index 456e576..b7c326f 100644 --- a/drivers/scsi/bfa/bfa_core.c +++ b/drivers/scsi/bfa/bfa_core.c @@ -775,7 +775,8 @@ bfa_intx(struct bfa_s *bfa) if (!intr) return BFA_TRUE; - bfa_msix_lpu_err(bfa, intr); + if (bfa->intr_enabled) + bfa_msix_lpu_err(bfa, intr); return BFA_TRUE; } @@ -803,11 +804,17 @@ bfa_isr_enable(struct bfa_s *bfa) writel(~umsk, bfa->iocfc.bfa_regs.intr_mask); bfa->iocfc.intr_mask = ~umsk; bfa_isr_mode_set(bfa, bfa->msix.nvecs != 0); + + /* + * Set the flag indicating successful enabling of interrupts + */ + bfa->intr_enabled = BFA_TRUE; } void bfa_isr_disable(struct bfa_s *bfa) { + bfa->intr_enabled = BFA_FALSE; bfa_isr_mode_set(bfa, BFA_FALSE); writel(-1L, bfa->iocfc.bfa_regs.intr_mask); bfa_msix_uninstall(bfa); @@ -1022,7 +1029,7 @@ bfa_iocfc_mem_claim(struct bfa_s *bfa, struct bfa_iocfc_cfg_s *cfg) { u8 *dm_kva = NULL; u64 dm_pa = 0; - int i, per_reqq_sz, per_rspq_sz, dbgsz; + int i, per_reqq_sz, per_rspq_sz; struct bfa_iocfc_s *iocfc = &bfa->iocfc; struct bfa_mem_dma_s *ioc_dma = BFA_MEM_IOC_DMA(bfa); struct bfa_mem_dma_s *iocfc_dma = BFA_MEM_IOCFC_DMA(bfa); @@ -1083,11 +1090,8 @@ bfa_iocfc_mem_claim(struct bfa_s *bfa, struct bfa_iocfc_cfg_s *cfg) BFA_CACHELINE_SZ); /* Claim IOCFC kva memory */ - dbgsz = (bfa_auto_recover) ? BFA_DBG_FWTRC_LEN : 0; - if (dbgsz > 0) { - bfa_ioc_debug_memclaim(&bfa->ioc, bfa_mem_kva_curp(iocfc)); - bfa_mem_kva_curp(iocfc) += dbgsz; - } + bfa_ioc_debug_memclaim(&bfa->ioc, bfa_mem_kva_curp(iocfc)); + bfa_mem_kva_curp(iocfc) += BFA_DBG_FWTRC_LEN; } /* @@ -1429,8 +1433,7 @@ bfa_iocfc_meminfo(struct bfa_iocfc_cfg_s *cfg, struct bfa_meminfo_s *meminfo, bfa_mem_dma_setup(meminfo, iocfc_dma, dm_len); /* kva memory setup for IOCFC */ - bfa_mem_kva_setup(meminfo, iocfc_kva, - ((bfa_auto_recover) ? BFA_DBG_FWTRC_LEN : 0)); + bfa_mem_kva_setup(meminfo, iocfc_kva, BFA_DBG_FWTRC_LEN); } /* diff --git a/drivers/scsi/bfa/bfa_cs.h b/drivers/scsi/bfa/bfa_cs.h index 12bfeed..91a8aa3 100644 --- a/drivers/scsi/bfa/bfa_cs.h +++ b/drivers/scsi/bfa/bfa_cs.h @@ -168,7 +168,7 @@ __bfa_trc32(struct bfa_trc_mod_s *trcm, int fileno, int line, u32 data) /* * bfa_q_deq - dequeue an element from head of the queue */ -#define bfa_q_deq(_q, _qe) { \ +#define bfa_q_deq(_q, _qe) do { \ if (!list_empty(_q)) { \ (*((struct list_head **) (_qe))) = bfa_q_next(_q); \ bfa_q_prev(bfa_q_next(*((struct list_head **) _qe))) = \ @@ -177,7 +177,7 @@ __bfa_trc32(struct bfa_trc_mod_s *trcm, int fileno, int line, u32 data) } else { \ *((struct list_head **) (_qe)) = (struct list_head *) NULL;\ } \ -} +} while (0) /* * bfa_q_deq_tail - dequeue an element from tail of the queue diff --git a/drivers/scsi/bfa/bfa_defs_fcs.h b/drivers/scsi/bfa/bfa_defs_fcs.h index 3bbc583..06f0a16 100644 --- a/drivers/scsi/bfa/bfa_defs_fcs.h +++ b/drivers/scsi/bfa/bfa_defs_fcs.h @@ -93,6 +93,7 @@ struct bfa_lport_cfg_s { wwn_t pwwn; /* port wwn */ wwn_t nwwn; /* node wwn */ struct bfa_lport_symname_s sym_name; /* vm port symbolic name */ + struct bfa_lport_symname_s node_sym_name; /* Node symbolic name */ enum bfa_lport_role roles; /* FCS port roles */ u32 rsvd; bfa_boolean_t preboot_vp; /* vport created from PBC */ @@ -192,6 +193,18 @@ struct bfa_lport_stats_s { u32 ns_gidft_unknown_rsp; u32 ns_gidft_alloc_wait; + u32 ns_rnnid_sent; + u32 ns_rnnid_accepts; + u32 ns_rnnid_rsp_err; + u32 ns_rnnid_rejects; + u32 ns_rnnid_alloc_wait; + + u32 ns_rsnn_nn_sent; + u32 ns_rsnn_nn_accepts; + u32 ns_rsnn_nn_rsp_err; + u32 ns_rsnn_nn_rejects; + u32 ns_rsnn_nn_alloc_wait; + /* * Mgmt Server stats */ @@ -410,6 +423,11 @@ struct bfa_rport_remote_link_stats_s { u32 icc; /* Invalid CRC Count */ }; +struct bfa_rport_qualifier_s { + wwn_t pwwn; /* Port WWN */ + u32 pid; /* port ID */ + u32 rsvd; +}; #define BFA_MAX_IO_INDEX 7 #define BFA_NO_IO_INDEX 9 diff --git a/drivers/scsi/bfa/bfa_fc.h b/drivers/scsi/bfa/bfa_fc.h index 8d0b88f..e0beb4d 100644 --- a/drivers/scsi/bfa/bfa_fc.h +++ b/drivers/scsi/bfa/bfa_fc.h @@ -1279,6 +1279,7 @@ enum { GS_GSPN_ID = 0x0118, /* Get symbolic PN on ID */ GS_RFT_ID = 0x0217, /* Register fc4type on ID */ GS_RSPN_ID = 0x0218, /* Register symbolic PN on ID */ + GS_RSNN_NN = 0x0239, /* Register symbolic NN on NN */ GS_RPN_ID = 0x0212, /* Register port name */ GS_RNN_ID = 0x0213, /* Register node name */ GS_RCS_ID = 0x0214, /* Register class of service */ @@ -1357,6 +1358,15 @@ struct fcgs_rspnid_req_s { }; /* + * RSNN_NN + */ +struct fcgs_rsnn_nn_req_s { + wwn_t node_name; /* Node name */ + u8 snn_len; /* symbolic node name length */ + u8 snn[256]; /* symbolic node name */ +}; + +/* * RPN_ID */ struct fcgs_rpnid_req_s { diff --git a/drivers/scsi/bfa/bfa_fcbuild.c b/drivers/scsi/bfa/bfa_fcbuild.c index 17b59b8..273cee9 100644 --- a/drivers/scsi/bfa/bfa_fcbuild.c +++ b/drivers/scsi/bfa/bfa_fcbuild.c @@ -1252,6 +1252,27 @@ fc_rspnid_build(struct fchs_s *fchs, void *pyld, u32 s_id, u16 ox_id, } u16 +fc_rsnn_nn_build(struct fchs_s *fchs, void *pyld, u32 s_id, + wwn_t node_name, u8 *name) +{ + struct ct_hdr_s *cthdr = (struct ct_hdr_s *) pyld; + struct fcgs_rsnn_nn_req_s *rsnn_nn = + (struct fcgs_rsnn_nn_req_s *) (cthdr + 1); + u32 d_id = bfa_hton3b(FC_NAME_SERVER); + + fc_gs_fchdr_build(fchs, d_id, s_id, 0); + fc_gs_cthdr_build(cthdr, s_id, GS_RSNN_NN); + + memset(rsnn_nn, 0, sizeof(struct fcgs_rsnn_nn_req_s)); + + rsnn_nn->node_name = node_name; + rsnn_nn->snn_len = (u8) strlen((char *)name); + strncpy((char *)rsnn_nn->snn, (char *)name, rsnn_nn->snn_len); + + return sizeof(struct fcgs_rsnn_nn_req_s) + sizeof(struct ct_hdr_s); +} + +u16 fc_gid_ft_build(struct fchs_s *fchs, void *pyld, u32 s_id, u8 fc4_type) { diff --git a/drivers/scsi/bfa/bfa_fcbuild.h b/drivers/scsi/bfa/bfa_fcbuild.h index 42cd9d4..03c753d 100644 --- a/drivers/scsi/bfa/bfa_fcbuild.h +++ b/drivers/scsi/bfa/bfa_fcbuild.h @@ -166,6 +166,8 @@ enum fc_parse_status fc_rrq_rsp_parse(struct fchs_s *buf, int len); u16 fc_rspnid_build(struct fchs_s *fchs, void *pld, u32 s_id, u16 ox_id, u8 *name); +u16 fc_rsnn_nn_build(struct fchs_s *fchs, void *pld, u32 s_id, + wwn_t node_name, u8 *name); u16 fc_rftid_build(struct fchs_s *fchs, void *pld, u32 s_id, u16 ox_id, enum bfa_lport_role role); diff --git a/drivers/scsi/bfa/bfa_fcpim.c b/drivers/scsi/bfa/bfa_fcpim.c index f0f80e2..1633963 100644 --- a/drivers/scsi/bfa/bfa_fcpim.c +++ b/drivers/scsi/bfa/bfa_fcpim.c @@ -1466,7 +1466,13 @@ bfa_status_t bfa_itnim_get_ioprofile(struct bfa_itnim_s *itnim, struct bfa_itnim_ioprofile_s *ioprofile) { - struct bfa_fcpim_s *fcpim = BFA_FCPIM(itnim->bfa); + struct bfa_fcpim_s *fcpim; + + if (!itnim) + return BFA_STATUS_NO_FCPIM_NEXUS; + + fcpim = BFA_FCPIM(itnim->bfa); + if (!fcpim->io_profile) return BFA_STATUS_IOPROFILE_OFF; @@ -1484,6 +1490,10 @@ void bfa_itnim_clear_stats(struct bfa_itnim_s *itnim) { int j; + + if (!itnim) + return; + memset(&itnim->stats, 0, sizeof(itnim->stats)); memset(&itnim->ioprofile, 0, sizeof(itnim->ioprofile)); for (j = 0; j < BFA_IOBUCKET_MAX; j++) diff --git a/drivers/scsi/bfa/bfa_fcs.c b/drivers/scsi/bfa/bfa_fcs.c index eaac57e..fd3e84d 100644 --- a/drivers/scsi/bfa/bfa_fcs.c +++ b/drivers/scsi/bfa/bfa_fcs.c @@ -76,6 +76,7 @@ bfa_fcs_attach(struct bfa_fcs_s *fcs, struct bfa_s *bfa, struct bfad_s *bfad, fcs->bfa = bfa; fcs->bfad = bfad; fcs->min_cfg = min_cfg; + fcs->num_rport_logins = 0; bfa->fcs = BFA_TRUE; fcbuild_init(); @@ -119,6 +120,18 @@ bfa_fcs_update_cfg(struct bfa_fcs_s *fcs) } /* + * Stop FCS operations. + */ +void +bfa_fcs_stop(struct bfa_fcs_s *fcs) +{ + bfa_wc_init(&fcs->wc, bfa_fcs_exit_comp, fcs); + bfa_wc_up(&fcs->wc); + bfa_fcs_fabric_modstop(fcs); + bfa_wc_wait(&fcs->wc); +} + +/* * fcs pbc vport initialization */ void @@ -153,6 +166,7 @@ bfa_fcs_driver_info_init(struct bfa_fcs_s *fcs, fcs->driver_info = *driver_info; bfa_fcs_fabric_psymb_init(&fcs->fabric); + bfa_fcs_fabric_nsymb_init(&fcs->fabric); } /* @@ -213,6 +227,8 @@ static void bfa_fcs_fabric_notify_offline(struct bfa_fcs_fabric_s *fabric); static void bfa_fcs_fabric_delay(void *cbarg); static void bfa_fcs_fabric_delete(struct bfa_fcs_fabric_s *fabric); static void bfa_fcs_fabric_delete_comp(void *cbarg); +static void bfa_fcs_fabric_stop(struct bfa_fcs_fabric_s *fabric); +static void bfa_fcs_fabric_stop_comp(void *cbarg); static void bfa_fcs_fabric_process_uf(struct bfa_fcs_fabric_s *fabric, struct fchs_s *fchs, u16 len); static void bfa_fcs_fabric_process_flogi(struct bfa_fcs_fabric_s *fabric, @@ -250,6 +266,10 @@ static void bfa_fcs_fabric_sm_isolated(struct bfa_fcs_fabric_s *fabric, enum bfa_fcs_fabric_event event); static void bfa_fcs_fabric_sm_deleting(struct bfa_fcs_fabric_s *fabric, enum bfa_fcs_fabric_event event); +static void bfa_fcs_fabric_sm_stopping(struct bfa_fcs_fabric_s *fabric, + enum bfa_fcs_fabric_event event); +static void bfa_fcs_fabric_sm_cleanup(struct bfa_fcs_fabric_s *fabric, + enum bfa_fcs_fabric_event event); /* * Beginning state before fabric creation. */ @@ -334,6 +354,11 @@ bfa_fcs_fabric_sm_linkdown(struct bfa_fcs_fabric_s *fabric, bfa_fcs_fabric_delete(fabric); break; + case BFA_FCS_FABRIC_SM_STOP: + bfa_sm_set_state(fabric, bfa_fcs_fabric_sm_cleanup); + bfa_fcs_fabric_stop(fabric); + break; + default: bfa_sm_fault(fabric->fcs, event); } @@ -585,6 +610,11 @@ bfa_fcs_fabric_sm_online(struct bfa_fcs_fabric_s *fabric, bfa_fcs_fabric_delete(fabric); break; + case BFA_FCS_FABRIC_SM_STOP: + bfa_sm_set_state(fabric, bfa_fcs_fabric_sm_stopping); + bfa_fcs_fabric_stop(fabric); + break; + case BFA_FCS_FABRIC_SM_AUTH_FAILED: bfa_sm_set_state(fabric, bfa_fcs_fabric_sm_auth_failed); bfa_sm_send_event(fabric->lps, BFA_LPS_SM_OFFLINE); @@ -682,7 +712,62 @@ bfa_fcs_fabric_sm_deleting(struct bfa_fcs_fabric_s *fabric, } } +/* + * Fabric is being stopped, awaiting vport stop completions. + */ +static void +bfa_fcs_fabric_sm_stopping(struct bfa_fcs_fabric_s *fabric, + enum bfa_fcs_fabric_event event) +{ + bfa_trc(fabric->fcs, fabric->bport.port_cfg.pwwn); + bfa_trc(fabric->fcs, event); + + switch (event) { + case BFA_FCS_FABRIC_SM_STOPCOMP: + bfa_sm_set_state(fabric, bfa_fcs_fabric_sm_cleanup); + bfa_sm_send_event(fabric->lps, BFA_LPS_SM_LOGOUT); + break; + + case BFA_FCS_FABRIC_SM_LINK_UP: + break; + + case BFA_FCS_FABRIC_SM_LINK_DOWN: + bfa_sm_set_state(fabric, bfa_fcs_fabric_sm_cleanup); + break; + + default: + bfa_sm_fault(fabric->fcs, event); + } +} + +/* + * Fabric is being stopped, cleanup without FLOGO + */ +static void +bfa_fcs_fabric_sm_cleanup(struct bfa_fcs_fabric_s *fabric, + enum bfa_fcs_fabric_event event) +{ + bfa_trc(fabric->fcs, fabric->bport.port_cfg.pwwn); + bfa_trc(fabric->fcs, event); + switch (event) { + case BFA_FCS_FABRIC_SM_STOPCOMP: + case BFA_FCS_FABRIC_SM_LOGOCOMP: + bfa_sm_set_state(fabric, bfa_fcs_fabric_sm_created); + bfa_wc_down(&(fabric->fcs)->wc); + break; + + case BFA_FCS_FABRIC_SM_LINK_DOWN: + /* + * Ignore - can get this event if we get notified about IOC down + * before the fabric completion callbk is done. + */ + break; + + default: + bfa_sm_fault(fabric->fcs, event); + } +} /* * fcs_fabric_private fabric private functions @@ -760,6 +845,44 @@ bfa_fcs_fabric_psymb_init(struct bfa_fcs_fabric_s *fabric) } /* + * Node Symbolic Name Creation for base port and all vports + */ +void +bfa_fcs_fabric_nsymb_init(struct bfa_fcs_fabric_s *fabric) +{ + struct bfa_lport_cfg_s *port_cfg = &fabric->bport.port_cfg; + char model[BFA_ADAPTER_MODEL_NAME_LEN] = {0}; + struct bfa_fcs_driver_info_s *driver_info = &fabric->fcs->driver_info; + + bfa_ioc_get_adapter_model(&fabric->fcs->bfa->ioc, model); + + /* Model name/number */ + strncpy((char *)&port_cfg->node_sym_name, model, + BFA_FCS_PORT_SYMBNAME_MODEL_SZ); + strncat((char *)&port_cfg->node_sym_name, + BFA_FCS_PORT_SYMBNAME_SEPARATOR, + sizeof(BFA_FCS_PORT_SYMBNAME_SEPARATOR)); + + /* Driver Version */ + strncat((char *)&port_cfg->node_sym_name, (char *)driver_info->version, + BFA_FCS_PORT_SYMBNAME_VERSION_SZ); + strncat((char *)&port_cfg->node_sym_name, + BFA_FCS_PORT_SYMBNAME_SEPARATOR, + sizeof(BFA_FCS_PORT_SYMBNAME_SEPARATOR)); + + /* Host machine name */ + strncat((char *)&port_cfg->node_sym_name, + (char *)driver_info->host_machine_name, + BFA_FCS_PORT_SYMBNAME_MACHINENAME_SZ); + strncat((char *)&port_cfg->node_sym_name, + BFA_FCS_PORT_SYMBNAME_SEPARATOR, + sizeof(BFA_FCS_PORT_SYMBNAME_SEPARATOR)); + + /* null terminate */ + port_cfg->node_sym_name.symname[BFA_SYMNAME_MAXLEN - 1] = 0; +} + +/* * bfa lps login completion callback */ void @@ -919,6 +1042,28 @@ bfa_fcs_fabric_delay(void *cbarg) } /* + * Stop all vports and wait for vport stop completions. + */ +static void +bfa_fcs_fabric_stop(struct bfa_fcs_fabric_s *fabric) +{ + struct bfa_fcs_vport_s *vport; + struct list_head *qe, *qen; + + bfa_wc_init(&fabric->stop_wc, bfa_fcs_fabric_stop_comp, fabric); + + list_for_each_safe(qe, qen, &fabric->vport_q) { + vport = (struct bfa_fcs_vport_s *) qe; + bfa_wc_up(&fabric->stop_wc); + bfa_fcs_vport_fcs_stop(vport); + } + + bfa_wc_up(&fabric->stop_wc); + bfa_fcs_lport_stop(&fabric->bport); + bfa_wc_wait(&fabric->stop_wc); +} + +/* * Computes operating BB_SCN value */ static u8 @@ -978,6 +1123,14 @@ bfa_fcs_fabric_delete_comp(void *cbarg) bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_DELCOMP); } +static void +bfa_fcs_fabric_stop_comp(void *cbarg) +{ + struct bfa_fcs_fabric_s *fabric = cbarg; + + bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_STOPCOMP); +} + /* * fcs_fabric_public fabric public functions */ @@ -1039,6 +1192,19 @@ bfa_fcs_fabric_modexit(struct bfa_fcs_s *fcs) } /* + * Fabric module stop -- stop FCS actions + */ +void +bfa_fcs_fabric_modstop(struct bfa_fcs_s *fcs) +{ + struct bfa_fcs_fabric_s *fabric; + + bfa_trc(fcs, 0); + fabric = &fcs->fabric; + bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_STOP); +} + +/* * Fabric module start -- kick starts FCS actions */ void @@ -1219,8 +1385,11 @@ bfa_fcs_fabric_uf_recv(struct bfa_fcs_fabric_s *fabric, struct fchs_s *fchs, return; } } - bfa_trc(fabric->fcs, els_cmd->els_code); - bfa_fcs_lport_uf_recv(&fabric->bport, fchs, len); + + if (!bfa_fcs_fabric_is_switched(fabric)) + bfa_fcs_lport_uf_recv(&fabric->bport, fchs, len); + + bfa_trc(fabric->fcs, fchs->type); } /* @@ -1294,7 +1463,7 @@ bfa_fcs_fabric_send_flogi_acc(struct bfa_fcs_fabric_s *fabric) u16 reqlen; struct fchs_s fchs; - fcxp = bfa_fcs_fcxp_alloc(fabric->fcs); + fcxp = bfa_fcs_fcxp_alloc(fabric->fcs, BFA_FALSE); /* * Do not expect this failure -- expect remote node to retry */ @@ -1387,6 +1556,13 @@ bfa_fcs_fabric_set_fabric_name(struct bfa_fcs_fabric_s *fabric, } } +void +bfa_cb_lps_flogo_comp(void *bfad, void *uarg) +{ + struct bfa_fcs_fabric_s *fabric = uarg; + bfa_sm_send_event(fabric, BFA_FCS_FABRIC_SM_LOGOCOMP); +} + /* * Returns FCS vf structure for a given vf_id. * diff --git a/drivers/scsi/bfa/bfa_fcs.h b/drivers/scsi/bfa/bfa_fcs.h index 51c9e13..6c4377c 100644 --- a/drivers/scsi/bfa/bfa_fcs.h +++ b/drivers/scsi/bfa/bfa_fcs.h @@ -62,9 +62,9 @@ struct bfa_fcs_s; #define N2N_LOCAL_PID 0x010000 #define N2N_REMOTE_PID 0x020000 #define BFA_FCS_RETRY_TIMEOUT 2000 +#define BFA_FCS_MAX_NS_RETRIES 5 #define BFA_FCS_PID_IS_WKA(pid) ((bfa_ntoh3b(pid) > 0xFFF000) ? 1 : 0) - - +#define BFA_FCS_MAX_RPORT_LOGINS 1024 struct bfa_fcs_lport_ns_s { bfa_sm_t sm; /* state machine */ @@ -72,6 +72,8 @@ struct bfa_fcs_lport_ns_s { struct bfa_fcs_lport_s *port; /* parent port */ struct bfa_fcxp_s *fcxp; struct bfa_fcxp_wqe_s fcxp_wqe; + u8 num_rnnid_retries; + u8 num_rsnn_nn_retries; }; @@ -205,6 +207,7 @@ struct bfa_fcs_fabric_s { struct bfa_lps_s *lps; /* lport login services */ u8 fabric_ip_addr[BFA_FCS_FABRIC_IPADDR_SZ]; /* attached fabric's ip addr */ + struct bfa_wc_s stop_wc; /* wait counter for stop */ }; #define bfa_fcs_fabric_npiv_capable(__f) ((__f)->is_npiv) @@ -264,6 +267,7 @@ struct bfa_fcs_fabric_s; #define bfa_fcs_lport_get_pwwn(_lport) ((_lport)->port_cfg.pwwn) #define bfa_fcs_lport_get_nwwn(_lport) ((_lport)->port_cfg.nwwn) #define bfa_fcs_lport_get_psym_name(_lport) ((_lport)->port_cfg.sym_name) +#define bfa_fcs_lport_get_nsym_name(_lport) ((_lport)->port_cfg.node_sym_name) #define bfa_fcs_lport_is_initiator(_lport) \ ((_lport)->port_cfg.roles & BFA_LPORT_ROLE_FCP_IM) #define bfa_fcs_lport_get_nrports(_lport) \ @@ -286,9 +290,8 @@ bfa_fcs_lport_get_drvport(struct bfa_fcs_lport_s *port) bfa_boolean_t bfa_fcs_lport_is_online(struct bfa_fcs_lport_s *port); struct bfa_fcs_lport_s *bfa_fcs_get_base_port(struct bfa_fcs_s *fcs); -void bfa_fcs_lport_get_rports(struct bfa_fcs_lport_s *port, - wwn_t rport_wwns[], int *nrports); - +void bfa_fcs_lport_get_rport_quals(struct bfa_fcs_lport_s *port, + struct bfa_rport_qualifier_s rport[], int *nrports); wwn_t bfa_fcs_lport_get_rport(struct bfa_fcs_lport_s *port, wwn_t wwn, int index, int nrports, bfa_boolean_t bwwn); @@ -324,12 +327,17 @@ void bfa_fcs_lport_init(struct bfa_fcs_lport_s *lport, void bfa_fcs_lport_online(struct bfa_fcs_lport_s *port); void bfa_fcs_lport_offline(struct bfa_fcs_lport_s *port); void bfa_fcs_lport_delete(struct bfa_fcs_lport_s *port); +void bfa_fcs_lport_stop(struct bfa_fcs_lport_s *port); struct bfa_fcs_rport_s *bfa_fcs_lport_get_rport_by_pid( struct bfa_fcs_lport_s *port, u32 pid); +struct bfa_fcs_rport_s *bfa_fcs_lport_get_rport_by_old_pid( + struct bfa_fcs_lport_s *port, u32 pid); struct bfa_fcs_rport_s *bfa_fcs_lport_get_rport_by_pwwn( struct bfa_fcs_lport_s *port, wwn_t pwwn); struct bfa_fcs_rport_s *bfa_fcs_lport_get_rport_by_nwwn( struct bfa_fcs_lport_s *port, wwn_t nwwn); +struct bfa_fcs_rport_s *bfa_fcs_lport_get_rport_by_qualifier( + struct bfa_fcs_lport_s *port, wwn_t pwwn, u32 pid); void bfa_fcs_lport_add_rport(struct bfa_fcs_lport_s *port, struct bfa_fcs_rport_s *rport); void bfa_fcs_lport_del_rport(struct bfa_fcs_lport_s *port, @@ -338,6 +346,8 @@ void bfa_fcs_lport_ns_init(struct bfa_fcs_lport_s *vport); void bfa_fcs_lport_ns_offline(struct bfa_fcs_lport_s *vport); void bfa_fcs_lport_ns_online(struct bfa_fcs_lport_s *vport); void bfa_fcs_lport_ns_query(struct bfa_fcs_lport_s *port); +void bfa_fcs_lport_ns_util_send_rspn_id(void *cbarg, + struct bfa_fcxp_s *fcxp_alloced); void bfa_fcs_lport_scn_init(struct bfa_fcs_lport_s *vport); void bfa_fcs_lport_scn_offline(struct bfa_fcs_lport_s *vport); void bfa_fcs_lport_scn_online(struct bfa_fcs_lport_s *vport); @@ -382,6 +392,7 @@ void bfa_fcs_vport_online(struct bfa_fcs_vport_s *vport); void bfa_fcs_vport_offline(struct bfa_fcs_vport_s *vport); void bfa_fcs_vport_delete_comp(struct bfa_fcs_vport_s *vport); void bfa_fcs_vport_fcs_delete(struct bfa_fcs_vport_s *vport); +void bfa_fcs_vport_fcs_stop(struct bfa_fcs_vport_s *vport); void bfa_fcs_vport_stop_comp(struct bfa_fcs_vport_s *vport); #define BFA_FCS_RPORT_DEF_DEL_TIMEOUT 90 /* in secs */ @@ -419,6 +430,7 @@ struct bfa_fcs_rport_s { struct bfa_fcs_s *fcs; /* fcs instance */ struct bfad_rport_s *rp_drv; /* driver peer instance */ u32 pid; /* port ID of rport */ + u32 old_pid; /* PID before rport goes offline */ u16 maxfrsize; /* maximum frame size */ __be16 reply_oxid; /* OX_ID of inbound requests */ enum fc_cos fc_cos; /* FC classes of service supp */ @@ -459,7 +471,7 @@ struct bfa_fcs_rport_s *bfa_fcs_rport_lookup(struct bfa_fcs_lport_s *port, struct bfa_fcs_rport_s *bfa_fcs_rport_lookup_by_nwwn( struct bfa_fcs_lport_s *port, wwn_t rnwwn); void bfa_fcs_rport_set_del_timeout(u8 rport_tmo); - +void bfa_fcs_rport_set_max_logins(u32 max_logins); void bfa_fcs_rport_uf_recv(struct bfa_fcs_rport_s *rport, struct fchs_s *fchs, u16 len); void bfa_fcs_rport_scn(struct bfa_fcs_rport_s *rport); @@ -505,12 +517,13 @@ struct bfa_fcs_itnim_s { struct bfa_fcxp_s *fcxp; /* FCXP in use */ struct bfa_itnim_stats_s stats; /* itn statistics */ }; -#define bfa_fcs_fcxp_alloc(__fcs) \ - bfa_fcxp_alloc(NULL, (__fcs)->bfa, 0, 0, NULL, NULL, NULL, NULL) - -#define bfa_fcs_fcxp_alloc_wait(__bfa, __wqe, __alloc_cbfn, __alloc_cbarg) \ - bfa_fcxp_alloc_wait(__bfa, __wqe, __alloc_cbfn, __alloc_cbarg, \ - NULL, 0, 0, NULL, NULL, NULL, NULL) +#define bfa_fcs_fcxp_alloc(__fcs, __req) \ + bfa_fcxp_req_rsp_alloc(NULL, (__fcs)->bfa, 0, 0, \ + NULL, NULL, NULL, NULL, __req) +#define bfa_fcs_fcxp_alloc_wait(__bfa, __wqe, __alloc_cbfn, \ + __alloc_cbarg, __req) \ + bfa_fcxp_req_rsp_alloc_wait(__bfa, __wqe, __alloc_cbfn, \ + __alloc_cbarg, NULL, 0, 0, NULL, NULL, NULL, NULL, __req) static inline struct bfad_port_s * bfa_fcs_itnim_get_drvport(struct bfa_fcs_itnim_s *itnim) @@ -592,7 +605,7 @@ bfa_status_t bfa_fcs_itnim_stats_clear(struct bfa_fcs_lport_s *port, struct bfa_fcs_itnim_s *bfa_fcs_itnim_create(struct bfa_fcs_rport_s *rport); void bfa_fcs_itnim_delete(struct bfa_fcs_itnim_s *itnim); void bfa_fcs_itnim_rport_offline(struct bfa_fcs_itnim_s *itnim); -void bfa_fcs_itnim_rport_online(struct bfa_fcs_itnim_s *itnim); +void bfa_fcs_itnim_brp_online(struct bfa_fcs_itnim_s *itnim); bfa_status_t bfa_fcs_itnim_get_online_state(struct bfa_fcs_itnim_s *itnim); void bfa_fcs_itnim_is_initiator(struct bfa_fcs_itnim_s *itnim); void bfa_fcs_fcpim_uf_recv(struct bfa_fcs_itnim_s *itnim, @@ -676,6 +689,7 @@ struct bfa_fcs_s { struct bfa_fcs_stats_s stats; /* FCS statistics */ struct bfa_wc_s wc; /* waiting counter */ int fcs_aen_seq; + u32 num_rport_logins; }; /* @@ -702,6 +716,9 @@ enum bfa_fcs_fabric_event { BFA_FCS_FABRIC_SM_DELCOMP = 14, /* all vports deleted event */ BFA_FCS_FABRIC_SM_LOOPBACK = 15, /* Received our own FLOGI */ BFA_FCS_FABRIC_SM_START = 16, /* from driver */ + BFA_FCS_FABRIC_SM_STOP = 17, /* Stop from driver */ + BFA_FCS_FABRIC_SM_STOPCOMP = 18, /* Stop completion */ + BFA_FCS_FABRIC_SM_LOGOCOMP = 19, /* FLOGO completion */ }; /* @@ -727,6 +744,26 @@ enum rport_event { RPSM_EVENT_ADDRESS_DISC = 16, /* Need to Discover rport's PID */ RPSM_EVENT_PRLO_RCVD = 17, /* PRLO from remote device */ RPSM_EVENT_PLOGI_RETRY = 18, /* Retry PLOGI continuously */ + RPSM_EVENT_FC4_FCS_ONLINE = 19, /*!< FC-4 FCS online complete */ +}; + +/* + * fcs_itnim_sm FCS itnim state machine events + */ +enum bfa_fcs_itnim_event { + BFA_FCS_ITNIM_SM_FCS_ONLINE = 1, /* rport online event */ + BFA_FCS_ITNIM_SM_OFFLINE = 2, /* rport offline */ + BFA_FCS_ITNIM_SM_FRMSENT = 3, /* prli frame is sent */ + BFA_FCS_ITNIM_SM_RSP_OK = 4, /* good response */ + BFA_FCS_ITNIM_SM_RSP_ERROR = 5, /* error response */ + BFA_FCS_ITNIM_SM_TIMEOUT = 6, /* delay timeout */ + BFA_FCS_ITNIM_SM_HCB_OFFLINE = 7, /* BFA online callback */ + BFA_FCS_ITNIM_SM_HCB_ONLINE = 8, /* BFA offline callback */ + BFA_FCS_ITNIM_SM_INITIATOR = 9, /* rport is initiator */ + BFA_FCS_ITNIM_SM_DELETE = 10, /* delete event from rport */ + BFA_FCS_ITNIM_SM_PRLO = 11, /* delete event from rport */ + BFA_FCS_ITNIM_SM_RSP_NOT_SUPP = 12, /* cmd not supported rsp */ + BFA_FCS_ITNIM_SM_HAL_ONLINE = 13, /*!< bfa rport online event */ }; /* @@ -741,6 +778,7 @@ void bfa_fcs_update_cfg(struct bfa_fcs_s *fcs); void bfa_fcs_driver_info_init(struct bfa_fcs_s *fcs, struct bfa_fcs_driver_info_s *driver_info); void bfa_fcs_exit(struct bfa_fcs_s *fcs); +void bfa_fcs_stop(struct bfa_fcs_s *fcs); /* * bfa fcs vf public functions @@ -766,11 +804,13 @@ void bfa_fcs_fabric_modstart(struct bfa_fcs_s *fcs); void bfa_fcs_fabric_uf_recv(struct bfa_fcs_fabric_s *fabric, struct fchs_s *fchs, u16 len); void bfa_fcs_fabric_psymb_init(struct bfa_fcs_fabric_s *fabric); +void bfa_fcs_fabric_nsymb_init(struct bfa_fcs_fabric_s *fabric); void bfa_fcs_fabric_set_fabric_name(struct bfa_fcs_fabric_s *fabric, wwn_t fabric_name); u16 bfa_fcs_fabric_get_switch_oui(struct bfa_fcs_fabric_s *fabric); void bfa_fcs_uf_attach(struct bfa_fcs_s *fcs); void bfa_fcs_port_attach(struct bfa_fcs_s *fcs); +void bfa_fcs_fabric_modstop(struct bfa_fcs_s *fcs); void bfa_fcs_fabric_sm_online(struct bfa_fcs_fabric_s *fabric, enum bfa_fcs_fabric_event event); void bfa_fcs_fabric_sm_loopback(struct bfa_fcs_fabric_s *fabric, diff --git a/drivers/scsi/bfa/bfa_fcs_fcpim.c b/drivers/scsi/bfa/bfa_fcs_fcpim.c index 9272840..6dc7926 100644 --- a/drivers/scsi/bfa/bfa_fcs_fcpim.c +++ b/drivers/scsi/bfa/bfa_fcs_fcpim.c @@ -40,25 +40,6 @@ static void bfa_fcs_itnim_prli_response(void *fcsarg, static void bfa_fcs_itnim_aen_post(struct bfa_fcs_itnim_s *itnim, enum bfa_itnim_aen_event event); -/* - * fcs_itnim_sm FCS itnim state machine events - */ - -enum bfa_fcs_itnim_event { - BFA_FCS_ITNIM_SM_ONLINE = 1, /* rport online event */ - BFA_FCS_ITNIM_SM_OFFLINE = 2, /* rport offline */ - BFA_FCS_ITNIM_SM_FRMSENT = 3, /* prli frame is sent */ - BFA_FCS_ITNIM_SM_RSP_OK = 4, /* good response */ - BFA_FCS_ITNIM_SM_RSP_ERROR = 5, /* error response */ - BFA_FCS_ITNIM_SM_TIMEOUT = 6, /* delay timeout */ - BFA_FCS_ITNIM_SM_HCB_OFFLINE = 7, /* BFA online callback */ - BFA_FCS_ITNIM_SM_HCB_ONLINE = 8, /* BFA offline callback */ - BFA_FCS_ITNIM_SM_INITIATOR = 9, /* rport is initiator */ - BFA_FCS_ITNIM_SM_DELETE = 10, /* delete event from rport */ - BFA_FCS_ITNIM_SM_PRLO = 11, /* delete event from rport */ - BFA_FCS_ITNIM_SM_RSP_NOT_SUPP = 12, /* cmd not supported rsp */ -}; - static void bfa_fcs_itnim_sm_offline(struct bfa_fcs_itnim_s *itnim, enum bfa_fcs_itnim_event event); static void bfa_fcs_itnim_sm_prli_send(struct bfa_fcs_itnim_s *itnim, @@ -69,6 +50,8 @@ static void bfa_fcs_itnim_sm_prli_retry(struct bfa_fcs_itnim_s *itnim, enum bfa_fcs_itnim_event event); static void bfa_fcs_itnim_sm_hcb_online(struct bfa_fcs_itnim_s *itnim, enum bfa_fcs_itnim_event event); +static void bfa_fcs_itnim_sm_hal_rport_online(struct bfa_fcs_itnim_s *itnim, + enum bfa_fcs_itnim_event event); static void bfa_fcs_itnim_sm_online(struct bfa_fcs_itnim_s *itnim, enum bfa_fcs_itnim_event event); static void bfa_fcs_itnim_sm_hcb_offline(struct bfa_fcs_itnim_s *itnim, @@ -99,7 +82,7 @@ bfa_fcs_itnim_sm_offline(struct bfa_fcs_itnim_s *itnim, bfa_trc(itnim->fcs, event); switch (event) { - case BFA_FCS_ITNIM_SM_ONLINE: + case BFA_FCS_ITNIM_SM_FCS_ONLINE: bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_prli_send); itnim->prli_retries = 0; bfa_fcs_itnim_send_prli(itnim, NULL); @@ -138,6 +121,7 @@ bfa_fcs_itnim_sm_prli_send(struct bfa_fcs_itnim_s *itnim, case BFA_FCS_ITNIM_SM_INITIATOR: bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_initiator); bfa_fcxp_walloc_cancel(itnim->fcs->bfa, &itnim->fcxp_wqe); + bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_FCS_ONLINE); break; case BFA_FCS_ITNIM_SM_OFFLINE: @@ -166,12 +150,13 @@ bfa_fcs_itnim_sm_prli(struct bfa_fcs_itnim_s *itnim, switch (event) { case BFA_FCS_ITNIM_SM_RSP_OK: - if (itnim->rport->scsi_function == BFA_RPORT_INITIATOR) { + if (itnim->rport->scsi_function == BFA_RPORT_INITIATOR) bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_initiator); - } else { - bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_hcb_online); - bfa_itnim_online(itnim->bfa_itnim, itnim->seq_rec); - } + else + bfa_sm_set_state(itnim, + bfa_fcs_itnim_sm_hal_rport_online); + + bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_FCS_ONLINE); break; case BFA_FCS_ITNIM_SM_RSP_ERROR: @@ -194,6 +179,7 @@ bfa_fcs_itnim_sm_prli(struct bfa_fcs_itnim_s *itnim, case BFA_FCS_ITNIM_SM_INITIATOR: bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_initiator); bfa_fcxp_discard(itnim->fcxp); + bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_FCS_ONLINE); break; case BFA_FCS_ITNIM_SM_DELETE: @@ -208,6 +194,44 @@ bfa_fcs_itnim_sm_prli(struct bfa_fcs_itnim_s *itnim, } static void +bfa_fcs_itnim_sm_hal_rport_online(struct bfa_fcs_itnim_s *itnim, + enum bfa_fcs_itnim_event event) +{ + bfa_trc(itnim->fcs, itnim->rport->pwwn); + bfa_trc(itnim->fcs, event); + + switch (event) { + case BFA_FCS_ITNIM_SM_HAL_ONLINE: + if (!itnim->bfa_itnim) + itnim->bfa_itnim = bfa_itnim_create(itnim->fcs->bfa, + itnim->rport->bfa_rport, itnim); + + if (itnim->bfa_itnim) { + bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_hcb_online); + bfa_itnim_online(itnim->bfa_itnim, itnim->seq_rec); + } else { + bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_offline); + bfa_sm_send_event(itnim->rport, RPSM_EVENT_DELETE); + } + + break; + + case BFA_FCS_ITNIM_SM_OFFLINE: + bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_offline); + bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_OFFLINE); + break; + + case BFA_FCS_ITNIM_SM_DELETE: + bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_offline); + bfa_fcs_itnim_free(itnim); + break; + + default: + bfa_sm_fault(itnim->fcs, event); + } +} + +static void bfa_fcs_itnim_sm_prli_retry(struct bfa_fcs_itnim_s *itnim, enum bfa_fcs_itnim_event event) { @@ -238,6 +262,7 @@ bfa_fcs_itnim_sm_prli_retry(struct bfa_fcs_itnim_s *itnim, case BFA_FCS_ITNIM_SM_INITIATOR: bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_initiator); bfa_timer_stop(&itnim->timer); + bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_FCS_ONLINE); break; case BFA_FCS_ITNIM_SM_DELETE: @@ -275,9 +300,8 @@ bfa_fcs_itnim_sm_hcb_online(struct bfa_fcs_itnim_s *itnim, break; case BFA_FCS_ITNIM_SM_OFFLINE: - bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_offline); + bfa_sm_set_state(itnim, bfa_fcs_itnim_sm_hcb_offline); bfa_itnim_offline(itnim->bfa_itnim); - bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_OFFLINE); break; case BFA_FCS_ITNIM_SM_DELETE: @@ -372,8 +396,14 @@ bfa_fcs_itnim_sm_initiator(struct bfa_fcs_itnim_s *itnim, bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_OFFLINE); break; + /* + * fcs_online is expected here for well known initiator ports + */ + case BFA_FCS_ITNIM_SM_FCS_ONLINE: + bfa_sm_send_event(itnim->rport, RPSM_EVENT_FC4_FCS_ONLINE); + break; + case BFA_FCS_ITNIM_SM_RSP_ERROR: - case BFA_FCS_ITNIM_SM_ONLINE: case BFA_FCS_ITNIM_SM_INITIATOR: break; @@ -426,11 +456,12 @@ bfa_fcs_itnim_send_prli(void *itnim_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(itnim->fcs, itnim->rport->pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { itnim->stats.fcxp_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &itnim->fcxp_wqe, - bfa_fcs_itnim_send_prli, itnim); + bfa_fcs_itnim_send_prli, itnim, BFA_TRUE); return; } itnim->fcxp = fcxp; @@ -483,7 +514,7 @@ bfa_fcs_itnim_prli_response(void *fcsarg, struct bfa_fcxp_s *fcxp, void *cbarg, if (prli_resp->parampage.servparams.initiator) { bfa_trc(itnim->fcs, prli_resp->parampage.type); itnim->rport->scsi_function = - BFA_RPORT_INITIATOR; + BFA_RPORT_INITIATOR; itnim->stats.prli_rsp_acc++; itnim->stats.initiator++; bfa_sm_send_event(itnim, @@ -531,7 +562,11 @@ bfa_fcs_itnim_timeout(void *arg) static void bfa_fcs_itnim_free(struct bfa_fcs_itnim_s *itnim) { - bfa_itnim_delete(itnim->bfa_itnim); + if (itnim->bfa_itnim) { + bfa_itnim_delete(itnim->bfa_itnim); + itnim->bfa_itnim = NULL; + } + bfa_fcb_itnim_free(itnim->fcs->bfad, itnim->itnim_drv); } @@ -552,7 +587,6 @@ bfa_fcs_itnim_create(struct bfa_fcs_rport_s *rport) struct bfa_fcs_lport_s *port = rport->port; struct bfa_fcs_itnim_s *itnim; struct bfad_itnim_s *itnim_drv; - struct bfa_itnim_s *bfa_itnim; /* * call bfad to allocate the itnim @@ -570,20 +604,7 @@ bfa_fcs_itnim_create(struct bfa_fcs_rport_s *rport) itnim->fcs = rport->fcs; itnim->itnim_drv = itnim_drv; - /* - * call BFA to create the itnim - */ - bfa_itnim = - bfa_itnim_create(port->fcs->bfa, rport->bfa_rport, itnim); - - if (bfa_itnim == NULL) { - bfa_trc(port->fcs, rport->pwwn); - bfa_fcb_itnim_free(port->fcs->bfad, itnim_drv); - WARN_ON(1); - return NULL; - } - - itnim->bfa_itnim = bfa_itnim; + itnim->bfa_itnim = NULL; itnim->seq_rec = BFA_FALSE; itnim->rec_support = BFA_FALSE; itnim->conf_comp = BFA_FALSE; @@ -613,20 +634,12 @@ bfa_fcs_itnim_delete(struct bfa_fcs_itnim_s *itnim) * Notification from rport that PLOGI is complete to initiate FC-4 session. */ void -bfa_fcs_itnim_rport_online(struct bfa_fcs_itnim_s *itnim) +bfa_fcs_itnim_brp_online(struct bfa_fcs_itnim_s *itnim) { itnim->stats.onlines++; - if (!BFA_FCS_PID_IS_WKA(itnim->rport->pid)) { - bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_ONLINE); - } else { - /* - * For well known addresses, we set the itnim to initiator - * state - */ - itnim->stats.initiator++; - bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_INITIATOR); - } + if (!BFA_FCS_PID_IS_WKA(itnim->rport->pid)) + bfa_sm_send_event(itnim, BFA_FCS_ITNIM_SM_HAL_ONLINE); } /* diff --git a/drivers/scsi/bfa/bfa_fcs_lport.c b/drivers/scsi/bfa/bfa_fcs_lport.c index bcc4966..3b75f6f 100644 --- a/drivers/scsi/bfa/bfa_fcs_lport.c +++ b/drivers/scsi/bfa/bfa_fcs_lport.c @@ -131,6 +131,8 @@ bfa_fcs_lport_sm_init(struct bfa_fcs_lport_s *port, /* If vport - send completion call back */ if (port->vport) bfa_fcs_vport_stop_comp(port->vport); + else + bfa_wc_down(&(port->fabric->stop_wc)); break; case BFA_FCS_PORT_SM_OFFLINE: @@ -166,6 +168,8 @@ bfa_fcs_lport_sm_online( /* If vport - send completion call back */ if (port->vport) bfa_fcs_vport_stop_comp(port->vport); + else + bfa_wc_down(&(port->fabric->stop_wc)); } else { bfa_sm_set_state(port, bfa_fcs_lport_sm_stopping); list_for_each_safe(qe, qen, &port->rport_q) { @@ -222,6 +226,8 @@ bfa_fcs_lport_sm_offline( /* If vport - send completion call back */ if (port->vport) bfa_fcs_vport_stop_comp(port->vport); + else + bfa_wc_down(&(port->fabric->stop_wc)); } else { bfa_sm_set_state(port, bfa_fcs_lport_sm_stopping); list_for_each_safe(qe, qen, &port->rport_q) { @@ -267,6 +273,8 @@ bfa_fcs_lport_sm_stopping(struct bfa_fcs_lport_s *port, /* If vport - send completion call back */ if (port->vport) bfa_fcs_vport_stop_comp(port->vport); + else + bfa_wc_down(&(port->fabric->stop_wc)); } break; @@ -340,7 +348,7 @@ bfa_fcs_lport_send_ls_rjt(struct bfa_fcs_lport_s *port, struct fchs_s *rx_fchs, bfa_trc(port->fcs, rx_fchs->d_id); bfa_trc(port->fcs, rx_fchs->s_id); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -370,7 +378,7 @@ bfa_fcs_lport_send_fcgs_rjt(struct bfa_fcs_lport_s *port, bfa_trc(port->fcs, rx_fchs->d_id); bfa_trc(port->fcs, rx_fchs->s_id); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -507,7 +515,7 @@ bfa_fcs_lport_echo(struct bfa_fcs_lport_s *port, struct fchs_s *rx_fchs, bfa_trc(port->fcs, rx_fchs->s_id); bfa_trc(port->fcs, rx_fchs->d_id); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -552,7 +560,7 @@ bfa_fcs_lport_rnid(struct bfa_fcs_lport_s *port, struct fchs_s *rx_fchs, bfa_trc(port->fcs, rx_fchs->d_id); bfa_trc(port->fcs, rx_len); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -684,7 +692,7 @@ bfa_fcs_lport_abts_acc(struct bfa_fcs_lport_s *port, struct fchs_s *rx_fchs) bfa_trc(port->fcs, rx_fchs->d_id); bfa_trc(port->fcs, rx_fchs->s_id); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -854,6 +862,25 @@ bfa_fcs_lport_get_rport_by_pid(struct bfa_fcs_lport_s *port, u32 pid) } /* + * OLD_PID based Lookup for a R-Port in the Port R-Port Queue + */ +struct bfa_fcs_rport_s * +bfa_fcs_lport_get_rport_by_old_pid(struct bfa_fcs_lport_s *port, u32 pid) +{ + struct bfa_fcs_rport_s *rport; + struct list_head *qe; + + list_for_each(qe, &port->rport_q) { + rport = (struct bfa_fcs_rport_s *) qe; + if (rport->old_pid == pid) + return rport; + } + + bfa_trc(port->fcs, pid); + return NULL; +} + +/* * PWWN based Lookup for a R-Port in the Port R-Port Queue */ struct bfa_fcs_rport_s * @@ -892,6 +919,26 @@ bfa_fcs_lport_get_rport_by_nwwn(struct bfa_fcs_lport_s *port, wwn_t nwwn) } /* + * PWWN & PID based Lookup for a R-Port in the Port R-Port Queue + */ +struct bfa_fcs_rport_s * +bfa_fcs_lport_get_rport_by_qualifier(struct bfa_fcs_lport_s *port, + wwn_t pwwn, u32 pid) +{ + struct bfa_fcs_rport_s *rport; + struct list_head *qe; + + list_for_each(qe, &port->rport_q) { + rport = (struct bfa_fcs_rport_s *) qe; + if (wwn_is_equal(rport->pwwn, pwwn) && rport->pid == pid) + return rport; + } + + bfa_trc(port->fcs, pwwn); + return NULL; +} + +/* * Called by rport module when new rports are discovered. */ void @@ -939,6 +986,16 @@ bfa_fcs_lport_offline(struct bfa_fcs_lport_s *port) } /* + * Called by fabric for base port and by vport for virtual ports + * when target mode driver is unloaded. + */ +void +bfa_fcs_lport_stop(struct bfa_fcs_lport_s *port) +{ + bfa_sm_send_event(port, BFA_FCS_PORT_SM_STOP); +} + +/* * Called by fabric to delete base lport and associated resources. * * Called by vport to delete lport and associated resources. Should call @@ -1657,10 +1714,11 @@ bfa_fcs_lport_fdmi_send_rhba(void *fdmi_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &fdmi->fcxp_wqe, - bfa_fcs_lport_fdmi_send_rhba, fdmi); + bfa_fcs_lport_fdmi_send_rhba, fdmi, BFA_TRUE); return; } fdmi->fcxp = fcxp; @@ -1931,10 +1989,11 @@ bfa_fcs_lport_fdmi_send_rprt(void *fdmi_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &fdmi->fcxp_wqe, - bfa_fcs_lport_fdmi_send_rprt, fdmi); + bfa_fcs_lport_fdmi_send_rprt, fdmi, BFA_TRUE); return; } fdmi->fcxp = fcxp; @@ -2146,10 +2205,11 @@ bfa_fcs_lport_fdmi_send_rpa(void *fdmi_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &fdmi->fcxp_wqe, - bfa_fcs_lport_fdmi_send_rpa, fdmi); + bfa_fcs_lport_fdmi_send_rpa, fdmi, BFA_TRUE); return; } fdmi->fcxp = fcxp; @@ -2736,10 +2796,11 @@ bfa_fcs_lport_ms_send_gmal(void *ms_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->pid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ms->fcxp_wqe, - bfa_fcs_lport_ms_send_gmal, ms); + bfa_fcs_lport_ms_send_gmal, ms, BFA_TRUE); return; } ms->fcxp = fcxp; @@ -2936,10 +2997,11 @@ bfa_fcs_lport_ms_send_gfn(void *ms_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->pid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ms->fcxp_wqe, - bfa_fcs_lport_ms_send_gfn, ms); + bfa_fcs_lport_ms_send_gfn, ms, BFA_TRUE); return; } ms->fcxp = fcxp; @@ -3012,11 +3074,12 @@ bfa_fcs_lport_ms_send_plogi(void *ms_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->pid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { port->stats.ms_plogi_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ms->fcxp_wqe, - bfa_fcs_lport_ms_send_plogi, ms); + bfa_fcs_lport_ms_send_plogi, ms, BFA_TRUE); return; } ms->fcxp = fcxp; @@ -3166,6 +3229,10 @@ static void bfa_fcs_lport_ns_send_rff_id(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced); static void bfa_fcs_lport_ns_send_gid_ft(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced); +static void bfa_fcs_lport_ns_send_rnn_id(void *ns_cbarg, + struct bfa_fcxp_s *fcxp_alloced); +static void bfa_fcs_lport_ns_send_rsnn_nn(void *ns_cbarg, + struct bfa_fcxp_s *fcxp_alloced); static void bfa_fcs_lport_ns_timeout(void *arg); static void bfa_fcs_lport_ns_plogi_response(void *fcsarg, struct bfa_fcxp_s *fcxp, @@ -3202,6 +3269,20 @@ static void bfa_fcs_lport_ns_gid_ft_response(void *fcsarg, u32 rsp_len, u32 resid_len, struct fchs_s *rsp_fchs); +static void bfa_fcs_lport_ns_rnn_id_response(void *fcsarg, + struct bfa_fcxp_s *fcxp, + void *cbarg, + bfa_status_t req_status, + u32 rsp_len, + u32 resid_len, + struct fchs_s *rsp_fchs); +static void bfa_fcs_lport_ns_rsnn_nn_response(void *fcsarg, + struct bfa_fcxp_s *fcxp, + void *cbarg, + bfa_status_t req_status, + u32 rsp_len, + u32 resid_len, + struct fchs_s *rsp_fchs); static void bfa_fcs_lport_ns_process_gidft_pids( struct bfa_fcs_lport_s *port, u32 *pid_buf, u32 n_pids); @@ -3226,6 +3307,8 @@ enum vport_ns_event { NSSM_EVENT_RFTID_SENT = 9, NSSM_EVENT_RFFID_SENT = 10, NSSM_EVENT_GIDFT_SENT = 11, + NSSM_EVENT_RNNID_SENT = 12, + NSSM_EVENT_RSNN_NN_SENT = 13, }; static void bfa_fcs_lport_ns_sm_offline(struct bfa_fcs_lport_ns_s *ns, @@ -3266,6 +3349,21 @@ static void bfa_fcs_lport_ns_sm_gid_ft_retry(struct bfa_fcs_lport_ns_s *ns, enum vport_ns_event event); static void bfa_fcs_lport_ns_sm_online(struct bfa_fcs_lport_ns_s *ns, enum vport_ns_event event); +static void bfa_fcs_lport_ns_sm_sending_rnn_id( + struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event); +static void bfa_fcs_lport_ns_sm_rnn_id(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event); +static void bfa_fcs_lport_ns_sm_rnn_id_retry(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event); +static void bfa_fcs_lport_ns_sm_sending_rsnn_nn( + struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event); +static void bfa_fcs_lport_ns_sm_rsnn_nn(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event); +static void bfa_fcs_lport_ns_sm_rsnn_nn_retry( + struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event); /* * Start in offline state - awaiting linkup */ @@ -3333,8 +3431,9 @@ bfa_fcs_lport_ns_sm_plogi(struct bfa_fcs_lport_ns_s *ns, break; case NSSM_EVENT_RSP_OK: - bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_sending_rspn_id); - bfa_fcs_lport_ns_send_rspn_id(ns, NULL); + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_sending_rnn_id); + ns->num_rnnid_retries = 0; + bfa_fcs_lport_ns_send_rnn_id(ns, NULL); break; case NSSM_EVENT_PORT_OFFLINE: @@ -3374,6 +3473,176 @@ bfa_fcs_lport_ns_sm_plogi_retry(struct bfa_fcs_lport_ns_s *ns, } static void +bfa_fcs_lport_ns_sm_sending_rnn_id(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event) +{ + bfa_trc(ns->port->fcs, ns->port->port_cfg.pwwn); + bfa_trc(ns->port->fcs, event); + + switch (event) { + case NSSM_EVENT_RNNID_SENT: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_rnn_id); + break; + + case NSSM_EVENT_PORT_OFFLINE: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_offline); + bfa_fcxp_walloc_cancel(BFA_FCS_GET_HAL_FROM_PORT(ns->port), + &ns->fcxp_wqe); + break; + default: + bfa_sm_fault(ns->port->fcs, event); + } +} + +static void +bfa_fcs_lport_ns_sm_rnn_id(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event) +{ + bfa_trc(ns->port->fcs, ns->port->port_cfg.pwwn); + bfa_trc(ns->port->fcs, event); + + switch (event) { + case NSSM_EVENT_RSP_OK: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_sending_rsnn_nn); + ns->num_rnnid_retries = 0; + ns->num_rsnn_nn_retries = 0; + bfa_fcs_lport_ns_send_rsnn_nn(ns, NULL); + break; + + case NSSM_EVENT_RSP_ERROR: + if (ns->num_rnnid_retries < BFA_FCS_MAX_NS_RETRIES) { + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_rnn_id_retry); + ns->port->stats.ns_retries++; + ns->num_rnnid_retries++; + bfa_timer_start(BFA_FCS_GET_HAL_FROM_PORT(ns->port), + &ns->timer, bfa_fcs_lport_ns_timeout, ns, + BFA_FCS_RETRY_TIMEOUT); + } else { + bfa_sm_set_state(ns, + bfa_fcs_lport_ns_sm_sending_rspn_id); + bfa_fcs_lport_ns_send_rspn_id(ns, NULL); + } + break; + + case NSSM_EVENT_PORT_OFFLINE: + bfa_fcxp_discard(ns->fcxp); + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_offline); + break; + + default: + bfa_sm_fault(ns->port->fcs, event); + } +} + +static void +bfa_fcs_lport_ns_sm_rnn_id_retry(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event) +{ + bfa_trc(ns->port->fcs, ns->port->port_cfg.pwwn); + bfa_trc(ns->port->fcs, event); + + switch (event) { + case NSSM_EVENT_TIMEOUT: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_sending_rnn_id); + bfa_fcs_lport_ns_send_rnn_id(ns, NULL); + break; + + case NSSM_EVENT_PORT_OFFLINE: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_offline); + bfa_timer_stop(&ns->timer); + break; + + default: + bfa_sm_fault(ns->port->fcs, event); + } +} + +static void +bfa_fcs_lport_ns_sm_sending_rsnn_nn(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event) +{ + bfa_trc(ns->port->fcs, ns->port->port_cfg.pwwn); + bfa_trc(ns->port->fcs, event); + + switch (event) { + case NSSM_EVENT_RSNN_NN_SENT: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_rsnn_nn); + break; + + case NSSM_EVENT_PORT_OFFLINE: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_offline); + bfa_fcxp_walloc_cancel(BFA_FCS_GET_HAL_FROM_PORT(ns->port), + &ns->fcxp_wqe); + break; + + default: + bfa_sm_fault(ns->port->fcs, event); + } +} + +static void +bfa_fcs_lport_ns_sm_rsnn_nn(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event) +{ + bfa_trc(ns->port->fcs, ns->port->port_cfg.pwwn); + bfa_trc(ns->port->fcs, event); + + switch (event) { + case NSSM_EVENT_RSP_OK: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_sending_rspn_id); + ns->num_rsnn_nn_retries = 0; + bfa_fcs_lport_ns_send_rspn_id(ns, NULL); + break; + + case NSSM_EVENT_RSP_ERROR: + if (ns->num_rsnn_nn_retries < BFA_FCS_MAX_NS_RETRIES) { + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_rsnn_nn_retry); + ns->port->stats.ns_retries++; + ns->num_rsnn_nn_retries++; + bfa_timer_start(BFA_FCS_GET_HAL_FROM_PORT(ns->port), + &ns->timer, bfa_fcs_lport_ns_timeout, + ns, BFA_FCS_RETRY_TIMEOUT); + } else { + bfa_sm_set_state(ns, + bfa_fcs_lport_ns_sm_sending_rspn_id); + bfa_fcs_lport_ns_send_rspn_id(ns, NULL); + } + break; + + case NSSM_EVENT_PORT_OFFLINE: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_offline); + bfa_fcxp_discard(ns->fcxp); + break; + + default: + bfa_sm_fault(ns->port->fcs, event); + } +} + +static void +bfa_fcs_lport_ns_sm_rsnn_nn_retry(struct bfa_fcs_lport_ns_s *ns, + enum vport_ns_event event) +{ + bfa_trc(ns->port->fcs, ns->port->port_cfg.pwwn); + bfa_trc(ns->port->fcs, event); + + switch (event) { + case NSSM_EVENT_TIMEOUT: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_sending_rsnn_nn); + bfa_fcs_lport_ns_send_rsnn_nn(ns, NULL); + break; + + case NSSM_EVENT_PORT_OFFLINE: + bfa_sm_set_state(ns, bfa_fcs_lport_ns_sm_offline); + bfa_timer_stop(&ns->timer); + break; + + default: + bfa_sm_fault(ns->port->fcs, event); + } +} + +static void bfa_fcs_lport_ns_sm_sending_rspn_id(struct bfa_fcs_lport_ns_s *ns, enum vport_ns_event event) { @@ -3770,11 +4039,12 @@ bfa_fcs_lport_ns_send_plogi(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->pid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { port->stats.ns_plogi_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, - bfa_fcs_lport_ns_send_plogi, ns); + bfa_fcs_lport_ns_send_plogi, ns, BFA_TRUE); return; } ns->fcxp = fcxp; @@ -3853,6 +4123,162 @@ bfa_fcs_lport_ns_plogi_response(void *fcsarg, struct bfa_fcxp_s *fcxp, } /* + * Register node name for port_id + */ +static void +bfa_fcs_lport_ns_send_rnn_id(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) +{ + struct bfa_fcs_lport_ns_s *ns = ns_cbarg; + struct bfa_fcs_lport_s *port = ns->port; + struct fchs_s fchs; + int len; + struct bfa_fcxp_s *fcxp; + + bfa_trc(port->fcs, port->port_cfg.pwwn); + + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); + if (!fcxp) { + port->stats.ns_rnnid_alloc_wait++; + bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, + bfa_fcs_lport_ns_send_rnn_id, ns, BFA_TRUE); + return; + } + + ns->fcxp = fcxp; + + len = fc_rnnid_build(&fchs, bfa_fcxp_get_reqbuf(fcxp), + bfa_fcs_lport_get_fcid(port), + bfa_fcs_lport_get_fcid(port), + bfa_fcs_lport_get_nwwn(port)); + + bfa_fcxp_send(fcxp, NULL, port->fabric->vf_id, port->lp_tag, BFA_FALSE, + FC_CLASS_3, len, &fchs, + bfa_fcs_lport_ns_rnn_id_response, (void *)ns, + FC_MAX_PDUSZ, FC_FCCT_TOV); + + port->stats.ns_rnnid_sent++; + bfa_sm_send_event(ns, NSSM_EVENT_RNNID_SENT); +} + +static void +bfa_fcs_lport_ns_rnn_id_response(void *fcsarg, struct bfa_fcxp_s *fcxp, + void *cbarg, bfa_status_t req_status, + u32 rsp_len, u32 resid_len, + struct fchs_s *rsp_fchs) + +{ + struct bfa_fcs_lport_ns_s *ns = (struct bfa_fcs_lport_ns_s *) cbarg; + struct bfa_fcs_lport_s *port = ns->port; + struct ct_hdr_s *cthdr = NULL; + + bfa_trc(port->fcs, port->port_cfg.pwwn); + + /* + * Sanity Checks + */ + if (req_status != BFA_STATUS_OK) { + bfa_trc(port->fcs, req_status); + port->stats.ns_rnnid_rsp_err++; + bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR); + return; + } + + cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp); + cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code); + + if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) { + port->stats.ns_rnnid_accepts++; + bfa_sm_send_event(ns, NSSM_EVENT_RSP_OK); + return; + } + + port->stats.ns_rnnid_rejects++; + bfa_trc(port->fcs, cthdr->reason_code); + bfa_trc(port->fcs, cthdr->exp_code); + bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR); +} + +/* + * Register the symbolic node name for a given node name. + */ +static void +bfa_fcs_lport_ns_send_rsnn_nn(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) +{ + struct bfa_fcs_lport_ns_s *ns = ns_cbarg; + struct bfa_fcs_lport_s *port = ns->port; + struct fchs_s fchs; + int len; + struct bfa_fcxp_s *fcxp; + u8 *nsymbl; + + bfa_trc(port->fcs, port->port_cfg.pwwn); + + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); + if (!fcxp) { + port->stats.ns_rsnn_nn_alloc_wait++; + bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, + bfa_fcs_lport_ns_send_rsnn_nn, ns, BFA_TRUE); + return; + } + ns->fcxp = fcxp; + + nsymbl = (u8 *) &(bfa_fcs_lport_get_nsym_name( + bfa_fcs_get_base_port(port->fcs))); + + len = fc_rsnn_nn_build(&fchs, bfa_fcxp_get_reqbuf(fcxp), + bfa_fcs_lport_get_fcid(port), + bfa_fcs_lport_get_nwwn(port), nsymbl); + + bfa_fcxp_send(fcxp, NULL, port->fabric->vf_id, port->lp_tag, BFA_FALSE, + FC_CLASS_3, len, &fchs, + bfa_fcs_lport_ns_rsnn_nn_response, (void *)ns, + FC_MAX_PDUSZ, FC_FCCT_TOV); + + port->stats.ns_rsnn_nn_sent++; + + bfa_sm_send_event(ns, NSSM_EVENT_RSNN_NN_SENT); +} + +static void +bfa_fcs_lport_ns_rsnn_nn_response(void *fcsarg, struct bfa_fcxp_s *fcxp, + void *cbarg, bfa_status_t req_status, + u32 rsp_len, u32 resid_len, + struct fchs_s *rsp_fchs) +{ + struct bfa_fcs_lport_ns_s *ns = (struct bfa_fcs_lport_ns_s *) cbarg; + struct bfa_fcs_lport_s *port = ns->port; + struct ct_hdr_s *cthdr = NULL; + + bfa_trc(port->fcs, port->port_cfg.pwwn); + + /* + * Sanity Checks + */ + if (req_status != BFA_STATUS_OK) { + bfa_trc(port->fcs, req_status); + port->stats.ns_rsnn_nn_rsp_err++; + bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR); + return; + } + + cthdr = (struct ct_hdr_s *) BFA_FCXP_RSP_PLD(fcxp); + cthdr->cmd_rsp_code = be16_to_cpu(cthdr->cmd_rsp_code); + + if (cthdr->cmd_rsp_code == CT_RSP_ACCEPT) { + port->stats.ns_rsnn_nn_accepts++; + bfa_sm_send_event(ns, NSSM_EVENT_RSP_OK); + return; + } + + port->stats.ns_rsnn_nn_rejects++; + bfa_trc(port->fcs, cthdr->reason_code); + bfa_trc(port->fcs, cthdr->exp_code); + bfa_sm_send_event(ns, NSSM_EVENT_RSP_ERROR); +} + +/* * Register the symbolic port name. */ static void @@ -3870,11 +4296,12 @@ bfa_fcs_lport_ns_send_rspn_id(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { port->stats.ns_rspnid_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, - bfa_fcs_lport_ns_send_rspn_id, ns); + bfa_fcs_lport_ns_send_rspn_id, ns, BFA_TRUE); return; } ns->fcxp = fcxp; @@ -3971,11 +4398,12 @@ bfa_fcs_lport_ns_send_rft_id(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { port->stats.ns_rftid_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, - bfa_fcs_lport_ns_send_rft_id, ns); + bfa_fcs_lport_ns_send_rft_id, ns, BFA_TRUE); return; } ns->fcxp = fcxp; @@ -4044,11 +4472,12 @@ bfa_fcs_lport_ns_send_rff_id(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { port->stats.ns_rffid_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, - bfa_fcs_lport_ns_send_rff_id, ns); + bfa_fcs_lport_ns_send_rff_id, ns, BFA_TRUE); return; } ns->fcxp = fcxp; @@ -4127,11 +4556,12 @@ bfa_fcs_lport_ns_send_gid_ft(void *ns_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->pid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { port->stats.ns_gidft_alloc_wait++; bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, - bfa_fcs_lport_ns_send_gid_ft, ns); + bfa_fcs_lport_ns_send_gid_ft, ns, BFA_TRUE); return; } ns->fcxp = fcxp; @@ -4261,6 +4691,10 @@ bfa_fcs_lport_ns_process_gidft_pids(struct bfa_fcs_lport_s *port, u32 *pid_buf, struct fcgs_gidft_resp_s *gidft_entry; struct bfa_fcs_rport_s *rport; u32 ii; + struct bfa_fcs_fabric_s *fabric = port->fabric; + struct bfa_fcs_vport_s *vport; + struct list_head *qe; + u8 found = 0; for (ii = 0; ii < n_pids; ii++) { gidft_entry = (struct fcgs_gidft_resp_s *) &pid_buf[ii]; @@ -4269,6 +4703,29 @@ bfa_fcs_lport_ns_process_gidft_pids(struct bfa_fcs_lport_s *port, u32 *pid_buf, continue; /* + * Ignore PID if it is of base port + * (Avoid vports discovering base port as remote port) + */ + if (gidft_entry->pid == fabric->bport.pid) + continue; + + /* + * Ignore PID if it is of vport created on the same base port + * (Avoid vport discovering every other vport created on the + * same port as remote port) + */ + list_for_each(qe, &fabric->vport_q) { + vport = (struct bfa_fcs_vport_s *) qe; + if (vport->lport.pid == gidft_entry->pid) + found = 1; + } + + if (found) { + found = 0; + continue; + } + + /* * Check if this rport already exists */ rport = bfa_fcs_lport_get_rport_by_pid(port, gidft_entry->pid); @@ -4335,7 +4792,8 @@ bfa_fcs_lport_ns_query(struct bfa_fcs_lport_s *port) struct bfa_fcs_lport_ns_s *ns = BFA_FCS_GET_NS_FROM_PORT(port); bfa_trc(port->fcs, port->pid); - bfa_sm_send_event(ns, NSSM_EVENT_NS_QUERY); + if (bfa_sm_cmp_state(ns, bfa_fcs_lport_ns_sm_online)) + bfa_sm_send_event(ns, NSSM_EVENT_NS_QUERY); } static void @@ -4355,6 +4813,70 @@ bfa_fcs_lport_ns_boot_target_disc(bfa_fcs_lport_t *port) } } +void +bfa_fcs_lport_ns_util_send_rspn_id(void *cbarg, struct bfa_fcxp_s *fcxp_alloced) +{ + struct bfa_fcs_lport_ns_s *ns = cbarg; + struct bfa_fcs_lport_s *port = ns->port; + struct fchs_s fchs; + struct bfa_fcxp_s *fcxp; + u8 symbl[256]; + u8 *psymbl = &symbl[0]; + int len; + + if (!bfa_sm_cmp_state(port, bfa_fcs_lport_sm_online)) + return; + + /* Avoid sending RSPN in the following states. */ + if (bfa_sm_cmp_state(ns, bfa_fcs_lport_ns_sm_offline) || + bfa_sm_cmp_state(ns, bfa_fcs_lport_ns_sm_plogi_sending) || + bfa_sm_cmp_state(ns, bfa_fcs_lport_ns_sm_plogi) || + bfa_sm_cmp_state(ns, bfa_fcs_lport_ns_sm_plogi_retry) || + bfa_sm_cmp_state(ns, bfa_fcs_lport_ns_sm_rspn_id_retry)) + return; + + memset(symbl, 0, sizeof(symbl)); + bfa_trc(port->fcs, port->port_cfg.pwwn); + + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); + if (!fcxp) { + port->stats.ns_rspnid_alloc_wait++; + bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &ns->fcxp_wqe, + bfa_fcs_lport_ns_util_send_rspn_id, ns, BFA_FALSE); + return; + } + + ns->fcxp = fcxp; + + if (port->vport) { + /* + * For Vports, we append the vport's port symbolic name + * to that of the base port. + */ + strncpy((char *)psymbl, (char *)&(bfa_fcs_lport_get_psym_name + (bfa_fcs_get_base_port(port->fcs))), + strlen((char *)&bfa_fcs_lport_get_psym_name( + bfa_fcs_get_base_port(port->fcs)))); + + /* Ensure we have a null terminating string. */ + ((char *)psymbl)[strlen((char *)&bfa_fcs_lport_get_psym_name( + bfa_fcs_get_base_port(port->fcs)))] = 0; + + strncat((char *)psymbl, + (char *)&(bfa_fcs_lport_get_psym_name(port)), + strlen((char *)&bfa_fcs_lport_get_psym_name(port))); + } + + len = fc_rspnid_build(&fchs, bfa_fcxp_get_reqbuf(fcxp), + bfa_fcs_lport_get_fcid(port), 0, psymbl); + + bfa_fcxp_send(fcxp, NULL, port->fabric->vf_id, port->lp_tag, BFA_FALSE, + FC_CLASS_3, len, &fchs, NULL, NULL, FC_MAX_PDUSZ, 0); + + port->stats.ns_rspnid_sent++; +} + /* * FCS SCN */ @@ -4529,10 +5051,11 @@ bfa_fcs_lport_scn_send_scr(void *scn_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(port->fcs, port->pid); bfa_trc(port->fcs, port->port_cfg.pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &scn->fcxp_wqe, - bfa_fcs_lport_scn_send_scr, scn); + bfa_fcs_lport_scn_send_scr, scn, BFA_TRUE); return; } scn->fcxp = fcxp; @@ -4614,7 +5137,7 @@ bfa_fcs_lport_scn_send_ls_acc(struct bfa_fcs_lport_s *port, bfa_trc(port->fcs, rx_fchs->s_id); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -4688,14 +5211,33 @@ static void bfa_fcs_lport_scn_portid_rscn(struct bfa_fcs_lport_s *port, u32 rpid) { struct bfa_fcs_rport_s *rport; + struct bfa_fcs_fabric_s *fabric = port->fabric; + struct bfa_fcs_vport_s *vport; + struct list_head *qe; bfa_trc(port->fcs, rpid); /* + * Ignore PID if it is of base port or of vports created on the + * same base port. It is to avoid vports discovering base port or + * other vports created on same base port as remote port + */ + if (rpid == fabric->bport.pid) + return; + + list_for_each(qe, &fabric->vport_q) { + vport = (struct bfa_fcs_vport_s *) qe; + if (vport->lport.pid == rpid) + return; + } + /* * If this is an unknown device, then it just came online. * Otherwise let rport handle the RSCN event. */ rport = bfa_fcs_lport_get_rport_by_pid(port, rpid); + if (!rport) + rport = bfa_fcs_lport_get_rport_by_old_pid(port, rpid); + if (rport == NULL) { /* * If min cfg mode is enabled, we donot need to @@ -4888,15 +5430,15 @@ bfa_fcs_lport_get_rport(struct bfa_fcs_lport_s *port, wwn_t wwn, int index, } void -bfa_fcs_lport_get_rports(struct bfa_fcs_lport_s *port, - wwn_t rport_wwns[], int *nrports) +bfa_fcs_lport_get_rport_quals(struct bfa_fcs_lport_s *port, + struct bfa_rport_qualifier_s rports[], int *nrports) { struct list_head *qh, *qe; struct bfa_fcs_rport_s *rport = NULL; int i; struct bfa_fcs_s *fcs; - if (port == NULL || rport_wwns == NULL || *nrports == 0) + if (port == NULL || rports == NULL || *nrports == 0) return; fcs = port->fcs; @@ -4916,7 +5458,13 @@ bfa_fcs_lport_get_rports(struct bfa_fcs_lport_s *port, continue; } - rport_wwns[i] = rport->pwwn; + if (!rport->pwwn && !rport->pid) { + qe = bfa_q_next(qe); + continue; + } + + rports[i].pwwn = rport->pwwn; + rports[i].pid = rport->pid; i++; qe = bfa_q_next(qe); @@ -5760,6 +6308,16 @@ bfa_fcs_vport_cleanup(struct bfa_fcs_vport_s *vport) { vport->vport_stats.fab_cleanup++; } + +/* + * Stop notification from fabric SM. To be invoked from within FCS. + */ +void +bfa_fcs_vport_fcs_stop(struct bfa_fcs_vport_s *vport) +{ + bfa_sm_send_event(vport, BFA_FCS_VPORT_SM_STOP); +} + /* * delete notification from fabric SM. To be invoked from within FCS. */ diff --git a/drivers/scsi/bfa/bfa_fcs_rport.c b/drivers/scsi/bfa/bfa_fcs_rport.c index fe0463a..cc43b2a 100644 --- a/drivers/scsi/bfa/bfa_fcs_rport.c +++ b/drivers/scsi/bfa/bfa_fcs_rport.c @@ -30,14 +30,22 @@ static u32 bfa_fcs_rport_del_timeout = BFA_FCS_RPORT_DEF_DEL_TIMEOUT * 1000; /* In millisecs */ /* + * bfa_fcs_rport_max_logins is max count of bfa_fcs_rports + * whereas DEF_CFG_NUM_RPORTS is max count of bfa_rports + */ +static u32 bfa_fcs_rport_max_logins = BFA_FCS_MAX_RPORT_LOGINS; + +/* * forward declarations */ static struct bfa_fcs_rport_s *bfa_fcs_rport_alloc( struct bfa_fcs_lport_s *port, wwn_t pwwn, u32 rpid); static void bfa_fcs_rport_free(struct bfa_fcs_rport_s *rport); static void bfa_fcs_rport_hal_online(struct bfa_fcs_rport_s *rport); -static void bfa_fcs_rport_online_action(struct bfa_fcs_rport_s *rport); -static void bfa_fcs_rport_offline_action(struct bfa_fcs_rport_s *rport); +static void bfa_fcs_rport_fcs_online_action(struct bfa_fcs_rport_s *rport); +static void bfa_fcs_rport_hal_online_action(struct bfa_fcs_rport_s *rport); +static void bfa_fcs_rport_fcs_offline_action(struct bfa_fcs_rport_s *rport); +static void bfa_fcs_rport_hal_offline_action(struct bfa_fcs_rport_s *rport); static void bfa_fcs_rport_update(struct bfa_fcs_rport_s *rport, struct fc_logi_s *plogi); static void bfa_fcs_rport_timeout(void *arg); @@ -76,6 +84,7 @@ static void bfa_fcs_rport_send_ls_rjt(struct bfa_fcs_rport_s *rport, static void bfa_fcs_rport_process_adisc(struct bfa_fcs_rport_s *rport, struct fchs_s *rx_fchs, u16 len); static void bfa_fcs_rport_send_prlo_acc(struct bfa_fcs_rport_s *rport); +static void bfa_fcs_rport_hal_offline(struct bfa_fcs_rport_s *rport); static void bfa_fcs_rport_sm_uninit(struct bfa_fcs_rport_s *rport, enum rport_event event); @@ -87,6 +96,8 @@ static void bfa_fcs_rport_sm_plogi_retry(struct bfa_fcs_rport_s *rport, enum rport_event event); static void bfa_fcs_rport_sm_plogi(struct bfa_fcs_rport_s *rport, enum rport_event event); +static void bfa_fcs_rport_sm_fc4_fcs_online(struct bfa_fcs_rport_s *rport, + enum rport_event event); static void bfa_fcs_rport_sm_hal_online(struct bfa_fcs_rport_s *rport, enum rport_event event); static void bfa_fcs_rport_sm_online(struct bfa_fcs_rport_s *rport, @@ -123,6 +134,10 @@ static void bfa_fcs_rport_sm_nsdisc_sent(struct bfa_fcs_rport_s *rport, enum rport_event event); static void bfa_fcs_rport_sm_nsdisc_sent(struct bfa_fcs_rport_s *rport, enum rport_event event); +static void bfa_fcs_rport_sm_fc4_off_delete(struct bfa_fcs_rport_s *rport, + enum rport_event event); +static void bfa_fcs_rport_sm_delete_pending(struct bfa_fcs_rport_s *rport, + enum rport_event event); static struct bfa_sm_table_s rport_sm_table[] = { {BFA_SM(bfa_fcs_rport_sm_uninit), BFA_RPORT_UNINIT}, @@ -130,6 +145,7 @@ static struct bfa_sm_table_s rport_sm_table[] = { {BFA_SM(bfa_fcs_rport_sm_plogiacc_sending), BFA_RPORT_ONLINE}, {BFA_SM(bfa_fcs_rport_sm_plogi_retry), BFA_RPORT_PLOGI_RETRY}, {BFA_SM(bfa_fcs_rport_sm_plogi), BFA_RPORT_PLOGI}, + {BFA_SM(bfa_fcs_rport_sm_fc4_fcs_online), BFA_RPORT_ONLINE}, {BFA_SM(bfa_fcs_rport_sm_hal_online), BFA_RPORT_ONLINE}, {BFA_SM(bfa_fcs_rport_sm_online), BFA_RPORT_ONLINE}, {BFA_SM(bfa_fcs_rport_sm_nsquery_sending), BFA_RPORT_NSQUERY}, @@ -167,8 +183,8 @@ bfa_fcs_rport_sm_uninit(struct bfa_fcs_rport_s *rport, enum rport_event event) break; case RPSM_EVENT_PLOGI_RCVD: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_plogiacc_sending); - bfa_fcs_rport_send_plogiacc(rport, NULL); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); + bfa_fcs_rport_fcs_online_action(rport); break; case RPSM_EVENT_PLOGI_COMP: @@ -252,8 +268,8 @@ bfa_fcs_rport_sm_plogiacc_sending(struct bfa_fcs_rport_s *rport, switch (event) { case RPSM_EVENT_FCXP_SENT: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); - bfa_fcs_rport_hal_online(rport); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); + bfa_fcs_rport_fcs_online_action(rport); break; case RPSM_EVENT_DELETE: @@ -348,9 +364,9 @@ bfa_fcs_rport_sm_plogi_retry(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_PLOGI_COMP: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); bfa_timer_stop(&rport->timer); - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); break; default: @@ -370,9 +386,9 @@ bfa_fcs_rport_sm_plogi(struct bfa_fcs_rport_s *rport, enum rport_event event) switch (event) { case RPSM_EVENT_ACCEPTED: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); rport->plogi_retries = 0; - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); break; case RPSM_EVENT_LOGO_RCVD: @@ -397,6 +413,7 @@ bfa_fcs_rport_sm_plogi(struct bfa_fcs_rport_s *rport, enum rport_event event) BFA_FCS_RETRY_TIMEOUT); } else { bfa_stats(rport->port, rport_del_max_plogi_retry); + rport->old_pid = rport->pid; rport->pid = 0; bfa_sm_set_state(rport, bfa_fcs_rport_sm_offline); bfa_timer_start(rport->fcs->bfa, &rport->timer, @@ -443,13 +460,77 @@ bfa_fcs_rport_sm_plogi(struct bfa_fcs_rport_s *rport, enum rport_event event) break; case RPSM_EVENT_PLOGI_COMP: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); + break; + + default: + bfa_sm_fault(rport->fcs, event); + } +} + +/* + * PLOGI is done. Await bfa_fcs_itnim to ascertain the scsi function + */ +static void +bfa_fcs_rport_sm_fc4_fcs_online(struct bfa_fcs_rport_s *rport, + enum rport_event event) +{ + bfa_trc(rport->fcs, rport->pwwn); + bfa_trc(rport->fcs, rport->pid); + bfa_trc(rport->fcs, event); + + switch (event) { + case RPSM_EVENT_FC4_FCS_ONLINE: + if (rport->scsi_function == BFA_RPORT_INITIATOR) { + if (!BFA_FCS_PID_IS_WKA(rport->pid)) + bfa_fcs_rpf_rport_online(rport); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_online); + break; + } + + if (!rport->bfa_rport) + rport->bfa_rport = + bfa_rport_create(rport->fcs->bfa, rport); + + if (rport->bfa_rport) { + bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_fcs_rport_hal_online(rport); + } else { + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); + bfa_fcs_rport_fcs_offline_action(rport); + } + break; + + case RPSM_EVENT_PLOGI_RCVD: + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); + rport->plogi_pending = BFA_TRUE; + bfa_fcs_rport_fcs_offline_action(rport); + break; + + case RPSM_EVENT_PLOGI_COMP: + case RPSM_EVENT_LOGO_IMP: + case RPSM_EVENT_ADDRESS_CHANGE: + case RPSM_EVENT_SCN: + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); + bfa_fcs_rport_fcs_offline_action(rport); + break; + + case RPSM_EVENT_LOGO_RCVD: + case RPSM_EVENT_PRLO_RCVD: + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); + bfa_fcs_rport_fcs_offline_action(rport); + break; + + case RPSM_EVENT_DELETE: + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); + bfa_fcs_rport_fcs_offline_action(rport); break; default: bfa_sm_fault(rport->fcs, event); + break; } } @@ -468,41 +549,34 @@ bfa_fcs_rport_sm_hal_online(struct bfa_fcs_rport_s *rport, switch (event) { case RPSM_EVENT_HCB_ONLINE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_online); - bfa_fcs_rport_online_action(rport); + bfa_fcs_rport_hal_online_action(rport); break; - case RPSM_EVENT_PRLO_RCVD: case RPSM_EVENT_PLOGI_COMP: break; + case RPSM_EVENT_PRLO_RCVD: case RPSM_EVENT_LOGO_RCVD: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_logorcv); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); + bfa_fcs_rport_fcs_offline_action(rport); break; + case RPSM_EVENT_SCN: case RPSM_EVENT_LOGO_IMP: case RPSM_EVENT_ADDRESS_CHANGE: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_offline); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); + bfa_fcs_rport_fcs_offline_action(rport); break; case RPSM_EVENT_PLOGI_RCVD: rport->plogi_pending = BFA_TRUE; - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_offline); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); + bfa_fcs_rport_fcs_offline_action(rport); break; case RPSM_EVENT_DELETE: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_logosend); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); - break; - - case RPSM_EVENT_SCN: - /* - * @todo - * Ignore SCN - PLOGI just completed, FC-4 login should detect - * device failures. - */ + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); + bfa_fcs_rport_fcs_offline_action(rport); break; default: @@ -537,18 +611,18 @@ bfa_fcs_rport_sm_online(struct bfa_fcs_rport_s *rport, enum rport_event event) case RPSM_EVENT_LOGO_IMP: case RPSM_EVENT_ADDRESS_CHANGE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_DELETE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_LOGO_RCVD: case RPSM_EVENT_PRLO_RCVD: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_PLOGI_COMP: @@ -579,7 +653,7 @@ bfa_fcs_rport_sm_nsquery_sending(struct bfa_fcs_rport_s *rport, case RPSM_EVENT_DELETE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_SCN: @@ -592,24 +666,16 @@ bfa_fcs_rport_sm_nsquery_sending(struct bfa_fcs_rport_s *rport, case RPSM_EVENT_PRLO_RCVD: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_LOGO_IMP: - rport->pid = 0; - bfa_sm_set_state(rport, bfa_fcs_rport_sm_offline); - bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_timer_start(rport->fcs->bfa, &rport->timer, - bfa_fcs_rport_timeout, rport, - bfa_fcs_rport_del_timeout); - break; - case RPSM_EVENT_PLOGI_RCVD: case RPSM_EVENT_ADDRESS_CHANGE: case RPSM_EVENT_PLOGI_COMP: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; default: @@ -642,14 +708,14 @@ bfa_fcs_rport_sm_nsquery(struct bfa_fcs_rport_s *rport, enum rport_event event) bfa_fcs_rport_send_nsdisc(rport, NULL); } else { bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); } break; case RPSM_EVENT_DELETE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_SCN: @@ -659,7 +725,7 @@ bfa_fcs_rport_sm_nsquery(struct bfa_fcs_rport_s *rport, enum rport_event event) case RPSM_EVENT_PRLO_RCVD: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_PLOGI_COMP: @@ -668,7 +734,7 @@ bfa_fcs_rport_sm_nsquery(struct bfa_fcs_rport_s *rport, enum rport_event event) case RPSM_EVENT_LOGO_IMP: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; default: @@ -696,21 +762,21 @@ bfa_fcs_rport_sm_adisc_sending(struct bfa_fcs_rport_s *rport, case RPSM_EVENT_DELETE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_LOGO_IMP: case RPSM_EVENT_ADDRESS_CHANGE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_LOGO_RCVD: case RPSM_EVENT_PRLO_RCVD: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_SCN: @@ -719,7 +785,7 @@ bfa_fcs_rport_sm_adisc_sending(struct bfa_fcs_rport_s *rport, case RPSM_EVENT_PLOGI_RCVD: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; default: @@ -756,13 +822,13 @@ bfa_fcs_rport_sm_adisc(struct bfa_fcs_rport_s *rport, enum rport_event event) case RPSM_EVENT_FAILED: case RPSM_EVENT_ADDRESS_CHANGE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_DELETE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_SCN: @@ -774,14 +840,14 @@ bfa_fcs_rport_sm_adisc(struct bfa_fcs_rport_s *rport, enum rport_event event) case RPSM_EVENT_LOGO_IMP: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_offline); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; case RPSM_EVENT_LOGO_RCVD: case RPSM_EVENT_PRLO_RCVD: bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logorcv); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_offline_action(rport); + bfa_fcs_rport_hal_offline_action(rport); break; default: @@ -803,13 +869,19 @@ bfa_fcs_rport_sm_fc4_logorcv(struct bfa_fcs_rport_s *rport, switch (event) { case RPSM_EVENT_FC4_OFFLINE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_logorcv); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + bfa_fcs_rport_hal_offline(rport); break; case RPSM_EVENT_DELETE: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_logosend); + if (rport->pid && (rport->prlo == BFA_TRUE)) + bfa_fcs_rport_send_prlo_acc(rport); + if (rport->pid && (rport->prlo == BFA_FALSE)) + bfa_fcs_rport_send_logo_acc(rport); + + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_off_delete); break; + case RPSM_EVENT_HCB_ONLINE: case RPSM_EVENT_LOGO_RCVD: case RPSM_EVENT_PRLO_RCVD: case RPSM_EVENT_ADDRESS_CHANGE: @@ -835,7 +907,20 @@ bfa_fcs_rport_sm_fc4_logosend(struct bfa_fcs_rport_s *rport, switch (event) { case RPSM_EVENT_FC4_OFFLINE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_logosend); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + bfa_fcs_rport_hal_offline(rport); + break; + + case RPSM_EVENT_LOGO_RCVD: + bfa_fcs_rport_send_logo_acc(rport); + case RPSM_EVENT_PRLO_RCVD: + if (rport->prlo == BFA_TRUE) + bfa_fcs_rport_send_prlo_acc(rport); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_off_delete); + break; + + case RPSM_EVENT_HCB_ONLINE: + case RPSM_EVENT_DELETE: + /* Rport is being deleted */ break; default: @@ -857,13 +942,23 @@ bfa_fcs_rport_sm_fc4_offline(struct bfa_fcs_rport_s *rport, switch (event) { case RPSM_EVENT_FC4_OFFLINE: bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_offline); - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + bfa_fcs_rport_hal_offline(rport); break; - case RPSM_EVENT_SCN: - case RPSM_EVENT_LOGO_IMP: case RPSM_EVENT_LOGO_RCVD: + /* + * Rport is going offline. Just ack the logo + */ + bfa_fcs_rport_send_logo_acc(rport); + break; + case RPSM_EVENT_PRLO_RCVD: + bfa_fcs_rport_send_prlo_acc(rport); + break; + + case RPSM_EVENT_HCB_ONLINE: + case RPSM_EVENT_SCN: + case RPSM_EVENT_LOGO_IMP: case RPSM_EVENT_ADDRESS_CHANGE: /* * rport is already going offline. @@ -907,24 +1002,23 @@ bfa_fcs_rport_sm_hcb_offline(struct bfa_fcs_rport_s *rport, */ case RPSM_EVENT_ADDRESS_CHANGE: - if (bfa_fcs_lport_is_online(rport->port)) { - if (bfa_fcs_fabric_is_switched(rport->port->fabric)) { - bfa_sm_set_state(rport, - bfa_fcs_rport_sm_nsdisc_sending); - rport->ns_retries = 0; - bfa_fcs_rport_send_nsdisc(rport, NULL); - } else { - bfa_sm_set_state(rport, - bfa_fcs_rport_sm_plogi_sending); - rport->plogi_retries = 0; - bfa_fcs_rport_send_plogi(rport, NULL); - } - } else { + if (!bfa_fcs_lport_is_online(rport->port)) { rport->pid = 0; bfa_sm_set_state(rport, bfa_fcs_rport_sm_offline); bfa_timer_start(rport->fcs->bfa, &rport->timer, bfa_fcs_rport_timeout, rport, bfa_fcs_rport_del_timeout); + break; + } + if (bfa_fcs_fabric_is_switched(rport->port->fabric)) { + bfa_sm_set_state(rport, + bfa_fcs_rport_sm_nsdisc_sending); + rport->ns_retries = 0; + bfa_fcs_rport_send_nsdisc(rport, NULL); + } else { + bfa_sm_set_state(rport, bfa_fcs_rport_sm_plogi_sending); + rport->plogi_retries = 0; + bfa_fcs_rport_send_plogi(rport, NULL); } break; @@ -1001,7 +1095,11 @@ bfa_fcs_rport_sm_hcb_logorcv(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_DELETE: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hcb_logosend); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_delete_pending); + if (rport->pid && (rport->prlo == BFA_TRUE)) + bfa_fcs_rport_send_prlo_acc(rport); + if (rport->pid && (rport->prlo == BFA_FALSE)) + bfa_fcs_rport_send_logo_acc(rport); break; case RPSM_EVENT_LOGO_IMP: @@ -1040,7 +1138,14 @@ bfa_fcs_rport_sm_hcb_logosend(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_LOGO_RCVD: + bfa_fcs_rport_send_logo_acc(rport); case RPSM_EVENT_PRLO_RCVD: + if (rport->prlo == BFA_TRUE) + bfa_fcs_rport_send_prlo_acc(rport); + + bfa_sm_set_state(rport, bfa_fcs_rport_sm_delete_pending); + break; + case RPSM_EVENT_ADDRESS_CHANGE: break; @@ -1072,7 +1177,11 @@ bfa_fcs_rport_sm_logo_sending(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_LOGO_RCVD: + bfa_fcs_rport_send_logo_acc(rport); case RPSM_EVENT_PRLO_RCVD: + if (rport->prlo == BFA_TRUE) + bfa_fcs_rport_send_prlo_acc(rport); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_uninit); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); bfa_fcs_rport_free(rport); @@ -1126,9 +1235,9 @@ bfa_fcs_rport_sm_offline(struct bfa_fcs_rport_s *rport, enum rport_event event) break; case RPSM_EVENT_PLOGI_COMP: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); bfa_timer_stop(&rport->timer); - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); break; case RPSM_EVENT_PLOGI_SEND: @@ -1190,9 +1299,9 @@ bfa_fcs_rport_sm_nsdisc_sending(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_PLOGI_COMP: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); bfa_fcxp_walloc_cancel(rport->fcs->bfa, &rport->fcxp_wqe); - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); break; default: @@ -1254,9 +1363,9 @@ bfa_fcs_rport_sm_nsdisc_retry(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_PLOGI_COMP: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); bfa_timer_stop(&rport->timer); - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); break; default: @@ -1296,6 +1405,7 @@ bfa_fcs_rport_sm_nsdisc_sent(struct bfa_fcs_rport_s *rport, bfa_fcs_rport_sm_nsdisc_sending); bfa_fcs_rport_send_nsdisc(rport, NULL); } else { + rport->old_pid = rport->pid; rport->pid = 0; bfa_sm_set_state(rport, bfa_fcs_rport_sm_offline); bfa_timer_start(rport->fcs->bfa, &rport->timer, @@ -1343,9 +1453,9 @@ bfa_fcs_rport_sm_nsdisc_sent(struct bfa_fcs_rport_s *rport, break; case RPSM_EVENT_PLOGI_COMP: - bfa_sm_set_state(rport, bfa_fcs_rport_sm_hal_online); + bfa_sm_set_state(rport, bfa_fcs_rport_sm_fc4_fcs_online); bfa_fcxp_discard(rport->fcxp); - bfa_fcs_rport_hal_online(rport); + bfa_fcs_rport_fcs_online_action(rport); break; default: @@ -1353,7 +1463,63 @@ bfa_fcs_rport_sm_nsdisc_sent(struct bfa_fcs_rport_s *rport, } } +/* + * Rport needs to be deleted + * waiting for ITNIM clean up to finish + */ +static void +bfa_fcs_rport_sm_fc4_off_delete(struct bfa_fcs_rport_s *rport, + enum rport_event event) +{ + bfa_trc(rport->fcs, rport->pwwn); + bfa_trc(rport->fcs, rport->pid); + bfa_trc(rport->fcs, event); + switch (event) { + case RPSM_EVENT_FC4_OFFLINE: + bfa_sm_set_state(rport, bfa_fcs_rport_sm_delete_pending); + bfa_fcs_rport_hal_offline(rport); + break; + + case RPSM_EVENT_DELETE: + case RPSM_EVENT_PLOGI_RCVD: + /* Ignore these events */ + break; + + default: + bfa_sm_fault(rport->fcs, event); + break; + } +} + +/* + * RPort needs to be deleted + * waiting for BFA/FW to finish current processing + */ +static void +bfa_fcs_rport_sm_delete_pending(struct bfa_fcs_rport_s *rport, + enum rport_event event) +{ + bfa_trc(rport->fcs, rport->pwwn); + bfa_trc(rport->fcs, rport->pid); + bfa_trc(rport->fcs, event); + + switch (event) { + case RPSM_EVENT_HCB_OFFLINE: + bfa_sm_set_state(rport, bfa_fcs_rport_sm_uninit); + bfa_fcs_rport_free(rport); + break; + + case RPSM_EVENT_DELETE: + case RPSM_EVENT_LOGO_IMP: + case RPSM_EVENT_PLOGI_RCVD: + /* Ignore these events */ + break; + + default: + bfa_sm_fault(rport->fcs, event); + } +} /* * fcs_rport_private FCS RPORT provate functions @@ -1370,10 +1536,11 @@ bfa_fcs_rport_send_plogi(void *rport_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(rport->fcs, rport->pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &rport->fcxp_wqe, - bfa_fcs_rport_send_plogi, rport); + bfa_fcs_rport_send_plogi, rport, BFA_TRUE); return; } rport->fcxp = fcxp; @@ -1490,10 +1657,11 @@ bfa_fcs_rport_send_plogiacc(void *rport_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(rport->fcs, rport->pwwn); bfa_trc(rport->fcs, rport->reply_oxid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &rport->fcxp_wqe, - bfa_fcs_rport_send_plogiacc, rport); + bfa_fcs_rport_send_plogiacc, rport, BFA_FALSE); return; } rport->fcxp = fcxp; @@ -1522,10 +1690,11 @@ bfa_fcs_rport_send_adisc(void *rport_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(rport->fcs, rport->pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &rport->fcxp_wqe, - bfa_fcs_rport_send_adisc, rport); + bfa_fcs_rport_send_adisc, rport, BFA_TRUE); return; } rport->fcxp = fcxp; @@ -1585,10 +1754,11 @@ bfa_fcs_rport_send_nsdisc(void *rport_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(rport->fcs, rport->pid); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &rport->fcxp_wqe, - bfa_fcs_rport_send_nsdisc, rport); + bfa_fcs_rport_send_nsdisc, rport, BFA_TRUE); return; } rport->fcxp = fcxp; @@ -1741,10 +1911,11 @@ bfa_fcs_rport_send_logo(void *rport_cbarg, struct bfa_fcxp_s *fcxp_alloced) port = rport->port; - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &rport->fcxp_wqe, - bfa_fcs_rport_send_logo, rport); + bfa_fcs_rport_send_logo, rport, BFA_FALSE); return; } rport->fcxp = fcxp; @@ -1778,7 +1949,7 @@ bfa_fcs_rport_send_logo_acc(void *rport_cbarg) port = rport->port; - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -1849,7 +2020,7 @@ bfa_fcs_rport_process_prli(struct bfa_fcs_rport_s *rport, bfa_fcs_itnim_is_initiator(rport->itnim); } - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -1886,7 +2057,7 @@ bfa_fcs_rport_process_rpsc(struct bfa_fcs_rport_s *rport, speeds.port_op_speed = fc_bfa_speed_to_rpsc_operspeed(pport_attr.speed); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -1920,7 +2091,7 @@ bfa_fcs_rport_process_adisc(struct bfa_fcs_rport_s *rport, */ if (bfa_fcs_itnim_get_online_state(rport->itnim) == BFA_STATUS_OK) { - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; @@ -1957,6 +2128,15 @@ bfa_fcs_rport_hal_online(struct bfa_fcs_rport_s *rport) bfa_rport_online(rport->bfa_rport, &rport_info); } +static void +bfa_fcs_rport_hal_offline(struct bfa_fcs_rport_s *rport) +{ + if (rport->bfa_rport) + bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_OFFLINE); + else + bfa_cb_rport_offline(rport); +} + static struct bfa_fcs_rport_s * bfa_fcs_rport_alloc(struct bfa_fcs_lport_s *port, wwn_t pwwn, u32 rpid) { @@ -1967,6 +2147,11 @@ bfa_fcs_rport_alloc(struct bfa_fcs_lport_s *port, wwn_t pwwn, u32 rpid) /* * allocate rport */ + if (fcs->num_rport_logins >= bfa_fcs_rport_max_logins) { + bfa_trc(fcs, rpid); + return NULL; + } + if (bfa_fcb_rport_alloc(fcs->bfad, &rport, &rport_drv) != BFA_STATUS_OK) { bfa_trc(fcs, rpid); @@ -1981,16 +2166,9 @@ bfa_fcs_rport_alloc(struct bfa_fcs_lport_s *port, wwn_t pwwn, u32 rpid) rport->rp_drv = rport_drv; rport->pid = rpid; rport->pwwn = pwwn; + rport->old_pid = 0; - /* - * allocate BFA rport - */ - rport->bfa_rport = bfa_rport_create(port->fcs->bfa, rport); - if (!rport->bfa_rport) { - bfa_trc(fcs, rpid); - kfree(rport_drv); - return NULL; - } + rport->bfa_rport = NULL; /* * allocate FC-4s @@ -2001,14 +2179,13 @@ bfa_fcs_rport_alloc(struct bfa_fcs_lport_s *port, wwn_t pwwn, u32 rpid) rport->itnim = bfa_fcs_itnim_create(rport); if (!rport->itnim) { bfa_trc(fcs, rpid); - bfa_sm_send_event(rport->bfa_rport, - BFA_RPORT_SM_DELETE); kfree(rport_drv); return NULL; } } bfa_fcs_lport_add_rport(port, rport); + fcs->num_rport_logins++; bfa_sm_set_state(rport, bfa_fcs_rport_sm_uninit); @@ -2024,20 +2201,28 @@ static void bfa_fcs_rport_free(struct bfa_fcs_rport_s *rport) { struct bfa_fcs_lport_s *port = rport->port; + struct bfa_fcs_s *fcs = port->fcs; /* * - delete FC-4s * - delete BFA rport * - remove from queue of rports */ + rport->plogi_pending = BFA_FALSE; + if (bfa_fcs_lport_is_initiator(port)) { bfa_fcs_itnim_delete(rport->itnim); if (rport->pid != 0 && !BFA_FCS_PID_IS_WKA(rport->pid)) bfa_fcs_rpf_rport_offline(rport); } - bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_DELETE); + if (rport->bfa_rport) { + bfa_sm_send_event(rport->bfa_rport, BFA_RPORT_SM_DELETE); + rport->bfa_rport = NULL; + } + bfa_fcs_lport_del_rport(port, rport); + fcs->num_rport_logins--; kfree(rport->rp_drv); } @@ -2071,7 +2256,18 @@ bfa_fcs_rport_aen_post(struct bfa_fcs_rport_s *rport, } static void -bfa_fcs_rport_online_action(struct bfa_fcs_rport_s *rport) +bfa_fcs_rport_fcs_online_action(struct bfa_fcs_rport_s *rport) +{ + if ((!rport->pid) || (!rport->pwwn)) { + bfa_trc(rport->fcs, rport->pid); + bfa_sm_fault(rport->fcs, rport->pid); + } + + bfa_sm_send_event(rport->itnim, BFA_FCS_ITNIM_SM_FCS_ONLINE); +} + +static void +bfa_fcs_rport_hal_online_action(struct bfa_fcs_rport_s *rport) { struct bfa_fcs_lport_s *port = rport->port; struct bfad_s *bfad = (struct bfad_s *)port->fcs->bfad; @@ -2086,7 +2282,7 @@ bfa_fcs_rport_online_action(struct bfa_fcs_rport_s *rport) } if (bfa_fcs_lport_is_initiator(port)) { - bfa_fcs_itnim_rport_online(rport->itnim); + bfa_fcs_itnim_brp_online(rport->itnim); if (!BFA_FCS_PID_IS_WKA(rport->pid)) bfa_fcs_rpf_rport_online(rport); }; @@ -2102,15 +2298,28 @@ bfa_fcs_rport_online_action(struct bfa_fcs_rport_s *rport) } static void -bfa_fcs_rport_offline_action(struct bfa_fcs_rport_s *rport) +bfa_fcs_rport_fcs_offline_action(struct bfa_fcs_rport_s *rport) +{ + if (!BFA_FCS_PID_IS_WKA(rport->pid)) + bfa_fcs_rpf_rport_offline(rport); + + bfa_fcs_itnim_rport_offline(rport->itnim); +} + +static void +bfa_fcs_rport_hal_offline_action(struct bfa_fcs_rport_s *rport) { struct bfa_fcs_lport_s *port = rport->port; struct bfad_s *bfad = (struct bfad_s *)port->fcs->bfad; char lpwwn_buf[BFA_STRING_32]; char rpwwn_buf[BFA_STRING_32]; + if (!rport->bfa_rport) { + bfa_fcs_rport_fcs_offline_action(rport); + return; + } + rport->stats.offlines++; - rport->plogi_pending = BFA_FALSE; wwn2str(lpwwn_buf, bfa_fcs_lport_get_pwwn(port)); wwn2str(rpwwn_buf, rport->pwwn); @@ -2340,7 +2549,6 @@ bfa_fcs_rport_scn(struct bfa_fcs_rport_s *rport) bfa_sm_send_event(rport, RPSM_EVENT_SCN); } - /* * brief * This routine BFA callback for bfa_rport_online() call. @@ -2508,7 +2716,7 @@ bfa_fcs_rport_send_prlo_acc(struct bfa_fcs_rport_s *rport) bfa_trc(rport->fcs, rport->pid); - fcxp = bfa_fcs_fcxp_alloc(port->fcs); + fcxp = bfa_fcs_fcxp_alloc(port->fcs, BFA_FALSE); if (!fcxp) return; len = fc_prlo_acc_build(&fchs, bfa_fcxp_get_reqbuf(fcxp), @@ -2534,7 +2742,7 @@ bfa_fcs_rport_send_ls_rjt(struct bfa_fcs_rport_s *rport, struct fchs_s *rx_fchs, bfa_trc(rport->fcs, rx_fchs->s_id); - fcxp = bfa_fcs_fcxp_alloc(rport->fcs); + fcxp = bfa_fcs_fcxp_alloc(rport->fcs, BFA_FALSE); if (!fcxp) return; @@ -2582,6 +2790,17 @@ bfa_fcs_rport_prlo(struct bfa_fcs_rport_s *rport, __be16 ox_id) bfa_sm_send_event(rport, RPSM_EVENT_PRLO_RCVD); } +/* + * Called by BFAD to set the max limit on number of bfa_fcs_rport allocation + * which limits number of concurrent logins to remote ports + */ +void +bfa_fcs_rport_set_max_logins(u32 max_logins) +{ + if (max_logins > 0) + bfa_fcs_rport_max_logins = max_logins; +} + void bfa_fcs_rport_get_attr(struct bfa_fcs_rport_s *rport, struct bfa_rport_attr_s *rport_attr) @@ -2605,9 +2824,11 @@ bfa_fcs_rport_get_attr(struct bfa_fcs_rport_s *rport, rport_attr->curr_speed = rport->rpf.rpsc_speed; rport_attr->assigned_speed = rport->rpf.assigned_speed; - qos_attr.qos_priority = rport->bfa_rport->qos_attr.qos_priority; - qos_attr.qos_flow_id = - cpu_to_be32(rport->bfa_rport->qos_attr.qos_flow_id); + if (rport->bfa_rport) { + qos_attr.qos_priority = rport->bfa_rport->qos_attr.qos_priority; + qos_attr.qos_flow_id = + cpu_to_be32(rport->bfa_rport->qos_attr.qos_flow_id); + } rport_attr->qos_attr = qos_attr; rport_attr->trl_enforced = BFA_FALSE; @@ -2940,10 +3161,11 @@ bfa_fcs_rpf_send_rpsc2(void *rpf_cbarg, struct bfa_fcxp_s *fcxp_alloced) bfa_trc(rport->fcs, rport->pwwn); - fcxp = fcxp_alloced ? fcxp_alloced : bfa_fcs_fcxp_alloc(port->fcs); + fcxp = fcxp_alloced ? fcxp_alloced : + bfa_fcs_fcxp_alloc(port->fcs, BFA_TRUE); if (!fcxp) { bfa_fcs_fcxp_alloc_wait(port->fcs->bfa, &rpf->fcxp_wqe, - bfa_fcs_rpf_send_rpsc2, rpf); + bfa_fcs_rpf_send_rpsc2, rpf, BFA_TRUE); return; } rpf->fcxp = fcxp; diff --git a/drivers/scsi/bfa/bfa_ioc.c b/drivers/scsi/bfa/bfa_ioc.c index 21ad290..75ca875 100644 --- a/drivers/scsi/bfa/bfa_ioc.c +++ b/drivers/scsi/bfa/bfa_ioc.c @@ -92,7 +92,6 @@ static void bfa_ioc_event_notify(struct bfa_ioc_s *ioc , enum bfa_ioc_event_e event); static void bfa_ioc_disable_comp(struct bfa_ioc_s *ioc); static void bfa_ioc_lpu_stop(struct bfa_ioc_s *ioc); -static void bfa_ioc_debug_save_ftrc(struct bfa_ioc_s *ioc); static void bfa_ioc_fail_notify(struct bfa_ioc_s *ioc); static void bfa_ioc_pf_fwmismatch(struct bfa_ioc_s *ioc); @@ -599,8 +598,9 @@ bfa_ioc_sm_fail(struct bfa_ioc_s *ioc, enum ioc_event event) break; case IOC_E_HWERROR: + case IOC_E_HWFAILED: /* - * HB failure notification, ignore. + * HB failure / HW error notification, ignore. */ break; default: @@ -632,6 +632,10 @@ bfa_ioc_sm_hwfail(struct bfa_ioc_s *ioc, enum ioc_event event) bfa_fsm_set_state(ioc, bfa_ioc_sm_uninit); break; + case IOC_E_HWERROR: + /* Ignore - already in hwfail state */ + break; + default: bfa_sm_fault(ioc, event); } @@ -1455,7 +1459,7 @@ bfa_ioc_fwver_cmp(struct bfa_ioc_s *ioc, struct bfi_ioc_image_hdr_s *fwhdr) bfa_cb_image_get_chunk(bfa_ioc_asic_gen(ioc), 0); for (i = 0; i < BFI_IOC_MD5SUM_SZ; i++) { - if (fwhdr->md5sum[i] != drv_fwhdr->md5sum[i]) { + if (fwhdr->md5sum[i] != cpu_to_le32(drv_fwhdr->md5sum[i])) { bfa_trc(ioc, i); bfa_trc(ioc, fwhdr->md5sum[i]); bfa_trc(ioc, drv_fwhdr->md5sum[i]); @@ -1480,7 +1484,7 @@ bfa_ioc_fwver_valid(struct bfa_ioc_s *ioc, u32 boot_env) drv_fwhdr = (struct bfi_ioc_image_hdr_s *) bfa_cb_image_get_chunk(bfa_ioc_asic_gen(ioc), 0); - if (fwhdr.signature != drv_fwhdr->signature) { + if (fwhdr.signature != cpu_to_le32(drv_fwhdr->signature)) { bfa_trc(ioc, fwhdr.signature); bfa_trc(ioc, drv_fwhdr->signature); return BFA_FALSE; @@ -1704,7 +1708,7 @@ bfa_ioc_download_fw(struct bfa_ioc_s *ioc, u32 boot_type, * write smem */ bfa_mem_write(ioc->ioc_regs.smem_page_start, loff, - fwimg[BFA_IOC_FLASH_OFFSET_IN_CHUNK(i)]); + cpu_to_le32(fwimg[BFA_IOC_FLASH_OFFSET_IN_CHUNK(i)])); loff += sizeof(u32); @@ -2260,6 +2264,12 @@ bfa_ioc_disable(struct bfa_ioc_s *ioc) bfa_fsm_send_event(ioc, IOC_E_DISABLE); } +void +bfa_ioc_suspend(struct bfa_ioc_s *ioc) +{ + ioc->dbg_fwsave_once = BFA_TRUE; + bfa_fsm_send_event(ioc, IOC_E_HWERROR); +} /* * Initialize memory for saving firmware trace. Driver must initialize @@ -2269,7 +2279,7 @@ void bfa_ioc_debug_memclaim(struct bfa_ioc_s *ioc, void *dbg_fwsave) { ioc->dbg_fwsave = dbg_fwsave; - ioc->dbg_fwsave_len = (ioc->iocpf.auto_recover) ? BFA_DBG_FWTRC_LEN : 0; + ioc->dbg_fwsave_len = BFA_DBG_FWTRC_LEN; } /* @@ -2856,7 +2866,7 @@ bfa_ioc_fw_stats_clear(struct bfa_ioc_s *ioc) /* * Save firmware trace if configured. */ -static void +void bfa_ioc_debug_save_ftrc(struct bfa_ioc_s *ioc) { int tlen; diff --git a/drivers/scsi/bfa/bfa_ioc.h b/drivers/scsi/bfa/bfa_ioc.h index 7b916e0..b2856f9 100644 --- a/drivers/scsi/bfa/bfa_ioc.h +++ b/drivers/scsi/bfa/bfa_ioc.h @@ -820,6 +820,7 @@ void bfa_ioc_attach(struct bfa_ioc_s *ioc, void *bfa, struct bfa_ioc_cbfn_s *cbfn, struct bfa_timer_mod_s *timer_mod); void bfa_ioc_auto_recover(bfa_boolean_t auto_recover); void bfa_ioc_detach(struct bfa_ioc_s *ioc); +void bfa_ioc_suspend(struct bfa_ioc_s *ioc); void bfa_ioc_pci_init(struct bfa_ioc_s *ioc, struct bfa_pcidev_s *pcidev, enum bfi_pcifn_class clscode); void bfa_ioc_mem_claim(struct bfa_ioc_s *ioc, u8 *dm_kva, u64 dm_pa); @@ -866,6 +867,7 @@ bfa_boolean_t bfa_ioc_fwver_cmp(struct bfa_ioc_s *ioc, void bfa_ioc_aen_post(struct bfa_ioc_s *ioc, enum bfa_ioc_aen_event event); bfa_status_t bfa_ioc_fw_stats_get(struct bfa_ioc_s *ioc, void *stats); bfa_status_t bfa_ioc_fw_stats_clear(struct bfa_ioc_s *ioc); +void bfa_ioc_debug_save_ftrc(struct bfa_ioc_s *ioc); /* * asic block configuration related APIs diff --git a/drivers/scsi/bfa/bfa_modules.h b/drivers/scsi/bfa/bfa_modules.h index 2d36e48..189fff7 100644 --- a/drivers/scsi/bfa/bfa_modules.h +++ b/drivers/scsi/bfa/bfa_modules.h @@ -121,6 +121,7 @@ struct bfa_s { bfa_boolean_t fcs; /* FCS is attached to BFA */ struct bfa_msix_s msix; int bfa_aen_seq; + bfa_boolean_t intr_enabled; /* Status of interrupts */ }; extern bfa_boolean_t bfa_auto_recover; diff --git a/drivers/scsi/bfa/bfa_svc.c b/drivers/scsi/bfa/bfa_svc.c index 2e856e6..b2538d6 100644 --- a/drivers/scsi/bfa/bfa_svc.c +++ b/drivers/scsi/bfa/bfa_svc.c @@ -440,9 +440,11 @@ claim_fcxps_mem(struct bfa_fcxp_mod_s *mod) fcxp = (struct bfa_fcxp_s *) bfa_mem_kva_curp(mod); memset(fcxp, 0, sizeof(struct bfa_fcxp_s) * mod->num_fcxps); - INIT_LIST_HEAD(&mod->fcxp_free_q); + INIT_LIST_HEAD(&mod->fcxp_req_free_q); + INIT_LIST_HEAD(&mod->fcxp_rsp_free_q); INIT_LIST_HEAD(&mod->fcxp_active_q); - INIT_LIST_HEAD(&mod->fcxp_unused_q); + INIT_LIST_HEAD(&mod->fcxp_req_unused_q); + INIT_LIST_HEAD(&mod->fcxp_rsp_unused_q); mod->fcxp_list = fcxp; @@ -450,7 +452,14 @@ claim_fcxps_mem(struct bfa_fcxp_mod_s *mod) fcxp->fcxp_mod = mod; fcxp->fcxp_tag = i; - list_add_tail(&fcxp->qe, &mod->fcxp_free_q); + if (i < (mod->num_fcxps / 2)) { + list_add_tail(&fcxp->qe, &mod->fcxp_req_free_q); + fcxp->req_rsp = BFA_TRUE; + } else { + list_add_tail(&fcxp->qe, &mod->fcxp_rsp_free_q); + fcxp->req_rsp = BFA_FALSE; + } + bfa_reqq_winit(&fcxp->reqq_wqe, bfa_fcxp_qresume, fcxp); fcxp->reqq_waiting = BFA_FALSE; @@ -514,7 +523,8 @@ bfa_fcxp_attach(struct bfa_s *bfa, void *bfad, struct bfa_iocfc_cfg_s *cfg, if (!cfg->drvcfg.min_cfg) mod->rsp_pld_sz = BFA_FCXP_MAX_LBUF_SZ; - INIT_LIST_HEAD(&mod->wait_q); + INIT_LIST_HEAD(&mod->req_wait_q); + INIT_LIST_HEAD(&mod->rsp_wait_q); claim_fcxps_mem(mod); } @@ -542,7 +552,8 @@ bfa_fcxp_iocdisable(struct bfa_s *bfa) struct list_head *qe, *qen; /* Enqueue unused fcxp resources to free_q */ - list_splice_tail_init(&mod->fcxp_unused_q, &mod->fcxp_free_q); + list_splice_tail_init(&mod->fcxp_req_unused_q, &mod->fcxp_req_free_q); + list_splice_tail_init(&mod->fcxp_rsp_unused_q, &mod->fcxp_rsp_free_q); list_for_each_safe(qe, qen, &mod->fcxp_active_q) { fcxp = (struct bfa_fcxp_s *) qe; @@ -559,11 +570,14 @@ bfa_fcxp_iocdisable(struct bfa_s *bfa) } static struct bfa_fcxp_s * -bfa_fcxp_get(struct bfa_fcxp_mod_s *fm) +bfa_fcxp_get(struct bfa_fcxp_mod_s *fm, bfa_boolean_t req) { struct bfa_fcxp_s *fcxp; - bfa_q_deq(&fm->fcxp_free_q, &fcxp); + if (req) + bfa_q_deq(&fm->fcxp_req_free_q, &fcxp); + else + bfa_q_deq(&fm->fcxp_rsp_free_q, &fcxp); if (fcxp) list_add_tail(&fcxp->qe, &fm->fcxp_active_q); @@ -642,7 +656,11 @@ bfa_fcxp_put(struct bfa_fcxp_s *fcxp) struct bfa_fcxp_mod_s *mod = fcxp->fcxp_mod; struct bfa_fcxp_wqe_s *wqe; - bfa_q_deq(&mod->wait_q, &wqe); + if (fcxp->req_rsp) + bfa_q_deq(&mod->req_wait_q, &wqe); + else + bfa_q_deq(&mod->rsp_wait_q, &wqe); + if (wqe) { bfa_trc(mod->bfa, fcxp->fcxp_tag); @@ -657,7 +675,11 @@ bfa_fcxp_put(struct bfa_fcxp_s *fcxp) WARN_ON(!bfa_q_is_on_q(&mod->fcxp_active_q, fcxp)); list_del(&fcxp->qe); - list_add_tail(&fcxp->qe, &mod->fcxp_free_q); + + if (fcxp->req_rsp) + list_add_tail(&fcxp->qe, &mod->fcxp_req_free_q); + else + list_add_tail(&fcxp->qe, &mod->fcxp_rsp_free_q); } static void @@ -900,21 +922,23 @@ bfa_fcxp_queue(struct bfa_fcxp_s *fcxp, struct bfi_fcxp_send_req_s *send_req) * Address (given the sge index). * @param[in] get_rsp_sglen function ptr to be called to get a response SG * len (given the sge index). + * @param[in] req Allocated FCXP is used to send req or rsp? + * request - BFA_TRUE, response - BFA_FALSE * * @return FCXP instance. NULL on failure. */ struct bfa_fcxp_s * -bfa_fcxp_alloc(void *caller, struct bfa_s *bfa, int nreq_sgles, - int nrsp_sgles, bfa_fcxp_get_sgaddr_t req_sga_cbfn, - bfa_fcxp_get_sglen_t req_sglen_cbfn, - bfa_fcxp_get_sgaddr_t rsp_sga_cbfn, - bfa_fcxp_get_sglen_t rsp_sglen_cbfn) +bfa_fcxp_req_rsp_alloc(void *caller, struct bfa_s *bfa, int nreq_sgles, + int nrsp_sgles, bfa_fcxp_get_sgaddr_t req_sga_cbfn, + bfa_fcxp_get_sglen_t req_sglen_cbfn, + bfa_fcxp_get_sgaddr_t rsp_sga_cbfn, + bfa_fcxp_get_sglen_t rsp_sglen_cbfn, bfa_boolean_t req) { struct bfa_fcxp_s *fcxp = NULL; WARN_ON(bfa == NULL); - fcxp = bfa_fcxp_get(BFA_FCXP_MOD(bfa)); + fcxp = bfa_fcxp_get(BFA_FCXP_MOD(bfa), req); if (fcxp == NULL) return NULL; @@ -1071,17 +1095,20 @@ bfa_fcxp_abort(struct bfa_fcxp_s *fcxp) } void -bfa_fcxp_alloc_wait(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe, +bfa_fcxp_req_rsp_alloc_wait(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe, bfa_fcxp_alloc_cbfn_t alloc_cbfn, void *alloc_cbarg, void *caller, int nreq_sgles, int nrsp_sgles, bfa_fcxp_get_sgaddr_t req_sga_cbfn, bfa_fcxp_get_sglen_t req_sglen_cbfn, bfa_fcxp_get_sgaddr_t rsp_sga_cbfn, - bfa_fcxp_get_sglen_t rsp_sglen_cbfn) + bfa_fcxp_get_sglen_t rsp_sglen_cbfn, bfa_boolean_t req) { struct bfa_fcxp_mod_s *mod = BFA_FCXP_MOD(bfa); - WARN_ON(!list_empty(&mod->fcxp_free_q)); + if (req) + WARN_ON(!list_empty(&mod->fcxp_req_free_q)); + else + WARN_ON(!list_empty(&mod->fcxp_rsp_free_q)); wqe->alloc_cbfn = alloc_cbfn; wqe->alloc_cbarg = alloc_cbarg; @@ -1094,7 +1121,10 @@ bfa_fcxp_alloc_wait(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe, wqe->rsp_sga_cbfn = rsp_sga_cbfn; wqe->rsp_sglen_cbfn = rsp_sglen_cbfn; - list_add_tail(&wqe->qe, &mod->wait_q); + if (req) + list_add_tail(&wqe->qe, &mod->req_wait_q); + else + list_add_tail(&wqe->qe, &mod->rsp_wait_q); } void @@ -1102,7 +1132,8 @@ bfa_fcxp_walloc_cancel(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe) { struct bfa_fcxp_mod_s *mod = BFA_FCXP_MOD(bfa); - WARN_ON(!bfa_q_is_on_q(&mod->wait_q, wqe)); + WARN_ON(!bfa_q_is_on_q(&mod->req_wait_q, wqe) || + !bfa_q_is_on_q(&mod->rsp_wait_q, wqe)); list_del(&wqe->qe); } @@ -1153,8 +1184,13 @@ bfa_fcxp_res_recfg(struct bfa_s *bfa, u16 num_fcxp_fw) int i; for (i = 0; i < (mod->num_fcxps - num_fcxp_fw); i++) { - bfa_q_deq_tail(&mod->fcxp_free_q, &qe); - list_add_tail(qe, &mod->fcxp_unused_q); + if (i < ((mod->num_fcxps - num_fcxp_fw) / 2)) { + bfa_q_deq_tail(&mod->fcxp_req_free_q, &qe); + list_add_tail(qe, &mod->fcxp_req_unused_q); + } else { + bfa_q_deq_tail(&mod->fcxp_rsp_free_q, &qe); + list_add_tail(qe, &mod->fcxp_rsp_unused_q); + } } } @@ -1404,11 +1440,11 @@ bfa_lps_sm_logout(struct bfa_lps_s *lps, enum bfa_lps_event event) switch (event) { case BFA_LPS_SM_FWRSP: + case BFA_LPS_SM_OFFLINE: bfa_sm_set_state(lps, bfa_lps_sm_init); bfa_lps_logout_comp(lps); break; - case BFA_LPS_SM_OFFLINE: case BFA_LPS_SM_DELETE: bfa_sm_set_state(lps, bfa_lps_sm_init); break; @@ -1786,6 +1822,8 @@ bfa_lps_logout_comp_cb(void *arg, bfa_boolean_t complete) if (lps->fdisc) bfa_cb_lps_fdisclogo_comp(lps->bfa->bfad, lps->uarg); + else + bfa_cb_lps_flogo_comp(lps->bfa->bfad, lps->uarg); } /* @@ -4237,6 +4275,10 @@ bfa_rport_sm_offline(struct bfa_rport_s *rp, enum bfa_rport_event event) bfa_sm_set_state(rp, bfa_rport_sm_iocdisable); break; + case BFA_RPORT_SM_OFFLINE: + bfa_rport_offline_cb(rp); + break; + default: bfa_stats(rp, sm_off_unexp); bfa_sm_fault(rp->bfa, event); @@ -4353,6 +4395,7 @@ bfa_rport_sm_offline_pending(struct bfa_rport_s *rp, case BFA_RPORT_SM_HWFAIL: bfa_stats(rp, sm_offp_hwf); bfa_sm_set_state(rp, bfa_rport_sm_iocdisable); + bfa_rport_offline_cb(rp); break; default: @@ -4731,8 +4774,10 @@ bfa_rport_speed(struct bfa_rport_s *rport, enum bfa_port_speed speed) WARN_ON(speed == 0); WARN_ON(speed == BFA_PORT_SPEED_AUTO); - rport->rport_info.speed = speed; - bfa_sm_send_event(rport, BFA_RPORT_SM_SET_SPEED); + if (rport) { + rport->rport_info.speed = speed; + bfa_sm_send_event(rport, BFA_RPORT_SM_SET_SPEED); + } } /* Set Rport LUN Mask */ diff --git a/drivers/scsi/bfa/bfa_svc.h b/drivers/scsi/bfa/bfa_svc.h index f300675..1abcf7c 100644 --- a/drivers/scsi/bfa/bfa_svc.h +++ b/drivers/scsi/bfa/bfa_svc.h @@ -97,10 +97,13 @@ struct bfa_fcxp_mod_s { struct bfa_s *bfa; /* backpointer to BFA */ struct bfa_fcxp_s *fcxp_list; /* array of FCXPs */ u16 num_fcxps; /* max num FCXP requests */ - struct list_head fcxp_free_q; /* free FCXPs */ - struct list_head fcxp_active_q; /* active FCXPs */ - struct list_head wait_q; /* wait queue for free fcxp */ - struct list_head fcxp_unused_q; /* unused fcxps */ + struct list_head fcxp_req_free_q; /* free FCXPs used for sending req */ + struct list_head fcxp_rsp_free_q; /* free FCXPs used for sending req */ + struct list_head fcxp_active_q; /* active FCXPs */ + struct list_head req_wait_q; /* wait queue for free req_fcxp */ + struct list_head rsp_wait_q; /* wait queue for free rsp_fcxp */ + struct list_head fcxp_req_unused_q; /* unused req_fcxps */ + struct list_head fcxp_rsp_unused_q; /* unused rsp_fcxps */ u32 req_pld_sz; u32 rsp_pld_sz; struct bfa_mem_dma_s dma_seg[BFA_FCXP_DMA_SEGS]; @@ -197,6 +200,7 @@ struct bfa_fcxp_s { struct bfa_cb_qe_s hcb_qe; /* comp: callback qelem */ struct bfa_reqq_wait_s reqq_wqe; bfa_boolean_t reqq_waiting; + bfa_boolean_t req_rsp; /* Used to track req/rsp fcxp */ }; struct bfa_fcxp_wqe_s { @@ -586,20 +590,22 @@ void bfa_rport_unset_lunmask(struct bfa_s *bfa, struct bfa_rport_s *rp); /* * bfa fcxp API functions */ -struct bfa_fcxp_s *bfa_fcxp_alloc(void *bfad_fcxp, struct bfa_s *bfa, +struct bfa_fcxp_s *bfa_fcxp_req_rsp_alloc(void *bfad_fcxp, struct bfa_s *bfa, int nreq_sgles, int nrsp_sgles, bfa_fcxp_get_sgaddr_t get_req_sga, bfa_fcxp_get_sglen_t get_req_sglen, bfa_fcxp_get_sgaddr_t get_rsp_sga, - bfa_fcxp_get_sglen_t get_rsp_sglen); -void bfa_fcxp_alloc_wait(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe, + bfa_fcxp_get_sglen_t get_rsp_sglen, + bfa_boolean_t req); +void bfa_fcxp_req_rsp_alloc_wait(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe, bfa_fcxp_alloc_cbfn_t alloc_cbfn, void *cbarg, void *bfad_fcxp, int nreq_sgles, int nrsp_sgles, bfa_fcxp_get_sgaddr_t get_req_sga, bfa_fcxp_get_sglen_t get_req_sglen, bfa_fcxp_get_sgaddr_t get_rsp_sga, - bfa_fcxp_get_sglen_t get_rsp_sglen); + bfa_fcxp_get_sglen_t get_rsp_sglen, + bfa_boolean_t req); void bfa_fcxp_walloc_cancel(struct bfa_s *bfa, struct bfa_fcxp_wqe_s *wqe); void bfa_fcxp_discard(struct bfa_fcxp_s *fcxp); @@ -658,6 +664,7 @@ u8 bfa_lps_get_fwtag(struct bfa_s *bfa, u8 lp_tag); u32 bfa_lps_get_base_pid(struct bfa_s *bfa); u8 bfa_lps_get_tag_from_pid(struct bfa_s *bfa, u32 pid); void bfa_cb_lps_flogi_comp(void *bfad, void *uarg, bfa_status_t status); +void bfa_cb_lps_flogo_comp(void *bfad, void *uarg); void bfa_cb_lps_fdisc_comp(void *bfad, void *uarg, bfa_status_t status); void bfa_cb_lps_fdisclogo_comp(void *bfad, void *uarg); void bfa_cb_lps_cvl_event(void *bfad, void *uarg); diff --git a/drivers/scsi/bfa/bfad.c b/drivers/scsi/bfa/bfad.c index 2c8f0c7..c374949 100644 --- a/drivers/scsi/bfa/bfad.c +++ b/drivers/scsi/bfa/bfad.c @@ -57,6 +57,7 @@ int pcie_max_read_reqsz; int bfa_debugfs_enable = 1; int msix_disable_cb = 0, msix_disable_ct = 0; int max_xfer_size = BFAD_MAX_SECTORS >> 1; +int max_rport_logins = BFA_FCS_MAX_RPORT_LOGINS; /* Firmware releated */ u32 bfi_image_cb_size, bfi_image_ct_size, bfi_image_ct2_size; @@ -148,6 +149,8 @@ MODULE_PARM_DESC(bfa_debugfs_enable, "Enables debugfs feature, default=1," module_param(max_xfer_size, int, S_IRUGO | S_IWUSR); MODULE_PARM_DESC(max_xfer_size, "default=32MB," " Range[64k|128k|256k|512k|1024k|2048k]"); +module_param(max_rport_logins, int, S_IRUGO | S_IWUSR); +MODULE_PARM_DESC(max_rport_logins, "Max number of logins to initiator and target rports on a port (physical/logical), default=1024"); static void bfad_sm_uninit(struct bfad_s *bfad, enum bfad_sm_event event); @@ -736,6 +739,9 @@ bfad_pci_init(struct pci_dev *pdev, struct bfad_s *bfad) } } + /* Enable PCIE Advanced Error Recovery (AER) if kernel supports */ + pci_enable_pcie_error_reporting(pdev); + bfad->pci_bar0_kva = pci_iomap(pdev, 0, pci_resource_len(pdev, 0)); bfad->pci_bar2_kva = pci_iomap(pdev, 2, pci_resource_len(pdev, 2)); @@ -806,6 +812,8 @@ bfad_pci_init(struct pci_dev *pdev, struct bfad_s *bfad) } } + pci_save_state(pdev); + return 0; out_release_region: @@ -822,6 +830,8 @@ bfad_pci_uninit(struct pci_dev *pdev, struct bfad_s *bfad) pci_iounmap(pdev, bfad->pci_bar0_kva); pci_iounmap(pdev, bfad->pci_bar2_kva); pci_release_regions(pdev); + /* Disable PCIE Advanced Error Recovery (AER) */ + pci_disable_pcie_error_reporting(pdev); pci_disable_device(pdev); pci_set_drvdata(pdev, NULL); } @@ -1258,6 +1268,16 @@ bfad_setup_intr(struct bfad_s *bfad) error = pci_enable_msix(bfad->pcidev, msix_entries, bfad->nvec); if (error) { + /* In CT1 & CT2, try to allocate just one vector */ + if (bfa_asic_id_ctc(pdev->device)) { + printk(KERN_WARNING "bfa %s: trying one msix " + "vector failed to allocate %d[%d]\n", + bfad->pci_name, bfad->nvec, error); + bfad->nvec = 1; + error = pci_enable_msix(bfad->pcidev, + msix_entries, bfad->nvec); + } + /* * Only error number of vector is available. * We don't have a mechanism to map multiple @@ -1267,12 +1287,13 @@ bfad_setup_intr(struct bfad_s *bfad) * vectors. Linux doesn't duplicate vectors * in the MSIX table for this case. */ - - printk(KERN_WARNING "bfad%d: " - "pci_enable_msix failed (%d)," - " use line based.\n", bfad->inst_no, error); - - goto line_based; + if (error) { + printk(KERN_WARNING "bfad%d: " + "pci_enable_msix failed (%d), " + "use line based.\n", + bfad->inst_no, error); + goto line_based; + } } /* Disable INTX in MSI-X mode */ @@ -1470,6 +1491,197 @@ bfad_pci_remove(struct pci_dev *pdev) kfree(bfad); } +/* + * PCI Error Recovery entry, error detected. + */ +static pci_ers_result_t +bfad_pci_error_detected(struct pci_dev *pdev, pci_channel_state_t state) +{ + struct bfad_s *bfad = pci_get_drvdata(pdev); + unsigned long flags; + pci_ers_result_t ret = PCI_ERS_RESULT_NONE; + + dev_printk(KERN_ERR, &pdev->dev, + "error detected state: %d - flags: 0x%x\n", + state, bfad->bfad_flags); + + switch (state) { + case pci_channel_io_normal: /* non-fatal error */ + spin_lock_irqsave(&bfad->bfad_lock, flags); + bfad->bfad_flags &= ~BFAD_EEH_BUSY; + /* Suspend/fail all bfa operations */ + bfa_ioc_suspend(&bfad->bfa.ioc); + spin_unlock_irqrestore(&bfad->bfad_lock, flags); + del_timer_sync(&bfad->hal_tmo); + ret = PCI_ERS_RESULT_CAN_RECOVER; + break; + case pci_channel_io_frozen: /* fatal error */ + init_completion(&bfad->comp); + spin_lock_irqsave(&bfad->bfad_lock, flags); + bfad->bfad_flags |= BFAD_EEH_BUSY; + /* Suspend/fail all bfa operations */ + bfa_ioc_suspend(&bfad->bfa.ioc); + bfa_fcs_stop(&bfad->bfa_fcs); + spin_unlock_irqrestore(&bfad->bfad_lock, flags); + wait_for_completion(&bfad->comp); + + bfad_remove_intr(bfad); + del_timer_sync(&bfad->hal_tmo); + pci_disable_device(pdev); + ret = PCI_ERS_RESULT_NEED_RESET; + break; + case pci_channel_io_perm_failure: /* PCI Card is DEAD */ + spin_lock_irqsave(&bfad->bfad_lock, flags); + bfad->bfad_flags |= BFAD_EEH_BUSY | + BFAD_EEH_PCI_CHANNEL_IO_PERM_FAILURE; + spin_unlock_irqrestore(&bfad->bfad_lock, flags); + + /* If the error_detected handler is called with the reason + * pci_channel_io_perm_failure - it will subsequently call + * pci_remove() entry point to remove the pci device from the + * system - So defer the cleanup to pci_remove(); cleaning up + * here causes inconsistent state during pci_remove(). + */ + ret = PCI_ERS_RESULT_DISCONNECT; + break; + default: + WARN_ON(1); + } + + return ret; +} + +int +restart_bfa(struct bfad_s *bfad) +{ + unsigned long flags; + struct pci_dev *pdev = bfad->pcidev; + + bfa_attach(&bfad->bfa, bfad, &bfad->ioc_cfg, + &bfad->meminfo, &bfad->hal_pcidev); + + /* Enable Interrupt and wait bfa_init completion */ + if (bfad_setup_intr(bfad)) { + dev_printk(KERN_WARNING, &pdev->dev, + "%s: bfad_setup_intr failed\n", bfad->pci_name); + bfa_sm_send_event(bfad, BFAD_E_INTR_INIT_FAILED); + return -1; + } + + init_completion(&bfad->comp); + spin_lock_irqsave(&bfad->bfad_lock, flags); + bfa_iocfc_init(&bfad->bfa); + spin_unlock_irqrestore(&bfad->bfad_lock, flags); + + /* Set up interrupt handler for each vectors */ + if ((bfad->bfad_flags & BFAD_MSIX_ON) && + bfad_install_msix_handler(bfad)) + dev_printk(KERN_WARNING, &pdev->dev, + "%s: install_msix failed.\n", bfad->pci_name); + + bfad_init_timer(bfad); + wait_for_completion(&bfad->comp); + bfad_drv_start(bfad); + + return 0; +} + +/* + * PCI Error Recovery entry, re-initialize the chip. + */ +static pci_ers_result_t +bfad_pci_slot_reset(struct pci_dev *pdev) +{ + struct bfad_s *bfad = pci_get_drvdata(pdev); + u8 byte; + + dev_printk(KERN_ERR, &pdev->dev, + "bfad_pci_slot_reset flags: 0x%x\n", bfad->bfad_flags); + + if (pci_enable_device(pdev)) { + dev_printk(KERN_ERR, &pdev->dev, "Cannot re-enable " + "PCI device after reset.\n"); + return PCI_ERS_RESULT_DISCONNECT; + } + + pci_restore_state(pdev); + + /* + * Read some byte (e.g. DMA max. payload size which can't + * be 0xff any time) to make sure - we did not hit another PCI error + * in the middle of recovery. If we did, then declare permanent failure. + */ + pci_read_config_byte(pdev, 0x68, &byte); + if (byte == 0xff) { + dev_printk(KERN_ERR, &pdev->dev, + "slot_reset failed ... got another PCI error !\n"); + goto out_disable_device; + } + + pci_save_state(pdev); + pci_set_master(pdev); + + if (pci_set_dma_mask(bfad->pcidev, DMA_BIT_MASK(64)) != 0) + if (pci_set_dma_mask(bfad->pcidev, DMA_BIT_MASK(32)) != 0) + goto out_disable_device; + + pci_cleanup_aer_uncorrect_error_status(pdev); + + if (restart_bfa(bfad) == -1) + goto out_disable_device; + + pci_enable_pcie_error_reporting(pdev); + dev_printk(KERN_WARNING, &pdev->dev, + "slot_reset completed flags: 0x%x!\n", bfad->bfad_flags); + + return PCI_ERS_RESULT_RECOVERED; + +out_disable_device: + pci_disable_device(pdev); + return PCI_ERS_RESULT_DISCONNECT; +} + +static pci_ers_result_t +bfad_pci_mmio_enabled(struct pci_dev *pdev) +{ + unsigned long flags; + struct bfad_s *bfad = pci_get_drvdata(pdev); + + dev_printk(KERN_INFO, &pdev->dev, "mmio_enabled\n"); + + /* Fetch FW diagnostic information */ + bfa_ioc_debug_save_ftrc(&bfad->bfa.ioc); + + /* Cancel all pending IOs */ + spin_lock_irqsave(&bfad->bfad_lock, flags); + init_completion(&bfad->comp); + bfa_fcs_stop(&bfad->bfa_fcs); + spin_unlock_irqrestore(&bfad->bfad_lock, flags); + wait_for_completion(&bfad->comp); + + bfad_remove_intr(bfad); + del_timer_sync(&bfad->hal_tmo); + pci_disable_device(pdev); + + return PCI_ERS_RESULT_NEED_RESET; +} + +static void +bfad_pci_resume(struct pci_dev *pdev) +{ + unsigned long flags; + struct bfad_s *bfad = pci_get_drvdata(pdev); + + dev_printk(KERN_WARNING, &pdev->dev, "resume\n"); + + /* wait until the link is online */ + bfad_rport_online_wait(bfad); + + spin_lock_irqsave(&bfad->bfad_lock, flags); + bfad->bfad_flags &= ~BFAD_EEH_BUSY; + spin_unlock_irqrestore(&bfad->bfad_lock, flags); +} + struct pci_device_id bfad_id_table[] = { { .vendor = BFA_PCI_VENDOR_ID_BROCADE, @@ -1513,11 +1725,22 @@ struct pci_device_id bfad_id_table[] = { MODULE_DEVICE_TABLE(pci, bfad_id_table); +/* + * PCI error recovery handlers. + */ +static struct pci_error_handlers bfad_err_handler = { + .error_detected = bfad_pci_error_detected, + .slot_reset = bfad_pci_slot_reset, + .mmio_enabled = bfad_pci_mmio_enabled, + .resume = bfad_pci_resume, +}; + static struct pci_driver bfad_pci_driver = { .name = BFAD_DRIVER_NAME, .id_table = bfad_id_table, .probe = bfad_pci_probe, .remove = __devexit_p(bfad_pci_remove), + .err_handler = &bfad_err_handler, }; /* @@ -1546,6 +1769,7 @@ bfad_init(void) bfa_auto_recover = ioc_auto_recover; bfa_fcs_rport_set_del_timeout(rport_del_timeout); + bfa_fcs_rport_set_max_logins(max_rport_logins); error = pci_register_driver(&bfad_pci_driver); if (error) { diff --git a/drivers/scsi/bfa/bfad_attr.c b/drivers/scsi/bfa/bfad_attr.c index b839274..72f5dc3 100644 --- a/drivers/scsi/bfa/bfad_attr.c +++ b/drivers/scsi/bfa/bfad_attr.c @@ -587,6 +587,37 @@ bfad_im_vport_disable(struct fc_vport *fc_vport, bool disable) return 0; } +void +bfad_im_vport_set_symbolic_name(struct fc_vport *fc_vport) +{ + struct bfad_vport_s *vport = (struct bfad_vport_s *)fc_vport->dd_data; + struct bfad_im_port_s *im_port = + (struct bfad_im_port_s *)vport->drv_port.im_port; + struct bfad_s *bfad = im_port->bfad; + struct Scsi_Host *vshost = vport->drv_port.im_port->shost; + char *sym_name = fc_vport->symbolic_name; + struct bfa_fcs_vport_s *fcs_vport; + wwn_t pwwn; + unsigned long flags; + + u64_to_wwn(fc_host_port_name(vshost), (u8 *)&pwwn); + + spin_lock_irqsave(&bfad->bfad_lock, flags); + fcs_vport = bfa_fcs_vport_lookup(&bfad->bfa_fcs, 0, pwwn); + spin_unlock_irqrestore(&bfad->bfad_lock, flags); + + if (fcs_vport == NULL) + return; + + spin_lock_irqsave(&bfad->bfad_lock, flags); + if (strlen(sym_name) > 0) { + strcpy(fcs_vport->lport.port_cfg.sym_name.symname, sym_name); + bfa_fcs_lport_ns_util_send_rspn_id( + BFA_FCS_GET_NS_FROM_PORT((&fcs_vport->lport)), NULL); + } + spin_unlock_irqrestore(&bfad->bfad_lock, flags); +} + struct fc_function_template bfad_im_fc_function_template = { /* Target dynamic attributes */ @@ -640,6 +671,7 @@ struct fc_function_template bfad_im_fc_function_template = { .vport_create = bfad_im_vport_create, .vport_delete = bfad_im_vport_delete, .vport_disable = bfad_im_vport_disable, + .set_vport_symbolic_name = bfad_im_vport_set_symbolic_name, .bsg_request = bfad_im_bsg_request, .bsg_timeout = bfad_im_bsg_timeout, }; @@ -792,6 +824,13 @@ bfad_im_model_desc_show(struct device *dev, struct device_attribute *attr, else if (nports == 2 && !bfa_ioc_is_cna(&bfad->bfa.ioc)) snprintf(model_descr, BFA_ADAPTER_MODEL_DESCR_LEN, "Brocade 16Gbps PCIe dual port FC HBA"); + } else if (!strcmp(model, "Brocade-1867")) { + if (nports == 1 && !bfa_ioc_is_cna(&bfad->bfa.ioc)) + snprintf(model_descr, BFA_ADAPTER_MODEL_DESCR_LEN, + "Brocade 16Gbps PCIe single port FC HBA for IBM"); + else if (nports == 2 && !bfa_ioc_is_cna(&bfad->bfa.ioc)) + snprintf(model_descr, BFA_ADAPTER_MODEL_DESCR_LEN, + "Brocade 16Gbps PCIe dual port FC HBA for IBM"); } else snprintf(model_descr, BFA_ADAPTER_MODEL_DESCR_LEN, "Invalid Model"); @@ -909,15 +948,16 @@ bfad_im_num_of_discovered_ports_show(struct device *dev, struct bfad_port_s *port = im_port->port; struct bfad_s *bfad = im_port->bfad; int nrports = 2048; - wwn_t *rports = NULL; + struct bfa_rport_qualifier_s *rports = NULL; unsigned long flags; - rports = kzalloc(sizeof(wwn_t) * nrports , GFP_ATOMIC); + rports = kzalloc(sizeof(struct bfa_rport_qualifier_s) * nrports, + GFP_ATOMIC); if (rports == NULL) return snprintf(buf, PAGE_SIZE, "Failed\n"); spin_lock_irqsave(&bfad->bfad_lock, flags); - bfa_fcs_lport_get_rports(port->fcs_port, rports, &nrports); + bfa_fcs_lport_get_rport_quals(port->fcs_port, rports, &nrports); spin_unlock_irqrestore(&bfad->bfad_lock, flags); kfree(rports); diff --git a/drivers/scsi/bfa/bfad_bsg.c b/drivers/scsi/bfa/bfad_bsg.c index 9c1495b3..0afa390 100644 --- a/drivers/scsi/bfa/bfad_bsg.c +++ b/drivers/scsi/bfa/bfad_bsg.c @@ -535,7 +535,8 @@ bfad_iocmd_lport_get_rports(struct bfad_s *bfad, void *cmd, if (bfad_chk_iocmd_sz(payload_len, sizeof(struct bfa_bsg_lport_get_rports_s), - sizeof(wwn_t) * iocmd->nrports) != BFA_STATUS_OK) { + sizeof(struct bfa_rport_qualifier_s) * iocmd->nrports) + != BFA_STATUS_OK) { iocmd->status = BFA_STATUS_VERSION_FAIL; return 0; } @@ -552,8 +553,9 @@ bfad_iocmd_lport_get_rports(struct bfad_s *bfad, void *cmd, goto out; } - bfa_fcs_lport_get_rports(fcs_port, (wwn_t *)iocmd_bufptr, - &iocmd->nrports); + bfa_fcs_lport_get_rport_quals(fcs_port, + (struct bfa_rport_qualifier_s *)iocmd_bufptr, + &iocmd->nrports); spin_unlock_irqrestore(&bfad->bfad_lock, flags); iocmd->status = BFA_STATUS_OK; out: @@ -578,7 +580,11 @@ bfad_iocmd_rport_get_attr(struct bfad_s *bfad, void *cmd) goto out; } - fcs_rport = bfa_fcs_rport_lookup(fcs_port, iocmd->rpwwn); + if (iocmd->pid) + fcs_rport = bfa_fcs_lport_get_rport_by_qualifier(fcs_port, + iocmd->rpwwn, iocmd->pid); + else + fcs_rport = bfa_fcs_rport_lookup(fcs_port, iocmd->rpwwn); if (fcs_rport == NULL) { bfa_trc(bfad, 0); spin_unlock_irqrestore(&bfad->bfad_lock, flags); @@ -671,9 +677,11 @@ bfad_iocmd_rport_get_stats(struct bfad_s *bfad, void *cmd) memcpy((void *)&iocmd->stats, (void *)&fcs_rport->stats, sizeof(struct bfa_rport_stats_s)); - memcpy((void *)&iocmd->stats.hal_stats, - (void *)&(bfa_fcs_rport_get_halrport(fcs_rport)->stats), - sizeof(struct bfa_rport_hal_stats_s)); + if (bfa_fcs_rport_get_halrport(fcs_rport)) { + memcpy((void *)&iocmd->stats.hal_stats, + (void *)&(bfa_fcs_rport_get_halrport(fcs_rport)->stats), + sizeof(struct bfa_rport_hal_stats_s)); + } spin_unlock_irqrestore(&bfad->bfad_lock, flags); iocmd->status = BFA_STATUS_OK; @@ -709,7 +717,8 @@ bfad_iocmd_rport_clr_stats(struct bfad_s *bfad, void *cmd) memset((char *)&fcs_rport->stats, 0, sizeof(struct bfa_rport_stats_s)); rport = bfa_fcs_rport_get_halrport(fcs_rport); - memset(&rport->stats, 0, sizeof(rport->stats)); + if (rport) + memset(&rport->stats, 0, sizeof(rport->stats)); spin_unlock_irqrestore(&bfad->bfad_lock, flags); iocmd->status = BFA_STATUS_OK; out: @@ -744,7 +753,8 @@ bfad_iocmd_rport_set_speed(struct bfad_s *bfad, void *cmd) fcs_rport->rpf.assigned_speed = iocmd->speed; /* Set this speed in f/w only if the RPSC speed is not available */ if (fcs_rport->rpf.rpsc_speed == BFA_PORT_SPEED_UNKNOWN) - bfa_rport_speed(fcs_rport->bfa_rport, iocmd->speed); + if (fcs_rport->bfa_rport) + bfa_rport_speed(fcs_rport->bfa_rport, iocmd->speed); spin_unlock_irqrestore(&bfad->bfad_lock, flags); iocmd->status = BFA_STATUS_OK; out: @@ -1030,9 +1040,10 @@ bfad_iocmd_itnim_get_iostats(struct bfad_s *bfad, void *cmd) iocmd->status = BFA_STATUS_UNKNOWN_RWWN; else { iocmd->status = BFA_STATUS_OK; - memcpy((void *)&iocmd->iostats, (void *) - &(bfa_fcs_itnim_get_halitn(itnim)->stats), - sizeof(struct bfa_itnim_iostats_s)); + if (bfa_fcs_itnim_get_halitn(itnim)) + memcpy((void *)&iocmd->iostats, (void *) + &(bfa_fcs_itnim_get_halitn(itnim)->stats), + sizeof(struct bfa_itnim_iostats_s)); } } spin_unlock_irqrestore(&bfad->bfad_lock, flags); @@ -2949,13 +2960,13 @@ bfad_fcxp_bsg_send(struct fc_bsg_job *job, struct bfad_fcxp *drv_fcxp, spin_lock_irqsave(&bfad->bfad_lock, flags); /* Allocate bfa_fcxp structure */ - hal_fcxp = bfa_fcxp_alloc(drv_fcxp, &bfad->bfa, + hal_fcxp = bfa_fcxp_req_rsp_alloc(drv_fcxp, &bfad->bfa, drv_fcxp->num_req_sgles, drv_fcxp->num_rsp_sgles, bfad_fcxp_get_req_sgaddr_cb, bfad_fcxp_get_req_sglen_cb, bfad_fcxp_get_rsp_sgaddr_cb, - bfad_fcxp_get_rsp_sglen_cb); + bfad_fcxp_get_rsp_sglen_cb, BFA_TRUE); if (!hal_fcxp) { bfa_trc(bfad, 0); spin_unlock_irqrestore(&bfad->bfad_lock, flags); diff --git a/drivers/scsi/bfa/bfad_bsg.h b/drivers/scsi/bfa/bfad_bsg.h index 17ad672..8c569dd 100644 --- a/drivers/scsi/bfa/bfad_bsg.h +++ b/drivers/scsi/bfa/bfad_bsg.h @@ -319,6 +319,8 @@ struct bfa_bsg_rport_attr_s { u16 vf_id; wwn_t pwwn; wwn_t rpwwn; + u32 pid; + u32 rsvd; struct bfa_rport_attr_s attr; }; diff --git a/drivers/scsi/bfa/bfad_drv.h b/drivers/scsi/bfa/bfad_drv.h index 7f74f1d..1840651 100644 --- a/drivers/scsi/bfa/bfad_drv.h +++ b/drivers/scsi/bfa/bfad_drv.h @@ -37,6 +37,7 @@ #include <linux/vmalloc.h> #include <linux/workqueue.h> #include <linux/bitops.h> +#include <linux/aer.h> #include <scsi/scsi.h> #include <scsi/scsi_host.h> #include <scsi/scsi_tcq.h> @@ -56,7 +57,7 @@ #ifdef BFA_DRIVER_VERSION #define BFAD_DRIVER_VERSION BFA_DRIVER_VERSION #else -#define BFAD_DRIVER_VERSION "3.0.23.0" +#define BFAD_DRIVER_VERSION "3.1.2.0" #endif #define BFAD_PROTO_NAME FCPI_NAME @@ -81,6 +82,8 @@ #define BFAD_FC4_PROBE_DONE 0x00000200 #define BFAD_PORT_DELETE 0x00000001 #define BFAD_INTX_ON 0x00000400 +#define BFAD_EEH_BUSY 0x00000800 +#define BFAD_EEH_PCI_CHANNEL_IO_PERM_FAILURE 0x00001000 /* * BFAD related definition */ diff --git a/drivers/scsi/bfa/bfad_im.c b/drivers/scsi/bfa/bfad_im.c index 2eebf8d..8f92732 100644 --- a/drivers/scsi/bfa/bfad_im.c +++ b/drivers/scsi/bfa/bfad_im.c @@ -1216,6 +1216,15 @@ bfad_im_queuecommand_lck(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd return 0; } + if (bfad->bfad_flags & BFAD_EEH_BUSY) { + if (bfad->bfad_flags & BFAD_EEH_PCI_CHANNEL_IO_PERM_FAILURE) + cmnd->result = DID_NO_CONNECT << 16; + else + cmnd->result = DID_REQUEUE << 16; + done(cmnd); + return 0; + } + sg_cnt = scsi_dma_map(cmnd); if (sg_cnt < 0) return SCSI_MLQUEUE_HOST_BUSY; diff --git a/drivers/scsi/bnx2fc/bnx2fc_io.c b/drivers/scsi/bnx2fc/bnx2fc_io.c index 73f231c..8d4626c 100644 --- a/drivers/scsi/bnx2fc/bnx2fc_io.c +++ b/drivers/scsi/bnx2fc/bnx2fc_io.c @@ -1807,7 +1807,7 @@ static void bnx2fc_parse_fcp_rsp(struct bnx2fc_cmd *io_req, fcp_sns_len = SCSI_SENSE_BUFFERSIZE; } - memset(sc_cmd->sense_buffer, 0, sizeof(sc_cmd->sense_buffer)); + memset(sc_cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE); if (fcp_sns_len) memcpy(sc_cmd->sense_buffer, rq_data, fcp_sns_len); diff --git a/drivers/scsi/constants.c b/drivers/scsi/constants.c index 450e011..76e4c03 100644 --- a/drivers/scsi/constants.c +++ b/drivers/scsi/constants.c @@ -1422,7 +1422,8 @@ static const char * const hostbyte_table[]={ "DID_OK", "DID_NO_CONNECT", "DID_BUS_BUSY", "DID_TIME_OUT", "DID_BAD_TARGET", "DID_ABORT", "DID_PARITY", "DID_ERROR", "DID_RESET", "DID_BAD_INTR", "DID_PASSTHROUGH", "DID_SOFT_ERROR", "DID_IMM_RETRY", "DID_REQUEUE", -"DID_TRANSPORT_DISRUPTED", "DID_TRANSPORT_FAILFAST" }; +"DID_TRANSPORT_DISRUPTED", "DID_TRANSPORT_FAILFAST", "DID_TARGET_FAILURE", +"DID_NEXUS_FAILURE" }; #define NUM_HOSTBYTE_STRS ARRAY_SIZE(hostbyte_table) static const char * const driverbyte_table[]={ diff --git a/drivers/scsi/device_handler/scsi_dh_alua.c b/drivers/scsi/device_handler/scsi_dh_alua.c index 08d80a6..6f4d8e6 100644 --- a/drivers/scsi/device_handler/scsi_dh_alua.c +++ b/drivers/scsi/device_handler/scsi_dh_alua.c @@ -641,8 +641,7 @@ static int alua_rtpg(struct scsi_device *sdev, struct alua_dh_data *h) h->state = TPGS_STATE_STANDBY; break; case TPGS_STATE_OFFLINE: - case TPGS_STATE_UNAVAILABLE: - /* Path unusable for unavailable/offline */ + /* Path unusable */ err = SCSI_DH_DEV_OFFLINED; break; default: diff --git a/drivers/scsi/device_handler/scsi_dh_rdac.c b/drivers/scsi/device_handler/scsi_dh_rdac.c index 20c4557f..69c915a 100644 --- a/drivers/scsi/device_handler/scsi_dh_rdac.c +++ b/drivers/scsi/device_handler/scsi_dh_rdac.c @@ -790,29 +790,19 @@ static const struct scsi_dh_devlist rdac_dev_list[] = { {"IBM", "1815"}, {"IBM", "1818"}, {"IBM", "3526"}, - {"SGI", "TP9400"}, - {"SGI", "TP9500"}, - {"SGI", "TP9700"}, + {"SGI", "TP9"}, {"SGI", "IS"}, {"STK", "OPENstorage D280"}, - {"SUN", "CSM200_R"}, - {"SUN", "LCSM100_I"}, - {"SUN", "LCSM100_S"}, - {"SUN", "LCSM100_E"}, - {"SUN", "LCSM100_F"}, - {"DELL", "MD3000"}, - {"DELL", "MD3000i"}, - {"DELL", "MD32xx"}, - {"DELL", "MD32xxi"}, - {"DELL", "MD36xxi"}, - {"DELL", "MD36xxf"}, - {"LSI", "INF-01-00"}, - {"ENGENIO", "INF-01-00"}, {"STK", "FLEXLINE 380"}, - {"SUN", "CSM100_R_FC"}, + {"SUN", "CSM"}, + {"SUN", "LCSM100"}, {"SUN", "STK6580_6780"}, {"SUN", "SUN_6180"}, {"SUN", "ArrayStorage"}, + {"DELL", "MD3"}, + {"NETAPP", "INF-01-00"}, + {"LSI", "INF-01-00"}, + {"ENGENIO", "INF-01-00"}, {NULL, NULL}, }; @@ -863,7 +853,7 @@ static int rdac_bus_attach(struct scsi_device *sdev) if (!scsi_dh_data) { sdev_printk(KERN_ERR, sdev, "%s: Attach failed\n", RDAC_NAME); - return 0; + return -ENOMEM; } scsi_dh_data->scsi_dh = &rdac_dh; diff --git a/drivers/scsi/hpsa.c b/drivers/scsi/hpsa.c index 2b4261c..4217e49 100644 --- a/drivers/scsi/hpsa.c +++ b/drivers/scsi/hpsa.c @@ -99,6 +99,15 @@ static const struct pci_device_id hpsa_pci_device_id[] = { {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSF, 0x103C, 0x3354}, {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSF, 0x103C, 0x3355}, {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSF, 0x103C, 0x3356}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1920}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1921}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1922}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1923}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1924}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1925}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1926}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSH, 0x103C, 0x1928}, + {PCI_VENDOR_ID_HP, PCI_DEVICE_ID_HP_CISSF, 0x103C, 0x334d}, {PCI_VENDOR_ID_HP, PCI_ANY_ID, PCI_ANY_ID, PCI_ANY_ID, PCI_CLASS_STORAGE_RAID << 8, 0xffff << 8, 0}, {0,} @@ -118,13 +127,22 @@ static struct board_type products[] = { {0x3249103C, "Smart Array P812", &SA5_access}, {0x324a103C, "Smart Array P712m", &SA5_access}, {0x324b103C, "Smart Array P711m", &SA5_access}, - {0x3350103C, "Smart Array", &SA5_access}, - {0x3351103C, "Smart Array", &SA5_access}, - {0x3352103C, "Smart Array", &SA5_access}, - {0x3353103C, "Smart Array", &SA5_access}, - {0x3354103C, "Smart Array", &SA5_access}, - {0x3355103C, "Smart Array", &SA5_access}, - {0x3356103C, "Smart Array", &SA5_access}, + {0x3350103C, "Smart Array P222", &SA5_access}, + {0x3351103C, "Smart Array P420", &SA5_access}, + {0x3352103C, "Smart Array P421", &SA5_access}, + {0x3353103C, "Smart Array P822", &SA5_access}, + {0x3354103C, "Smart Array P420i", &SA5_access}, + {0x3355103C, "Smart Array P220i", &SA5_access}, + {0x3356103C, "Smart Array P721m", &SA5_access}, + {0x1920103C, "Smart Array", &SA5_access}, + {0x1921103C, "Smart Array", &SA5_access}, + {0x1922103C, "Smart Array", &SA5_access}, + {0x1923103C, "Smart Array", &SA5_access}, + {0x1924103C, "Smart Array", &SA5_access}, + {0x1925103C, "Smart Array", &SA5_access}, + {0x1926103C, "Smart Array", &SA5_access}, + {0x1928103C, "Smart Array", &SA5_access}, + {0x334d103C, "Smart Array P822se", &SA5_access}, {0xFFFF103C, "Unknown Smart Array", &SA5_access}, }; @@ -2610,7 +2628,7 @@ static int hpsa_eh_abort_handler(struct scsi_cmnd *sc) /* not in reqQ, if also not in cmpQ, must have already completed */ found = hpsa_find_cmd_in_queue(h, sc, &h->cmpQ); if (!found) { - dev_dbg(&h->pdev->dev, "%s Request FAILED (not known to driver).\n", + dev_dbg(&h->pdev->dev, "%s Request SUCCEEDED (not known to driver).\n", msg); return SUCCESS; } @@ -3266,7 +3284,7 @@ static void fill_cmd(struct CommandList *c, u8 cmd, struct ctlr_info *h, c->Request.Timeout = 0; /* Don't time out */ memset(&c->Request.CDB[0], 0, sizeof(c->Request.CDB)); c->Request.CDB[0] = cmd; - c->Request.CDB[1] = 0x03; /* Reset target above */ + c->Request.CDB[1] = HPSA_RESET_TYPE_LUN; /* If bytes 4-7 are zero, it means reset the */ /* LunID device */ c->Request.CDB[4] = 0x00; @@ -3338,7 +3356,8 @@ static void __iomem *remap_pci_mem(ulong base, ulong size) { ulong page_base = ((ulong) base) & PAGE_MASK; ulong page_offs = ((ulong) base) - page_base; - void __iomem *page_remapped = ioremap(page_base, page_offs + size); + void __iomem *page_remapped = ioremap_nocache(page_base, + page_offs + size); return page_remapped ? (page_remapped + page_offs) : NULL; } diff --git a/drivers/scsi/ibmvscsi/Makefile b/drivers/scsi/ibmvscsi/Makefile index ff5b5c5..cb150d1 100644 --- a/drivers/scsi/ibmvscsi/Makefile +++ b/drivers/scsi/ibmvscsi/Makefile @@ -1,7 +1,3 @@ -obj-$(CONFIG_SCSI_IBMVSCSI) += ibmvscsic.o - -ibmvscsic-y += ibmvscsi.o -ibmvscsic-$(CONFIG_PPC_PSERIES) += rpa_vscsi.o - +obj-$(CONFIG_SCSI_IBMVSCSI) += ibmvscsi.o obj-$(CONFIG_SCSI_IBMVSCSIS) += ibmvstgt.o obj-$(CONFIG_SCSI_IBMVFC) += ibmvfc.o diff --git a/drivers/scsi/ibmvscsi/ibmvfc.c b/drivers/scsi/ibmvscsi/ibmvfc.c index 134a0ae..5e8d51b 100644 --- a/drivers/scsi/ibmvscsi/ibmvfc.c +++ b/drivers/scsi/ibmvscsi/ibmvfc.c @@ -2242,6 +2242,21 @@ static int ibmvfc_match_key(struct ibmvfc_event *evt, void *key) } /** + * ibmvfc_match_evt - Match function for specified event + * @evt: ibmvfc event struct + * @match: event to match + * + * Returns: + * 1 if event matches key / 0 if event does not match key + **/ +static int ibmvfc_match_evt(struct ibmvfc_event *evt, void *match) +{ + if (evt == match) + return 1; + return 0; +} + +/** * ibmvfc_abort_task_set - Abort outstanding commands to the device * @sdev: scsi device to abort commands * @@ -2322,7 +2337,20 @@ static int ibmvfc_abort_task_set(struct scsi_device *sdev) if (rc) { sdev_printk(KERN_INFO, sdev, "Cancel failed, resetting host\n"); ibmvfc_reset_host(vhost); - rsp_rc = 0; + rsp_rc = -EIO; + rc = ibmvfc_wait_for_ops(vhost, sdev->hostdata, ibmvfc_match_key); + + if (rc == SUCCESS) + rsp_rc = 0; + + rc = ibmvfc_wait_for_ops(vhost, evt, ibmvfc_match_evt); + if (rc != SUCCESS) { + spin_lock_irqsave(vhost->host->host_lock, flags); + ibmvfc_hard_reset_host(vhost); + spin_unlock_irqrestore(vhost->host->host_lock, flags); + rsp_rc = 0; + } + goto out; } } @@ -2597,8 +2625,10 @@ static void ibmvfc_handle_async(struct ibmvfc_async_crq *crq, case IBMVFC_AE_SCN_FABRIC: case IBMVFC_AE_SCN_DOMAIN: vhost->events_to_log |= IBMVFC_AE_RSCN; - vhost->delay_init = 1; - __ibmvfc_reset_host(vhost); + if (vhost->state < IBMVFC_HALTED) { + vhost->delay_init = 1; + __ibmvfc_reset_host(vhost); + } break; case IBMVFC_AE_SCN_NPORT: case IBMVFC_AE_SCN_GROUP: diff --git a/drivers/scsi/ibmvscsi/ibmvfc.h b/drivers/scsi/ibmvscsi/ibmvfc.h index 834c37f..3be8af6 100644 --- a/drivers/scsi/ibmvscsi/ibmvfc.h +++ b/drivers/scsi/ibmvscsi/ibmvfc.h @@ -29,8 +29,8 @@ #include "viosrp.h" #define IBMVFC_NAME "ibmvfc" -#define IBMVFC_DRIVER_VERSION "1.0.9" -#define IBMVFC_DRIVER_DATE "(August 5, 2010)" +#define IBMVFC_DRIVER_VERSION "1.0.10" +#define IBMVFC_DRIVER_DATE "(August 24, 2012)" #define IBMVFC_DEFAULT_TIMEOUT 60 #define IBMVFC_ADISC_CANCEL_TIMEOUT 45 diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.c b/drivers/scsi/ibmvscsi/ibmvscsi.c index 3a6c474..ef9a54c 100644 --- a/drivers/scsi/ibmvscsi/ibmvscsi.c +++ b/drivers/scsi/ibmvscsi/ibmvscsi.c @@ -93,13 +93,13 @@ static int max_requests = IBMVSCSI_MAX_REQUESTS_DEFAULT; static int max_events = IBMVSCSI_MAX_REQUESTS_DEFAULT + 2; static int fast_fail = 1; static int client_reserve = 1; +static char partition_name[97] = "UNKNOWN"; +static unsigned int partition_number = -1; static struct scsi_transport_template *ibmvscsi_transport_template; #define IBMVSCSI_VERSION "1.5.9" -static struct ibmvscsi_ops *ibmvscsi_ops; - MODULE_DESCRIPTION("IBM Virtual SCSI"); MODULE_AUTHOR("Dave Boutcher"); MODULE_LICENSE("GPL"); @@ -118,6 +118,316 @@ MODULE_PARM_DESC(fast_fail, "Enable fast fail. [Default=1]"); module_param_named(client_reserve, client_reserve, int, S_IRUGO ); MODULE_PARM_DESC(client_reserve, "Attempt client managed reserve/release"); +static void ibmvscsi_handle_crq(struct viosrp_crq *crq, + struct ibmvscsi_host_data *hostdata); + +/* ------------------------------------------------------------ + * Routines for managing the command/response queue + */ +/** + * ibmvscsi_handle_event: - Interrupt handler for crq events + * @irq: number of irq to handle, not used + * @dev_instance: ibmvscsi_host_data of host that received interrupt + * + * Disables interrupts and schedules srp_task + * Always returns IRQ_HANDLED + */ +static irqreturn_t ibmvscsi_handle_event(int irq, void *dev_instance) +{ + struct ibmvscsi_host_data *hostdata = + (struct ibmvscsi_host_data *)dev_instance; + vio_disable_interrupts(to_vio_dev(hostdata->dev)); + tasklet_schedule(&hostdata->srp_task); + return IRQ_HANDLED; +} + +/** + * release_crq_queue: - Deallocates data and unregisters CRQ + * @queue: crq_queue to initialize and register + * @host_data: ibmvscsi_host_data of host + * + * Frees irq, deallocates a page for messages, unmaps dma, and unregisters + * the crq with the hypervisor. + */ +static void ibmvscsi_release_crq_queue(struct crq_queue *queue, + struct ibmvscsi_host_data *hostdata, + int max_requests) +{ + long rc = 0; + struct vio_dev *vdev = to_vio_dev(hostdata->dev); + free_irq(vdev->irq, (void *)hostdata); + tasklet_kill(&hostdata->srp_task); + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); + } while ((rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); + dma_unmap_single(hostdata->dev, + queue->msg_token, + queue->size * sizeof(*queue->msgs), DMA_BIDIRECTIONAL); + free_page((unsigned long)queue->msgs); +} + +/** + * crq_queue_next_crq: - Returns the next entry in message queue + * @queue: crq_queue to use + * + * Returns pointer to next entry in queue, or NULL if there are no new + * entried in the CRQ. + */ +static struct viosrp_crq *crq_queue_next_crq(struct crq_queue *queue) +{ + struct viosrp_crq *crq; + unsigned long flags; + + spin_lock_irqsave(&queue->lock, flags); + crq = &queue->msgs[queue->cur]; + if (crq->valid & 0x80) { + if (++queue->cur == queue->size) + queue->cur = 0; + } else + crq = NULL; + spin_unlock_irqrestore(&queue->lock, flags); + + return crq; +} + +/** + * ibmvscsi_send_crq: - Send a CRQ + * @hostdata: the adapter + * @word1: the first 64 bits of the data + * @word2: the second 64 bits of the data + */ +static int ibmvscsi_send_crq(struct ibmvscsi_host_data *hostdata, + u64 word1, u64 word2) +{ + struct vio_dev *vdev = to_vio_dev(hostdata->dev); + + return plpar_hcall_norets(H_SEND_CRQ, vdev->unit_address, word1, word2); +} + +/** + * ibmvscsi_task: - Process srps asynchronously + * @data: ibmvscsi_host_data of host + */ +static void ibmvscsi_task(void *data) +{ + struct ibmvscsi_host_data *hostdata = (struct ibmvscsi_host_data *)data; + struct vio_dev *vdev = to_vio_dev(hostdata->dev); + struct viosrp_crq *crq; + int done = 0; + + while (!done) { + /* Pull all the valid messages off the CRQ */ + while ((crq = crq_queue_next_crq(&hostdata->queue)) != NULL) { + ibmvscsi_handle_crq(crq, hostdata); + crq->valid = 0x00; + } + + vio_enable_interrupts(vdev); + crq = crq_queue_next_crq(&hostdata->queue); + if (crq != NULL) { + vio_disable_interrupts(vdev); + ibmvscsi_handle_crq(crq, hostdata); + crq->valid = 0x00; + } else { + done = 1; + } + } +} + +static void gather_partition_info(void) +{ + struct device_node *rootdn; + + const char *ppartition_name; + const unsigned int *p_number_ptr; + + /* Retrieve information about this partition */ + rootdn = of_find_node_by_path("/"); + if (!rootdn) { + return; + } + + ppartition_name = of_get_property(rootdn, "ibm,partition-name", NULL); + if (ppartition_name) + strncpy(partition_name, ppartition_name, + sizeof(partition_name)); + p_number_ptr = of_get_property(rootdn, "ibm,partition-no", NULL); + if (p_number_ptr) + partition_number = *p_number_ptr; + of_node_put(rootdn); +} + +static void set_adapter_info(struct ibmvscsi_host_data *hostdata) +{ + memset(&hostdata->madapter_info, 0x00, + sizeof(hostdata->madapter_info)); + + dev_info(hostdata->dev, "SRP_VERSION: %s\n", SRP_VERSION); + strcpy(hostdata->madapter_info.srp_version, SRP_VERSION); + + strncpy(hostdata->madapter_info.partition_name, partition_name, + sizeof(hostdata->madapter_info.partition_name)); + + hostdata->madapter_info.partition_number = partition_number; + + hostdata->madapter_info.mad_version = 1; + hostdata->madapter_info.os_type = 2; +} + +/** + * reset_crq_queue: - resets a crq after a failure + * @queue: crq_queue to initialize and register + * @hostdata: ibmvscsi_host_data of host + * + */ +static int ibmvscsi_reset_crq_queue(struct crq_queue *queue, + struct ibmvscsi_host_data *hostdata) +{ + int rc = 0; + struct vio_dev *vdev = to_vio_dev(hostdata->dev); + + /* Close the CRQ */ + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); + } while ((rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); + + /* Clean out the queue */ + memset(queue->msgs, 0x00, PAGE_SIZE); + queue->cur = 0; + + set_adapter_info(hostdata); + + /* And re-open it again */ + rc = plpar_hcall_norets(H_REG_CRQ, + vdev->unit_address, + queue->msg_token, PAGE_SIZE); + if (rc == 2) { + /* Adapter is good, but other end is not ready */ + dev_warn(hostdata->dev, "Partner adapter not ready\n"); + } else if (rc != 0) { + dev_warn(hostdata->dev, "couldn't register crq--rc 0x%x\n", rc); + } + return rc; +} + +/** + * initialize_crq_queue: - Initializes and registers CRQ with hypervisor + * @queue: crq_queue to initialize and register + * @hostdata: ibmvscsi_host_data of host + * + * Allocates a page for messages, maps it for dma, and registers + * the crq with the hypervisor. + * Returns zero on success. + */ +static int ibmvscsi_init_crq_queue(struct crq_queue *queue, + struct ibmvscsi_host_data *hostdata, + int max_requests) +{ + int rc; + int retrc; + struct vio_dev *vdev = to_vio_dev(hostdata->dev); + + queue->msgs = (struct viosrp_crq *)get_zeroed_page(GFP_KERNEL); + + if (!queue->msgs) + goto malloc_failed; + queue->size = PAGE_SIZE / sizeof(*queue->msgs); + + queue->msg_token = dma_map_single(hostdata->dev, queue->msgs, + queue->size * sizeof(*queue->msgs), + DMA_BIDIRECTIONAL); + + if (dma_mapping_error(hostdata->dev, queue->msg_token)) + goto map_failed; + + gather_partition_info(); + set_adapter_info(hostdata); + + retrc = rc = plpar_hcall_norets(H_REG_CRQ, + vdev->unit_address, + queue->msg_token, PAGE_SIZE); + if (rc == H_RESOURCE) + /* maybe kexecing and resource is busy. try a reset */ + rc = ibmvscsi_reset_crq_queue(queue, + hostdata); + + if (rc == 2) { + /* Adapter is good, but other end is not ready */ + dev_warn(hostdata->dev, "Partner adapter not ready\n"); + retrc = 0; + } else if (rc != 0) { + dev_warn(hostdata->dev, "Error %d opening adapter\n", rc); + goto reg_crq_failed; + } + + queue->cur = 0; + spin_lock_init(&queue->lock); + + tasklet_init(&hostdata->srp_task, (void *)ibmvscsi_task, + (unsigned long)hostdata); + + if (request_irq(vdev->irq, + ibmvscsi_handle_event, + 0, "ibmvscsi", (void *)hostdata) != 0) { + dev_err(hostdata->dev, "couldn't register irq 0x%x\n", + vdev->irq); + goto req_irq_failed; + } + + rc = vio_enable_interrupts(vdev); + if (rc != 0) { + dev_err(hostdata->dev, "Error %d enabling interrupts!!!\n", rc); + goto req_irq_failed; + } + + return retrc; + + req_irq_failed: + tasklet_kill(&hostdata->srp_task); + rc = 0; + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); + } while ((rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); + reg_crq_failed: + dma_unmap_single(hostdata->dev, + queue->msg_token, + queue->size * sizeof(*queue->msgs), DMA_BIDIRECTIONAL); + map_failed: + free_page((unsigned long)queue->msgs); + malloc_failed: + return -1; +} + +/** + * reenable_crq_queue: - reenables a crq after + * @queue: crq_queue to initialize and register + * @hostdata: ibmvscsi_host_data of host + * + */ +static int ibmvscsi_reenable_crq_queue(struct crq_queue *queue, + struct ibmvscsi_host_data *hostdata) +{ + int rc = 0; + struct vio_dev *vdev = to_vio_dev(hostdata->dev); + + /* Re-enable the CRQ */ + do { + if (rc) + msleep(100); + rc = plpar_hcall_norets(H_ENABLE_CRQ, vdev->unit_address); + } while ((rc == H_IN_PROGRESS) || (rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); + + if (rc) + dev_err(hostdata->dev, "Error %d enabling adapter\n", rc); + return rc; +} + /* ------------------------------------------------------------ * Routines for the event pool and event structs */ @@ -611,7 +921,7 @@ static int ibmvscsi_send_srp_event(struct srp_event_struct *evt_struct, } if ((rc = - ibmvscsi_ops->send_crq(hostdata, crq_as_u64[0], crq_as_u64[1])) != 0) { + ibmvscsi_send_crq(hostdata, crq_as_u64[0], crq_as_u64[1])) != 0) { list_del(&evt_struct->list); del_timer(&evt_struct->timer); @@ -1420,8 +1730,8 @@ static int ibmvscsi_eh_host_reset_handler(struct scsi_cmnd *cmd) * @hostdata: ibmvscsi_host_data of host * */ -void ibmvscsi_handle_crq(struct viosrp_crq *crq, - struct ibmvscsi_host_data *hostdata) +static void ibmvscsi_handle_crq(struct viosrp_crq *crq, + struct ibmvscsi_host_data *hostdata) { long rc; unsigned long flags; @@ -1433,8 +1743,8 @@ void ibmvscsi_handle_crq(struct viosrp_crq *crq, case 0x01: /* Initialization message */ dev_info(hostdata->dev, "partner initialized\n"); /* Send back a response */ - if ((rc = ibmvscsi_ops->send_crq(hostdata, - 0xC002000000000000LL, 0)) == 0) { + rc = ibmvscsi_send_crq(hostdata, 0xC002000000000000LL, 0); + if (rc == 0) { /* Now login */ init_adapter(hostdata); } else { @@ -1541,6 +1851,9 @@ static int ibmvscsi_do_host_config(struct ibmvscsi_host_data *hostdata, host_config = &evt_struct->iu.mad.host_config; + /* The transport length field is only 16-bit */ + length = min(0xffff, length); + /* Set up a lun reset SRP command */ memset(host_config, 0x00, sizeof(*host_config)); host_config->common.type = VIOSRP_HOST_CONFIG_TYPE; @@ -1840,17 +2153,17 @@ static void ibmvscsi_do_work(struct ibmvscsi_host_data *hostdata) smp_rmb(); hostdata->reset_crq = 0; - rc = ibmvscsi_ops->reset_crq_queue(&hostdata->queue, hostdata); + rc = ibmvscsi_reset_crq_queue(&hostdata->queue, hostdata); if (!rc) - rc = ibmvscsi_ops->send_crq(hostdata, 0xC001000000000000LL, 0); + rc = ibmvscsi_send_crq(hostdata, 0xC001000000000000LL, 0); vio_enable_interrupts(to_vio_dev(hostdata->dev)); } else if (hostdata->reenable_crq) { smp_rmb(); action = "enable"; - rc = ibmvscsi_ops->reenable_crq_queue(&hostdata->queue, hostdata); + rc = ibmvscsi_reenable_crq_queue(&hostdata->queue, hostdata); hostdata->reenable_crq = 0; if (!rc) - rc = ibmvscsi_ops->send_crq(hostdata, 0xC001000000000000LL, 0); + rc = ibmvscsi_send_crq(hostdata, 0xC001000000000000LL, 0); } else return; @@ -1944,7 +2257,7 @@ static int ibmvscsi_probe(struct vio_dev *vdev, const struct vio_device_id *id) goto init_crq_failed; } - rc = ibmvscsi_ops->init_crq_queue(&hostdata->queue, hostdata, max_events); + rc = ibmvscsi_init_crq_queue(&hostdata->queue, hostdata, max_events); if (rc != 0 && rc != H_RESOURCE) { dev_err(&vdev->dev, "couldn't initialize crq. rc=%d\n", rc); goto kill_kthread; @@ -1974,7 +2287,7 @@ static int ibmvscsi_probe(struct vio_dev *vdev, const struct vio_device_id *id) * to fail if the other end is not acive. In that case we don't * want to scan */ - if (ibmvscsi_ops->send_crq(hostdata, 0xC001000000000000LL, 0) == 0 + if (ibmvscsi_send_crq(hostdata, 0xC001000000000000LL, 0) == 0 || rc == H_RESOURCE) { /* * Wait around max init_timeout secs for the adapter to finish @@ -2002,7 +2315,7 @@ static int ibmvscsi_probe(struct vio_dev *vdev, const struct vio_device_id *id) add_host_failed: release_event_pool(&hostdata->pool, hostdata); init_pool_failed: - ibmvscsi_ops->release_crq_queue(&hostdata->queue, hostdata, max_events); + ibmvscsi_release_crq_queue(&hostdata->queue, hostdata, max_events); kill_kthread: kthread_stop(hostdata->work_thread); init_crq_failed: @@ -2018,7 +2331,7 @@ static int ibmvscsi_remove(struct vio_dev *vdev) struct ibmvscsi_host_data *hostdata = dev_get_drvdata(&vdev->dev); unmap_persist_bufs(hostdata); release_event_pool(&hostdata->pool, hostdata); - ibmvscsi_ops->release_crq_queue(&hostdata->queue, hostdata, + ibmvscsi_release_crq_queue(&hostdata->queue, hostdata, max_events); kthread_stop(hostdata->work_thread); @@ -2039,7 +2352,10 @@ static int ibmvscsi_remove(struct vio_dev *vdev) static int ibmvscsi_resume(struct device *dev) { struct ibmvscsi_host_data *hostdata = dev_get_drvdata(dev); - return ibmvscsi_ops->resume(hostdata); + vio_disable_interrupts(to_vio_dev(hostdata->dev)); + tasklet_schedule(&hostdata->srp_task); + + return 0; } /** @@ -2076,9 +2392,7 @@ int __init ibmvscsi_module_init(void) driver_template.can_queue = max_requests; max_events = max_requests + 2; - if (firmware_has_feature(FW_FEATURE_VIO)) - ibmvscsi_ops = &rpavscsi_ops; - else + if (!firmware_has_feature(FW_FEATURE_VIO)) return -ENODEV; ibmvscsi_transport_template = diff --git a/drivers/scsi/ibmvscsi/ibmvscsi.h b/drivers/scsi/ibmvscsi/ibmvscsi.h index c503e17..7d64867 100644 --- a/drivers/scsi/ibmvscsi/ibmvscsi.h +++ b/drivers/scsi/ibmvscsi/ibmvscsi.h @@ -107,26 +107,4 @@ struct ibmvscsi_host_data { dma_addr_t adapter_info_addr; }; -/* routines for managing a command/response queue */ -void ibmvscsi_handle_crq(struct viosrp_crq *crq, - struct ibmvscsi_host_data *hostdata); - -struct ibmvscsi_ops { - int (*init_crq_queue)(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata, - int max_requests); - void (*release_crq_queue)(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata, - int max_requests); - int (*reset_crq_queue)(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata); - int (*reenable_crq_queue)(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata); - int (*send_crq)(struct ibmvscsi_host_data *hostdata, - u64 word1, u64 word2); - int (*resume) (struct ibmvscsi_host_data *hostdata); -}; - -extern struct ibmvscsi_ops rpavscsi_ops; - #endif /* IBMVSCSI_H */ diff --git a/drivers/scsi/ibmvscsi/rpa_vscsi.c b/drivers/scsi/ibmvscsi/rpa_vscsi.c deleted file mode 100644 index f48ae01..0000000 --- a/drivers/scsi/ibmvscsi/rpa_vscsi.c +++ /dev/null @@ -1,368 +0,0 @@ -/* ------------------------------------------------------------ - * rpa_vscsi.c - * (C) Copyright IBM Corporation 1994, 2003 - * Authors: Colin DeVilbiss (devilbis@us.ibm.com) - * Santiago Leon (santil@us.ibm.com) - * - * This program is free software; you can redistribute it and/or modify - * it under the terms of the GNU General Public License as published by - * the Free Software Foundation; either version 2 of the License, or - * (at your option) any later version. - * - * This program is distributed in the hope that it will be useful, - * but WITHOUT ANY WARRANTY; without even the implied warranty of - * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the - * GNU General Public License for more details. - * - * You should have received a copy of the GNU General Public License - * along with this program; if not, write to the Free Software - * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 - * USA - * - * ------------------------------------------------------------ - * RPA-specific functions of the SCSI host adapter for Virtual I/O devices - * - * This driver allows the Linux SCSI peripheral drivers to directly - * access devices in the hosting partition, either on an iSeries - * hypervisor system or a converged hypervisor system. - */ - -#include <asm/vio.h> -#include <asm/prom.h> -#include <asm/iommu.h> -#include <asm/hvcall.h> -#include <linux/delay.h> -#include <linux/dma-mapping.h> -#include <linux/gfp.h> -#include <linux/interrupt.h> -#include "ibmvscsi.h" - -static char partition_name[97] = "UNKNOWN"; -static unsigned int partition_number = -1; - -/* ------------------------------------------------------------ - * Routines for managing the command/response queue - */ -/** - * rpavscsi_handle_event: - Interrupt handler for crq events - * @irq: number of irq to handle, not used - * @dev_instance: ibmvscsi_host_data of host that received interrupt - * - * Disables interrupts and schedules srp_task - * Always returns IRQ_HANDLED - */ -static irqreturn_t rpavscsi_handle_event(int irq, void *dev_instance) -{ - struct ibmvscsi_host_data *hostdata = - (struct ibmvscsi_host_data *)dev_instance; - vio_disable_interrupts(to_vio_dev(hostdata->dev)); - tasklet_schedule(&hostdata->srp_task); - return IRQ_HANDLED; -} - -/** - * release_crq_queue: - Deallocates data and unregisters CRQ - * @queue: crq_queue to initialize and register - * @host_data: ibmvscsi_host_data of host - * - * Frees irq, deallocates a page for messages, unmaps dma, and unregisters - * the crq with the hypervisor. - */ -static void rpavscsi_release_crq_queue(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata, - int max_requests) -{ - long rc = 0; - struct vio_dev *vdev = to_vio_dev(hostdata->dev); - free_irq(vdev->irq, (void *)hostdata); - tasklet_kill(&hostdata->srp_task); - do { - if (rc) - msleep(100); - rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); - } while ((rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); - dma_unmap_single(hostdata->dev, - queue->msg_token, - queue->size * sizeof(*queue->msgs), DMA_BIDIRECTIONAL); - free_page((unsigned long)queue->msgs); -} - -/** - * crq_queue_next_crq: - Returns the next entry in message queue - * @queue: crq_queue to use - * - * Returns pointer to next entry in queue, or NULL if there are no new - * entried in the CRQ. - */ -static struct viosrp_crq *crq_queue_next_crq(struct crq_queue *queue) -{ - struct viosrp_crq *crq; - unsigned long flags; - - spin_lock_irqsave(&queue->lock, flags); - crq = &queue->msgs[queue->cur]; - if (crq->valid & 0x80) { - if (++queue->cur == queue->size) - queue->cur = 0; - } else - crq = NULL; - spin_unlock_irqrestore(&queue->lock, flags); - - return crq; -} - -/** - * rpavscsi_send_crq: - Send a CRQ - * @hostdata: the adapter - * @word1: the first 64 bits of the data - * @word2: the second 64 bits of the data - */ -static int rpavscsi_send_crq(struct ibmvscsi_host_data *hostdata, - u64 word1, u64 word2) -{ - struct vio_dev *vdev = to_vio_dev(hostdata->dev); - - return plpar_hcall_norets(H_SEND_CRQ, vdev->unit_address, word1, word2); -} - -/** - * rpavscsi_task: - Process srps asynchronously - * @data: ibmvscsi_host_data of host - */ -static void rpavscsi_task(void *data) -{ - struct ibmvscsi_host_data *hostdata = (struct ibmvscsi_host_data *)data; - struct vio_dev *vdev = to_vio_dev(hostdata->dev); - struct viosrp_crq *crq; - int done = 0; - - while (!done) { - /* Pull all the valid messages off the CRQ */ - while ((crq = crq_queue_next_crq(&hostdata->queue)) != NULL) { - ibmvscsi_handle_crq(crq, hostdata); - crq->valid = 0x00; - } - - vio_enable_interrupts(vdev); - if ((crq = crq_queue_next_crq(&hostdata->queue)) != NULL) { - vio_disable_interrupts(vdev); - ibmvscsi_handle_crq(crq, hostdata); - crq->valid = 0x00; - } else { - done = 1; - } - } -} - -static void gather_partition_info(void) -{ - struct device_node *rootdn; - - const char *ppartition_name; - const unsigned int *p_number_ptr; - - /* Retrieve information about this partition */ - rootdn = of_find_node_by_path("/"); - if (!rootdn) { - return; - } - - ppartition_name = of_get_property(rootdn, "ibm,partition-name", NULL); - if (ppartition_name) - strncpy(partition_name, ppartition_name, - sizeof(partition_name)); - p_number_ptr = of_get_property(rootdn, "ibm,partition-no", NULL); - if (p_number_ptr) - partition_number = *p_number_ptr; - of_node_put(rootdn); -} - -static void set_adapter_info(struct ibmvscsi_host_data *hostdata) -{ - memset(&hostdata->madapter_info, 0x00, - sizeof(hostdata->madapter_info)); - - dev_info(hostdata->dev, "SRP_VERSION: %s\n", SRP_VERSION); - strcpy(hostdata->madapter_info.srp_version, SRP_VERSION); - - strncpy(hostdata->madapter_info.partition_name, partition_name, - sizeof(hostdata->madapter_info.partition_name)); - - hostdata->madapter_info.partition_number = partition_number; - - hostdata->madapter_info.mad_version = 1; - hostdata->madapter_info.os_type = 2; -} - -/** - * reset_crq_queue: - resets a crq after a failure - * @queue: crq_queue to initialize and register - * @hostdata: ibmvscsi_host_data of host - * - */ -static int rpavscsi_reset_crq_queue(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata) -{ - int rc = 0; - struct vio_dev *vdev = to_vio_dev(hostdata->dev); - - /* Close the CRQ */ - do { - if (rc) - msleep(100); - rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); - } while ((rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); - - /* Clean out the queue */ - memset(queue->msgs, 0x00, PAGE_SIZE); - queue->cur = 0; - - set_adapter_info(hostdata); - - /* And re-open it again */ - rc = plpar_hcall_norets(H_REG_CRQ, - vdev->unit_address, - queue->msg_token, PAGE_SIZE); - if (rc == 2) { - /* Adapter is good, but other end is not ready */ - dev_warn(hostdata->dev, "Partner adapter not ready\n"); - } else if (rc != 0) { - dev_warn(hostdata->dev, "couldn't register crq--rc 0x%x\n", rc); - } - return rc; -} - -/** - * initialize_crq_queue: - Initializes and registers CRQ with hypervisor - * @queue: crq_queue to initialize and register - * @hostdata: ibmvscsi_host_data of host - * - * Allocates a page for messages, maps it for dma, and registers - * the crq with the hypervisor. - * Returns zero on success. - */ -static int rpavscsi_init_crq_queue(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata, - int max_requests) -{ - int rc; - int retrc; - struct vio_dev *vdev = to_vio_dev(hostdata->dev); - - queue->msgs = (struct viosrp_crq *)get_zeroed_page(GFP_KERNEL); - - if (!queue->msgs) - goto malloc_failed; - queue->size = PAGE_SIZE / sizeof(*queue->msgs); - - queue->msg_token = dma_map_single(hostdata->dev, queue->msgs, - queue->size * sizeof(*queue->msgs), - DMA_BIDIRECTIONAL); - - if (dma_mapping_error(hostdata->dev, queue->msg_token)) - goto map_failed; - - gather_partition_info(); - set_adapter_info(hostdata); - - retrc = rc = plpar_hcall_norets(H_REG_CRQ, - vdev->unit_address, - queue->msg_token, PAGE_SIZE); - if (rc == H_RESOURCE) - /* maybe kexecing and resource is busy. try a reset */ - rc = rpavscsi_reset_crq_queue(queue, - hostdata); - - if (rc == 2) { - /* Adapter is good, but other end is not ready */ - dev_warn(hostdata->dev, "Partner adapter not ready\n"); - retrc = 0; - } else if (rc != 0) { - dev_warn(hostdata->dev, "Error %d opening adapter\n", rc); - goto reg_crq_failed; - } - - queue->cur = 0; - spin_lock_init(&queue->lock); - - tasklet_init(&hostdata->srp_task, (void *)rpavscsi_task, - (unsigned long)hostdata); - - if (request_irq(vdev->irq, - rpavscsi_handle_event, - 0, "ibmvscsi", (void *)hostdata) != 0) { - dev_err(hostdata->dev, "couldn't register irq 0x%x\n", - vdev->irq); - goto req_irq_failed; - } - - rc = vio_enable_interrupts(vdev); - if (rc != 0) { - dev_err(hostdata->dev, "Error %d enabling interrupts!!!\n", rc); - goto req_irq_failed; - } - - return retrc; - - req_irq_failed: - tasklet_kill(&hostdata->srp_task); - rc = 0; - do { - if (rc) - msleep(100); - rc = plpar_hcall_norets(H_FREE_CRQ, vdev->unit_address); - } while ((rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); - reg_crq_failed: - dma_unmap_single(hostdata->dev, - queue->msg_token, - queue->size * sizeof(*queue->msgs), DMA_BIDIRECTIONAL); - map_failed: - free_page((unsigned long)queue->msgs); - malloc_failed: - return -1; -} - -/** - * reenable_crq_queue: - reenables a crq after - * @queue: crq_queue to initialize and register - * @hostdata: ibmvscsi_host_data of host - * - */ -static int rpavscsi_reenable_crq_queue(struct crq_queue *queue, - struct ibmvscsi_host_data *hostdata) -{ - int rc = 0; - struct vio_dev *vdev = to_vio_dev(hostdata->dev); - - /* Re-enable the CRQ */ - do { - if (rc) - msleep(100); - rc = plpar_hcall_norets(H_ENABLE_CRQ, vdev->unit_address); - } while ((rc == H_IN_PROGRESS) || (rc == H_BUSY) || (H_IS_LONG_BUSY(rc))); - - if (rc) - dev_err(hostdata->dev, "Error %d enabling adapter\n", rc); - return rc; -} - -/** - * rpavscsi_resume: - resume after suspend - * @hostdata: ibmvscsi_host_data of host - * - */ -static int rpavscsi_resume(struct ibmvscsi_host_data *hostdata) -{ - vio_disable_interrupts(to_vio_dev(hostdata->dev)); - tasklet_schedule(&hostdata->srp_task); - return 0; -} - -struct ibmvscsi_ops rpavscsi_ops = { - .init_crq_queue = rpavscsi_init_crq_queue, - .release_crq_queue = rpavscsi_release_crq_queue, - .reset_crq_queue = rpavscsi_reset_crq_queue, - .reenable_crq_queue = rpavscsi_reenable_crq_queue, - .send_crq = rpavscsi_send_crq, - .resume = rpavscsi_resume, -}; diff --git a/drivers/scsi/ipr.c b/drivers/scsi/ipr.c index 45e192a..e3f29f6 100644 --- a/drivers/scsi/ipr.c +++ b/drivers/scsi/ipr.c @@ -566,6 +566,23 @@ static void ipr_trc_hook(struct ipr_cmnd *ipr_cmd, #endif /** + * ipr_lock_and_done - Acquire lock and complete command + * @ipr_cmd: ipr command struct + * + * Return value: + * none + **/ +static void ipr_lock_and_done(struct ipr_cmnd *ipr_cmd) +{ + unsigned long lock_flags; + struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg; + + spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags); + ipr_cmd->done(ipr_cmd); + spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); +} + +/** * ipr_reinit_ipr_cmnd - Re-initialize an IPR Cmnd block for reuse * @ipr_cmd: ipr command struct * @@ -611,34 +628,50 @@ static void ipr_reinit_ipr_cmnd(struct ipr_cmnd *ipr_cmd) * Return value: * none **/ -static void ipr_init_ipr_cmnd(struct ipr_cmnd *ipr_cmd) +static void ipr_init_ipr_cmnd(struct ipr_cmnd *ipr_cmd, + void (*fast_done) (struct ipr_cmnd *)) { ipr_reinit_ipr_cmnd(ipr_cmd); ipr_cmd->u.scratch = 0; ipr_cmd->sibling = NULL; + ipr_cmd->fast_done = fast_done; init_timer(&ipr_cmd->timer); } /** - * ipr_get_free_ipr_cmnd - Get a free IPR Cmnd block + * __ipr_get_free_ipr_cmnd - Get a free IPR Cmnd block * @ioa_cfg: ioa config struct * * Return value: * pointer to ipr command struct **/ static -struct ipr_cmnd *ipr_get_free_ipr_cmnd(struct ipr_ioa_cfg *ioa_cfg) +struct ipr_cmnd *__ipr_get_free_ipr_cmnd(struct ipr_ioa_cfg *ioa_cfg) { struct ipr_cmnd *ipr_cmd; ipr_cmd = list_entry(ioa_cfg->free_q.next, struct ipr_cmnd, queue); list_del(&ipr_cmd->queue); - ipr_init_ipr_cmnd(ipr_cmd); return ipr_cmd; } /** + * ipr_get_free_ipr_cmnd - Get a free IPR Cmnd block and initialize it + * @ioa_cfg: ioa config struct + * + * Return value: + * pointer to ipr command struct + **/ +static +struct ipr_cmnd *ipr_get_free_ipr_cmnd(struct ipr_ioa_cfg *ioa_cfg) +{ + struct ipr_cmnd *ipr_cmd = __ipr_get_free_ipr_cmnd(ioa_cfg); + ipr_init_ipr_cmnd(ipr_cmd, ipr_lock_and_done); + return ipr_cmd; +} + +/** * ipr_mask_and_clear_interrupts - Mask all and clear specified interrupts * @ioa_cfg: ioa config struct * @clr_ints: interrupts to clear @@ -5116,8 +5149,9 @@ static irqreturn_t ipr_isr(int irq, void *devp) u16 cmd_index; int num_hrrq = 0; int irq_none = 0; - struct ipr_cmnd *ipr_cmd; + struct ipr_cmnd *ipr_cmd, *temp; irqreturn_t rc = IRQ_NONE; + LIST_HEAD(doneq); spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags); @@ -5138,8 +5172,8 @@ static irqreturn_t ipr_isr(int irq, void *devp) if (unlikely(cmd_index >= IPR_NUM_CMD_BLKS)) { ipr_isr_eh(ioa_cfg, "Invalid response handle from IOA"); - spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); - return IRQ_HANDLED; + rc = IRQ_HANDLED; + goto unlock_out; } ipr_cmd = ioa_cfg->ipr_cmnd_list[cmd_index]; @@ -5148,9 +5182,7 @@ static irqreturn_t ipr_isr(int irq, void *devp) ipr_trc_hook(ipr_cmd, IPR_TRACE_FINISH, ioasc); - list_del(&ipr_cmd->queue); - del_timer(&ipr_cmd->timer); - ipr_cmd->done(ipr_cmd); + list_move_tail(&ipr_cmd->queue, &doneq); rc = IRQ_HANDLED; @@ -5180,8 +5212,8 @@ static irqreturn_t ipr_isr(int irq, void *devp) } else if (num_hrrq == IPR_MAX_HRRQ_RETRIES && int_reg & IPR_PCII_HRRQ_UPDATED) { ipr_isr_eh(ioa_cfg, "Error clearing HRRQ"); - spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); - return IRQ_HANDLED; + rc = IRQ_HANDLED; + goto unlock_out; } else break; } @@ -5189,7 +5221,14 @@ static irqreturn_t ipr_isr(int irq, void *devp) if (unlikely(rc == IRQ_NONE)) rc = ipr_handle_other_interrupt(ioa_cfg, int_reg); +unlock_out: spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); + list_for_each_entry_safe(ipr_cmd, temp, &doneq, queue) { + list_del(&ipr_cmd->queue); + del_timer(&ipr_cmd->timer); + ipr_cmd->fast_done(ipr_cmd); + } + return rc; } @@ -5770,21 +5809,28 @@ static void ipr_scsi_done(struct ipr_cmnd *ipr_cmd) struct ipr_ioa_cfg *ioa_cfg = ipr_cmd->ioa_cfg; struct scsi_cmnd *scsi_cmd = ipr_cmd->scsi_cmd; u32 ioasc = be32_to_cpu(ipr_cmd->s.ioasa.hdr.ioasc); + unsigned long lock_flags; scsi_set_resid(scsi_cmd, be32_to_cpu(ipr_cmd->s.ioasa.hdr.residual_data_len)); if (likely(IPR_IOASC_SENSE_KEY(ioasc) == 0)) { - scsi_dma_unmap(ipr_cmd->scsi_cmd); + scsi_dma_unmap(scsi_cmd); + + spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags); list_add_tail(&ipr_cmd->queue, &ioa_cfg->free_q); scsi_cmd->scsi_done(scsi_cmd); - } else + spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); + } else { + spin_lock_irqsave(ioa_cfg->host->host_lock, lock_flags); ipr_erp_start(ioa_cfg, ipr_cmd); + spin_unlock_irqrestore(ioa_cfg->host->host_lock, lock_flags); + } } /** * ipr_queuecommand - Queue a mid-layer request + * @shost: scsi host struct * @scsi_cmd: scsi command struct - * @done: done function * * This function queues a request generated by the mid-layer. * @@ -5793,61 +5839,61 @@ static void ipr_scsi_done(struct ipr_cmnd *ipr_cmd) * SCSI_MLQUEUE_DEVICE_BUSY if device is busy * SCSI_MLQUEUE_HOST_BUSY if host is busy **/ -static int ipr_queuecommand_lck(struct scsi_cmnd *scsi_cmd, - void (*done) (struct scsi_cmnd *)) +static int ipr_queuecommand(struct Scsi_Host *shost, + struct scsi_cmnd *scsi_cmd) { struct ipr_ioa_cfg *ioa_cfg; struct ipr_resource_entry *res; struct ipr_ioarcb *ioarcb; struct ipr_cmnd *ipr_cmd; - int rc = 0; + unsigned long lock_flags; + int rc; - scsi_cmd->scsi_done = done; - ioa_cfg = (struct ipr_ioa_cfg *)scsi_cmd->device->host->hostdata; - res = scsi_cmd->device->hostdata; + ioa_cfg = (struct ipr_ioa_cfg *)shost->hostdata; + + spin_lock_irqsave(shost->host_lock, lock_flags); scsi_cmd->result = (DID_OK << 16); + res = scsi_cmd->device->hostdata; /* * We are currently blocking all devices due to a host reset * We have told the host to stop giving us new requests, but * ERP ops don't count. FIXME */ - if (unlikely(!ioa_cfg->allow_cmds && !ioa_cfg->ioa_is_dead)) + if (unlikely(!ioa_cfg->allow_cmds && !ioa_cfg->ioa_is_dead)) { + spin_unlock_irqrestore(shost->host_lock, lock_flags); return SCSI_MLQUEUE_HOST_BUSY; + } /* * FIXME - Create scsi_set_host_offline interface * and the ioa_is_dead check can be removed */ if (unlikely(ioa_cfg->ioa_is_dead || !res)) { - memset(scsi_cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE); - scsi_cmd->result = (DID_NO_CONNECT << 16); - scsi_cmd->scsi_done(scsi_cmd); - return 0; + spin_unlock_irqrestore(shost->host_lock, lock_flags); + goto err_nodev; + } + + if (ipr_is_gata(res) && res->sata_port) { + rc = ata_sas_queuecmd(scsi_cmd, res->sata_port->ap); + spin_unlock_irqrestore(shost->host_lock, lock_flags); + return rc; } - if (ipr_is_gata(res) && res->sata_port) - return ata_sas_queuecmd(scsi_cmd, res->sata_port->ap); + ipr_cmd = __ipr_get_free_ipr_cmnd(ioa_cfg); + spin_unlock_irqrestore(shost->host_lock, lock_flags); - ipr_cmd = ipr_get_free_ipr_cmnd(ioa_cfg); + ipr_init_ipr_cmnd(ipr_cmd, ipr_scsi_done); ioarcb = &ipr_cmd->ioarcb; - list_add_tail(&ipr_cmd->queue, &ioa_cfg->pending_q); memcpy(ioarcb->cmd_pkt.cdb, scsi_cmd->cmnd, scsi_cmd->cmd_len); ipr_cmd->scsi_cmd = scsi_cmd; - ioarcb->res_handle = res->res_handle; - ipr_cmd->done = ipr_scsi_done; - ipr_trc_hook(ipr_cmd, IPR_TRACE_START, IPR_GET_RES_PHYS_LOC(res)); + ipr_cmd->done = ipr_scsi_eh_done; if (ipr_is_gscsi(res) || ipr_is_vset_device(res)) { if (scsi_cmd->underflow == 0) ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_ULEN_CHK; - if (res->needs_sync_complete) { - ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_SYNC_COMPLETE; - res->needs_sync_complete = 0; - } - ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_NO_LINK_DESC; if (ipr_is_gscsi(res)) ioarcb->cmd_pkt.flags_lo |= IPR_FLAGS_LO_DELAY_AFTER_RST; @@ -5859,23 +5905,46 @@ static int ipr_queuecommand_lck(struct scsi_cmnd *scsi_cmd, (!ipr_is_gscsi(res) || scsi_cmd->cmnd[0] == IPR_QUERY_RSRC_STATE)) ioarcb->cmd_pkt.request_type = IPR_RQTYPE_IOACMD; - if (likely(rc == 0)) { - if (ioa_cfg->sis64) - rc = ipr_build_ioadl64(ioa_cfg, ipr_cmd); - else - rc = ipr_build_ioadl(ioa_cfg, ipr_cmd); - } + if (ioa_cfg->sis64) + rc = ipr_build_ioadl64(ioa_cfg, ipr_cmd); + else + rc = ipr_build_ioadl(ioa_cfg, ipr_cmd); - if (unlikely(rc != 0)) { - list_move_tail(&ipr_cmd->queue, &ioa_cfg->free_q); + spin_lock_irqsave(shost->host_lock, lock_flags); + if (unlikely(rc || (!ioa_cfg->allow_cmds && !ioa_cfg->ioa_is_dead))) { + list_add_tail(&ipr_cmd->queue, &ioa_cfg->free_q); + spin_unlock_irqrestore(shost->host_lock, lock_flags); + if (!rc) + scsi_dma_unmap(scsi_cmd); return SCSI_MLQUEUE_HOST_BUSY; } + if (unlikely(ioa_cfg->ioa_is_dead)) { + list_add_tail(&ipr_cmd->queue, &ioa_cfg->free_q); + spin_unlock_irqrestore(shost->host_lock, lock_flags); + scsi_dma_unmap(scsi_cmd); + goto err_nodev; + } + + ioarcb->res_handle = res->res_handle; + if (res->needs_sync_complete) { + ioarcb->cmd_pkt.flags_hi |= IPR_FLAGS_HI_SYNC_COMPLETE; + res->needs_sync_complete = 0; + } + list_add_tail(&ipr_cmd->queue, &ioa_cfg->pending_q); + ipr_trc_hook(ipr_cmd, IPR_TRACE_START, IPR_GET_RES_PHYS_LOC(res)); ipr_send_command(ipr_cmd); + spin_unlock_irqrestore(shost->host_lock, lock_flags); return 0; -} -static DEF_SCSI_QCMD(ipr_queuecommand) +err_nodev: + spin_lock_irqsave(shost->host_lock, lock_flags); + memset(scsi_cmd->sense_buffer, 0, SCSI_SENSE_BUFFERSIZE); + scsi_cmd->result = (DID_NO_CONNECT << 16); + scsi_cmd->scsi_done(scsi_cmd); + spin_unlock_irqrestore(shost->host_lock, lock_flags); + return 0; +} /** * ipr_ioctl - IOCTL handler @@ -8775,8 +8844,7 @@ static int __devinit ipr_probe_ioa(struct pci_dev *pdev, ioa_cfg = (struct ipr_ioa_cfg *)host->hostdata; memset(ioa_cfg, 0, sizeof(struct ipr_ioa_cfg)); - ata_host_init(&ioa_cfg->ata_host, &pdev->dev, - sata_port_info.flags, &ipr_sata_ops); + ata_host_init(&ioa_cfg->ata_host, &pdev->dev, &ipr_sata_ops); ioa_cfg->ipr_chip = ipr_get_chip_info(dev_id); diff --git a/drivers/scsi/ipr.h b/drivers/scsi/ipr.h index 153b8bd..c8a137f 100644 --- a/drivers/scsi/ipr.h +++ b/drivers/scsi/ipr.h @@ -38,8 +38,8 @@ /* * Literals */ -#define IPR_DRIVER_VERSION "2.5.3" -#define IPR_DRIVER_DATE "(March 10, 2012)" +#define IPR_DRIVER_VERSION "2.5.4" +#define IPR_DRIVER_DATE "(July 11, 2012)" /* * IPR_MAX_CMD_PER_LUN: This defines the maximum number of outstanding @@ -1525,6 +1525,7 @@ struct ipr_cmnd { struct ata_queued_cmd *qc; struct completion completion; struct timer_list timer; + void (*fast_done) (struct ipr_cmnd *); void (*done) (struct ipr_cmnd *); int (*job_step) (struct ipr_cmnd *); int (*job_step_failed) (struct ipr_cmnd *); diff --git a/drivers/scsi/isci/host.c b/drivers/scsi/isci/host.c index b334fdc..609dafd 100644 --- a/drivers/scsi/isci/host.c +++ b/drivers/scsi/isci/host.c @@ -1044,7 +1044,7 @@ static enum sci_status sci_controller_start(struct isci_host *ihost, return SCI_SUCCESS; } -void isci_host_scan_start(struct Scsi_Host *shost) +void isci_host_start(struct Scsi_Host *shost) { struct isci_host *ihost = SHOST_TO_SAS_HA(shost)->lldd_ha; unsigned long tmo = sci_controller_get_suggested_start_timeout(ihost); @@ -1079,7 +1079,6 @@ static void sci_controller_completion_handler(struct isci_host *ihost) void ireq_done(struct isci_host *ihost, struct isci_request *ireq, struct sas_task *task) { - task->lldd_task = NULL; if (!test_bit(IREQ_ABORT_PATH_ACTIVE, &ireq->flags) && !(task->task_state_flags & SAS_TASK_STATE_ABORTED)) { if (test_bit(IREQ_COMPLETE_IN_TARGET, &ireq->flags)) { @@ -1087,16 +1086,19 @@ void ireq_done(struct isci_host *ihost, struct isci_request *ireq, struct sas_ta dev_dbg(&ihost->pdev->dev, "%s: Normal - ireq/task = %p/%p\n", __func__, ireq, task); - + task->lldd_task = NULL; task->task_done(task); } else { dev_dbg(&ihost->pdev->dev, "%s: Error - ireq/task = %p/%p\n", __func__, ireq, task); - + if (sas_protocol_ata(task->task_proto)) + task->lldd_task = NULL; sas_task_abort(task); } - } + } else + task->lldd_task = NULL; + if (test_and_clear_bit(IREQ_ABORT_PATH_ACTIVE, &ireq->flags)) wake_up_all(&ihost->eventq); @@ -1120,10 +1122,16 @@ void isci_host_completion_routine(unsigned long data) sci_controller_completion_handler(ihost); spin_unlock_irq(&ihost->scic_lock); - /* the coalesence timeout doubles at each encoding step, so + /* + * we subtract SCI_MAX_PORTS to account for the number of dummy TCs + * issued for hardware issue workaround + */ + active = isci_tci_active(ihost) - SCI_MAX_PORTS; + + /* + * the coalesence timeout doubles at each encoding step, so * update it based on the ilog2 value of the outstanding requests */ - active = isci_tci_active(ihost); writel(SMU_ICC_GEN_VAL(NUMBER, active) | SMU_ICC_GEN_VAL(TIMER, ISCI_COALESCE_BASE + ilog2(active)), &ihost->smu_registers->interrupt_coalesce_control); @@ -1973,7 +1981,7 @@ static void sci_controller_afe_initialization(struct isci_host *ihost) } for (phy_id = 0; phy_id < SCI_MAX_PHYS; phy_id++) { - struct scu_afe_transceiver *xcvr = &afe->scu_afe_xcvr[phy_id]; + struct scu_afe_transceiver __iomem *xcvr = &afe->scu_afe_xcvr[phy_id]; const struct sci_phy_oem_params *oem_phy = &oem->phys[phy_id]; int cable_length_long = is_long_cable(phy_id, cable_selection_mask); diff --git a/drivers/scsi/isci/host.h b/drivers/scsi/isci/host.h index 9ab58e0..4911310 100644 --- a/drivers/scsi/isci/host.h +++ b/drivers/scsi/isci/host.h @@ -473,7 +473,7 @@ void sci_controller_remote_device_stopped(struct isci_host *ihost, enum sci_status sci_controller_continue_io(struct isci_request *ireq); int isci_host_scan_finished(struct Scsi_Host *, unsigned long); -void isci_host_scan_start(struct Scsi_Host *); +void isci_host_start(struct Scsi_Host *); u16 isci_alloc_tag(struct isci_host *ihost); enum sci_status isci_free_tag(struct isci_host *ihost, u16 io_tag); void isci_tci_free(struct isci_host *ihost, u16 tci); diff --git a/drivers/scsi/isci/init.c b/drivers/scsi/isci/init.c index 9be45a2..b74050b 100644 --- a/drivers/scsi/isci/init.c +++ b/drivers/scsi/isci/init.c @@ -156,7 +156,7 @@ static struct scsi_host_template isci_sht = { .target_alloc = sas_target_alloc, .slave_configure = sas_slave_configure, .scan_finished = isci_host_scan_finished, - .scan_start = isci_host_scan_start, + .scan_start = isci_host_start, .change_queue_depth = sas_change_queue_depth, .change_queue_type = sas_change_queue_type, .bios_param = sas_bios_param, @@ -644,7 +644,6 @@ static int __devinit isci_pci_probe(struct pci_dev *pdev, const struct pci_devic orom->hdr.version)) { dev_warn(&pdev->dev, "[%d]: invalid oem parameters detected, falling back to firmware\n", i); - devm_kfree(&pdev->dev, orom); orom = NULL; break; } @@ -722,11 +721,67 @@ static void __devexit isci_pci_remove(struct pci_dev *pdev) } } +#ifdef CONFIG_PM +static int isci_suspend(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct isci_host *ihost; + int i; + + for_each_isci_host(i, ihost, pdev) { + sas_suspend_ha(&ihost->sas_ha); + isci_host_deinit(ihost); + } + + pci_save_state(pdev); + pci_disable_device(pdev); + pci_set_power_state(pdev, PCI_D3hot); + + return 0; +} + +static int isci_resume(struct device *dev) +{ + struct pci_dev *pdev = to_pci_dev(dev); + struct isci_host *ihost; + int rc, i; + + pci_set_power_state(pdev, PCI_D0); + pci_restore_state(pdev); + + rc = pcim_enable_device(pdev); + if (rc) { + dev_err(&pdev->dev, + "enabling device failure after resume(%d)\n", rc); + return rc; + } + + pci_set_master(pdev); + + for_each_isci_host(i, ihost, pdev) { + sas_prep_resume_ha(&ihost->sas_ha); + + isci_host_init(ihost); + isci_host_start(ihost->sas_ha.core.shost); + wait_for_start(ihost); + + sas_resume_ha(&ihost->sas_ha); + } + + return 0; +} + +static SIMPLE_DEV_PM_OPS(isci_pm_ops, isci_suspend, isci_resume); +#endif + static struct pci_driver isci_pci_driver = { .name = DRV_NAME, .id_table = isci_id_table, .probe = isci_pci_probe, .remove = __devexit_p(isci_pci_remove), +#ifdef CONFIG_PM + .driver.pm = &isci_pm_ops, +#endif }; static __init int isci_init(void) diff --git a/drivers/scsi/isci/phy.c b/drivers/scsi/isci/phy.c index 18f43d4..cb87b2e 100644 --- a/drivers/scsi/isci/phy.c +++ b/drivers/scsi/isci/phy.c @@ -169,7 +169,7 @@ sci_phy_link_layer_initialization(struct isci_phy *iphy, phy_cap.gen1_no_ssc = 1; if (ihost->oem_parameters.controller.do_enable_ssc) { struct scu_afe_registers __iomem *afe = &ihost->scu_registers->afe; - struct scu_afe_transceiver *xcvr = &afe->scu_afe_xcvr[phy_idx]; + struct scu_afe_transceiver __iomem *xcvr = &afe->scu_afe_xcvr[phy_idx]; struct isci_pci_info *pci_info = to_pci_info(ihost->pdev); bool en_sas = false; bool en_sata = false; @@ -1205,6 +1205,7 @@ static void scu_link_layer_start_oob(struct isci_phy *iphy) /** Reset OOB sequence - start */ val = readl(&ll->phy_configuration); val &= ~(SCU_SAS_PCFG_GEN_BIT(OOB_RESET) | + SCU_SAS_PCFG_GEN_BIT(OOB_ENABLE) | SCU_SAS_PCFG_GEN_BIT(HARD_RESET)); writel(val, &ll->phy_configuration); readl(&ll->phy_configuration); /* flush */ @@ -1236,6 +1237,7 @@ static void scu_link_layer_tx_hard_reset( * to the starting state. */ phy_configuration_value = readl(&iphy->link_layer_registers->phy_configuration); + phy_configuration_value &= ~(SCU_SAS_PCFG_GEN_BIT(OOB_ENABLE)); phy_configuration_value |= (SCU_SAS_PCFG_GEN_BIT(HARD_RESET) | SCU_SAS_PCFG_GEN_BIT(OOB_RESET)); diff --git a/drivers/scsi/isci/probe_roms.c b/drivers/scsi/isci/probe_roms.c index 4d95654..8ac646e 100644 --- a/drivers/scsi/isci/probe_roms.c +++ b/drivers/scsi/isci/probe_roms.c @@ -104,7 +104,6 @@ struct isci_orom *isci_request_oprom(struct pci_dev *pdev) if (i >= len) { dev_err(&pdev->dev, "oprom parse error\n"); - devm_kfree(&pdev->dev, rom); rom = NULL; } pci_unmap_biosrom(oprom); diff --git a/drivers/scsi/isci/remote_node_context.h b/drivers/scsi/isci/remote_node_context.h index a703b9c..c7ee81d 100644 --- a/drivers/scsi/isci/remote_node_context.h +++ b/drivers/scsi/isci/remote_node_context.h @@ -212,7 +212,7 @@ enum sci_status sci_remote_node_context_destruct(struct sci_remote_node_context scics_sds_remote_node_context_callback callback, void *callback_parameter); enum sci_status sci_remote_node_context_suspend(struct sci_remote_node_context *sci_rnc, - u32 suspend_type, + enum sci_remote_node_suspension_reasons reason, u32 suspension_code); enum sci_status sci_remote_node_context_resume(struct sci_remote_node_context *sci_rnc, scics_sds_remote_node_context_callback cb_fn, diff --git a/drivers/scsi/iscsi_tcp.c b/drivers/scsi/iscsi_tcp.c index 9220861..1b91ca0 100644 --- a/drivers/scsi/iscsi_tcp.c +++ b/drivers/scsi/iscsi_tcp.c @@ -55,7 +55,7 @@ static struct scsi_transport_template *iscsi_sw_tcp_scsi_transport; static struct scsi_host_template iscsi_sw_tcp_sht; static struct iscsi_transport iscsi_sw_tcp_transport; -static unsigned int iscsi_max_lun = 512; +static unsigned int iscsi_max_lun = ~0; module_param_named(max_lun, iscsi_max_lun, uint, S_IRUGO); static int iscsi_sw_tcp_dbg; diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c index a59fcdc..bdb81cd 100644 --- a/drivers/scsi/libsas/sas_ata.c +++ b/drivers/scsi/libsas/sas_ata.c @@ -580,10 +580,7 @@ int sas_ata_init(struct domain_device *found_dev) struct ata_port *ap; int rc; - ata_host_init(&found_dev->sata_dev.ata_host, - ha->dev, - sata_port_info.flags, - &sas_sata_ops); + ata_host_init(&found_dev->sata_dev.ata_host, ha->dev, &sas_sata_ops); ap = ata_sas_port_alloc(&found_dev->sata_dev.ata_host, &sata_port_info, shost); @@ -700,6 +697,92 @@ void sas_probe_sata(struct asd_sas_port *port) if (ata_dev_disabled(sas_to_ata_dev(dev))) sas_fail_probe(dev, __func__, -ENODEV); } + +} + +static bool sas_ata_flush_pm_eh(struct asd_sas_port *port, const char *func) +{ + struct domain_device *dev, *n; + bool retry = false; + + list_for_each_entry_safe(dev, n, &port->dev_list, dev_list_node) { + int rc; + + if (!dev_is_sata(dev)) + continue; + + sas_ata_wait_eh(dev); + rc = dev->sata_dev.pm_result; + if (rc == -EAGAIN) + retry = true; + else if (rc) { + /* since we don't have a + * ->port_{suspend|resume} routine in our + * ata_port ops, and no entanglements with + * acpi, suspend should just be mechanical trip + * through eh, catch cases where these + * assumptions are invalidated + */ + WARN_ONCE(1, "failed %s %s error: %d\n", func, + dev_name(&dev->rphy->dev), rc); + } + + /* if libata failed to power manage the device, tear it down */ + if (ata_dev_disabled(sas_to_ata_dev(dev))) + sas_fail_probe(dev, func, -ENODEV); + } + + return retry; +} + +void sas_suspend_sata(struct asd_sas_port *port) +{ + struct domain_device *dev; + + retry: + mutex_lock(&port->ha->disco_mutex); + list_for_each_entry(dev, &port->dev_list, dev_list_node) { + struct sata_device *sata; + + if (!dev_is_sata(dev)) + continue; + + sata = &dev->sata_dev; + if (sata->ap->pm_mesg.event == PM_EVENT_SUSPEND) + continue; + + sata->pm_result = -EIO; + ata_sas_port_async_suspend(sata->ap, &sata->pm_result); + } + mutex_unlock(&port->ha->disco_mutex); + + if (sas_ata_flush_pm_eh(port, __func__)) + goto retry; +} + +void sas_resume_sata(struct asd_sas_port *port) +{ + struct domain_device *dev; + + retry: + mutex_lock(&port->ha->disco_mutex); + list_for_each_entry(dev, &port->dev_list, dev_list_node) { + struct sata_device *sata; + + if (!dev_is_sata(dev)) + continue; + + sata = &dev->sata_dev; + if (sata->ap->pm_mesg.event == PM_EVENT_ON) + continue; + + sata->pm_result = -EIO; + ata_sas_port_async_resume(sata->ap, &sata->pm_result); + } + mutex_unlock(&port->ha->disco_mutex); + + if (sas_ata_flush_pm_eh(port, __func__)) + goto retry; } /** diff --git a/drivers/scsi/libsas/sas_discover.c b/drivers/scsi/libsas/sas_discover.c index 3e9dc1a..a0c3003 100644 --- a/drivers/scsi/libsas/sas_discover.c +++ b/drivers/scsi/libsas/sas_discover.c @@ -24,6 +24,7 @@ #include <linux/scatterlist.h> #include <linux/slab.h> +#include <linux/async.h> #include <scsi/scsi_host.h> #include <scsi/scsi_eh.h> #include "sas_internal.h" @@ -180,16 +181,18 @@ int sas_notify_lldd_dev_found(struct domain_device *dev) struct Scsi_Host *shost = sas_ha->core.shost; struct sas_internal *i = to_sas_internal(shost->transportt); - if (i->dft->lldd_dev_found) { - res = i->dft->lldd_dev_found(dev); - if (res) { - printk("sas: driver on pcidev %s cannot handle " - "device %llx, error:%d\n", - dev_name(sas_ha->dev), - SAS_ADDR(dev->sas_addr), res); - } - kref_get(&dev->kref); + if (!i->dft->lldd_dev_found) + return 0; + + res = i->dft->lldd_dev_found(dev); + if (res) { + printk("sas: driver on pcidev %s cannot handle " + "device %llx, error:%d\n", + dev_name(sas_ha->dev), + SAS_ADDR(dev->sas_addr), res); } + set_bit(SAS_DEV_FOUND, &dev->state); + kref_get(&dev->kref); return res; } @@ -200,7 +203,10 @@ void sas_notify_lldd_dev_gone(struct domain_device *dev) struct Scsi_Host *shost = sas_ha->core.shost; struct sas_internal *i = to_sas_internal(shost->transportt); - if (i->dft->lldd_dev_gone) { + if (!i->dft->lldd_dev_gone) + return; + + if (test_and_clear_bit(SAS_DEV_FOUND, &dev->state)) { i->dft->lldd_dev_gone(dev); sas_put_device(dev); } @@ -234,6 +240,47 @@ static void sas_probe_devices(struct work_struct *work) } } +static void sas_suspend_devices(struct work_struct *work) +{ + struct asd_sas_phy *phy; + struct domain_device *dev; + struct sas_discovery_event *ev = to_sas_discovery_event(work); + struct asd_sas_port *port = ev->port; + struct Scsi_Host *shost = port->ha->core.shost; + struct sas_internal *si = to_sas_internal(shost->transportt); + + clear_bit(DISCE_SUSPEND, &port->disc.pending); + + sas_suspend_sata(port); + + /* lldd is free to forget the domain_device across the + * suspension, we force the issue here to keep the reference + * counts aligned + */ + list_for_each_entry(dev, &port->dev_list, dev_list_node) + sas_notify_lldd_dev_gone(dev); + + /* we are suspending, so we know events are disabled and + * phy_list is not being mutated + */ + list_for_each_entry(phy, &port->phy_list, port_phy_el) { + if (si->dft->lldd_port_formed) + si->dft->lldd_port_deformed(phy); + phy->suspended = 1; + port->suspended = 1; + } +} + +static void sas_resume_devices(struct work_struct *work) +{ + struct sas_discovery_event *ev = to_sas_discovery_event(work); + struct asd_sas_port *port = ev->port; + + clear_bit(DISCE_RESUME, &port->disc.pending); + + sas_resume_sata(port); +} + /** * sas_discover_end_dev -- discover an end device (SSP, etc) * @end: pointer to domain device of interest @@ -530,6 +577,8 @@ void sas_init_disc(struct sas_discovery *disc, struct asd_sas_port *port) [DISCE_DISCOVER_DOMAIN] = sas_discover_domain, [DISCE_REVALIDATE_DOMAIN] = sas_revalidate_domain, [DISCE_PROBE] = sas_probe_devices, + [DISCE_SUSPEND] = sas_suspend_devices, + [DISCE_RESUME] = sas_resume_devices, [DISCE_DESTRUCT] = sas_destruct_devices, }; diff --git a/drivers/scsi/libsas/sas_dump.c b/drivers/scsi/libsas/sas_dump.c index fc46093..cd6f99c 100644 --- a/drivers/scsi/libsas/sas_dump.c +++ b/drivers/scsi/libsas/sas_dump.c @@ -41,6 +41,7 @@ static const char *sas_phye_str[] = { [1] = "PHYE_OOB_DONE", [2] = "PHYE_OOB_ERROR", [3] = "PHYE_SPINUP_HOLD", + [4] = "PHYE_RESUME_TIMEOUT", }; void sas_dprint_porte(int phyid, enum port_event pe) diff --git a/drivers/scsi/libsas/sas_event.c b/drivers/scsi/libsas/sas_event.c index 789c4d8..aadbd53 100644 --- a/drivers/scsi/libsas/sas_event.c +++ b/drivers/scsi/libsas/sas_event.c @@ -134,7 +134,7 @@ static void notify_port_event(struct asd_sas_phy *phy, enum port_event event) &phy->port_events[event].work, ha); } -static void notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) +void sas_notify_phy_event(struct asd_sas_phy *phy, enum phy_event event) { struct sas_ha_struct *ha = phy->ha; @@ -159,7 +159,7 @@ int sas_init_events(struct sas_ha_struct *sas_ha) sas_ha->notify_ha_event = notify_ha_event; sas_ha->notify_port_event = notify_port_event; - sas_ha->notify_phy_event = notify_phy_event; + sas_ha->notify_phy_event = sas_notify_phy_event; return 0; } diff --git a/drivers/scsi/libsas/sas_init.c b/drivers/scsi/libsas/sas_init.c index 014297c..dbc8a79 100644 --- a/drivers/scsi/libsas/sas_init.c +++ b/drivers/scsi/libsas/sas_init.c @@ -178,7 +178,7 @@ Undo_phys: return error; } -int sas_unregister_ha(struct sas_ha_struct *sas_ha) +static void sas_disable_events(struct sas_ha_struct *sas_ha) { /* Set the state to unregistered to avoid further unchained * events to be queued, and flush any in-progress drainers @@ -189,7 +189,11 @@ int sas_unregister_ha(struct sas_ha_struct *sas_ha) spin_unlock_irq(&sas_ha->lock); __sas_drain_work(sas_ha); mutex_unlock(&sas_ha->drain_mutex); +} +int sas_unregister_ha(struct sas_ha_struct *sas_ha) +{ + sas_disable_events(sas_ha); sas_unregister_ports(sas_ha); /* flush unregistration work */ @@ -381,6 +385,90 @@ int sas_set_phy_speed(struct sas_phy *phy, return ret; } +void sas_prep_resume_ha(struct sas_ha_struct *ha) +{ + int i; + + set_bit(SAS_HA_REGISTERED, &ha->state); + + /* clear out any stale link events/data from the suspension path */ + for (i = 0; i < ha->num_phys; i++) { + struct asd_sas_phy *phy = ha->sas_phy[i]; + + memset(phy->attached_sas_addr, 0, SAS_ADDR_SIZE); + phy->port_events_pending = 0; + phy->phy_events_pending = 0; + phy->frame_rcvd_size = 0; + } +} +EXPORT_SYMBOL(sas_prep_resume_ha); + +static int phys_suspended(struct sas_ha_struct *ha) +{ + int i, rc = 0; + + for (i = 0; i < ha->num_phys; i++) { + struct asd_sas_phy *phy = ha->sas_phy[i]; + + if (phy->suspended) + rc++; + } + + return rc; +} + +void sas_resume_ha(struct sas_ha_struct *ha) +{ + const unsigned long tmo = msecs_to_jiffies(25000); + int i; + + /* deform ports on phys that did not resume + * at this point we may be racing the phy coming back (as posted + * by the lldd). So we post the event and once we are in the + * libsas context check that the phy remains suspended before + * tearing it down. + */ + i = phys_suspended(ha); + if (i) + dev_info(ha->dev, "waiting up to 25 seconds for %d phy%s to resume\n", + i, i > 1 ? "s" : ""); + wait_event_timeout(ha->eh_wait_q, phys_suspended(ha) == 0, tmo); + for (i = 0; i < ha->num_phys; i++) { + struct asd_sas_phy *phy = ha->sas_phy[i]; + + if (phy->suspended) { + dev_warn(&phy->phy->dev, "resume timeout\n"); + sas_notify_phy_event(phy, PHYE_RESUME_TIMEOUT); + } + } + + /* all phys are back up or timed out, turn on i/o so we can + * flush out disks that did not return + */ + scsi_unblock_requests(ha->core.shost); + sas_drain_work(ha); +} +EXPORT_SYMBOL(sas_resume_ha); + +void sas_suspend_ha(struct sas_ha_struct *ha) +{ + int i; + + sas_disable_events(ha); + scsi_block_requests(ha->core.shost); + for (i = 0; i < ha->num_phys; i++) { + struct asd_sas_port *port = ha->sas_port[i]; + + sas_discover_event(port, DISCE_SUSPEND); + } + + /* flush suspend events while unregistered */ + mutex_lock(&ha->drain_mutex); + __sas_drain_work(ha); + mutex_unlock(&ha->drain_mutex); +} +EXPORT_SYMBOL(sas_suspend_ha); + static void sas_phy_release(struct sas_phy *phy) { kfree(phy->hostdata); diff --git a/drivers/scsi/libsas/sas_internal.h b/drivers/scsi/libsas/sas_internal.h index 507e4cf..1de6796 100644 --- a/drivers/scsi/libsas/sas_internal.h +++ b/drivers/scsi/libsas/sas_internal.h @@ -89,6 +89,7 @@ int sas_smp_phy_control(struct domain_device *dev, int phy_id, enum phy_func phy_func, struct sas_phy_linkrates *); int sas_smp_get_phy_events(struct sas_phy *phy); +void sas_notify_phy_event(struct asd_sas_phy *phy, enum phy_event event); void sas_device_set_phy(struct domain_device *dev, struct sas_port *port); struct domain_device *sas_find_dev_by_rphy(struct sas_rphy *rphy); struct domain_device *sas_ex_to_ata(struct domain_device *ex_dev, int phy_id); diff --git a/drivers/scsi/libsas/sas_phy.c b/drivers/scsi/libsas/sas_phy.c index 521422e..cdee446c 100644 --- a/drivers/scsi/libsas/sas_phy.c +++ b/drivers/scsi/libsas/sas_phy.c @@ -94,6 +94,25 @@ static void sas_phye_spinup_hold(struct work_struct *work) i->dft->lldd_control_phy(phy, PHY_FUNC_RELEASE_SPINUP_HOLD, NULL); } +static void sas_phye_resume_timeout(struct work_struct *work) +{ + struct asd_sas_event *ev = to_asd_sas_event(work); + struct asd_sas_phy *phy = ev->phy; + + clear_bit(PHYE_RESUME_TIMEOUT, &phy->phy_events_pending); + + /* phew, lldd got the phy back in the nick of time */ + if (!phy->suspended) { + dev_info(&phy->phy->dev, "resume timeout cancelled\n"); + return; + } + + phy->error = 0; + phy->suspended = 0; + sas_deform_port(phy, 1); +} + + /* ---------- Phy class registration ---------- */ int sas_register_phys(struct sas_ha_struct *sas_ha) @@ -105,6 +124,8 @@ int sas_register_phys(struct sas_ha_struct *sas_ha) [PHYE_OOB_DONE] = sas_phye_oob_done, [PHYE_OOB_ERROR] = sas_phye_oob_error, [PHYE_SPINUP_HOLD] = sas_phye_spinup_hold, + [PHYE_RESUME_TIMEOUT] = sas_phye_resume_timeout, + }; static const work_func_t sas_port_event_fns[PORT_NUM_EVENTS] = { diff --git a/drivers/scsi/libsas/sas_port.c b/drivers/scsi/libsas/sas_port.c index e884a8c..1398b71 100644 --- a/drivers/scsi/libsas/sas_port.c +++ b/drivers/scsi/libsas/sas_port.c @@ -39,6 +39,49 @@ static bool phy_is_wideport_member(struct asd_sas_port *port, struct asd_sas_phy return true; } +static void sas_resume_port(struct asd_sas_phy *phy) +{ + struct domain_device *dev; + struct asd_sas_port *port = phy->port; + struct sas_ha_struct *sas_ha = phy->ha; + struct sas_internal *si = to_sas_internal(sas_ha->core.shost->transportt); + + if (si->dft->lldd_port_formed) + si->dft->lldd_port_formed(phy); + + if (port->suspended) + port->suspended = 0; + else { + /* we only need to handle "link returned" actions once */ + return; + } + + /* if the port came back: + * 1/ presume every device came back + * 2/ force the next revalidation to check all expander phys + */ + list_for_each_entry(dev, &port->dev_list, dev_list_node) { + int i, rc; + + rc = sas_notify_lldd_dev_found(dev); + if (rc) { + sas_unregister_dev(port, dev); + continue; + } + + if (dev->dev_type == EDGE_DEV || dev->dev_type == FANOUT_DEV) { + dev->ex_dev.ex_change_count = -1; + for (i = 0; i < dev->ex_dev.num_phys; i++) { + struct ex_phy *phy = &dev->ex_dev.ex_phy[i]; + + phy->phy_change_count = -1; + } + } + } + + sas_discover_event(port, DISCE_RESUME); +} + /** * sas_form_port -- add this phy to a port * @phy: the phy of interest @@ -58,7 +101,14 @@ static void sas_form_port(struct asd_sas_phy *phy) if (port) { if (!phy_is_wideport_member(port, phy)) sas_deform_port(phy, 0); - else { + else if (phy->suspended) { + phy->suspended = 0; + sas_resume_port(phy); + + /* phy came back, try to cancel the timeout */ + wake_up(&sas_ha->eh_wait_q); + return; + } else { SAS_DPRINTK("%s: phy%d belongs to port%d already(%d)!\n", __func__, phy->id, phy->port->id, phy->port->num_phys); diff --git a/drivers/scsi/lpfc/lpfc.h b/drivers/scsi/lpfc/lpfc.h index a65c05a..a184c24 100644 --- a/drivers/scsi/lpfc/lpfc.h +++ b/drivers/scsi/lpfc/lpfc.h @@ -73,6 +73,8 @@ struct lpfc_sli2_slim; #define LPFC_HB_MBOX_INTERVAL 5 /* Heart beat interval in seconds. */ #define LPFC_HB_MBOX_TIMEOUT 30 /* Heart beat timeout in seconds. */ +#define LPFC_LOOK_AHEAD_OFF 0 /* Look ahead logic is turned off */ + /* Error Attention event polling interval */ #define LPFC_ERATT_POLL_INTERVAL 5 /* EATT poll interval in seconds */ @@ -684,6 +686,7 @@ struct lpfc_hba { #define LPFC_FCF_FOV 1 /* Fast fcf failover */ #define LPFC_FCF_PRIORITY 2 /* Priority fcf failover */ uint32_t cfg_fcf_failover_policy; + uint32_t cfg_fcp_io_sched; uint32_t cfg_cr_delay; uint32_t cfg_cr_count; uint32_t cfg_multi_ring_support; @@ -695,6 +698,7 @@ struct lpfc_hba { uint32_t cfg_fcp_imax; uint32_t cfg_fcp_wq_count; uint32_t cfg_fcp_eq_count; + uint32_t cfg_fcp_io_channel; uint32_t cfg_sg_seg_cnt; uint32_t cfg_prot_sg_seg_cnt; uint32_t cfg_sg_dma_buf_size; @@ -732,7 +736,7 @@ struct lpfc_hba { uint32_t hbq_count; /* Count of configured HBQs */ struct hbq_s hbqs[LPFC_MAX_HBQS]; /* local copy of hbq indicies */ - uint32_t fcp_qidx; /* next work queue to post work to */ + atomic_t fcp_qidx; /* next work queue to post work to */ unsigned long pci_bar0_map; /* Physical address for PCI BAR0 */ unsigned long pci_bar1_map; /* Physical address for PCI BAR1 */ diff --git a/drivers/scsi/lpfc/lpfc_attr.c b/drivers/scsi/lpfc/lpfc_attr.c index adef5bb..b032562 100644 --- a/drivers/scsi/lpfc/lpfc_attr.c +++ b/drivers/scsi/lpfc/lpfc_attr.c @@ -3643,18 +3643,25 @@ lpfc_fcp_imax_store(struct device *dev, struct device_attribute *attr, struct lpfc_hba *phba = vport->phba; int val = 0, i; + /* fcp_imax is only valid for SLI4 */ + if (phba->sli_rev != LPFC_SLI_REV4) + return -EINVAL; + /* Sanity check on user data */ if (!isdigit(buf[0])) return -EINVAL; if (sscanf(buf, "%i", &val) != 1) return -EINVAL; - /* Value range is [636,651042] */ - if (val < LPFC_MIM_IMAX || val > LPFC_DMULT_CONST) + /* + * Value range for the HBA is [5000,5000000] + * The value for each EQ depends on how many EQs are configured. + */ + if (val < LPFC_MIN_IMAX || val > LPFC_MAX_IMAX) return -EINVAL; phba->cfg_fcp_imax = (uint32_t)val; - for (i = 0; i < phba->cfg_fcp_eq_count; i += LPFC_MAX_EQ_DELAY) + for (i = 0; i < phba->cfg_fcp_io_channel; i += LPFC_MAX_EQ_DELAY) lpfc_modify_fcp_eq_delay(phba, i); return strlen(buf); @@ -3662,13 +3669,14 @@ lpfc_fcp_imax_store(struct device *dev, struct device_attribute *attr, /* # lpfc_fcp_imax: The maximum number of fast-path FCP interrupts per second +# for the HBA. # -# Value range is [636,651042]. Default value is 10000. +# Value range is [5,000 to 5,000,000]. Default value is 50,000. */ -static int lpfc_fcp_imax = LPFC_FP_DEF_IMAX; +static int lpfc_fcp_imax = LPFC_DEF_IMAX; module_param(lpfc_fcp_imax, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(lpfc_fcp_imax, - "Set the maximum number of fast-path FCP interrupts per second"); + "Set the maximum number of FCP interrupts per second per HBA"); lpfc_param_show(fcp_imax) /** @@ -3687,14 +3695,19 @@ lpfc_param_show(fcp_imax) static int lpfc_fcp_imax_init(struct lpfc_hba *phba, int val) { - if (val >= LPFC_MIM_IMAX && val <= LPFC_DMULT_CONST) { + if (phba->sli_rev != LPFC_SLI_REV4) { + phba->cfg_fcp_imax = 0; + return 0; + } + + if (val >= LPFC_MIN_IMAX && val <= LPFC_MAX_IMAX) { phba->cfg_fcp_imax = val; return 0; } lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "3016 fcp_imax: %d out of range, using default\n", val); - phba->cfg_fcp_imax = LPFC_FP_DEF_IMAX; + phba->cfg_fcp_imax = LPFC_DEF_IMAX; return 0; } @@ -3765,6 +3778,16 @@ static DEVICE_ATTR(lpfc_max_scsicmpl_time, S_IRUGO | S_IWUSR, LPFC_ATTR_R(ack0, 0, 0, 1, "Enable ACK0 support"); /* +# lpfc_fcp_io_sched: Determine scheduling algrithmn for issuing FCP cmds +# range is [0,1]. Default value is 0. +# For [0], FCP commands are issued to Work Queues ina round robin fashion. +# For [1], FCP commands are issued to a Work Queue associated with the +# current CPU. +*/ +LPFC_ATTR_RW(fcp_io_sched, 0, 0, 1, "Determine scheduling algrithmn for " + "issuing commands [0] - Round Robin, [1] - Current CPU"); + +/* # lpfc_cr_delay & lpfc_cr_count: Default values for I/O colaesing # cr_delay (msec) or cr_count outstanding commands. cr_delay can take # value [0,63]. cr_count can take value [1,255]. Default value of cr_delay @@ -3844,21 +3867,33 @@ LPFC_ATTR_R(use_msi, 2, 0, 2, "Use Message Signaled Interrupts (1) or " /* # lpfc_fcp_wq_count: Set the number of fast-path FCP work queues +# This parameter is ignored and will eventually be depricated # -# Value range is [1,31]. Default value is 4. +# Value range is [1,7]. Default value is 4. */ -LPFC_ATTR_R(fcp_wq_count, LPFC_FP_WQN_DEF, LPFC_FP_WQN_MIN, LPFC_FP_WQN_MAX, +LPFC_ATTR_R(fcp_wq_count, LPFC_FCP_IO_CHAN_DEF, LPFC_FCP_IO_CHAN_MIN, + LPFC_FCP_IO_CHAN_MAX, "Set the number of fast-path FCP work queues, if possible"); /* -# lpfc_fcp_eq_count: Set the number of fast-path FCP event queues +# lpfc_fcp_eq_count: Set the number of FCP EQ/CQ/WQ IO channels # -# Value range is [1,7]. Default value is 1. +# Value range is [1,7]. Default value is 4. */ -LPFC_ATTR_R(fcp_eq_count, LPFC_FP_EQN_DEF, LPFC_FP_EQN_MIN, LPFC_FP_EQN_MAX, +LPFC_ATTR_R(fcp_eq_count, LPFC_FCP_IO_CHAN_DEF, LPFC_FCP_IO_CHAN_MIN, + LPFC_FCP_IO_CHAN_MAX, "Set the number of fast-path FCP event queues, if possible"); /* +# lpfc_fcp_io_channel: Set the number of FCP EQ/CQ/WQ IO channels +# +# Value range is [1,7]. Default value is 4. +*/ +LPFC_ATTR_R(fcp_io_channel, LPFC_FCP_IO_CHAN_DEF, LPFC_FCP_IO_CHAN_MIN, + LPFC_FCP_IO_CHAN_MAX, + "Set the number of FCP I/O channels"); + +/* # lpfc_enable_hba_reset: Allow or prevent HBA resets to the hardware. # 0 = HBA resets disabled # 1 = HBA resets enabled (default) @@ -3883,6 +3918,17 @@ LPFC_ATTR_R(enable_hba_heartbeat, 0, 0, 1, "Enable HBA Heartbeat."); LPFC_ATTR_R(enable_bg, 0, 0, 1, "Enable BlockGuard Support"); /* +# lpfc_fcp_look_ahead: Look ahead for completions in FCP start routine +# 0 = disabled (default) +# 1 = enabled +# Value range is [0,1]. Default value is 0. +*/ +unsigned int lpfc_fcp_look_ahead = LPFC_LOOK_AHEAD_OFF; + +module_param(lpfc_fcp_look_ahead, uint, S_IRUGO); +MODULE_PARM_DESC(lpfc_fcp_look_ahead, "Look ahead for completions"); + +/* # lpfc_prot_mask: i # - Bit mask of host protection capabilities used to register with the # SCSI mid-layer @@ -3976,6 +4022,7 @@ struct device_attribute *lpfc_hba_attrs[] = { &dev_attr_lpfc_topology, &dev_attr_lpfc_scan_down, &dev_attr_lpfc_link_speed, + &dev_attr_lpfc_fcp_io_sched, &dev_attr_lpfc_cr_delay, &dev_attr_lpfc_cr_count, &dev_attr_lpfc_multi_ring_support, @@ -4002,6 +4049,7 @@ struct device_attribute *lpfc_hba_attrs[] = { &dev_attr_lpfc_fcp_imax, &dev_attr_lpfc_fcp_wq_count, &dev_attr_lpfc_fcp_eq_count, + &dev_attr_lpfc_fcp_io_channel, &dev_attr_lpfc_enable_bg, &dev_attr_lpfc_soft_wwnn, &dev_attr_lpfc_soft_wwpn, @@ -4964,6 +5012,7 @@ struct fc_function_template lpfc_vport_transport_functions = { void lpfc_get_cfgparam(struct lpfc_hba *phba) { + lpfc_fcp_io_sched_init(phba, lpfc_fcp_io_sched); lpfc_cr_delay_init(phba, lpfc_cr_delay); lpfc_cr_count_init(phba, lpfc_cr_count); lpfc_multi_ring_support_init(phba, lpfc_multi_ring_support); @@ -4980,6 +5029,7 @@ lpfc_get_cfgparam(struct lpfc_hba *phba) lpfc_fcp_imax_init(phba, lpfc_fcp_imax); lpfc_fcp_wq_count_init(phba, lpfc_fcp_wq_count); lpfc_fcp_eq_count_init(phba, lpfc_fcp_eq_count); + lpfc_fcp_io_channel_init(phba, lpfc_fcp_io_channel); lpfc_enable_hba_reset_init(phba, lpfc_enable_hba_reset); lpfc_enable_hba_heartbeat_init(phba, lpfc_enable_hba_heartbeat); lpfc_enable_bg_init(phba, lpfc_enable_bg); diff --git a/drivers/scsi/lpfc/lpfc_bsg.c b/drivers/scsi/lpfc/lpfc_bsg.c index 253d9a8..f7368eb 100644 --- a/drivers/scsi/lpfc/lpfc_bsg.c +++ b/drivers/scsi/lpfc/lpfc_bsg.c @@ -195,7 +195,7 @@ lpfc_bsg_send_mgmt_cmd_cmp(struct lpfc_hba *phba, if (rsp->ulpStatus) { if (rsp->ulpStatus == IOSTAT_LOCAL_REJECT) { - switch (rsp->un.ulpWord[4] & 0xff) { + switch (rsp->un.ulpWord[4] & IOERR_PARAM_MASK) { case IOERR_SEQUENCE_TIMEOUT: rc = -ETIMEDOUT; break; @@ -1234,7 +1234,7 @@ lpfc_issue_ct_rsp_cmp(struct lpfc_hba *phba, if (rsp->ulpStatus) { if (rsp->ulpStatus == IOSTAT_LOCAL_REJECT) { - switch (rsp->un.ulpWord[4] & 0xff) { + switch (rsp->un.ulpWord[4] & IOERR_PARAM_MASK) { case IOERR_SEQUENCE_TIMEOUT: rc = -ETIMEDOUT; break; @@ -1714,6 +1714,8 @@ lpfc_sli4_bsg_set_link_diag_state(struct lpfc_hba *phba, uint32_t diag) phba->sli4_hba.lnk_info.lnk_no); link_diag_state = &pmboxq->u.mqe.un.link_diag_state; + bf_set(lpfc_mbx_set_diag_state_diag_bit_valid, &link_diag_state->u.req, + LPFC_DIAG_STATE_DIAG_BIT_VALID_CHANGE); bf_set(lpfc_mbx_set_diag_state_link_num, &link_diag_state->u.req, phba->sli4_hba.lnk_info.lnk_no); bf_set(lpfc_mbx_set_diag_state_link_type, &link_diag_state->u.req, @@ -4796,7 +4798,7 @@ lpfc_bsg_menlo_cmd_cmp(struct lpfc_hba *phba, menlo_resp->xri = rsp->ulpContext; if (rsp->ulpStatus) { if (rsp->ulpStatus == IOSTAT_LOCAL_REJECT) { - switch (rsp->un.ulpWord[4] & 0xff) { + switch (rsp->un.ulpWord[4] & IOERR_PARAM_MASK) { case IOERR_SEQUENCE_TIMEOUT: rc = -ETIMEDOUT; break; diff --git a/drivers/scsi/lpfc/lpfc_crtn.h b/drivers/scsi/lpfc/lpfc_crtn.h index 8a2a514..e470c48 100644 --- a/drivers/scsi/lpfc/lpfc_crtn.h +++ b/drivers/scsi/lpfc/lpfc_crtn.h @@ -196,8 +196,7 @@ irqreturn_t lpfc_sli_intr_handler(int, void *); irqreturn_t lpfc_sli_sp_intr_handler(int, void *); irqreturn_t lpfc_sli_fp_intr_handler(int, void *); irqreturn_t lpfc_sli4_intr_handler(int, void *); -irqreturn_t lpfc_sli4_sp_intr_handler(int, void *); -irqreturn_t lpfc_sli4_fp_intr_handler(int, void *); +irqreturn_t lpfc_sli4_hba_intr_handler(int, void *); void lpfc_read_rev(struct lpfc_hba *, LPFC_MBOXQ_t *); void lpfc_sli4_swap_str(struct lpfc_hba *, LPFC_MBOXQ_t *); @@ -391,6 +390,7 @@ extern spinlock_t pgcnt_lock; extern unsigned int pgcnt; extern unsigned int lpfc_prot_mask; extern unsigned char lpfc_prot_guard; +extern unsigned int lpfc_fcp_look_ahead; /* Interface exported by fabric iocb scheduler */ void lpfc_fabric_abort_nport(struct lpfc_nodelist *); @@ -457,6 +457,8 @@ int lpfc_sli4_queue_create(struct lpfc_hba *); void lpfc_sli4_queue_destroy(struct lpfc_hba *); void lpfc_sli4_abts_err_handler(struct lpfc_hba *, struct lpfc_nodelist *, struct sli4_wcqe_xri_aborted *); +void lpfc_sli_abts_recover_port(struct lpfc_vport *, + struct lpfc_nodelist *); int lpfc_hba_init_link_fc_topology(struct lpfc_hba *, uint32_t, uint32_t); int lpfc_issue_reg_vfi(struct lpfc_vport *); int lpfc_issue_unreg_vfi(struct lpfc_vport *); diff --git a/drivers/scsi/lpfc/lpfc_ct.c b/drivers/scsi/lpfc/lpfc_ct.c index 93e96b3..7ffabb7 100644 --- a/drivers/scsi/lpfc/lpfc_ct.c +++ b/drivers/scsi/lpfc/lpfc_ct.c @@ -104,7 +104,8 @@ lpfc_ct_unsol_event(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, if (unlikely(icmd->ulpStatus == IOSTAT_NEED_BUFFER)) { lpfc_sli_hbqbuf_add_hbqs(phba, LPFC_ELS_HBQ); } else if ((icmd->ulpStatus == IOSTAT_LOCAL_REJECT) && - ((icmd->un.ulpWord[4] & 0xff) == IOERR_RCV_BUFFER_WAITING)) { + ((icmd->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_RCV_BUFFER_WAITING)) { /* Not enough posted buffers; Try posting more buffers */ phba->fc_stat.NoRcvBuf++; if (!(phba->sli3_options & LPFC_SLI3_HBQ_ENABLED)) @@ -633,7 +634,8 @@ lpfc_cmpl_ct_cmd_gid_ft(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, /* Check for retry */ if (vport->fc_ns_retry < LPFC_MAX_NS_RETRY) { if (irsp->ulpStatus != IOSTAT_LOCAL_REJECT || - irsp->un.ulpWord[4] != IOERR_NO_RESOURCES) + (irsp->un.ulpWord[4] && IOERR_PARAM_MASK) != + IOERR_NO_RESOURCES) vport->fc_ns_retry++; /* CT command is being retried */ @@ -783,7 +785,9 @@ lpfc_cmpl_ct_cmd_gff_id(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, if (cmdiocb->retry < LPFC_MAX_NS_RETRY) { retry = 1; if (irsp->ulpStatus == IOSTAT_LOCAL_REJECT) { - switch (irsp->un.ulpWord[4]) { + switch ((irsp->un.ulpWord[4] & + IOERR_PARAM_MASK)) { + case IOERR_NO_RESOURCES: /* We don't increment the retry * count for this case. @@ -908,8 +912,10 @@ lpfc_cmpl_ct(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, cmdcode, irsp->ulpStatus, irsp->un.ulpWord[4]); if ((irsp->ulpStatus == IOSTAT_LOCAL_REJECT) && - ((irsp->un.ulpWord[4] == IOERR_SLI_DOWN) || - (irsp->un.ulpWord[4] == IOERR_SLI_ABORTED))) + (((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_SLI_DOWN) || + ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_SLI_ABORTED))) goto out; retry = cmdiocb->retry; diff --git a/drivers/scsi/lpfc/lpfc_debugfs.c b/drivers/scsi/lpfc/lpfc_debugfs.c index 3217d63..f63f5ff 100644 --- a/drivers/scsi/lpfc/lpfc_debugfs.c +++ b/drivers/scsi/lpfc/lpfc_debugfs.c @@ -490,9 +490,11 @@ lpfc_debugfs_dumpHostSlim_data(struct lpfc_hba *phba, char *buf, int size) len += snprintf(buf+len, size-len, "Ring %d: CMD GetInx:%d (Max:%d Next:%d " "Local:%d flg:x%x) RSP PutInx:%d Max:%d\n", - i, pgpp->cmdGetInx, pring->numCiocb, - pring->next_cmdidx, pring->local_getidx, - pring->flag, pgpp->rspPutInx, pring->numRiocb); + i, pgpp->cmdGetInx, pring->sli.sli3.numCiocb, + pring->sli.sli3.next_cmdidx, + pring->sli.sli3.local_getidx, + pring->flag, pgpp->rspPutInx, + pring->sli.sli3.numRiocb); } if (phba->sli_rev <= LPFC_SLI_REV3) { @@ -557,6 +559,9 @@ lpfc_debugfs_nodelist_data(struct lpfc_vport *vport, char *buf, int size) case NLP_STE_PRLI_ISSUE: statep = "PRLI "; break; + case NLP_STE_LOGO_ISSUE: + statep = "LOGO "; + break; case NLP_STE_UNMAPPED_NODE: statep = "UNMAP "; break; @@ -581,8 +586,13 @@ lpfc_debugfs_nodelist_data(struct lpfc_vport *vport, char *buf, int size) "WWNN %02x:%02x:%02x:%02x:%02x:%02x:%02x:%02x ", *name, *(name+1), *(name+2), *(name+3), *(name+4), *(name+5), *(name+6), *(name+7)); - len += snprintf(buf+len, size-len, "RPI:%03d flag:x%08x ", - ndlp->nlp_rpi, ndlp->nlp_flag); + if (ndlp->nlp_flag & NLP_RPI_REGISTERED) + len += snprintf(buf+len, size-len, "RPI:%03d ", + ndlp->nlp_rpi); + else + len += snprintf(buf+len, size-len, "RPI:none "); + len += snprintf(buf+len, size-len, "flag:x%08x ", + ndlp->nlp_flag); if (!ndlp->nlp_type) len += snprintf(buf+len, size-len, "UNKNOWN_TYPE "); if (ndlp->nlp_type & NLP_FC_NODE) @@ -1999,207 +2009,298 @@ lpfc_idiag_queinfo_read(struct file *file, char __user *buf, size_t nbytes, { struct lpfc_debug *debug = file->private_data; struct lpfc_hba *phba = (struct lpfc_hba *)debug->i_private; - int len = 0, fcp_qidx; + int len = 0; char *pbuffer; + int x, cnt; + int max_cnt; + struct lpfc_queue *qp = NULL; + if (!debug->buffer) debug->buffer = kmalloc(LPFC_QUE_INFO_GET_BUF_SIZE, GFP_KERNEL); if (!debug->buffer) return 0; pbuffer = debug->buffer; + max_cnt = LPFC_QUE_INFO_GET_BUF_SIZE - 128; if (*ppos) return 0; - /* Get slow-path event queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Slow-path EQ information:\n"); - if (phba->sli4_hba.sp_eq) { - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tEQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n\n", - phba->sli4_hba.sp_eq->queue_id, - phba->sli4_hba.sp_eq->entry_count, - phba->sli4_hba.sp_eq->entry_size, - phba->sli4_hba.sp_eq->host_index, - phba->sli4_hba.sp_eq->hba_index); - } + spin_lock_irq(&phba->hbalock); + + /* Fast-path event queue */ + if (phba->sli4_hba.hba_eq && phba->cfg_fcp_io_channel) { + cnt = phba->cfg_fcp_io_channel; + + for (x = 0; x < cnt; x++) { + + /* Fast-path EQ */ + qp = phba->sli4_hba.hba_eq[x]; + if (!qp) + goto proc_cq; + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\nHBA EQ info: " + "EQ-STAT[max:x%x noE:x%x " + "bs:x%x proc:x%llx]\n", + qp->q_cnt_1, qp->q_cnt_2, + qp->q_cnt_3, (unsigned long long)qp->q_cnt_4); + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "EQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, + qp->entry_count, + qp->entry_size, + qp->host_index, + qp->hba_index); + + + /* Reset max counter */ + qp->EQ_max_eqe = 0; + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; +proc_cq: + /* Fast-path FCP CQ */ + qp = phba->sli4_hba.fcp_cq[x]; + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\tFCP CQ info: "); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "AssocEQID[%02d]: " + "CQ STAT[max:x%x relw:x%x " + "xabt:x%x wq:x%llx]\n", + qp->assoc_qid, + qp->q_cnt_1, qp->q_cnt_2, + qp->q_cnt_3, (unsigned long long)qp->q_cnt_4); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\tCQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, qp->entry_count, + qp->entry_size, qp->host_index, + qp->hba_index); + - /* Get fast-path event queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Fast-path EQ information:\n"); - if (phba->sli4_hba.fp_eq) { - for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_eq_count; - fcp_qidx++) { - if (phba->sli4_hba.fp_eq[fcp_qidx]) { + /* Reset max counter */ + qp->CQ_max_cqe = 0; + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; + + /* Fast-path FCP WQ */ + qp = phba->sli4_hba.fcp_wq[x]; + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tFCP WQ info: "); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "AssocCQID[%02d]: " + "WQ-STAT[oflow:x%x posted:x%llx]\n", + qp->assoc_qid, + qp->q_cnt_1, (unsigned long long)qp->q_cnt_4); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tWQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, + qp->entry_count, + qp->entry_size, + qp->host_index, + qp->hba_index); + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; + + if (x) + continue; + + /* Only EQ 0 has slow path CQs configured */ + + /* Slow-path mailbox CQ */ + qp = phba->sli4_hba.mbx_cq; + if (qp) { + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\tMBX CQ info: "); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "AssocEQID[%02d]: " + "CQ-STAT[mbox:x%x relw:x%x " + "xabt:x%x wq:x%llx]\n", + qp->assoc_qid, + qp->q_cnt_1, qp->q_cnt_2, + qp->q_cnt_3, + (unsigned long long)qp->q_cnt_4); len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tEQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n", - phba->sli4_hba.fp_eq[fcp_qidx]->queue_id, - phba->sli4_hba.fp_eq[fcp_qidx]->entry_count, - phba->sli4_hba.fp_eq[fcp_qidx]->entry_size, - phba->sli4_hba.fp_eq[fcp_qidx]->host_index, - phba->sli4_hba.fp_eq[fcp_qidx]->hba_index); + "\tCQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, qp->entry_count, + qp->entry_size, qp->host_index, + qp->hba_index); + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; } - } - } - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); - - /* Get mailbox complete queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Slow-path MBX CQ information:\n"); - if (phba->sli4_hba.mbx_cq) { - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated EQID[%02d]:\n", - phba->sli4_hba.mbx_cq->assoc_qid); - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tCQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n\n", - phba->sli4_hba.mbx_cq->queue_id, - phba->sli4_hba.mbx_cq->entry_count, - phba->sli4_hba.mbx_cq->entry_size, - phba->sli4_hba.mbx_cq->host_index, - phba->sli4_hba.mbx_cq->hba_index); - } - /* Get slow-path complete queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Slow-path ELS CQ information:\n"); - if (phba->sli4_hba.els_cq) { - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated EQID[%02d]:\n", - phba->sli4_hba.els_cq->assoc_qid); - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tCQID [%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n\n", - phba->sli4_hba.els_cq->queue_id, - phba->sli4_hba.els_cq->entry_count, - phba->sli4_hba.els_cq->entry_size, - phba->sli4_hba.els_cq->host_index, - phba->sli4_hba.els_cq->hba_index); - } + /* Slow-path MBOX MQ */ + qp = phba->sli4_hba.mbx_wq; + if (qp) { + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tMBX MQ info: "); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "AssocCQID[%02d]:\n", + phba->sli4_hba.mbx_wq->assoc_qid); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tWQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, qp->entry_count, + qp->entry_size, qp->host_index, + qp->hba_index); + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; + } - /* Get fast-path complete queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Fast-path FCP CQ information:\n"); - fcp_qidx = 0; - if (phba->sli4_hba.fcp_cq) { - do { - if (phba->sli4_hba.fcp_cq[fcp_qidx]) { + /* Slow-path ELS response CQ */ + qp = phba->sli4_hba.els_cq; + if (qp) { len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated EQID[%02d]:\n", - phba->sli4_hba.fcp_cq[fcp_qidx]->assoc_qid); + "\tELS CQ info: "); len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tCQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n", - phba->sli4_hba.fcp_cq[fcp_qidx]->queue_id, - phba->sli4_hba.fcp_cq[fcp_qidx]->entry_count, - phba->sli4_hba.fcp_cq[fcp_qidx]->entry_size, - phba->sli4_hba.fcp_cq[fcp_qidx]->host_index, - phba->sli4_hba.fcp_cq[fcp_qidx]->hba_index); + "AssocEQID[%02d]: " + "CQ-STAT[max:x%x relw:x%x " + "xabt:x%x wq:x%llx]\n", + qp->assoc_qid, + qp->q_cnt_1, qp->q_cnt_2, + qp->q_cnt_3, + (unsigned long long)qp->q_cnt_4); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\tCQID [%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, qp->entry_count, + qp->entry_size, qp->host_index, + qp->hba_index); + + /* Reset max counter */ + qp->CQ_max_cqe = 0; + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; } - } while (++fcp_qidx < phba->cfg_fcp_eq_count); - len += snprintf(pbuffer+len, - LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); - } - /* Get mailbox queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Slow-path MBX MQ information:\n"); - if (phba->sli4_hba.mbx_wq) { - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated CQID[%02d]:\n", - phba->sli4_hba.mbx_wq->assoc_qid); - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tWQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n\n", - phba->sli4_hba.mbx_wq->queue_id, - phba->sli4_hba.mbx_wq->entry_count, - phba->sli4_hba.mbx_wq->entry_size, - phba->sli4_hba.mbx_wq->host_index, - phba->sli4_hba.mbx_wq->hba_index); - } + /* Slow-path ELS WQ */ + qp = phba->sli4_hba.els_wq; + if (qp) { + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tELS WQ info: "); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "AssocCQID[%02d]: " + " WQ-STAT[oflow:x%x " + "posted:x%llx]\n", + qp->assoc_qid, + qp->q_cnt_1, + (unsigned long long)qp->q_cnt_4); + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tWQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]", + qp->queue_id, qp->entry_count, + qp->entry_size, qp->host_index, + qp->hba_index); + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + if (len >= max_cnt) + goto too_big; + } - /* Get slow-path work queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Slow-path ELS WQ information:\n"); - if (phba->sli4_hba.els_wq) { - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated CQID[%02d]:\n", - phba->sli4_hba.els_wq->assoc_qid); - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tWQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n\n", - phba->sli4_hba.els_wq->queue_id, - phba->sli4_hba.els_wq->entry_count, - phba->sli4_hba.els_wq->entry_size, - phba->sli4_hba.els_wq->host_index, - phba->sli4_hba.els_wq->hba_index); - } + if (phba->sli4_hba.hdr_rq && phba->sli4_hba.dat_rq) { + /* Slow-path RQ header */ + qp = phba->sli4_hba.hdr_rq; - /* Get fast-path work queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Fast-path FCP WQ information:\n"); - if (phba->sli4_hba.fcp_wq) { - for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_wq_count; - fcp_qidx++) { - if (!phba->sli4_hba.fcp_wq[fcp_qidx]) - continue; - len += snprintf(pbuffer+len, + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tRQ info: "); + len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated CQID[%02d]:\n", - phba->sli4_hba.fcp_wq[fcp_qidx]->assoc_qid); - len += snprintf(pbuffer+len, + "AssocCQID[%02d]: " + "RQ-STAT[nopost:x%x nobuf:x%x " + "trunc:x%x rcv:x%llx]\n", + qp->assoc_qid, + qp->q_cnt_1, qp->q_cnt_2, + qp->q_cnt_3, + (unsigned long long)qp->q_cnt_4); + len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tWQID[%02d], " - "QE-COUNT[%04d], WQE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n", - phba->sli4_hba.fcp_wq[fcp_qidx]->queue_id, - phba->sli4_hba.fcp_wq[fcp_qidx]->entry_count, - phba->sli4_hba.fcp_wq[fcp_qidx]->entry_size, - phba->sli4_hba.fcp_wq[fcp_qidx]->host_index, - phba->sli4_hba.fcp_wq[fcp_qidx]->hba_index); + "\t\tHQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]\n", + qp->queue_id, + qp->entry_count, + qp->entry_size, + qp->host_index, + qp->hba_index); + + /* Slow-path RQ data */ + qp = phba->sli4_hba.dat_rq; + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, + "\t\tDQID[%02d], " + "QE-CNT[%04d], QE-SIZE[%04d], " + "HOST-IDX[%04d], PORT-IDX[%04d]\n", + qp->queue_id, + qp->entry_count, + qp->entry_size, + qp->host_index, + qp->hba_index); + + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); + } } - len += snprintf(pbuffer+len, - LPFC_QUE_INFO_GET_BUF_SIZE-len, "\n"); } - /* Get receive queue information */ - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Slow-path RQ information:\n"); - if (phba->sli4_hba.hdr_rq && phba->sli4_hba.dat_rq) { - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "Associated CQID[%02d]:\n", - phba->sli4_hba.hdr_rq->assoc_qid); - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tHQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n", - phba->sli4_hba.hdr_rq->queue_id, - phba->sli4_hba.hdr_rq->entry_count, - phba->sli4_hba.hdr_rq->entry_size, - phba->sli4_hba.hdr_rq->host_index, - phba->sli4_hba.hdr_rq->hba_index); - len += snprintf(pbuffer+len, LPFC_QUE_INFO_GET_BUF_SIZE-len, - "\tDQID[%02d], " - "QE-COUNT[%04d], QE-SIZE[%04d], " - "HOST-INDEX[%04d], PORT-INDEX[%04d]\n", - phba->sli4_hba.dat_rq->queue_id, - phba->sli4_hba.dat_rq->entry_count, - phba->sli4_hba.dat_rq->entry_size, - phba->sli4_hba.dat_rq->host_index, - phba->sli4_hba.dat_rq->hba_index); - } + spin_unlock_irq(&phba->hbalock); + return simple_read_from_buffer(buf, nbytes, ppos, pbuffer, len); + +too_big: + len += snprintf(pbuffer+len, + LPFC_QUE_INFO_GET_BUF_SIZE-len, "Truncated ...\n"); + spin_unlock_irq(&phba->hbalock); return simple_read_from_buffer(buf, nbytes, ppos, pbuffer, len); } @@ -2408,31 +2509,21 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf, switch (quetp) { case LPFC_IDIAG_EQ: - /* Slow-path event queue */ - if (phba->sli4_hba.sp_eq && - phba->sli4_hba.sp_eq->queue_id == queid) { - /* Sanity check */ - rc = lpfc_idiag_que_param_check( - phba->sli4_hba.sp_eq, index, count); - if (rc) - goto error_out; - idiag.ptr_private = phba->sli4_hba.sp_eq; - goto pass_check; - } - /* Fast-path event queue */ - if (phba->sli4_hba.fp_eq) { - for (qidx = 0; qidx < phba->cfg_fcp_eq_count; qidx++) { - if (phba->sli4_hba.fp_eq[qidx] && - phba->sli4_hba.fp_eq[qidx]->queue_id == + /* HBA event queue */ + if (phba->sli4_hba.hba_eq) { + for (qidx = 0; qidx < phba->cfg_fcp_io_channel; + qidx++) { + if (phba->sli4_hba.hba_eq[qidx] && + phba->sli4_hba.hba_eq[qidx]->queue_id == queid) { /* Sanity check */ rc = lpfc_idiag_que_param_check( - phba->sli4_hba.fp_eq[qidx], + phba->sli4_hba.hba_eq[qidx], index, count); if (rc) goto error_out; idiag.ptr_private = - phba->sli4_hba.fp_eq[qidx]; + phba->sli4_hba.hba_eq[qidx]; goto pass_check; } } @@ -2479,7 +2570,7 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf, phba->sli4_hba.fcp_cq[qidx]; goto pass_check; } - } while (++qidx < phba->cfg_fcp_eq_count); + } while (++qidx < phba->cfg_fcp_io_channel); } goto error_out; break; @@ -2511,7 +2602,8 @@ lpfc_idiag_queacc_write(struct file *file, const char __user *buf, } /* FCP work queue */ if (phba->sli4_hba.fcp_wq) { - for (qidx = 0; qidx < phba->cfg_fcp_wq_count; qidx++) { + for (qidx = 0; qidx < phba->cfg_fcp_io_channel; + qidx++) { if (!phba->sli4_hba.fcp_wq[qidx]) continue; if (phba->sli4_hba.fcp_wq[qidx]->queue_id == @@ -4490,7 +4582,7 @@ lpfc_debug_dump_all_queues(struct lpfc_hba *phba) lpfc_debug_dump_mbx_wq(phba); lpfc_debug_dump_els_wq(phba); - for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_wq_count; fcp_wqidx++) + for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_io_channel; fcp_wqidx++) lpfc_debug_dump_fcp_wq(phba, fcp_wqidx); lpfc_debug_dump_hdr_rq(phba); @@ -4501,14 +4593,12 @@ lpfc_debug_dump_all_queues(struct lpfc_hba *phba) lpfc_debug_dump_mbx_cq(phba); lpfc_debug_dump_els_cq(phba); - for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_wq_count; fcp_wqidx++) + for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_io_channel; fcp_wqidx++) lpfc_debug_dump_fcp_cq(phba, fcp_wqidx); /* * Dump Event Queues (EQs) */ - lpfc_debug_dump_sp_eq(phba); - - for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_wq_count; fcp_wqidx++) - lpfc_debug_dump_fcp_eq(phba, fcp_wqidx); + for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_io_channel; fcp_wqidx++) + lpfc_debug_dump_hba_eq(phba, fcp_wqidx); } diff --git a/drivers/scsi/lpfc/lpfc_debugfs.h b/drivers/scsi/lpfc/lpfc_debugfs.h index afe368f..8b2b6a3 100644 --- a/drivers/scsi/lpfc/lpfc_debugfs.h +++ b/drivers/scsi/lpfc/lpfc_debugfs.h @@ -36,6 +36,9 @@ /* dumpHostSlim output buffer size */ #define LPFC_DUMPHOSTSLIM_SIZE 4096 +/* dumpSLIqinfo output buffer size */ +#define LPFC_DUMPSLIQINFO_SIZE 4096 + /* hbqinfo output buffer size */ #define LPFC_HBQINFO_SIZE 8192 @@ -366,7 +369,7 @@ static inline void lpfc_debug_dump_fcp_wq(struct lpfc_hba *phba, int fcp_wqidx) { /* sanity check */ - if (fcp_wqidx >= phba->cfg_fcp_wq_count) + if (fcp_wqidx >= phba->cfg_fcp_io_channel) return; printk(KERN_ERR "FCP WQ: WQ[Idx:%d|Qid:%d]\n", @@ -388,15 +391,15 @@ lpfc_debug_dump_fcp_cq(struct lpfc_hba *phba, int fcp_wqidx) int fcp_cqidx, fcp_cqid; /* sanity check */ - if (fcp_wqidx >= phba->cfg_fcp_wq_count) + if (fcp_wqidx >= phba->cfg_fcp_io_channel) return; fcp_cqid = phba->sli4_hba.fcp_wq[fcp_wqidx]->assoc_qid; - for (fcp_cqidx = 0; fcp_cqidx < phba->cfg_fcp_eq_count; fcp_cqidx++) + for (fcp_cqidx = 0; fcp_cqidx < phba->cfg_fcp_io_channel; fcp_cqidx++) if (phba->sli4_hba.fcp_cq[fcp_cqidx]->queue_id == fcp_cqid) break; if (phba->intr_type == MSIX) { - if (fcp_cqidx >= phba->cfg_fcp_eq_count) + if (fcp_cqidx >= phba->cfg_fcp_io_channel) return; } else { if (fcp_cqidx > 0) @@ -410,7 +413,7 @@ lpfc_debug_dump_fcp_cq(struct lpfc_hba *phba, int fcp_wqidx) } /** - * lpfc_debug_dump_fcp_eq - dump all entries from a fcp work queue's evt queue + * lpfc_debug_dump_hba_eq - dump all entries from a fcp work queue's evt queue * @phba: Pointer to HBA context object. * @fcp_wqidx: Index to a FCP work queue. * @@ -418,36 +421,30 @@ lpfc_debug_dump_fcp_cq(struct lpfc_hba *phba, int fcp_wqidx) * associated to the FCP work queue specified by the @fcp_wqidx. **/ static inline void -lpfc_debug_dump_fcp_eq(struct lpfc_hba *phba, int fcp_wqidx) +lpfc_debug_dump_hba_eq(struct lpfc_hba *phba, int fcp_wqidx) { struct lpfc_queue *qdesc; int fcp_eqidx, fcp_eqid; int fcp_cqidx, fcp_cqid; /* sanity check */ - if (fcp_wqidx >= phba->cfg_fcp_wq_count) + if (fcp_wqidx >= phba->cfg_fcp_io_channel) return; fcp_cqid = phba->sli4_hba.fcp_wq[fcp_wqidx]->assoc_qid; - for (fcp_cqidx = 0; fcp_cqidx < phba->cfg_fcp_eq_count; fcp_cqidx++) + for (fcp_cqidx = 0; fcp_cqidx < phba->cfg_fcp_io_channel; fcp_cqidx++) if (phba->sli4_hba.fcp_cq[fcp_cqidx]->queue_id == fcp_cqid) break; if (phba->intr_type == MSIX) { - if (fcp_cqidx >= phba->cfg_fcp_eq_count) + if (fcp_cqidx >= phba->cfg_fcp_io_channel) return; } else { if (fcp_cqidx > 0) return; } - if (phba->cfg_fcp_eq_count == 0) { - fcp_eqidx = -1; - fcp_eqid = phba->sli4_hba.sp_eq->queue_id; - qdesc = phba->sli4_hba.sp_eq; - } else { - fcp_eqidx = fcp_cqidx; - fcp_eqid = phba->sli4_hba.fp_eq[fcp_eqidx]->queue_id; - qdesc = phba->sli4_hba.fp_eq[fcp_eqidx]; - } + fcp_eqidx = fcp_cqidx; + fcp_eqid = phba->sli4_hba.hba_eq[fcp_eqidx]->queue_id; + qdesc = phba->sli4_hba.hba_eq[fcp_eqidx]; printk(KERN_ERR "FCP EQ: WQ[Idx:%d|Qid:%d]->CQ[Idx:%d|Qid:%d]->" "EQ[Idx:%d|Qid:%d]\n", @@ -543,25 +540,6 @@ lpfc_debug_dump_mbx_cq(struct lpfc_hba *phba) } /** - * lpfc_debug_dump_sp_eq - dump all entries from slow-path event queue - * @phba: Pointer to HBA context object. - * - * This function dumps all entries from the slow-path event queue. - **/ -static inline void -lpfc_debug_dump_sp_eq(struct lpfc_hba *phba) -{ - printk(KERN_ERR "SP EQ: WQ[Qid:%d/Qid:%d]->CQ[Qid:%d/Qid:%d]->" - "EQ[Qid:%d]:\n", - phba->sli4_hba.mbx_wq->queue_id, - phba->sli4_hba.els_wq->queue_id, - phba->sli4_hba.mbx_cq->queue_id, - phba->sli4_hba.els_cq->queue_id, - phba->sli4_hba.sp_eq->queue_id); - lpfc_debug_dump_q(phba->sli4_hba.sp_eq); -} - -/** * lpfc_debug_dump_wq_by_id - dump all entries from a work queue by queue id * @phba: Pointer to HBA context object. * @qid: Work queue identifier. @@ -574,10 +552,10 @@ lpfc_debug_dump_wq_by_id(struct lpfc_hba *phba, int qid) { int wq_idx; - for (wq_idx = 0; wq_idx < phba->cfg_fcp_wq_count; wq_idx++) + for (wq_idx = 0; wq_idx < phba->cfg_fcp_io_channel; wq_idx++) if (phba->sli4_hba.fcp_wq[wq_idx]->queue_id == qid) break; - if (wq_idx < phba->cfg_fcp_wq_count) { + if (wq_idx < phba->cfg_fcp_io_channel) { printk(KERN_ERR "FCP WQ[Idx:%d|Qid:%d]\n", wq_idx, qid); lpfc_debug_dump_q(phba->sli4_hba.fcp_wq[wq_idx]); return; @@ -644,9 +622,9 @@ lpfc_debug_dump_cq_by_id(struct lpfc_hba *phba, int qid) do { if (phba->sli4_hba.fcp_cq[cq_idx]->queue_id == qid) break; - } while (++cq_idx < phba->cfg_fcp_eq_count); + } while (++cq_idx < phba->cfg_fcp_io_channel); - if (cq_idx < phba->cfg_fcp_eq_count) { + if (cq_idx < phba->cfg_fcp_io_channel) { printk(KERN_ERR "FCP CQ[Idx:%d|Qid:%d]\n", cq_idx, qid); lpfc_debug_dump_q(phba->sli4_hba.fcp_cq[cq_idx]); return; @@ -677,21 +655,17 @@ lpfc_debug_dump_eq_by_id(struct lpfc_hba *phba, int qid) { int eq_idx; - for (eq_idx = 0; eq_idx < phba->cfg_fcp_eq_count; eq_idx++) { - if (phba->sli4_hba.fp_eq[eq_idx]->queue_id == qid) + for (eq_idx = 0; eq_idx < phba->cfg_fcp_io_channel; eq_idx++) { + if (phba->sli4_hba.hba_eq[eq_idx]->queue_id == qid) break; } - if (eq_idx < phba->cfg_fcp_eq_count) { + if (eq_idx < phba->cfg_fcp_io_channel) { printk(KERN_ERR "FCP EQ[Idx:%d|Qid:%d]\n", eq_idx, qid); - lpfc_debug_dump_q(phba->sli4_hba.fp_eq[eq_idx]); + lpfc_debug_dump_q(phba->sli4_hba.hba_eq[eq_idx]); return; } - if (phba->sli4_hba.sp_eq->queue_id == qid) { - printk(KERN_ERR "SP EQ[|Qid:%d]\n", qid); - lpfc_debug_dump_q(phba->sli4_hba.sp_eq); - } } void lpfc_debug_dump_all_queues(struct lpfc_hba *); diff --git a/drivers/scsi/lpfc/lpfc_disc.h b/drivers/scsi/lpfc/lpfc_disc.h index 1d84b63..af49fb0 100644 --- a/drivers/scsi/lpfc/lpfc_disc.h +++ b/drivers/scsi/lpfc/lpfc_disc.h @@ -145,6 +145,7 @@ struct lpfc_node_rrq { #define NLP_RCV_PLOGI 0x00080000 /* Rcv'ed PLOGI from remote system */ #define NLP_LOGO_ACC 0x00100000 /* Process LOGO after ACC completes */ #define NLP_TGT_NO_SCSIID 0x00200000 /* good PRLI but no binding for scsid */ +#define NLP_ISSUE_LOGO 0x00400000 /* waiting to issue a LOGO */ #define NLP_ACC_REGLOGIN 0x01000000 /* Issue Reg Login after successful ACC */ #define NLP_NPR_ADISC 0x02000000 /* Issue ADISC when dq'ed from @@ -201,10 +202,11 @@ struct lpfc_node_rrq { #define NLP_STE_ADISC_ISSUE 0x2 /* ADISC was sent to NL_PORT */ #define NLP_STE_REG_LOGIN_ISSUE 0x3 /* REG_LOGIN was issued for NL_PORT */ #define NLP_STE_PRLI_ISSUE 0x4 /* PRLI was sent to NL_PORT */ -#define NLP_STE_UNMAPPED_NODE 0x5 /* PRLI completed from NL_PORT */ -#define NLP_STE_MAPPED_NODE 0x6 /* Identified as a FCP Target */ -#define NLP_STE_NPR_NODE 0x7 /* NPort disappeared */ -#define NLP_STE_MAX_STATE 0x8 +#define NLP_STE_LOGO_ISSUE 0x5 /* LOGO was sent to NL_PORT */ +#define NLP_STE_UNMAPPED_NODE 0x6 /* PRLI completed from NL_PORT */ +#define NLP_STE_MAPPED_NODE 0x7 /* Identified as a FCP Target */ +#define NLP_STE_NPR_NODE 0x8 /* NPort disappeared */ +#define NLP_STE_MAX_STATE 0x9 #define NLP_STE_FREED_NODE 0xff /* node entry was freed to MEM_NLP */ /* For UNUSED_NODE state, the node has just been allocated. diff --git a/drivers/scsi/lpfc/lpfc_els.c b/drivers/scsi/lpfc/lpfc_els.c index d54ae19..cfe533b 100644 --- a/drivers/scsi/lpfc/lpfc_els.c +++ b/drivers/scsi/lpfc/lpfc_els.c @@ -962,7 +962,8 @@ lpfc_cmpl_els_flogi(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, if ((phba->fcoe_cvl_eventtag_attn == phba->fcoe_cvl_eventtag) && (irsp->ulpStatus == IOSTAT_LOCAL_REJECT) && - (irsp->un.ulpWord[4] == IOERR_SLI_ABORTED)) + ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_SLI_ABORTED)) goto stop_rr_fcf_flogi; else phba->fcoe_cvl_eventtag_attn = @@ -1108,8 +1109,10 @@ flogifail: /* Start discovery */ lpfc_disc_start(vport); } else if (((irsp->ulpStatus != IOSTAT_LOCAL_REJECT) || - ((irsp->un.ulpWord[4] != IOERR_SLI_ABORTED) && - (irsp->un.ulpWord[4] != IOERR_SLI_DOWN))) && + (((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) != + IOERR_SLI_ABORTED) && + ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) != + IOERR_SLI_DOWN))) && (phba->link_state != LPFC_CLEAR_LA)) { /* If FLOGI failed enable link interrupt. */ lpfc_issue_clear_la(phba, vport); @@ -1476,6 +1479,10 @@ lpfc_plogi_confirm_nport(struct lpfc_hba *phba, uint32_t *prsp, return ndlp; memset(&rrq.xri_bitmap, 0, sizeof(new_ndlp->active_rrqs.xri_bitmap)); + lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, + "3178 PLOGI confirm: ndlp %p x%x: new_ndlp %p\n", + ndlp, ndlp->nlp_DID, new_ndlp); + if (!new_ndlp) { rc = memcmp(&ndlp->nlp_portname, name, sizeof(struct lpfc_name)); @@ -1527,6 +1534,9 @@ lpfc_plogi_confirm_nport(struct lpfc_hba *phba, uint32_t *prsp, /* The new_ndlp is replacing ndlp totally, so we need * to put ndlp on UNUSED list and try to free it. */ + lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, + "3179 PLOGI confirm NEW: %x %x\n", + new_ndlp->nlp_DID, keepDID); /* Fix up the rport accordingly */ rport = ndlp->rport; @@ -1559,23 +1569,34 @@ lpfc_plogi_confirm_nport(struct lpfc_hba *phba, uint32_t *prsp, lpfc_drop_node(vport, ndlp); } else { + lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, + "3180 PLOGI confirm SWAP: %x %x\n", + new_ndlp->nlp_DID, keepDID); + lpfc_unreg_rpi(vport, ndlp); + /* Two ndlps cannot have the same did */ ndlp->nlp_DID = keepDID; if (phba->sli_rev == LPFC_SLI_REV4) memcpy(&ndlp->active_rrqs.xri_bitmap, &rrq.xri_bitmap, sizeof(ndlp->active_rrqs.xri_bitmap)); + /* Since we are swapping the ndlp passed in with the new one - * and the did has already been swapped, copy over the - * state and names. + * and the did has already been swapped, copy over state. + * The new WWNs are already in new_ndlp since thats what + * we looked it up by in the begining of this routine. */ - memcpy(&new_ndlp->nlp_portname, &ndlp->nlp_portname, - sizeof(struct lpfc_name)); - memcpy(&new_ndlp->nlp_nodename, &ndlp->nlp_nodename, - sizeof(struct lpfc_name)); new_ndlp->nlp_state = ndlp->nlp_state; - lpfc_nlp_set_state(vport, ndlp, NLP_STE_NPR_NODE); + + /* Since we are switching over to the new_ndlp, the old + * ndlp should be put in the NPR state, unless we have + * already started re-discovery on it. + */ + if ((ndlp->nlp_state == NLP_STE_UNMAPPED_NODE) || + (ndlp->nlp_state == NLP_STE_MAPPED_NODE)) + lpfc_nlp_set_state(vport, ndlp, NLP_STE_NPR_NODE); + /* Fix up the rport accordingly */ rport = ndlp->rport; if (rport) { @@ -2367,6 +2388,8 @@ lpfc_cmpl_els_logo(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, IOCB_t *irsp; struct lpfc_sli *psli; struct lpfcMboxq *mbox; + unsigned long flags; + uint32_t skip_recovery = 0; psli = &phba->sli; /* we pass cmdiocb to state machine which needs rspiocb as well */ @@ -2381,47 +2404,52 @@ lpfc_cmpl_els_logo(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, "LOGO cmpl: status:x%x/x%x did:x%x", irsp->ulpStatus, irsp->un.ulpWord[4], ndlp->nlp_DID); + /* LOGO completes to NPort <nlp_DID> */ lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, "0105 LOGO completes to NPort x%x " "Data: x%x x%x x%x x%x\n", ndlp->nlp_DID, irsp->ulpStatus, irsp->un.ulpWord[4], irsp->ulpTimeout, vport->num_disc_nodes); - /* Check to see if link went down during discovery */ - if (lpfc_els_chk_latt(vport)) + + if (lpfc_els_chk_latt(vport)) { + skip_recovery = 1; goto out; + } + /* Check to see if link went down during discovery */ if (ndlp->nlp_flag & NLP_TARGET_REMOVE) { /* NLP_EVT_DEVICE_RM should unregister the RPI * which should abort all outstanding IOs. */ lpfc_disc_state_machine(vport, ndlp, cmdiocb, NLP_EVT_DEVICE_RM); + skip_recovery = 1; goto out; } if (irsp->ulpStatus) { /* Check for retry */ - if (lpfc_els_retry(phba, cmdiocb, rspiocb)) + if (lpfc_els_retry(phba, cmdiocb, rspiocb)) { /* ELS command is being retried */ + skip_recovery = 1; goto out; + } /* LOGO failed */ lpfc_printf_vlog(vport, KERN_ERR, LOG_ELS, "2756 LOGO failure DID:%06X Status:x%x/x%x\n", ndlp->nlp_DID, irsp->ulpStatus, irsp->un.ulpWord[4]); /* Do not call DSM for lpfc_els_abort'ed ELS cmds */ - if (lpfc_error_lost_link(irsp)) + if (lpfc_error_lost_link(irsp)) { + skip_recovery = 1; goto out; - else - lpfc_disc_state_machine(vport, ndlp, cmdiocb, - NLP_EVT_CMPL_LOGO); - } else - /* Good status, call state machine. - * This will unregister the rpi if needed. - */ - lpfc_disc_state_machine(vport, ndlp, cmdiocb, - NLP_EVT_CMPL_LOGO); + } + } + + /* Call state machine. This will unregister the rpi if needed. */ + lpfc_disc_state_machine(vport, ndlp, cmdiocb, NLP_EVT_CMPL_LOGO); + out: lpfc_els_free_iocb(phba, cmdiocb); /* If we are in pt2pt mode, we could rcv new S_ID on PLOGI */ @@ -2436,9 +2464,30 @@ out: if (lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT) == MBX_NOT_FINISHED) { mempool_free(mbox, phba->mbox_mem_pool); + skip_recovery = 1; } } } + + /* + * If the node is a target, the handling attempts to recover the port. + * For any other port type, the rpi is unregistered as an implicit + * LOGO. + */ + if ((ndlp->nlp_type & NLP_FCP_TARGET) && (skip_recovery == 0)) { + lpfc_cancel_retry_delay_tmo(vport, ndlp); + spin_lock_irqsave(shost->host_lock, flags); + ndlp->nlp_flag |= NLP_NPR_2B_DISC; + spin_unlock_irqrestore(shost->host_lock, flags); + + lpfc_printf_vlog(vport, KERN_INFO, LOG_ELS, + "3187 LOGO completes to NPort x%x: Start " + "Recovery Data: x%x x%x x%x x%x\n", + ndlp->nlp_DID, irsp->ulpStatus, + irsp->un.ulpWord[4], irsp->ulpTimeout, + vport->num_disc_nodes); + lpfc_disc_start(vport); + } return; } @@ -2501,10 +2550,27 @@ lpfc_issue_els_logo(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, "Issue LOGO: did:x%x", ndlp->nlp_DID, 0, 0); + /* + * If we are issuing a LOGO, we may try to recover the remote NPort + * by issuing a PLOGI later. Even though we issue ELS cmds by the + * VPI, if we have a valid RPI, and that RPI gets unreg'ed while + * that ELS command is in-flight, the HBA returns a IOERR_INVALID_RPI + * for that ELS cmd. To avoid this situation, lets get rid of the + * RPI right now, before any ELS cmds are sent. + */ + spin_lock_irq(shost->host_lock); + ndlp->nlp_flag |= NLP_ISSUE_LOGO; + spin_unlock_irq(shost->host_lock); + if (lpfc_unreg_rpi(vport, ndlp)) { + lpfc_els_free_iocb(phba, elsiocb); + return 0; + } + phba->fc_stat.elsXmitLOGO++; elsiocb->iocb_cmpl = lpfc_cmpl_els_logo; spin_lock_irq(shost->host_lock); ndlp->nlp_flag |= NLP_LOGO_SND; + ndlp->nlp_flag &= ~NLP_ISSUE_LOGO; spin_unlock_irq(shost->host_lock); rc = lpfc_sli_issue_iocb(phba, LPFC_ELS_RING, elsiocb, 0); @@ -2920,7 +2986,7 @@ lpfc_els_retry_delay_handler(struct lpfc_nodelist *ndlp) case ELS_CMD_LOGO: if (!lpfc_issue_els_logo(vport, ndlp, retry)) { ndlp->nlp_prev_state = ndlp->nlp_state; - lpfc_nlp_set_state(vport, ndlp, NLP_STE_NPR_NODE); + lpfc_nlp_set_state(vport, ndlp, NLP_STE_LOGO_ISSUE); } break; case ELS_CMD_FDISC: @@ -3007,7 +3073,7 @@ lpfc_els_retry(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, } break; case IOSTAT_LOCAL_REJECT: - switch ((irsp->un.ulpWord[4] & 0xff)) { + switch ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK)) { case IOERR_LOOP_OPEN_FAILURE: if (cmd == ELS_CMD_FLOGI) { if (PCI_DEVICE_ID_HORNET == @@ -3214,7 +3280,8 @@ lpfc_els_retry(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, if (((cmd == ELS_CMD_PLOGI) || (cmd == ELS_CMD_ADISC)) && ((irsp->ulpStatus != IOSTAT_LOCAL_REJECT) || - ((irsp->un.ulpWord[4] & 0xff) != IOERR_NO_RESOURCES))) { + ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) != + IOERR_NO_RESOURCES))) { /* Don't reset timer for no resources */ /* If discovery / RSCN timer is running, reset it */ @@ -3273,7 +3340,7 @@ lpfc_els_retry(struct lpfc_hba *phba, struct lpfc_iocbq *cmdiocb, return 1; case ELS_CMD_LOGO: ndlp->nlp_prev_state = ndlp->nlp_state; - lpfc_nlp_set_state(vport, ndlp, NLP_STE_NPR_NODE); + lpfc_nlp_set_state(vport, ndlp, NLP_STE_LOGO_ISSUE); lpfc_issue_els_logo(vport, ndlp, cmdiocb->retry); return 1; } @@ -3533,13 +3600,17 @@ lpfc_mbx_cmpl_dflt_rpi(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb) lpfc_mbuf_free(phba, mp->virt, mp->phys); kfree(mp); mempool_free(pmb, phba->mbox_mem_pool); - if (ndlp && NLP_CHK_NODE_ACT(ndlp)) { - lpfc_nlp_put(ndlp); - /* This is the end of the default RPI cleanup logic for this - * ndlp. If no other discovery threads are using this ndlp. - * we should free all resources associated with it. - */ - lpfc_nlp_not_used(ndlp); + if (ndlp) { + if (NLP_CHK_NODE_ACT(ndlp)) { + lpfc_nlp_put(ndlp); + /* This is the end of the default RPI cleanup logic for + * this ndlp. If no other discovery threads are using + * this ndlp, free all resources associated with it. + */ + lpfc_nlp_not_used(ndlp); + } else { + lpfc_drop_node(ndlp->vport, ndlp); + } } return; @@ -6803,7 +6874,8 @@ lpfc_els_unsol_event(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, if (icmd->ulpStatus == IOSTAT_NEED_BUFFER) { lpfc_sli_hbqbuf_add_hbqs(phba, LPFC_ELS_HBQ); } else if (icmd->ulpStatus == IOSTAT_LOCAL_REJECT && - (icmd->un.ulpWord[4] & 0xff) == IOERR_RCV_BUFFER_WAITING) { + (icmd->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_RCV_BUFFER_WAITING) { phba->fc_stat.NoRcvBuf++; /* Not enough posted buffers; Try posting more buffers */ if (!(phba->sli3_options & LPFC_SLI3_HBQ_ENABLED)) @@ -7985,3 +8057,47 @@ lpfc_sli4_els_xri_aborted(struct lpfc_hba *phba, spin_unlock_irqrestore(&phba->hbalock, iflag); return; } + +/* lpfc_sli_abts_recover_port - Recover a port that failed a BLS_ABORT req. + * @vport: pointer to virtual port object. + * @ndlp: nodelist pointer for the impacted node. + * + * The driver calls this routine in response to an SLI4 XRI ABORT CQE + * or an SLI3 ASYNC_STATUS_CN event from the port. For either event, + * the driver is required to send a LOGO to the remote node before it + * attempts to recover its login to the remote node. + */ +void +lpfc_sli_abts_recover_port(struct lpfc_vport *vport, + struct lpfc_nodelist *ndlp) +{ + struct Scsi_Host *shost; + struct lpfc_hba *phba; + unsigned long flags = 0; + + shost = lpfc_shost_from_vport(vport); + phba = vport->phba; + if (ndlp->nlp_state != NLP_STE_MAPPED_NODE) { + lpfc_printf_log(phba, KERN_INFO, + LOG_SLI, "3093 No rport recovery needed. " + "rport in state 0x%x\n", ndlp->nlp_state); + return; + } + lpfc_printf_log(phba, KERN_INFO, LOG_SLI, + "3094 Start rport recovery on shost id 0x%x " + "fc_id 0x%06x vpi 0x%x rpi 0x%x state 0x%x " + "flags 0x%x\n", + shost->host_no, ndlp->nlp_DID, + vport->vpi, ndlp->nlp_rpi, ndlp->nlp_state, + ndlp->nlp_flag); + /* + * The rport is not responding. Remove the FCP-2 flag to prevent + * an ADISC in the follow-up recovery code. + */ + spin_lock_irqsave(shost->host_lock, flags); + ndlp->nlp_fcp_info &= ~NLP_FCP_2_DEVICE; + spin_unlock_irqrestore(shost->host_lock, flags); + lpfc_issue_els_logo(vport, ndlp, 0); + lpfc_nlp_set_state(vport, ndlp, NLP_STE_LOGO_ISSUE); +} + diff --git a/drivers/scsi/lpfc/lpfc_hbadisc.c b/drivers/scsi/lpfc/lpfc_hbadisc.c index 9b4f929..e9845d2 100644 --- a/drivers/scsi/lpfc/lpfc_hbadisc.c +++ b/drivers/scsi/lpfc/lpfc_hbadisc.c @@ -123,6 +123,10 @@ lpfc_dev_loss_tmo_callbk(struct fc_rport *rport) "rport devlosscb: sid:x%x did:x%x flg:x%x", ndlp->nlp_sid, ndlp->nlp_DID, ndlp->nlp_flag); + lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_NODE, + "3181 dev_loss_callbk x%06x, rport %p flg x%x\n", + ndlp->nlp_DID, ndlp->rport, ndlp->nlp_flag); + /* Don't defer this if we are in the process of deleting the vport * or unloading the driver. The unload will cleanup the node * appropriately we just need to cleanup the ndlp rport info here. @@ -142,6 +146,15 @@ lpfc_dev_loss_tmo_callbk(struct fc_rport *rport) if (ndlp->nlp_state == NLP_STE_MAPPED_NODE) return; + if (ndlp->nlp_type & NLP_FABRIC) { + + /* If the WWPN of the rport and ndlp don't match, ignore it */ + if (rport->port_name != wwn_to_u64(ndlp->nlp_portname.u.wwn)) { + put_device(&rport->dev); + return; + } + } + evtp = &ndlp->dev_loss_evt; if (!list_empty(&evtp->evt_listp)) @@ -202,6 +215,10 @@ lpfc_dev_loss_tmo_handler(struct lpfc_nodelist *ndlp) "rport devlosstmo:did:x%x type:x%x id:x%x", ndlp->nlp_DID, ndlp->nlp_type, rport->scsi_target_id); + lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_NODE, + "3182 dev_loss_tmo_handler x%06x, rport %p flg x%x\n", + ndlp->nlp_DID, ndlp->rport, ndlp->nlp_flag); + /* Don't defer this if we are in the process of deleting the vport * or unloading the driver. The unload will cleanup the node * appropriately we just need to cleanup the ndlp rport info here. @@ -3492,7 +3509,7 @@ lpfc_create_static_vport(struct lpfc_hba *phba) LPFC_MBOXQ_t *pmb = NULL; MAILBOX_t *mb; struct static_vport_info *vport_info; - int rc = 0, i; + int mbx_wait_rc = 0, i; struct fc_vport_identifiers vport_id; struct fc_vport *new_fc_vport; struct Scsi_Host *shost; @@ -3509,7 +3526,7 @@ lpfc_create_static_vport(struct lpfc_hba *phba) " allocate mailbox memory\n"); return; } - + memset(pmb, 0, sizeof(LPFC_MBOXQ_t)); mb = &pmb->u.mb; vport_info = kzalloc(sizeof(struct static_vport_info), GFP_KERNEL); @@ -3523,24 +3540,31 @@ lpfc_create_static_vport(struct lpfc_hba *phba) vport_buff = (uint8_t *) vport_info; do { + /* free dma buffer from previous round */ + if (pmb->context1) { + mp = (struct lpfc_dmabuf *)pmb->context1; + lpfc_mbuf_free(phba, mp->virt, mp->phys); + kfree(mp); + } if (lpfc_dump_static_vport(phba, pmb, offset)) goto out; pmb->vport = phba->pport; - rc = lpfc_sli_issue_mbox_wait(phba, pmb, LPFC_MBOX_TMO); + mbx_wait_rc = lpfc_sli_issue_mbox_wait(phba, pmb, + LPFC_MBOX_TMO); - if ((rc != MBX_SUCCESS) || mb->mbxStatus) { + if ((mbx_wait_rc != MBX_SUCCESS) || mb->mbxStatus) { lpfc_printf_log(phba, KERN_WARNING, LOG_INIT, "0544 lpfc_create_static_vport failed to" " issue dump mailbox command ret 0x%x " "status 0x%x\n", - rc, mb->mbxStatus); + mbx_wait_rc, mb->mbxStatus); goto out; } if (phba->sli_rev == LPFC_SLI_REV4) { byte_count = pmb->u.mqe.un.mb_words[5]; - mp = (struct lpfc_dmabuf *) pmb->context2; + mp = (struct lpfc_dmabuf *)pmb->context1; if (byte_count > sizeof(struct static_vport_info) - offset) byte_count = sizeof(struct static_vport_info) @@ -3604,9 +3628,9 @@ lpfc_create_static_vport(struct lpfc_hba *phba) out: kfree(vport_info); - if (rc != MBX_TIMEOUT) { - if (pmb->context2) { - mp = (struct lpfc_dmabuf *) pmb->context2; + if (mbx_wait_rc != MBX_TIMEOUT) { + if (pmb->context1) { + mp = (struct lpfc_dmabuf *)pmb->context1; lpfc_mbuf_free(phba, mp->virt, mp->phys); kfree(mp); } @@ -3834,6 +3858,10 @@ lpfc_register_remote_port(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) if (rport_ids.roles != FC_RPORT_ROLE_UNKNOWN) fc_remote_port_rolechg(rport, rport_ids.roles); + lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_NODE, + "3183 rport register x%06x, rport %p role x%x\n", + ndlp->nlp_DID, rport, rport_ids.roles); + if ((rport->scsi_target_id != -1) && (rport->scsi_target_id < LPFC_MAX_TARGET)) { ndlp->nlp_sid = rport->scsi_target_id; @@ -3850,6 +3878,10 @@ lpfc_unregister_remote_port(struct lpfc_nodelist *ndlp) "rport delete: did:x%x flg:x%x type x%x", ndlp->nlp_DID, ndlp->nlp_flag, ndlp->nlp_type); + lpfc_printf_vlog(ndlp->vport, KERN_INFO, LOG_NODE, + "3184 rport unregister x%06x, rport %p\n", + ndlp->nlp_DID, rport); + fc_remote_port_delete(rport); return; @@ -3964,6 +3996,7 @@ lpfc_nlp_state_name(char *buffer, size_t size, int state) [NLP_STE_ADISC_ISSUE] = "ADISC", [NLP_STE_REG_LOGIN_ISSUE] = "REGLOGIN", [NLP_STE_PRLI_ISSUE] = "PRLI", + [NLP_STE_LOGO_ISSUE] = "LOGO", [NLP_STE_UNMAPPED_NODE] = "UNMAPPED", [NLP_STE_MAPPED_NODE] = "MAPPED", [NLP_STE_NPR_NODE] = "NPR", @@ -4330,6 +4363,26 @@ lpfc_no_rpi(struct lpfc_hba *phba, struct lpfc_nodelist *ndlp) return 0; } +/** + * lpfc_nlp_logo_unreg - Unreg mailbox completion handler before LOGO + * @phba: Pointer to HBA context object. + * @pmb: Pointer to mailbox object. + * + * This function will issue an ELS LOGO command after completing + * the UNREG_RPI. + **/ +void +lpfc_nlp_logo_unreg(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb) +{ + struct lpfc_vport *vport = pmb->vport; + struct lpfc_nodelist *ndlp; + + ndlp = (struct lpfc_nodelist *)(pmb->context1); + if (!ndlp) + return; + lpfc_issue_els_logo(vport, ndlp, 0); +} + /* * Free rpi associated with LPFC_NODELIST entry. * This routine is called from lpfc_freenode(), when we are removing @@ -4354,9 +4407,16 @@ lpfc_unreg_rpi(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) rpi = ndlp->nlp_rpi; if (phba->sli_rev == LPFC_SLI_REV4) rpi = phba->sli4_hba.rpi_ids[ndlp->nlp_rpi]; + lpfc_unreg_login(phba, vport->vpi, rpi, mbox); mbox->vport = vport; - mbox->mbox_cmpl = lpfc_sli_def_mbox_cmpl; + if (ndlp->nlp_flag & NLP_ISSUE_LOGO) { + mbox->context1 = ndlp; + mbox->mbox_cmpl = lpfc_nlp_logo_unreg; + } else { + mbox->mbox_cmpl = lpfc_sli_def_mbox_cmpl; + } + rc = lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT); if (rc == MBX_NOT_FINISHED) mempool_free(mbox, phba->mbox_mem_pool); @@ -4499,9 +4559,13 @@ lpfc_cleanup_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) lpfc_disable_node(vport, ndlp); } + + /* Don't need to clean up REG_LOGIN64 cmds for Default RPI cleanup */ + /* cleanup any ndlp on mbox q waiting for reglogin cmpl */ if ((mb = phba->sli.mbox_active)) { if ((mb->u.mb.mbxCommand == MBX_REG_LOGIN64) && + !(mb->mbox_flag & LPFC_MBX_IMED_UNREG) && (ndlp == (struct lpfc_nodelist *) mb->context2)) { mb->context2 = NULL; mb->mbox_cmpl = lpfc_sli_def_mbox_cmpl; @@ -4512,6 +4576,7 @@ lpfc_cleanup_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) /* Cleanup REG_LOGIN completions which are not yet processed */ list_for_each_entry(mb, &phba->sli.mboxq_cmpl, list) { if ((mb->u.mb.mbxCommand != MBX_REG_LOGIN64) || + (mb->mbox_flag & LPFC_MBX_IMED_UNREG) || (ndlp != (struct lpfc_nodelist *) mb->context2)) continue; @@ -4521,6 +4586,7 @@ lpfc_cleanup_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) list_for_each_entry_safe(mb, nextmb, &phba->sli.mboxq, list) { if ((mb->u.mb.mbxCommand == MBX_REG_LOGIN64) && + !(mb->mbox_flag & LPFC_MBX_IMED_UNREG) && (ndlp == (struct lpfc_nodelist *) mb->context2)) { mp = (struct lpfc_dmabuf *) (mb->context1); if (mp) { @@ -4585,7 +4651,7 @@ lpfc_nlp_remove(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp) mbox->mbox_flag |= LPFC_MBX_IMED_UNREG; mbox->mbox_cmpl = lpfc_mbx_cmpl_dflt_rpi; mbox->vport = vport; - mbox->context2 = NULL; + mbox->context2 = ndlp; rc =lpfc_sli_issue_mbox(phba, mbox, MBX_NOWAIT); if (rc == MBX_NOT_FINISHED) { mempool_free(mbox, phba->mbox_mem_pool); @@ -5365,9 +5431,17 @@ __lpfc_find_node(struct lpfc_vport *vport, node_filter filter, void *param) struct lpfc_nodelist *ndlp; list_for_each_entry(ndlp, &vport->fc_nodes, nlp_listp) { - if (filter(ndlp, param)) + if (filter(ndlp, param)) { + lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE, + "3185 FIND node filter %p DID " + "Data: x%p x%x x%x\n", + filter, ndlp, ndlp->nlp_DID, + ndlp->nlp_flag); return ndlp; + } } + lpfc_printf_vlog(vport, KERN_INFO, LOG_NODE, + "3186 FIND node filter %p NOT FOUND.\n", filter); return NULL; } diff --git a/drivers/scsi/lpfc/lpfc_hw.h b/drivers/scsi/lpfc/lpfc_hw.h index 41bb1d2..7398ca8 100644 --- a/drivers/scsi/lpfc/lpfc_hw.h +++ b/drivers/scsi/lpfc/lpfc_hw.h @@ -1188,8 +1188,8 @@ typedef struct { */ /* Number of rings currently used and available. */ -#define MAX_CONFIGURED_RINGS 3 -#define MAX_RINGS 4 +#define MAX_SLI3_CONFIGURED_RINGS 3 +#define MAX_SLI3_RINGS 4 /* IOCB / Mailbox is owned by FireFly */ #define OWN_CHIP 1 @@ -1251,6 +1251,8 @@ typedef struct { #define PCI_VENDOR_ID_SERVERENGINE 0x19a2 #define PCI_DEVICE_ID_TIGERSHARK 0x0704 #define PCI_DEVICE_ID_TOMCAT 0x0714 +#define PCI_DEVICE_ID_SKYHAWK 0x0724 +#define PCI_DEVICE_ID_SKYHAWK_VF 0x072c #define JEDEC_ID_ADDRESS 0x0080001c #define FIREFLY_JEDEC_ID 0x1ACC @@ -1458,6 +1460,7 @@ typedef struct { /* FireFly BIU registers */ #define MBX_UNREG_FCFI 0xA2 #define MBX_INIT_VFI 0xA3 #define MBX_INIT_VPI 0xA4 +#define MBX_ACCESS_VDATA 0xA5 #define MBX_AUTH_PORT 0xF8 #define MBX_SECURITY_MGMT 0xF9 @@ -2991,7 +2994,7 @@ typedef struct _PCB { uint32_t pgpAddrLow; uint32_t pgpAddrHigh; - SLI2_RDSC rdsc[MAX_RINGS]; + SLI2_RDSC rdsc[MAX_SLI3_RINGS]; } PCB_t; /* NEW_FEATURE */ @@ -3101,18 +3104,18 @@ struct lpfc_pgp { struct sli2_desc { uint32_t unused1[16]; - struct lpfc_hgp host[MAX_RINGS]; - struct lpfc_pgp port[MAX_RINGS]; + struct lpfc_hgp host[MAX_SLI3_RINGS]; + struct lpfc_pgp port[MAX_SLI3_RINGS]; }; struct sli3_desc { - struct lpfc_hgp host[MAX_RINGS]; + struct lpfc_hgp host[MAX_SLI3_RINGS]; uint32_t reserved[8]; uint32_t hbq_put[16]; }; struct sli3_pgp { - struct lpfc_pgp port[MAX_RINGS]; + struct lpfc_pgp port[MAX_SLI3_RINGS]; uint32_t hbq_get[16]; }; @@ -3242,6 +3245,7 @@ typedef struct { #define IOERR_SLI_DOWN 0x101 /* ulpStatus - Driver defined */ #define IOERR_SLI_BRESET 0x102 #define IOERR_SLI_ABORTED 0x103 +#define IOERR_PARAM_MASK 0x1ff } PARM_ERR; typedef union { diff --git a/drivers/scsi/lpfc/lpfc_hw4.h b/drivers/scsi/lpfc/lpfc_hw4.h index 953603a..834b699 100644 --- a/drivers/scsi/lpfc/lpfc_hw4.h +++ b/drivers/scsi/lpfc/lpfc_hw4.h @@ -187,11 +187,17 @@ struct lpfc_sli_intf { /* Active interrupt test count */ #define LPFC_ACT_INTR_CNT 4 +/* Algrithmns for scheduling FCP commands to WQs */ +#define LPFC_FCP_SCHED_ROUND_ROBIN 0 +#define LPFC_FCP_SCHED_BY_CPU 1 + /* Delay Multiplier constant */ #define LPFC_DMULT_CONST 651042 -#define LPFC_MIM_IMAX 636 -#define LPFC_FP_DEF_IMAX 10000 -#define LPFC_SP_DEF_IMAX 10000 + +/* Configuration of Interrupts / sec for entire HBA port */ +#define LPFC_MIN_IMAX 5000 +#define LPFC_MAX_IMAX 5000000 +#define LPFC_DEF_IMAX 50000 /* PORT_CAPABILITIES constants. */ #define LPFC_MAX_SUPPORTED_PAGES 8 @@ -338,7 +344,7 @@ struct lpfc_cqe { * Define mask value for xri_aborted and wcqe completed CQE extended status. * Currently, extended status is limited to 9 bits (0x0 -> 0x103) . */ -#define WCQE_PARAM_MASK 0x1FF; +#define WCQE_PARAM_MASK 0x1FF /* completion queue entry for wqe completions */ struct lpfc_wcqe_complete { @@ -880,13 +886,19 @@ struct mbox_header { #define LPFC_MBOX_OPCODE_EQ_DESTROY 0x37 #define LPFC_MBOX_OPCODE_QUERY_FW_CFG 0x3A #define LPFC_MBOX_OPCODE_FUNCTION_RESET 0x3D +#define LPFC_MBOX_OPCODE_SET_PHYSICAL_LINK_CONFIG 0x3E +#define LPFC_MBOX_OPCODE_SET_BOOT_CONFIG 0x43 #define LPFC_MBOX_OPCODE_GET_PORT_NAME 0x4D #define LPFC_MBOX_OPCODE_MQ_CREATE_EXT 0x5A +#define LPFC_MBOX_OPCODE_GET_VPD_DATA 0x5B +#define LPFC_MBOX_OPCODE_SEND_ACTIVATION 0x73 +#define LPFC_MBOX_OPCODE_RESET_LICENSES 0x74 #define LPFC_MBOX_OPCODE_GET_RSRC_EXTENT_INFO 0x9A #define LPFC_MBOX_OPCODE_GET_ALLOC_RSRC_EXTENT 0x9B #define LPFC_MBOX_OPCODE_ALLOC_RSRC_EXTENT 0x9C #define LPFC_MBOX_OPCODE_DEALLOC_RSRC_EXTENT 0x9D #define LPFC_MBOX_OPCODE_GET_FUNCTION_CONFIG 0xA0 +#define LPFC_MBOX_OPCODE_GET_PROFILE_CAPACITIES 0xA1 #define LPFC_MBOX_OPCODE_GET_PROFILE_CONFIG 0xA4 #define LPFC_MBOX_OPCODE_SET_PROFILE_CONFIG 0xA5 #define LPFC_MBOX_OPCODE_GET_PROFILE_LIST 0xA6 @@ -1382,6 +1394,11 @@ struct lpfc_mbx_set_link_diag_state { #define lpfc_mbx_set_diag_state_diag_SHIFT 0 #define lpfc_mbx_set_diag_state_diag_MASK 0x00000001 #define lpfc_mbx_set_diag_state_diag_WORD word0 +#define lpfc_mbx_set_diag_state_diag_bit_valid_SHIFT 2 +#define lpfc_mbx_set_diag_state_diag_bit_valid_MASK 0x00000001 +#define lpfc_mbx_set_diag_state_diag_bit_valid_WORD word0 +#define LPFC_DIAG_STATE_DIAG_BIT_VALID_NO_CHANGE 0 +#define LPFC_DIAG_STATE_DIAG_BIT_VALID_CHANGE 1 #define lpfc_mbx_set_diag_state_link_num_SHIFT 16 #define lpfc_mbx_set_diag_state_link_num_MASK 0x0000003F #define lpfc_mbx_set_diag_state_link_num_WORD word0 @@ -2556,7 +2573,7 @@ struct lpfc_mbx_get_sli4_parameters { }; struct lpfc_rscr_desc_generic { -#define LPFC_RSRC_DESC_WSIZE 18 +#define LPFC_RSRC_DESC_WSIZE 22 uint32_t desc[LPFC_RSRC_DESC_WSIZE]; }; @@ -2566,6 +2583,9 @@ struct lpfc_rsrc_desc_pcie { #define lpfc_rsrc_desc_pcie_type_MASK 0x000000ff #define lpfc_rsrc_desc_pcie_type_WORD word0 #define LPFC_RSRC_DESC_TYPE_PCIE 0x40 +#define lpfc_rsrc_desc_pcie_length_SHIFT 8 +#define lpfc_rsrc_desc_pcie_length_MASK 0x000000ff +#define lpfc_rsrc_desc_pcie_length_WORD word0 uint32_t word1; #define lpfc_rsrc_desc_pcie_pfnum_SHIFT 0 #define lpfc_rsrc_desc_pcie_pfnum_MASK 0x000000ff @@ -2593,6 +2613,12 @@ struct lpfc_rsrc_desc_fcfcoe { #define lpfc_rsrc_desc_fcfcoe_type_MASK 0x000000ff #define lpfc_rsrc_desc_fcfcoe_type_WORD word0 #define LPFC_RSRC_DESC_TYPE_FCFCOE 0x43 +#define lpfc_rsrc_desc_fcfcoe_length_SHIFT 8 +#define lpfc_rsrc_desc_fcfcoe_length_MASK 0x000000ff +#define lpfc_rsrc_desc_fcfcoe_length_WORD word0 +#define LPFC_RSRC_DESC_TYPE_FCFCOE_V0_RSVD 0 +#define LPFC_RSRC_DESC_TYPE_FCFCOE_V0_LENGTH 72 +#define LPFC_RSRC_DESC_TYPE_FCFCOE_V1_LENGTH 88 uint32_t word1; #define lpfc_rsrc_desc_fcfcoe_vfnum_SHIFT 0 #define lpfc_rsrc_desc_fcfcoe_vfnum_MASK 0x000000ff @@ -2651,6 +2677,12 @@ struct lpfc_rsrc_desc_fcfcoe { #define lpfc_rsrc_desc_fcfcoe_eq_cnt_SHIFT 16 #define lpfc_rsrc_desc_fcfcoe_eq_cnt_MASK 0x0000ffff #define lpfc_rsrc_desc_fcfcoe_eq_cnt_WORD word13 +/* extended FC/FCoE Resource Descriptor when length = 88 bytes */ + uint32_t bw_min; + uint32_t bw_max; + uint32_t iops_min; + uint32_t iops_max; + uint32_t reserved[4]; }; struct lpfc_func_cfg { diff --git a/drivers/scsi/lpfc/lpfc_init.c b/drivers/scsi/lpfc/lpfc_init.c index 628a703..8a55a58 100644 --- a/drivers/scsi/lpfc/lpfc_init.c +++ b/drivers/scsi/lpfc/lpfc_init.c @@ -480,11 +480,11 @@ lpfc_config_port_post(struct lpfc_hba *phba) phba->link_state = LPFC_LINK_DOWN; /* Only process IOCBs on ELS ring till hba_state is READY */ - if (psli->ring[psli->extra_ring].cmdringaddr) + if (psli->ring[psli->extra_ring].sli.sli3.cmdringaddr) psli->ring[psli->extra_ring].flag |= LPFC_STOP_IOCB_EVENT; - if (psli->ring[psli->fcp_ring].cmdringaddr) + if (psli->ring[psli->fcp_ring].sli.sli3.cmdringaddr) psli->ring[psli->fcp_ring].flag |= LPFC_STOP_IOCB_EVENT; - if (psli->ring[psli->next_ring].cmdringaddr) + if (psli->ring[psli->next_ring].sli.sli3.cmdringaddr) psli->ring[psli->next_ring].flag |= LPFC_STOP_IOCB_EVENT; /* Post receive buffers for desired rings */ @@ -2059,6 +2059,11 @@ lpfc_get_hba_model_desc(struct lpfc_hba *phba, uint8_t *mdp, uint8_t *descp) oneConnect = 1; m = (typeof(m)){"OCe15100", "PCIe", "FCoE"}; break; + case PCI_DEVICE_ID_SKYHAWK: + case PCI_DEVICE_ID_SKYHAWK_VF: + oneConnect = 1; + m = (typeof(m)){"OCe14000", "PCIe", "FCoE"}; + break; default: m = (typeof(m)){"Unknown", "", ""}; break; @@ -4546,6 +4551,13 @@ lpfc_sli_driver_resource_setup(struct lpfc_hba *phba) phba->cfg_sg_seg_cnt = LPFC_DEFAULT_MENLO_SG_SEG_CNT; } + if (!phba->sli.ring) + phba->sli.ring = (struct lpfc_sli_ring *) + kzalloc(LPFC_SLI3_MAX_RING * + sizeof(struct lpfc_sli_ring), GFP_KERNEL); + if (!phba->sli.ring) + return -ENOMEM; + /* * Since the sg_tablesize is module parameter, the sg_dma_buf_size * used to create the sg_dma_buf_pool must be dynamically calculated. @@ -4690,6 +4702,10 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba) /* Get all the module params for configuring this host */ lpfc_get_cfgparam(phba); phba->max_vpi = LPFC_MAX_VPI; + + /* Eventually cfg_fcp_eq_count / cfg_fcp_wq_count will be depricated */ + phba->cfg_fcp_io_channel = phba->cfg_fcp_eq_count; + /* This will be set to correct value after the read_config mbox */ phba->max_vports = 0; @@ -4705,6 +4721,16 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba) sges_per_segment = 2; /* + * For SLI4, instead of using ring 0 (LPFC_FCP_RING) for FCP commands + * we will associate a new ring, for each FCP fastpath EQ/CQ/WQ tuple. + */ + if (!phba->sli.ring) + phba->sli.ring = kzalloc( + (LPFC_SLI3_MAX_RING + phba->cfg_fcp_io_channel) * + sizeof(struct lpfc_sli_ring), GFP_KERNEL); + if (!phba->sli.ring) + return -ENOMEM; + /* * Since the sg_tablesize is module parameter, the sg_dma_buf_size * used to create the sg_dma_buf_pool must be dynamically calculated. * 2 segments are added since the IOCB needs a command and response bde. @@ -4909,21 +4935,15 @@ lpfc_sli4_driver_resource_setup(struct lpfc_hba *phba) goto out_remove_rpi_hdrs; } - /* - * The cfg_fcp_eq_count can be zero whenever there is exactly one - * interrupt vector. This is not an error - */ - if (phba->cfg_fcp_eq_count) { - phba->sli4_hba.fcp_eq_hdl = - kzalloc((sizeof(struct lpfc_fcp_eq_hdl) * - phba->cfg_fcp_eq_count), GFP_KERNEL); - if (!phba->sli4_hba.fcp_eq_hdl) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "2572 Failed allocate memory for " - "fast-path per-EQ handle array\n"); - rc = -ENOMEM; - goto out_free_fcf_rr_bmask; - } + phba->sli4_hba.fcp_eq_hdl = + kzalloc((sizeof(struct lpfc_fcp_eq_hdl) * + phba->cfg_fcp_io_channel), GFP_KERNEL); + if (!phba->sli4_hba.fcp_eq_hdl) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "2572 Failed allocate memory for " + "fast-path per-EQ handle array\n"); + rc = -ENOMEM; + goto out_free_fcf_rr_bmask; } phba->sli4_hba.msix_entries = kzalloc((sizeof(struct msix_entry) * @@ -5550,6 +5570,10 @@ lpfc_hba_free(struct lpfc_hba *phba) /* Release the driver assigned board number */ idr_remove(&lpfc_hba_index, phba->brd_no); + /* Free memory allocated with sli rings */ + kfree(phba->sli.ring); + phba->sli.ring = NULL; + kfree(phba); return; } @@ -6275,8 +6299,9 @@ lpfc_sli4_read_config(struct lpfc_hba *phba) uint32_t shdr_status, shdr_add_status; struct lpfc_mbx_get_func_cfg *get_func_cfg; struct lpfc_rsrc_desc_fcfcoe *desc; + char *pdesc_0; uint32_t desc_count; - int length, i, rc = 0; + int length, i, rc = 0, rc2; pmb = (LPFC_MBOXQ_t *) mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL); if (!pmb) { @@ -6388,18 +6413,17 @@ lpfc_sli4_read_config(struct lpfc_hba *phba) LPFC_MBOX_OPCODE_GET_FUNCTION_CONFIG, length, LPFC_SLI4_MBX_EMBED); - rc = lpfc_sli_issue_mbox(phba, pmb, MBX_POLL); + rc2 = lpfc_sli_issue_mbox(phba, pmb, MBX_POLL); shdr = (union lpfc_sli4_cfg_shdr *) &pmb->u.mqe.un.sli4_config.header.cfg_shdr; shdr_status = bf_get(lpfc_mbox_hdr_status, &shdr->response); shdr_add_status = bf_get(lpfc_mbox_hdr_add_status, &shdr->response); - if (rc || shdr_status || shdr_add_status) { + if (rc2 || shdr_status || shdr_add_status) { lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "3026 Mailbox failed , mbxCmd x%x " "GET_FUNCTION_CONFIG, mbxStatus x%x\n", bf_get(lpfc_mqe_command, &pmb->u.mqe), bf_get(lpfc_mqe_status, &pmb->u.mqe)); - rc = -EIO; goto read_cfg_out; } @@ -6407,11 +6431,18 @@ lpfc_sli4_read_config(struct lpfc_hba *phba) get_func_cfg = &pmb->u.mqe.un.get_func_cfg; desc_count = get_func_cfg->func_cfg.rsrc_desc_count; + pdesc_0 = (char *)&get_func_cfg->func_cfg.desc[0]; + desc = (struct lpfc_rsrc_desc_fcfcoe *)pdesc_0; + length = bf_get(lpfc_rsrc_desc_fcfcoe_length, desc); + if (length == LPFC_RSRC_DESC_TYPE_FCFCOE_V0_RSVD) + length = LPFC_RSRC_DESC_TYPE_FCFCOE_V0_LENGTH; + else if (length != LPFC_RSRC_DESC_TYPE_FCFCOE_V1_LENGTH) + goto read_cfg_out; + for (i = 0; i < LPFC_RSRC_DESC_MAX_NUM; i++) { - desc = (struct lpfc_rsrc_desc_fcfcoe *) - &get_func_cfg->func_cfg.desc[i]; + desc = (struct lpfc_rsrc_desc_fcfcoe *)(pdesc_0 + length * i); if (LPFC_RSRC_DESC_TYPE_FCFCOE == - bf_get(lpfc_rsrc_desc_pcie_type, desc)) { + bf_get(lpfc_rsrc_desc_fcfcoe_type, desc)) { phba->sli4_hba.iov.pf_number = bf_get(lpfc_rsrc_desc_fcfcoe_pfnum, desc); phba->sli4_hba.iov.vf_number = @@ -6425,13 +6456,11 @@ lpfc_sli4_read_config(struct lpfc_hba *phba) "3027 GET_FUNCTION_CONFIG: pf_number:%d, " "vf_number:%d\n", phba->sli4_hba.iov.pf_number, phba->sli4_hba.iov.vf_number); - else { + else lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "3028 GET_FUNCTION_CONFIG: failed to find " "Resrouce Descriptor:x%x\n", LPFC_RSRC_DESC_TYPE_FCFCOE); - rc = -EIO; - } read_cfg_out: mempool_free(pmb, phba->mbox_mem_pool); @@ -6512,53 +6541,40 @@ lpfc_setup_endian_order(struct lpfc_hba *phba) static int lpfc_sli4_queue_verify(struct lpfc_hba *phba) { - int cfg_fcp_wq_count; - int cfg_fcp_eq_count; + int cfg_fcp_io_channel; + uint32_t cpu; + uint32_t i = 0; + /* - * Sanity check for confiugred queue parameters against the run-time + * Sanity check for configured queue parameters against the run-time * device parameters */ - /* Sanity check on FCP fast-path WQ parameters */ - cfg_fcp_wq_count = phba->cfg_fcp_wq_count; - if (cfg_fcp_wq_count > - (phba->sli4_hba.max_cfg_param.max_wq - LPFC_SP_WQN_DEF)) { - cfg_fcp_wq_count = phba->sli4_hba.max_cfg_param.max_wq - - LPFC_SP_WQN_DEF; - if (cfg_fcp_wq_count < LPFC_FP_WQN_MIN) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "2581 Not enough WQs (%d) from " - "the pci function for supporting " - "FCP WQs (%d)\n", - phba->sli4_hba.max_cfg_param.max_wq, - phba->cfg_fcp_wq_count); - goto out_error; - } + /* Sanity check on HBA EQ parameters */ + cfg_fcp_io_channel = phba->cfg_fcp_io_channel; + + /* It doesn't make sense to have more io channels then CPUs */ + for_each_online_cpu(cpu) { + i++; + } + if (i < cfg_fcp_io_channel) { lpfc_printf_log(phba, KERN_WARNING, LOG_INIT, - "2582 Not enough WQs (%d) from the pci " - "function for supporting the requested " - "FCP WQs (%d), the actual FCP WQs can " - "be supported: %d\n", - phba->sli4_hba.max_cfg_param.max_wq, - phba->cfg_fcp_wq_count, cfg_fcp_wq_count); - } - /* The actual number of FCP work queues adopted */ - phba->cfg_fcp_wq_count = cfg_fcp_wq_count; - - /* Sanity check on FCP fast-path EQ parameters */ - cfg_fcp_eq_count = phba->cfg_fcp_eq_count; - if (cfg_fcp_eq_count > - (phba->sli4_hba.max_cfg_param.max_eq - LPFC_SP_EQN_DEF)) { - cfg_fcp_eq_count = phba->sli4_hba.max_cfg_param.max_eq - - LPFC_SP_EQN_DEF; - if (cfg_fcp_eq_count < LPFC_FP_EQN_MIN) { + "3188 Reducing IO channels to match number of " + "CPUs: from %d to %d\n", cfg_fcp_io_channel, i); + cfg_fcp_io_channel = i; + } + + if (cfg_fcp_io_channel > + phba->sli4_hba.max_cfg_param.max_eq) { + cfg_fcp_io_channel = phba->sli4_hba.max_cfg_param.max_eq; + if (cfg_fcp_io_channel < LPFC_FCP_IO_CHAN_MIN) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "2574 Not enough EQs (%d) from the " "pci function for supporting FCP " "EQs (%d)\n", phba->sli4_hba.max_cfg_param.max_eq, - phba->cfg_fcp_eq_count); + phba->cfg_fcp_io_channel); goto out_error; } lpfc_printf_log(phba, KERN_WARNING, LOG_INIT, @@ -6567,22 +6583,16 @@ lpfc_sli4_queue_verify(struct lpfc_hba *phba) "FCP EQs (%d), the actual FCP EQs can " "be supported: %d\n", phba->sli4_hba.max_cfg_param.max_eq, - phba->cfg_fcp_eq_count, cfg_fcp_eq_count); - } - /* It does not make sense to have more EQs than WQs */ - if (cfg_fcp_eq_count > phba->cfg_fcp_wq_count) { - lpfc_printf_log(phba, KERN_WARNING, LOG_INIT, - "2593 The FCP EQ count(%d) cannot be greater " - "than the FCP WQ count(%d), limiting the " - "FCP EQ count to %d\n", cfg_fcp_eq_count, - phba->cfg_fcp_wq_count, - phba->cfg_fcp_wq_count); - cfg_fcp_eq_count = phba->cfg_fcp_wq_count; + phba->cfg_fcp_io_channel, cfg_fcp_io_channel); } + + /* Eventually cfg_fcp_eq_count / cfg_fcp_wq_count will be depricated */ + /* The actual number of FCP event queues adopted */ - phba->cfg_fcp_eq_count = cfg_fcp_eq_count; - /* The overall number of event queues used */ - phba->sli4_hba.cfg_eqn = phba->cfg_fcp_eq_count + LPFC_SP_EQN_DEF; + phba->cfg_fcp_eq_count = cfg_fcp_io_channel; + phba->cfg_fcp_wq_count = cfg_fcp_io_channel; + phba->cfg_fcp_io_channel = cfg_fcp_io_channel; + phba->sli4_hba.cfg_eqn = cfg_fcp_io_channel; /* Get EQ depth from module parameter, fake the default for now */ phba->sli4_hba.eq_esize = LPFC_EQE_SIZE_4B; @@ -6615,50 +6625,104 @@ int lpfc_sli4_queue_create(struct lpfc_hba *phba) { struct lpfc_queue *qdesc; - int fcp_eqidx, fcp_cqidx, fcp_wqidx; + int idx; /* - * Create Event Queues (EQs) + * Create HBA Record arrays. */ + if (!phba->cfg_fcp_io_channel) + return -ERANGE; - /* Create slow path event queue */ - qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.eq_esize, - phba->sli4_hba.eq_ecount); - if (!qdesc) { + phba->sli4_hba.mq_esize = LPFC_MQE_SIZE; + phba->sli4_hba.mq_ecount = LPFC_MQE_DEF_COUNT; + phba->sli4_hba.wq_esize = LPFC_WQE_SIZE; + phba->sli4_hba.wq_ecount = LPFC_WQE_DEF_COUNT; + phba->sli4_hba.rq_esize = LPFC_RQE_SIZE; + phba->sli4_hba.rq_ecount = LPFC_RQE_DEF_COUNT; + + phba->sli4_hba.hba_eq = kzalloc((sizeof(struct lpfc_queue *) * + phba->cfg_fcp_io_channel), GFP_KERNEL); + if (!phba->sli4_hba.hba_eq) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "2576 Failed allocate memory for " + "fast-path EQ record array\n"); + goto out_error; + } + + phba->sli4_hba.fcp_cq = kzalloc((sizeof(struct lpfc_queue *) * + phba->cfg_fcp_io_channel), GFP_KERNEL); + if (!phba->sli4_hba.fcp_cq) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0496 Failed allocate slow-path EQ\n"); + "2577 Failed allocate memory for fast-path " + "CQ record array\n"); + goto out_error; + } + + phba->sli4_hba.fcp_wq = kzalloc((sizeof(struct lpfc_queue *) * + phba->cfg_fcp_io_channel), GFP_KERNEL); + if (!phba->sli4_hba.fcp_wq) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "2578 Failed allocate memory for fast-path " + "WQ record array\n"); goto out_error; } - phba->sli4_hba.sp_eq = qdesc; /* - * Create fast-path FCP Event Queue(s). The cfg_fcp_eq_count can be - * zero whenever there is exactly one interrupt vector. This is not - * an error. + * Since the first EQ can have multiple CQs associated with it, + * this array is used to quickly see if we have a FCP fast-path + * CQ match. */ - if (phba->cfg_fcp_eq_count) { - phba->sli4_hba.fp_eq = kzalloc((sizeof(struct lpfc_queue *) * - phba->cfg_fcp_eq_count), GFP_KERNEL); - if (!phba->sli4_hba.fp_eq) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "2576 Failed allocate memory for " - "fast-path EQ record array\n"); - goto out_free_sp_eq; - } + phba->sli4_hba.fcp_cq_map = kzalloc((sizeof(uint16_t) * + phba->cfg_fcp_io_channel), GFP_KERNEL); + if (!phba->sli4_hba.fcp_cq_map) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "2545 Failed allocate memory for fast-path " + "CQ map\n"); + goto out_error; } - for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_eq_count; fcp_eqidx++) { + + /* + * Create HBA Event Queues (EQs). The cfg_fcp_io_channel specifies + * how many EQs to create. + */ + for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++) { + + /* Create EQs */ qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.eq_esize, phba->sli4_hba.eq_ecount); if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0497 Failed allocate fast-path EQ\n"); - goto out_free_fp_eq; + "0497 Failed allocate EQ (%d)\n", idx); + goto out_error; } - phba->sli4_hba.fp_eq[fcp_eqidx] = qdesc; + phba->sli4_hba.hba_eq[idx] = qdesc; + + /* Create Fast Path FCP CQs */ + qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.cq_esize, + phba->sli4_hba.cq_ecount); + if (!qdesc) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "0499 Failed allocate fast-path FCP " + "CQ (%d)\n", idx); + goto out_error; + } + phba->sli4_hba.fcp_cq[idx] = qdesc; + + /* Create Fast Path FCP WQs */ + qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.wq_esize, + phba->sli4_hba.wq_ecount); + if (!qdesc) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "0503 Failed allocate fast-path FCP " + "WQ (%d)\n", idx); + goto out_error; + } + phba->sli4_hba.fcp_wq[idx] = qdesc; } + /* - * Create Complete Queues (CQs) + * Create Slow Path Completion Queues (CQs) */ /* Create slow-path Mailbox Command Complete Queue */ @@ -6667,7 +6731,7 @@ lpfc_sli4_queue_create(struct lpfc_hba *phba) if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0500 Failed allocate slow-path mailbox CQ\n"); - goto out_free_fp_eq; + goto out_error; } phba->sli4_hba.mbx_cq = qdesc; @@ -6677,59 +6741,29 @@ lpfc_sli4_queue_create(struct lpfc_hba *phba) if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0501 Failed allocate slow-path ELS CQ\n"); - goto out_free_mbx_cq; + goto out_error; } phba->sli4_hba.els_cq = qdesc; /* - * Create fast-path FCP Completion Queue(s), one-to-one with FCP EQs. - * If there are no FCP EQs then create exactly one FCP CQ. + * Create Slow Path Work Queues (WQs) */ - if (phba->cfg_fcp_eq_count) - phba->sli4_hba.fcp_cq = kzalloc((sizeof(struct lpfc_queue *) * - phba->cfg_fcp_eq_count), - GFP_KERNEL); - else - phba->sli4_hba.fcp_cq = kzalloc(sizeof(struct lpfc_queue *), - GFP_KERNEL); - if (!phba->sli4_hba.fcp_cq) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "2577 Failed allocate memory for fast-path " - "CQ record array\n"); - goto out_free_els_cq; - } - fcp_cqidx = 0; - do { - qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.cq_esize, - phba->sli4_hba.cq_ecount); - if (!qdesc) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0499 Failed allocate fast-path FCP " - "CQ (%d)\n", fcp_cqidx); - goto out_free_fcp_cq; - } - phba->sli4_hba.fcp_cq[fcp_cqidx] = qdesc; - } while (++fcp_cqidx < phba->cfg_fcp_eq_count); /* Create Mailbox Command Queue */ - phba->sli4_hba.mq_esize = LPFC_MQE_SIZE; - phba->sli4_hba.mq_ecount = LPFC_MQE_DEF_COUNT; qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.mq_esize, phba->sli4_hba.mq_ecount); if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0505 Failed allocate slow-path MQ\n"); - goto out_free_fcp_cq; + goto out_error; } phba->sli4_hba.mbx_wq = qdesc; /* - * Create all the Work Queues (WQs) + * Create ELS Work Queues */ - phba->sli4_hba.wq_esize = LPFC_WQE_SIZE; - phba->sli4_hba.wq_ecount = LPFC_WQE_DEF_COUNT; /* Create slow-path ELS Work Queue */ qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.wq_esize, @@ -6737,36 +6771,13 @@ lpfc_sli4_queue_create(struct lpfc_hba *phba) if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0504 Failed allocate slow-path ELS WQ\n"); - goto out_free_mbx_wq; + goto out_error; } phba->sli4_hba.els_wq = qdesc; - /* Create fast-path FCP Work Queue(s) */ - phba->sli4_hba.fcp_wq = kzalloc((sizeof(struct lpfc_queue *) * - phba->cfg_fcp_wq_count), GFP_KERNEL); - if (!phba->sli4_hba.fcp_wq) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "2578 Failed allocate memory for fast-path " - "WQ record array\n"); - goto out_free_els_wq; - } - for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_wq_count; fcp_wqidx++) { - qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.wq_esize, - phba->sli4_hba.wq_ecount); - if (!qdesc) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0503 Failed allocate fast-path FCP " - "WQ (%d)\n", fcp_wqidx); - goto out_free_fcp_wq; - } - phba->sli4_hba.fcp_wq[fcp_wqidx] = qdesc; - } - /* * Create Receive Queue (RQ) */ - phba->sli4_hba.rq_esize = LPFC_RQE_SIZE; - phba->sli4_hba.rq_ecount = LPFC_RQE_DEF_COUNT; /* Create Receive Queue for header */ qdesc = lpfc_sli4_queue_alloc(phba, phba->sli4_hba.rq_esize, @@ -6774,7 +6785,7 @@ lpfc_sli4_queue_create(struct lpfc_hba *phba) if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0506 Failed allocate receive HRQ\n"); - goto out_free_fcp_wq; + goto out_error; } phba->sli4_hba.hdr_rq = qdesc; @@ -6784,52 +6795,14 @@ lpfc_sli4_queue_create(struct lpfc_hba *phba) if (!qdesc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0507 Failed allocate receive DRQ\n"); - goto out_free_hdr_rq; + goto out_error; } phba->sli4_hba.dat_rq = qdesc; return 0; -out_free_hdr_rq: - lpfc_sli4_queue_free(phba->sli4_hba.hdr_rq); - phba->sli4_hba.hdr_rq = NULL; -out_free_fcp_wq: - for (--fcp_wqidx; fcp_wqidx >= 0; fcp_wqidx--) { - lpfc_sli4_queue_free(phba->sli4_hba.fcp_wq[fcp_wqidx]); - phba->sli4_hba.fcp_wq[fcp_wqidx] = NULL; - } - kfree(phba->sli4_hba.fcp_wq); - phba->sli4_hba.fcp_wq = NULL; -out_free_els_wq: - lpfc_sli4_queue_free(phba->sli4_hba.els_wq); - phba->sli4_hba.els_wq = NULL; -out_free_mbx_wq: - lpfc_sli4_queue_free(phba->sli4_hba.mbx_wq); - phba->sli4_hba.mbx_wq = NULL; -out_free_fcp_cq: - for (--fcp_cqidx; fcp_cqidx >= 0; fcp_cqidx--) { - lpfc_sli4_queue_free(phba->sli4_hba.fcp_cq[fcp_cqidx]); - phba->sli4_hba.fcp_cq[fcp_cqidx] = NULL; - } - kfree(phba->sli4_hba.fcp_cq); - phba->sli4_hba.fcp_cq = NULL; -out_free_els_cq: - lpfc_sli4_queue_free(phba->sli4_hba.els_cq); - phba->sli4_hba.els_cq = NULL; -out_free_mbx_cq: - lpfc_sli4_queue_free(phba->sli4_hba.mbx_cq); - phba->sli4_hba.mbx_cq = NULL; -out_free_fp_eq: - for (--fcp_eqidx; fcp_eqidx >= 0; fcp_eqidx--) { - lpfc_sli4_queue_free(phba->sli4_hba.fp_eq[fcp_eqidx]); - phba->sli4_hba.fp_eq[fcp_eqidx] = NULL; - } - kfree(phba->sli4_hba.fp_eq); - phba->sli4_hba.fp_eq = NULL; -out_free_sp_eq: - lpfc_sli4_queue_free(phba->sli4_hba.sp_eq); - phba->sli4_hba.sp_eq = NULL; out_error: + lpfc_sli4_queue_destroy(phba); return -ENOMEM; } @@ -6848,58 +6821,86 @@ out_error: void lpfc_sli4_queue_destroy(struct lpfc_hba *phba) { - int fcp_qidx; + int idx; + + if (phba->sli4_hba.hba_eq != NULL) { + /* Release HBA event queue */ + for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++) { + if (phba->sli4_hba.hba_eq[idx] != NULL) { + lpfc_sli4_queue_free( + phba->sli4_hba.hba_eq[idx]); + phba->sli4_hba.hba_eq[idx] = NULL; + } + } + kfree(phba->sli4_hba.hba_eq); + phba->sli4_hba.hba_eq = NULL; + } + + if (phba->sli4_hba.fcp_cq != NULL) { + /* Release FCP completion queue */ + for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++) { + if (phba->sli4_hba.fcp_cq[idx] != NULL) { + lpfc_sli4_queue_free( + phba->sli4_hba.fcp_cq[idx]); + phba->sli4_hba.fcp_cq[idx] = NULL; + } + } + kfree(phba->sli4_hba.fcp_cq); + phba->sli4_hba.fcp_cq = NULL; + } + + if (phba->sli4_hba.fcp_wq != NULL) { + /* Release FCP work queue */ + for (idx = 0; idx < phba->cfg_fcp_io_channel; idx++) { + if (phba->sli4_hba.fcp_wq[idx] != NULL) { + lpfc_sli4_queue_free( + phba->sli4_hba.fcp_wq[idx]); + phba->sli4_hba.fcp_wq[idx] = NULL; + } + } + kfree(phba->sli4_hba.fcp_wq); + phba->sli4_hba.fcp_wq = NULL; + } + + /* Release FCP CQ mapping array */ + if (phba->sli4_hba.fcp_cq_map != NULL) { + kfree(phba->sli4_hba.fcp_cq_map); + phba->sli4_hba.fcp_cq_map = NULL; + } /* Release mailbox command work queue */ - lpfc_sli4_queue_free(phba->sli4_hba.mbx_wq); - phba->sli4_hba.mbx_wq = NULL; + if (phba->sli4_hba.mbx_wq != NULL) { + lpfc_sli4_queue_free(phba->sli4_hba.mbx_wq); + phba->sli4_hba.mbx_wq = NULL; + } /* Release ELS work queue */ - lpfc_sli4_queue_free(phba->sli4_hba.els_wq); - phba->sli4_hba.els_wq = NULL; - - /* Release FCP work queue */ - if (phba->sli4_hba.fcp_wq != NULL) - for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_wq_count; - fcp_qidx++) - lpfc_sli4_queue_free(phba->sli4_hba.fcp_wq[fcp_qidx]); - kfree(phba->sli4_hba.fcp_wq); - phba->sli4_hba.fcp_wq = NULL; + if (phba->sli4_hba.els_wq != NULL) { + lpfc_sli4_queue_free(phba->sli4_hba.els_wq); + phba->sli4_hba.els_wq = NULL; + } /* Release unsolicited receive queue */ - lpfc_sli4_queue_free(phba->sli4_hba.hdr_rq); - phba->sli4_hba.hdr_rq = NULL; - lpfc_sli4_queue_free(phba->sli4_hba.dat_rq); - phba->sli4_hba.dat_rq = NULL; + if (phba->sli4_hba.hdr_rq != NULL) { + lpfc_sli4_queue_free(phba->sli4_hba.hdr_rq); + phba->sli4_hba.hdr_rq = NULL; + } + if (phba->sli4_hba.dat_rq != NULL) { + lpfc_sli4_queue_free(phba->sli4_hba.dat_rq); + phba->sli4_hba.dat_rq = NULL; + } /* Release ELS complete queue */ - lpfc_sli4_queue_free(phba->sli4_hba.els_cq); - phba->sli4_hba.els_cq = NULL; + if (phba->sli4_hba.els_cq != NULL) { + lpfc_sli4_queue_free(phba->sli4_hba.els_cq); + phba->sli4_hba.els_cq = NULL; + } /* Release mailbox command complete queue */ - lpfc_sli4_queue_free(phba->sli4_hba.mbx_cq); - phba->sli4_hba.mbx_cq = NULL; - - /* Release FCP response complete queue */ - fcp_qidx = 0; - if (phba->sli4_hba.fcp_cq != NULL) - do - lpfc_sli4_queue_free(phba->sli4_hba.fcp_cq[fcp_qidx]); - while (++fcp_qidx < phba->cfg_fcp_eq_count); - kfree(phba->sli4_hba.fcp_cq); - phba->sli4_hba.fcp_cq = NULL; - - /* Release fast-path event queue */ - if (phba->sli4_hba.fp_eq != NULL) - for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_eq_count; - fcp_qidx++) - lpfc_sli4_queue_free(phba->sli4_hba.fp_eq[fcp_qidx]); - kfree(phba->sli4_hba.fp_eq); - phba->sli4_hba.fp_eq = NULL; - - /* Release slow-path event queue */ - lpfc_sli4_queue_free(phba->sli4_hba.sp_eq); - phba->sli4_hba.sp_eq = NULL; + if (phba->sli4_hba.mbx_cq != NULL) { + lpfc_sli4_queue_free(phba->sli4_hba.mbx_cq); + phba->sli4_hba.mbx_cq = NULL; + } return; } @@ -6919,61 +6920,124 @@ lpfc_sli4_queue_destroy(struct lpfc_hba *phba) int lpfc_sli4_queue_setup(struct lpfc_hba *phba) { + struct lpfc_sli *psli = &phba->sli; + struct lpfc_sli_ring *pring; int rc = -ENOMEM; int fcp_eqidx, fcp_cqidx, fcp_wqidx; int fcp_cq_index = 0; /* - * Set up Event Queues (EQs) + * Set up HBA Event Queues (EQs) */ - /* Set up slow-path event queue */ - if (!phba->sli4_hba.sp_eq) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0520 Slow-path EQ not allocated\n"); - goto out_error; - } - rc = lpfc_eq_create(phba, phba->sli4_hba.sp_eq, - LPFC_SP_DEF_IMAX); - if (rc) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0521 Failed setup of slow-path EQ: " - "rc = 0x%x\n", rc); - goto out_error; - } - lpfc_printf_log(phba, KERN_INFO, LOG_INIT, - "2583 Slow-path EQ setup: queue-id=%d\n", - phba->sli4_hba.sp_eq->queue_id); - - /* Set up fast-path event queue */ - if (phba->cfg_fcp_eq_count && !phba->sli4_hba.fp_eq) { + /* Set up HBA event queue */ + if (phba->cfg_fcp_io_channel && !phba->sli4_hba.hba_eq) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "3147 Fast-path EQs not allocated\n"); rc = -ENOMEM; - goto out_destroy_sp_eq; + goto out_error; } - for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_eq_count; fcp_eqidx++) { - if (!phba->sli4_hba.fp_eq[fcp_eqidx]) { + for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_io_channel; fcp_eqidx++) { + if (!phba->sli4_hba.hba_eq[fcp_eqidx]) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0522 Fast-path EQ (%d) not " "allocated\n", fcp_eqidx); rc = -ENOMEM; - goto out_destroy_fp_eq; + goto out_destroy_hba_eq; } - rc = lpfc_eq_create(phba, phba->sli4_hba.fp_eq[fcp_eqidx], - phba->cfg_fcp_imax); + rc = lpfc_eq_create(phba, phba->sli4_hba.hba_eq[fcp_eqidx], + (phba->cfg_fcp_imax / phba->cfg_fcp_io_channel)); if (rc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0523 Failed setup of fast-path EQ " "(%d), rc = 0x%x\n", fcp_eqidx, rc); - goto out_destroy_fp_eq; + goto out_destroy_hba_eq; } lpfc_printf_log(phba, KERN_INFO, LOG_INIT, - "2584 Fast-path EQ setup: " + "2584 HBA EQ setup: " "queue[%d]-id=%d\n", fcp_eqidx, - phba->sli4_hba.fp_eq[fcp_eqidx]->queue_id); + phba->sli4_hba.hba_eq[fcp_eqidx]->queue_id); } + /* Set up fast-path FCP Response Complete Queue */ + if (!phba->sli4_hba.fcp_cq) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "3148 Fast-path FCP CQ array not " + "allocated\n"); + rc = -ENOMEM; + goto out_destroy_hba_eq; + } + + for (fcp_cqidx = 0; fcp_cqidx < phba->cfg_fcp_io_channel; fcp_cqidx++) { + if (!phba->sli4_hba.fcp_cq[fcp_cqidx]) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "0526 Fast-path FCP CQ (%d) not " + "allocated\n", fcp_cqidx); + rc = -ENOMEM; + goto out_destroy_fcp_cq; + } + rc = lpfc_cq_create(phba, phba->sli4_hba.fcp_cq[fcp_cqidx], + phba->sli4_hba.hba_eq[fcp_cqidx], LPFC_WCQ, LPFC_FCP); + if (rc) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "0527 Failed setup of fast-path FCP " + "CQ (%d), rc = 0x%x\n", fcp_cqidx, rc); + goto out_destroy_fcp_cq; + } + + /* Setup fcp_cq_map for fast lookup */ + phba->sli4_hba.fcp_cq_map[fcp_cqidx] = + phba->sli4_hba.fcp_cq[fcp_cqidx]->queue_id; + + lpfc_printf_log(phba, KERN_INFO, LOG_INIT, + "2588 FCP CQ setup: cq[%d]-id=%d, " + "parent seq[%d]-id=%d\n", + fcp_cqidx, + phba->sli4_hba.fcp_cq[fcp_cqidx]->queue_id, + fcp_cqidx, + phba->sli4_hba.hba_eq[fcp_cqidx]->queue_id); + } + + /* Set up fast-path FCP Work Queue */ + if (!phba->sli4_hba.fcp_wq) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "3149 Fast-path FCP WQ array not " + "allocated\n"); + rc = -ENOMEM; + goto out_destroy_fcp_cq; + } + + for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_io_channel; fcp_wqidx++) { + if (!phba->sli4_hba.fcp_wq[fcp_wqidx]) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "0534 Fast-path FCP WQ (%d) not " + "allocated\n", fcp_wqidx); + rc = -ENOMEM; + goto out_destroy_fcp_wq; + } + rc = lpfc_wq_create(phba, phba->sli4_hba.fcp_wq[fcp_wqidx], + phba->sli4_hba.fcp_cq[fcp_wqidx], + LPFC_FCP); + if (rc) { + lpfc_printf_log(phba, KERN_ERR, LOG_INIT, + "0535 Failed setup of fast-path FCP " + "WQ (%d), rc = 0x%x\n", fcp_wqidx, rc); + goto out_destroy_fcp_wq; + } + + /* Bind this WQ to the next FCP ring */ + pring = &psli->ring[MAX_SLI3_CONFIGURED_RINGS + fcp_wqidx]; + pring->sli.sli4.wqp = (void *)phba->sli4_hba.fcp_wq[fcp_wqidx]; + phba->sli4_hba.fcp_cq[fcp_wqidx]->pring = pring; + + lpfc_printf_log(phba, KERN_INFO, LOG_INIT, + "2591 FCP WQ setup: wq[%d]-id=%d, " + "parent cq[%d]-id=%d\n", + fcp_wqidx, + phba->sli4_hba.fcp_wq[fcp_wqidx]->queue_id, + fcp_cq_index, + phba->sli4_hba.fcp_cq[fcp_wqidx]->queue_id); + } /* * Set up Complete Queues (CQs) */ @@ -6983,20 +7047,20 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0528 Mailbox CQ not allocated\n"); rc = -ENOMEM; - goto out_destroy_fp_eq; + goto out_destroy_fcp_wq; } - rc = lpfc_cq_create(phba, phba->sli4_hba.mbx_cq, phba->sli4_hba.sp_eq, - LPFC_MCQ, LPFC_MBOX); + rc = lpfc_cq_create(phba, phba->sli4_hba.mbx_cq, + phba->sli4_hba.hba_eq[0], LPFC_MCQ, LPFC_MBOX); if (rc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0529 Failed setup of slow-path mailbox CQ: " "rc = 0x%x\n", rc); - goto out_destroy_fp_eq; + goto out_destroy_fcp_wq; } lpfc_printf_log(phba, KERN_INFO, LOG_INIT, "2585 MBX CQ setup: cq-id=%d, parent eq-id=%d\n", phba->sli4_hba.mbx_cq->queue_id, - phba->sli4_hba.sp_eq->queue_id); + phba->sli4_hba.hba_eq[0]->queue_id); /* Set up slow-path ELS Complete Queue */ if (!phba->sli4_hba.els_cq) { @@ -7005,8 +7069,8 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) rc = -ENOMEM; goto out_destroy_mbx_cq; } - rc = lpfc_cq_create(phba, phba->sli4_hba.els_cq, phba->sli4_hba.sp_eq, - LPFC_WCQ, LPFC_ELS); + rc = lpfc_cq_create(phba, phba->sli4_hba.els_cq, + phba->sli4_hba.hba_eq[0], LPFC_WCQ, LPFC_ELS); if (rc) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0531 Failed setup of slow-path ELS CQ: " @@ -7016,52 +7080,7 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) lpfc_printf_log(phba, KERN_INFO, LOG_INIT, "2586 ELS CQ setup: cq-id=%d, parent eq-id=%d\n", phba->sli4_hba.els_cq->queue_id, - phba->sli4_hba.sp_eq->queue_id); - - /* Set up fast-path FCP Response Complete Queue */ - if (!phba->sli4_hba.fcp_cq) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "3148 Fast-path FCP CQ array not " - "allocated\n"); - rc = -ENOMEM; - goto out_destroy_els_cq; - } - fcp_cqidx = 0; - do { - if (!phba->sli4_hba.fcp_cq[fcp_cqidx]) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0526 Fast-path FCP CQ (%d) not " - "allocated\n", fcp_cqidx); - rc = -ENOMEM; - goto out_destroy_fcp_cq; - } - if (phba->cfg_fcp_eq_count) - rc = lpfc_cq_create(phba, - phba->sli4_hba.fcp_cq[fcp_cqidx], - phba->sli4_hba.fp_eq[fcp_cqidx], - LPFC_WCQ, LPFC_FCP); - else - rc = lpfc_cq_create(phba, - phba->sli4_hba.fcp_cq[fcp_cqidx], - phba->sli4_hba.sp_eq, - LPFC_WCQ, LPFC_FCP); - if (rc) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0527 Failed setup of fast-path FCP " - "CQ (%d), rc = 0x%x\n", fcp_cqidx, rc); - goto out_destroy_fcp_cq; - } - lpfc_printf_log(phba, KERN_INFO, LOG_INIT, - "2588 FCP CQ setup: cq[%d]-id=%d, " - "parent %seq[%d]-id=%d\n", - fcp_cqidx, - phba->sli4_hba.fcp_cq[fcp_cqidx]->queue_id, - (phba->cfg_fcp_eq_count) ? "" : "sp_", - fcp_cqidx, - (phba->cfg_fcp_eq_count) ? - phba->sli4_hba.fp_eq[fcp_cqidx]->queue_id : - phba->sli4_hba.sp_eq->queue_id); - } while (++fcp_cqidx < phba->cfg_fcp_eq_count); + phba->sli4_hba.hba_eq[0]->queue_id); /* * Set up all the Work Queues (WQs) @@ -7072,7 +7091,7 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0538 Slow-path MQ not allocated\n"); rc = -ENOMEM; - goto out_destroy_fcp_cq; + goto out_destroy_els_cq; } rc = lpfc_mq_create(phba, phba->sli4_hba.mbx_wq, phba->sli4_hba.mbx_cq, LPFC_MBOX); @@ -7080,7 +7099,7 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0539 Failed setup of slow-path MQ: " "rc = 0x%x\n", rc); - goto out_destroy_fcp_cq; + goto out_destroy_els_cq; } lpfc_printf_log(phba, KERN_INFO, LOG_INIT, "2589 MBX MQ setup: wq-id=%d, parent cq-id=%d\n", @@ -7102,49 +7121,17 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) "rc = 0x%x\n", rc); goto out_destroy_mbx_wq; } + + /* Bind this WQ to the ELS ring */ + pring = &psli->ring[LPFC_ELS_RING]; + pring->sli.sli4.wqp = (void *)phba->sli4_hba.els_wq; + phba->sli4_hba.els_cq->pring = pring; + lpfc_printf_log(phba, KERN_INFO, LOG_INIT, "2590 ELS WQ setup: wq-id=%d, parent cq-id=%d\n", phba->sli4_hba.els_wq->queue_id, phba->sli4_hba.els_cq->queue_id); - /* Set up fast-path FCP Work Queue */ - if (!phba->sli4_hba.fcp_wq) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "3149 Fast-path FCP WQ array not " - "allocated\n"); - rc = -ENOMEM; - goto out_destroy_els_wq; - } - for (fcp_wqidx = 0; fcp_wqidx < phba->cfg_fcp_wq_count; fcp_wqidx++) { - if (!phba->sli4_hba.fcp_wq[fcp_wqidx]) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0534 Fast-path FCP WQ (%d) not " - "allocated\n", fcp_wqidx); - rc = -ENOMEM; - goto out_destroy_fcp_wq; - } - rc = lpfc_wq_create(phba, phba->sli4_hba.fcp_wq[fcp_wqidx], - phba->sli4_hba.fcp_cq[fcp_cq_index], - LPFC_FCP); - if (rc) { - lpfc_printf_log(phba, KERN_ERR, LOG_INIT, - "0535 Failed setup of fast-path FCP " - "WQ (%d), rc = 0x%x\n", fcp_wqidx, rc); - goto out_destroy_fcp_wq; - } - lpfc_printf_log(phba, KERN_INFO, LOG_INIT, - "2591 FCP WQ setup: wq[%d]-id=%d, " - "parent cq[%d]-id=%d\n", - fcp_wqidx, - phba->sli4_hba.fcp_wq[fcp_wqidx]->queue_id, - fcp_cq_index, - phba->sli4_hba.fcp_cq[fcp_cq_index]->queue_id); - /* Round robin FCP Work Queue's Completion Queue assignment */ - if (phba->cfg_fcp_eq_count) - fcp_cq_index = ((fcp_cq_index + 1) % - phba->cfg_fcp_eq_count); - } - /* * Create Receive Queue (RQ) */ @@ -7152,7 +7139,7 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) lpfc_printf_log(phba, KERN_ERR, LOG_INIT, "0540 Receive Queue not allocated\n"); rc = -ENOMEM; - goto out_destroy_fcp_wq; + goto out_destroy_els_wq; } lpfc_rq_adjust_repost(phba, phba->sli4_hba.hdr_rq, LPFC_ELS_HBQ); @@ -7175,25 +7162,23 @@ lpfc_sli4_queue_setup(struct lpfc_hba *phba) phba->sli4_hba.els_cq->queue_id); return 0; -out_destroy_fcp_wq: - for (--fcp_wqidx; fcp_wqidx >= 0; fcp_wqidx--) - lpfc_wq_destroy(phba, phba->sli4_hba.fcp_wq[fcp_wqidx]); out_destroy_els_wq: lpfc_wq_destroy(phba, phba->sli4_hba.els_wq); out_destroy_mbx_wq: lpfc_mq_destroy(phba, phba->sli4_hba.mbx_wq); -out_destroy_fcp_cq: - for (--fcp_cqidx; fcp_cqidx >= 0; fcp_cqidx--) - lpfc_cq_destroy(phba, phba->sli4_hba.fcp_cq[fcp_cqidx]); out_destroy_els_cq: lpfc_cq_destroy(phba, phba->sli4_hba.els_cq); out_destroy_mbx_cq: lpfc_cq_destroy(phba, phba->sli4_hba.mbx_cq); -out_destroy_fp_eq: +out_destroy_fcp_wq: + for (--fcp_wqidx; fcp_wqidx >= 0; fcp_wqidx--) + lpfc_wq_destroy(phba, phba->sli4_hba.fcp_wq[fcp_wqidx]); +out_destroy_fcp_cq: + for (--fcp_cqidx; fcp_cqidx >= 0; fcp_cqidx--) + lpfc_cq_destroy(phba, phba->sli4_hba.fcp_cq[fcp_cqidx]); +out_destroy_hba_eq: for (--fcp_eqidx; fcp_eqidx >= 0; fcp_eqidx--) - lpfc_eq_destroy(phba, phba->sli4_hba.fp_eq[fcp_eqidx]); -out_destroy_sp_eq: - lpfc_eq_destroy(phba, phba->sli4_hba.sp_eq); + lpfc_eq_destroy(phba, phba->sli4_hba.hba_eq[fcp_eqidx]); out_error: return rc; } @@ -7222,27 +7207,27 @@ lpfc_sli4_queue_unset(struct lpfc_hba *phba) /* Unset unsolicited receive queue */ lpfc_rq_destroy(phba, phba->sli4_hba.hdr_rq, phba->sli4_hba.dat_rq); /* Unset FCP work queue */ - for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_wq_count; fcp_qidx++) - lpfc_wq_destroy(phba, phba->sli4_hba.fcp_wq[fcp_qidx]); + if (phba->sli4_hba.fcp_wq) { + for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_io_channel; + fcp_qidx++) + lpfc_wq_destroy(phba, phba->sli4_hba.fcp_wq[fcp_qidx]); + } /* Unset mailbox command complete queue */ lpfc_cq_destroy(phba, phba->sli4_hba.mbx_cq); /* Unset ELS complete queue */ lpfc_cq_destroy(phba, phba->sli4_hba.els_cq); /* Unset FCP response complete queue */ if (phba->sli4_hba.fcp_cq) { - fcp_qidx = 0; - do { + for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_io_channel; + fcp_qidx++) lpfc_cq_destroy(phba, phba->sli4_hba.fcp_cq[fcp_qidx]); - } while (++fcp_qidx < phba->cfg_fcp_eq_count); } /* Unset fast-path event queue */ - if (phba->sli4_hba.fp_eq) { - for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_eq_count; + if (phba->sli4_hba.hba_eq) { + for (fcp_qidx = 0; fcp_qidx < phba->cfg_fcp_io_channel; fcp_qidx++) - lpfc_eq_destroy(phba, phba->sli4_hba.fp_eq[fcp_qidx]); + lpfc_eq_destroy(phba, phba->sli4_hba.hba_eq[fcp_qidx]); } - /* Unset slow-path event queue */ - lpfc_eq_destroy(phba, phba->sli4_hba.sp_eq); } /** @@ -7590,10 +7575,11 @@ lpfc_sli4_send_nop_mbox_cmds(struct lpfc_hba *phba, uint32_t cnt) /* Set up NOP SLI4_CONFIG mailbox-ioctl command */ length = (sizeof(struct lpfc_mbx_nop) - sizeof(struct lpfc_sli4_cfg_mhdr)); - lpfc_sli4_config(phba, mboxq, LPFC_MBOX_SUBSYSTEM_COMMON, - LPFC_MBOX_OPCODE_NOP, length, LPFC_SLI4_MBX_EMBED); for (cmdsent = 0; cmdsent < cnt; cmdsent++) { + lpfc_sli4_config(phba, mboxq, LPFC_MBOX_SUBSYSTEM_COMMON, + LPFC_MBOX_OPCODE_NOP, length, + LPFC_SLI4_MBX_EMBED); if (!phba->sli4_hba.intr_enable) rc = lpfc_sli_issue_mbox(phba, mboxq, MBX_POLL); else { @@ -8133,33 +8119,22 @@ enable_msix_vectors: "message=%d\n", index, phba->sli4_hba.msix_entries[index].vector, phba->sli4_hba.msix_entries[index].entry); + /* * Assign MSI-X vectors to interrupt handlers */ - if (vectors > 1) - rc = request_irq(phba->sli4_hba.msix_entries[0].vector, - &lpfc_sli4_sp_intr_handler, IRQF_SHARED, - LPFC_SP_DRIVER_HANDLER_NAME, phba); - else - /* All Interrupts need to be handled by one EQ */ - rc = request_irq(phba->sli4_hba.msix_entries[0].vector, - &lpfc_sli4_intr_handler, IRQF_SHARED, - LPFC_DRIVER_NAME, phba); - if (rc) { - lpfc_printf_log(phba, KERN_WARNING, LOG_INIT, - "0485 MSI-X slow-path request_irq failed " - "(%d)\n", rc); - goto msi_fail_out; - } + for (index = 0; index < vectors; index++) { + memset(&phba->sli4_hba.handler_name[index], 0, 16); + sprintf((char *)&phba->sli4_hba.handler_name[index], + LPFC_DRIVER_HANDLER_NAME"%d", index); - /* The rest of the vector(s) are associated to fast-path handler(s) */ - for (index = 1; index < vectors; index++) { - phba->sli4_hba.fcp_eq_hdl[index - 1].idx = index - 1; - phba->sli4_hba.fcp_eq_hdl[index - 1].phba = phba; + phba->sli4_hba.fcp_eq_hdl[index].idx = index; + phba->sli4_hba.fcp_eq_hdl[index].phba = phba; + atomic_set(&phba->sli4_hba.fcp_eq_hdl[index].fcp_eq_in_use, 1); rc = request_irq(phba->sli4_hba.msix_entries[index].vector, - &lpfc_sli4_fp_intr_handler, IRQF_SHARED, - LPFC_FP_DRIVER_HANDLER_NAME, - &phba->sli4_hba.fcp_eq_hdl[index - 1]); + &lpfc_sli4_hba_intr_handler, IRQF_SHARED, + (char *)&phba->sli4_hba.handler_name[index], + &phba->sli4_hba.fcp_eq_hdl[index]); if (rc) { lpfc_printf_log(phba, KERN_WARNING, LOG_INIT, "0486 MSI-X fast-path (%d) " @@ -8173,12 +8148,9 @@ enable_msix_vectors: cfg_fail_out: /* free the irq already requested */ - for (--index; index >= 1; index--) - free_irq(phba->sli4_hba.msix_entries[index - 1].vector, - &phba->sli4_hba.fcp_eq_hdl[index - 1]); - - /* free the irq already requested */ - free_irq(phba->sli4_hba.msix_entries[0].vector, phba); + for (--index; index >= 0; index--) + free_irq(phba->sli4_hba.msix_entries[index].vector, + &phba->sli4_hba.fcp_eq_hdl[index]); msi_fail_out: /* Unconfigure MSI-X capability structure */ @@ -8199,11 +8171,9 @@ lpfc_sli4_disable_msix(struct lpfc_hba *phba) int index; /* Free up MSI-X multi-message vectors */ - free_irq(phba->sli4_hba.msix_entries[0].vector, phba); - - for (index = 1; index < phba->sli4_hba.msix_vec_nr; index++) + for (index = 0; index < phba->sli4_hba.msix_vec_nr; index++) free_irq(phba->sli4_hba.msix_entries[index].vector, - &phba->sli4_hba.fcp_eq_hdl[index - 1]); + &phba->sli4_hba.fcp_eq_hdl[index]); /* Disable MSI-X */ pci_disable_msix(phba->pcidev); @@ -8249,7 +8219,7 @@ lpfc_sli4_enable_msi(struct lpfc_hba *phba) return rc; } - for (index = 0; index < phba->cfg_fcp_eq_count; index++) { + for (index = 0; index < phba->cfg_fcp_io_channel; index++) { phba->sli4_hba.fcp_eq_hdl[index].idx = index; phba->sli4_hba.fcp_eq_hdl[index].phba = phba; } @@ -8329,10 +8299,12 @@ lpfc_sli4_enable_intr(struct lpfc_hba *phba, uint32_t cfg_mode) /* Indicate initialization to INTx mode */ phba->intr_type = INTx; intr_mode = 0; - for (index = 0; index < phba->cfg_fcp_eq_count; + for (index = 0; index < phba->cfg_fcp_io_channel; index++) { phba->sli4_hba.fcp_eq_hdl[index].idx = index; phba->sli4_hba.fcp_eq_hdl[index].phba = phba; + atomic_set(&phba->sli4_hba.fcp_eq_hdl[index]. + fcp_eq_in_use, 1); } } } @@ -9449,7 +9421,7 @@ lpfc_pci_probe_one_s4(struct pci_dev *pdev, const struct pci_device_id *pid) int error; uint32_t cfg_mode, intr_mode; int mcnt; - int adjusted_fcp_eq_count; + int adjusted_fcp_io_channel; const struct firmware *fw; uint8_t file_name[16]; @@ -9552,13 +9524,13 @@ lpfc_pci_probe_one_s4(struct pci_dev *pdev, const struct pci_device_id *pid) } /* Default to single EQ for non-MSI-X */ if (phba->intr_type != MSIX) - adjusted_fcp_eq_count = 0; + adjusted_fcp_io_channel = 1; else if (phba->sli4_hba.msix_vec_nr < - phba->cfg_fcp_eq_count + 1) - adjusted_fcp_eq_count = phba->sli4_hba.msix_vec_nr - 1; + phba->cfg_fcp_io_channel) + adjusted_fcp_io_channel = phba->sli4_hba.msix_vec_nr; else - adjusted_fcp_eq_count = phba->cfg_fcp_eq_count; - phba->cfg_fcp_eq_count = adjusted_fcp_eq_count; + adjusted_fcp_io_channel = phba->cfg_fcp_io_channel; + phba->cfg_fcp_io_channel = adjusted_fcp_io_channel; /* Set up SLI-4 HBA */ if (lpfc_sli4_hba_setup(phba)) { lpfc_printf_log(phba, KERN_ERR, LOG_INIT, @@ -9694,6 +9666,7 @@ lpfc_pci_remove_one_s4(struct pci_dev *pdev) * buffers are released to their corresponding pools here. */ lpfc_scsi_free(phba); + lpfc_sli4_driver_resource_unset(phba); /* Unmap adapter Control and Doorbell registers */ @@ -10420,6 +10393,10 @@ static struct pci_device_id lpfc_id_table[] = { PCI_ANY_ID, PCI_ANY_ID, }, {PCI_VENDOR_ID_EMULEX, PCI_DEVICE_ID_LANCER_FCOE_VF, PCI_ANY_ID, PCI_ANY_ID, }, + {PCI_VENDOR_ID_EMULEX, PCI_DEVICE_ID_SKYHAWK, + PCI_ANY_ID, PCI_ANY_ID, }, + {PCI_VENDOR_ID_EMULEX, PCI_DEVICE_ID_SKYHAWK_VF, + PCI_ANY_ID, PCI_ANY_ID, }, { 0 } }; diff --git a/drivers/scsi/lpfc/lpfc_mbox.c b/drivers/scsi/lpfc/lpfc_mbox.c index 20336f0..efc9cd9 100644 --- a/drivers/scsi/lpfc/lpfc_mbox.c +++ b/drivers/scsi/lpfc/lpfc_mbox.c @@ -92,7 +92,7 @@ lpfc_dump_static_vport(struct lpfc_hba *phba, LPFC_MBOXQ_t *pmb, memset(mp->virt, 0, LPFC_BPL_SIZE); INIT_LIST_HEAD(&mp->list); /* save address for completion */ - pmb->context2 = (uint8_t *) mp; + pmb->context1 = (uint8_t *)mp; mb->un.varWords[3] = putPaddrLow(mp->phys); mb->un.varWords[4] = putPaddrHigh(mp->phys); mb->un.varDmp.sli4_length = sizeof(struct static_vport_info); @@ -950,44 +950,47 @@ lpfc_config_pcb_setup(struct lpfc_hba * phba) for (i = 0; i < psli->num_rings; i++) { pring = &psli->ring[i]; - pring->sizeCiocb = phba->sli_rev == 3 ? SLI3_IOCB_CMD_SIZE: + pring->sli.sli3.sizeCiocb = + phba->sli_rev == 3 ? SLI3_IOCB_CMD_SIZE : SLI2_IOCB_CMD_SIZE; - pring->sizeRiocb = phba->sli_rev == 3 ? SLI3_IOCB_RSP_SIZE: + pring->sli.sli3.sizeRiocb = + phba->sli_rev == 3 ? SLI3_IOCB_RSP_SIZE : SLI2_IOCB_RSP_SIZE; /* A ring MUST have both cmd and rsp entries defined to be valid */ - if ((pring->numCiocb == 0) || (pring->numRiocb == 0)) { + if ((pring->sli.sli3.numCiocb == 0) || + (pring->sli.sli3.numRiocb == 0)) { pcbp->rdsc[i].cmdEntries = 0; pcbp->rdsc[i].rspEntries = 0; pcbp->rdsc[i].cmdAddrHigh = 0; pcbp->rdsc[i].rspAddrHigh = 0; pcbp->rdsc[i].cmdAddrLow = 0; pcbp->rdsc[i].rspAddrLow = 0; - pring->cmdringaddr = NULL; - pring->rspringaddr = NULL; + pring->sli.sli3.cmdringaddr = NULL; + pring->sli.sli3.rspringaddr = NULL; continue; } /* Command ring setup for ring */ - pring->cmdringaddr = (void *)&phba->IOCBs[iocbCnt]; - pcbp->rdsc[i].cmdEntries = pring->numCiocb; + pring->sli.sli3.cmdringaddr = (void *)&phba->IOCBs[iocbCnt]; + pcbp->rdsc[i].cmdEntries = pring->sli.sli3.numCiocb; offset = (uint8_t *) &phba->IOCBs[iocbCnt] - (uint8_t *) phba->slim2p.virt; pdma_addr = phba->slim2p.phys + offset; pcbp->rdsc[i].cmdAddrHigh = putPaddrHigh(pdma_addr); pcbp->rdsc[i].cmdAddrLow = putPaddrLow(pdma_addr); - iocbCnt += pring->numCiocb; + iocbCnt += pring->sli.sli3.numCiocb; /* Response ring setup for ring */ - pring->rspringaddr = (void *) &phba->IOCBs[iocbCnt]; + pring->sli.sli3.rspringaddr = (void *) &phba->IOCBs[iocbCnt]; - pcbp->rdsc[i].rspEntries = pring->numRiocb; + pcbp->rdsc[i].rspEntries = pring->sli.sli3.numRiocb; offset = (uint8_t *)&phba->IOCBs[iocbCnt] - (uint8_t *)phba->slim2p.virt; pdma_addr = phba->slim2p.phys + offset; pcbp->rdsc[i].rspAddrHigh = putPaddrHigh(pdma_addr); pcbp->rdsc[i].rspAddrLow = putPaddrLow(pdma_addr); - iocbCnt += pring->numRiocb; + iocbCnt += pring->sli.sli3.numRiocb; } } @@ -1609,12 +1612,15 @@ lpfc_mbox_tmo_val(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq) switch (mbox->mbxCommand) { case MBX_WRITE_NV: /* 0x03 */ + case MBX_DUMP_MEMORY: /* 0x17 */ case MBX_UPDATE_CFG: /* 0x1B */ case MBX_DOWN_LOAD: /* 0x1C */ case MBX_DEL_LD_ENTRY: /* 0x1D */ + case MBX_WRITE_VPARMS: /* 0x32 */ case MBX_LOAD_AREA: /* 0x81 */ case MBX_WRITE_WWN: /* 0x98 */ case MBX_LOAD_EXP_ROM: /* 0x9C */ + case MBX_ACCESS_VDATA: /* 0xA5 */ return LPFC_MBOX_TMO_FLASH_CMD; case MBX_SLI4_CONFIG: /* 0x9b */ subsys = lpfc_sli_config_mbox_subsys_get(phba, mboxq); @@ -1625,11 +1631,17 @@ lpfc_mbox_tmo_val(struct lpfc_hba *phba, LPFC_MBOXQ_t *mboxq) case LPFC_MBOX_OPCODE_WRITE_OBJECT: case LPFC_MBOX_OPCODE_READ_OBJECT_LIST: case LPFC_MBOX_OPCODE_DELETE_OBJECT: - case LPFC_MBOX_OPCODE_GET_FUNCTION_CONFIG: case LPFC_MBOX_OPCODE_GET_PROFILE_LIST: case LPFC_MBOX_OPCODE_SET_ACT_PROFILE: + case LPFC_MBOX_OPCODE_GET_PROFILE_CONFIG: case LPFC_MBOX_OPCODE_SET_PROFILE_CONFIG: case LPFC_MBOX_OPCODE_GET_FACTORY_PROFILE_CONFIG: + case LPFC_MBOX_OPCODE_GET_PROFILE_CAPACITIES: + case LPFC_MBOX_OPCODE_SEND_ACTIVATION: + case LPFC_MBOX_OPCODE_RESET_LICENSES: + case LPFC_MBOX_OPCODE_SET_BOOT_CONFIG: + case LPFC_MBOX_OPCODE_GET_VPD_DATA: + case LPFC_MBOX_OPCODE_SET_PHYSICAL_LINK_CONFIG: return LPFC_MBOX_SLI4_CONFIG_EXTENDED_TMO; } } diff --git a/drivers/scsi/lpfc/lpfc_mem.c b/drivers/scsi/lpfc/lpfc_mem.c index ade763d3..cd86069 100644 --- a/drivers/scsi/lpfc/lpfc_mem.c +++ b/drivers/scsi/lpfc/lpfc_mem.c @@ -1,7 +1,7 @@ /******************************************************************* * This file is part of the Emulex Linux Device Driver for * * Fibre Channel Host Bus Adapters. * - * Copyright (C) 2004-2009 Emulex. All rights reserved. * + * Copyright (C) 2004-2012 Emulex. All rights reserved. * * EMULEX and SLI are trademarks of Emulex. * * www.emulex.com * * Portions Copyright (C) 2004-2005 Christoph Hellwig * @@ -194,6 +194,10 @@ lpfc_mem_free(struct lpfc_hba *phba) pci_pool_destroy(phba->lpfc_hbq_pool); phba->lpfc_hbq_pool = NULL; + if (phba->rrq_pool) + mempool_destroy(phba->rrq_pool); + phba->rrq_pool = NULL; + /* Free NLP memory pool */ mempool_destroy(phba->nlp_mem_pool); phba->nlp_mem_pool = NULL; diff --git a/drivers/scsi/lpfc/lpfc_nportdisc.c b/drivers/scsi/lpfc/lpfc_nportdisc.c index 9133a97..d8fadcb 100644 --- a/drivers/scsi/lpfc/lpfc_nportdisc.c +++ b/drivers/scsi/lpfc/lpfc_nportdisc.c @@ -1778,6 +1778,117 @@ lpfc_device_recov_prli_issue(struct lpfc_vport *vport, } static uint32_t +lpfc_rcv_plogi_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + struct lpfc_iocbq *cmdiocb = (struct lpfc_iocbq *)arg; + struct ls_rjt stat; + + memset(&stat, 0, sizeof(struct ls_rjt)); + stat.un.b.lsRjtRsnCode = LSRJT_UNABLE_TPC; + stat.un.b.lsRjtRsnCodeExp = LSEXP_NOTHING_MORE; + lpfc_els_rsp_reject(vport, stat.un.lsRjtError, cmdiocb, ndlp, NULL); + return ndlp->nlp_state; +} + +static uint32_t +lpfc_rcv_prli_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + struct lpfc_iocbq *cmdiocb = (struct lpfc_iocbq *)arg; + struct ls_rjt stat; + + memset(&stat, 0, sizeof(struct ls_rjt)); + stat.un.b.lsRjtRsnCode = LSRJT_UNABLE_TPC; + stat.un.b.lsRjtRsnCodeExp = LSEXP_NOTHING_MORE; + lpfc_els_rsp_reject(vport, stat.un.lsRjtError, cmdiocb, ndlp, NULL); + return ndlp->nlp_state; +} + +static uint32_t +lpfc_rcv_logo_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + struct Scsi_Host *shost = lpfc_shost_from_vport(vport); + struct lpfc_iocbq *cmdiocb = (struct lpfc_iocbq *)arg; + + spin_lock_irq(shost->host_lock); + ndlp->nlp_flag &= NLP_LOGO_ACC; + spin_unlock_irq(shost->host_lock); + lpfc_els_rsp_acc(vport, ELS_CMD_ACC, cmdiocb, ndlp, NULL); + return ndlp->nlp_state; +} + +static uint32_t +lpfc_rcv_padisc_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + struct lpfc_iocbq *cmdiocb = (struct lpfc_iocbq *)arg; + struct ls_rjt stat; + + memset(&stat, 0, sizeof(struct ls_rjt)); + stat.un.b.lsRjtRsnCode = LSRJT_UNABLE_TPC; + stat.un.b.lsRjtRsnCodeExp = LSEXP_NOTHING_MORE; + lpfc_els_rsp_reject(vport, stat.un.lsRjtError, cmdiocb, ndlp, NULL); + return ndlp->nlp_state; +} + +static uint32_t +lpfc_rcv_prlo_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + struct lpfc_iocbq *cmdiocb = (struct lpfc_iocbq *)arg; + struct ls_rjt stat; + + memset(&stat, 0, sizeof(struct ls_rjt)); + stat.un.b.lsRjtRsnCode = LSRJT_UNABLE_TPC; + stat.un.b.lsRjtRsnCodeExp = LSEXP_NOTHING_MORE; + lpfc_els_rsp_reject(vport, stat.un.lsRjtError, cmdiocb, ndlp, NULL); + return ndlp->nlp_state; +} + +static uint32_t +lpfc_cmpl_logo_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + struct Scsi_Host *shost = lpfc_shost_from_vport(vport); + + ndlp->nlp_prev_state = NLP_STE_LOGO_ISSUE; + lpfc_nlp_set_state(vport, ndlp, NLP_STE_NPR_NODE); + spin_lock_irq(shost->host_lock); + ndlp->nlp_flag &= ~(NLP_NODEV_REMOVE | NLP_NPR_2B_DISC); + spin_unlock_irq(shost->host_lock); + lpfc_disc_set_adisc(vport, ndlp); + return ndlp->nlp_state; +} + +static uint32_t +lpfc_device_rm_logo_issue(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + /* + * Take no action. If a LOGO is outstanding, then possibly DevLoss has + * timed out and is calling for Device Remove. In this case, the LOGO + * must be allowed to complete in state LOGO_ISSUE so that the rpi + * and other NLP flags are correctly cleaned up. + */ + return ndlp->nlp_state; +} + +static uint32_t +lpfc_device_recov_logo_issue(struct lpfc_vport *vport, + struct lpfc_nodelist *ndlp, + void *arg, uint32_t evt) +{ + /* + * Device Recovery events have no meaning for a node with a LOGO + * outstanding. The LOGO has to complete first and handle the + * node from that point. + */ + return ndlp->nlp_state; +} + +static uint32_t lpfc_rcv_plogi_unmap_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, void *arg, uint32_t evt) { @@ -2083,6 +2194,8 @@ lpfc_cmpl_logo_npr_node(struct lpfc_vport *vport, struct lpfc_nodelist *ndlp, void *arg, uint32_t evt) { struct Scsi_Host *shost = lpfc_shost_from_vport(vport); + + /* For the fabric port just clear the fc flags. */ if (ndlp->nlp_DID == Fabric_DID) { spin_lock_irq(shost->host_lock); vport->fc_flag &= ~(FC_FABRIC | FC_PUBLIC_LOOP); @@ -2297,6 +2410,20 @@ static uint32_t (*lpfc_disc_action[NLP_STE_MAX_STATE * NLP_EVT_MAX_EVENT]) lpfc_device_rm_prli_issue, /* DEVICE_RM */ lpfc_device_recov_prli_issue, /* DEVICE_RECOVERY */ + lpfc_rcv_plogi_logo_issue, /* RCV_PLOGI LOGO_ISSUE */ + lpfc_rcv_prli_logo_issue, /* RCV_PRLI */ + lpfc_rcv_logo_logo_issue, /* RCV_LOGO */ + lpfc_rcv_padisc_logo_issue, /* RCV_ADISC */ + lpfc_rcv_padisc_logo_issue, /* RCV_PDISC */ + lpfc_rcv_prlo_logo_issue, /* RCV_PRLO */ + lpfc_cmpl_plogi_illegal, /* CMPL_PLOGI */ + lpfc_disc_illegal, /* CMPL_PRLI */ + lpfc_cmpl_logo_logo_issue, /* CMPL_LOGO */ + lpfc_disc_illegal, /* CMPL_ADISC */ + lpfc_disc_illegal, /* CMPL_REG_LOGIN */ + lpfc_device_rm_logo_issue, /* DEVICE_RM */ + lpfc_device_recov_logo_issue, /* DEVICE_RECOVERY */ + lpfc_rcv_plogi_unmap_node, /* RCV_PLOGI UNMAPPED_NODE */ lpfc_rcv_prli_unmap_node, /* RCV_PRLI */ lpfc_rcv_logo_unmap_node, /* RCV_LOGO */ diff --git a/drivers/scsi/lpfc/lpfc_scsi.c b/drivers/scsi/lpfc/lpfc_scsi.c index 925975d..64013f3 100644 --- a/drivers/scsi/lpfc/lpfc_scsi.c +++ b/drivers/scsi/lpfc/lpfc_scsi.c @@ -60,12 +60,6 @@ static char *dif_op_str[] = { "PROT_WRITE_PASS", }; -static char *dif_grd_str[] = { - "NO_GUARD", - "DIF_CRC", - "DIX_IP", -}; - struct scsi_dif_tuple { __be16 guard_tag; /* Checksum */ __be16 app_tag; /* Opaque storage */ @@ -3482,9 +3476,15 @@ lpfc_handle_fcp_err(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, } lp = (uint32_t *)cmnd->sense_buffer; - if (!scsi_status && (resp_info & RESID_UNDER) && - vport->cfg_log_verbose & LOG_FCP_UNDER) - logit = LOG_FCP_UNDER; + /* special handling for under run conditions */ + if (!scsi_status && (resp_info & RESID_UNDER)) { + /* don't log under runs if fcp set... */ + if (vport->cfg_log_verbose & LOG_FCP) + logit = LOG_FCP_ERROR; + /* unless operator says so */ + if (vport->cfg_log_verbose & LOG_FCP_UNDER) + logit = LOG_FCP_UNDER; + } lpfc_printf_vlog(vport, KERN_WARNING, logit, "9024 FCP command x%x failed: x%x SNS x%x x%x " @@ -3552,11 +3552,11 @@ lpfc_handle_fcp_err(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, /* * Check SLI validation that all the transfer was actually done - * (fcpi_parm should be zero). Apply check only to reads. + * (fcpi_parm should be zero). */ - } else if (fcpi_parm && (cmnd->sc_data_direction == DMA_FROM_DEVICE)) { + } else if (fcpi_parm) { lpfc_printf_vlog(vport, KERN_WARNING, LOG_FCP | LOG_FCP_ERROR, - "9029 FCP Read Check Error Data: " + "9029 FCP Data Transfer Check Error: " "x%x x%x x%x x%x x%x\n", be32_to_cpu(fcpcmd->fcpDl), be32_to_cpu(fcprsp->rspResId), @@ -3615,7 +3615,7 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn, cmd = lpfc_cmd->pCmd; shost = cmd->device->host; - lpfc_cmd->result = pIocbOut->iocb.un.ulpWord[4]; + lpfc_cmd->result = (pIocbOut->iocb.un.ulpWord[4] & IOERR_PARAM_MASK); lpfc_cmd->status = pIocbOut->iocb.ulpStatus; /* pick up SLI4 exhange busy status from HBA */ lpfc_cmd->exch_busy = pIocbOut->iocb_flag & LPFC_EXCHANGE_BUSY; @@ -3660,10 +3660,10 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn, lpfc_cmd->status = IOSTAT_DRIVER_REJECT; else if (lpfc_cmd->status >= IOSTAT_CNT) lpfc_cmd->status = IOSTAT_DEFAULT; - if (lpfc_cmd->status == IOSTAT_FCP_RSP_ERROR - && !lpfc_cmd->fcp_rsp->rspStatus3 - && (lpfc_cmd->fcp_rsp->rspStatus2 & RESID_UNDER) - && !(phba->cfg_log_verbose & LOG_FCP_UNDER)) + if (lpfc_cmd->status == IOSTAT_FCP_RSP_ERROR && + !lpfc_cmd->fcp_rsp->rspStatus3 && + (lpfc_cmd->fcp_rsp->rspStatus2 & RESID_UNDER) && + !(vport->cfg_log_verbose & LOG_FCP_UNDER)) logit = 0; else logit = LOG_FCP | LOG_FCP_UNDER; @@ -3829,12 +3829,15 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn, cmd->scsi_done(cmd); if (phba->cfg_poll & ENABLE_FCP_RING_POLLING) { + spin_lock_irq(&phba->hbalock); + lpfc_cmd->pCmd = NULL; + spin_unlock_irq(&phba->hbalock); + /* * If there is a thread waiting for command completion * wake up the thread. */ spin_lock_irqsave(shost->host_lock, flags); - lpfc_cmd->pCmd = NULL; if (lpfc_cmd->waitq) wake_up(lpfc_cmd->waitq); spin_unlock_irqrestore(shost->host_lock, flags); @@ -3868,12 +3871,15 @@ lpfc_scsi_cmd_iocb_cmpl(struct lpfc_hba *phba, struct lpfc_iocbq *pIocbIn, } } + spin_lock_irq(&phba->hbalock); + lpfc_cmd->pCmd = NULL; + spin_unlock_irq(&phba->hbalock); + /* * If there is a thread waiting for command completion * wake up the thread. */ spin_lock_irqsave(shost->host_lock, flags); - lpfc_cmd->pCmd = NULL; if (lpfc_cmd->waitq) wake_up(lpfc_cmd->waitq); spin_unlock_irqrestore(shost->host_lock, flags); @@ -3919,6 +3925,8 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, struct lpfc_iocbq *piocbq = &(lpfc_cmd->cur_iocbq); int datadir = scsi_cmnd->sc_data_direction; char tag[2]; + uint8_t *ptr; + bool sli4; if (!pnode || !NLP_CHK_NODE_ACT(pnode)) return; @@ -3930,8 +3938,13 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, int_to_scsilun(lpfc_cmd->pCmd->device->lun, &lpfc_cmd->fcp_cmnd->fcp_lun); - memset(&fcp_cmnd->fcpCdb[0], 0, LPFC_FCP_CDB_LEN); - memcpy(&fcp_cmnd->fcpCdb[0], scsi_cmnd->cmnd, scsi_cmnd->cmd_len); + ptr = &fcp_cmnd->fcpCdb[0]; + memcpy(ptr, scsi_cmnd->cmnd, scsi_cmnd->cmd_len); + if (scsi_cmnd->cmd_len < LPFC_FCP_CDB_LEN) { + ptr += scsi_cmnd->cmd_len; + memset(ptr, 0, (LPFC_FCP_CDB_LEN - scsi_cmnd->cmd_len)); + } + if (scsi_populate_tag_msg(scsi_cmnd, tag)) { switch (tag[0]) { case HEAD_OF_QUEUE_TAG: @@ -3947,6 +3960,8 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, } else fcp_cmnd->fcpCntl1 = 0; + sli4 = (phba->sli_rev == LPFC_SLI_REV4); + /* * There are three possibilities here - use scatter-gather segment, use * the single mapping, or neither. Start the lpfc command prep by @@ -3956,11 +3971,12 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, if (scsi_sg_count(scsi_cmnd)) { if (datadir == DMA_TO_DEVICE) { iocb_cmd->ulpCommand = CMD_FCP_IWRITE64_CR; - if (phba->sli_rev < LPFC_SLI_REV4) { + if (sli4) + iocb_cmd->ulpPU = PARM_READ_CHECK; + else { iocb_cmd->un.fcpi.fcpi_parm = 0; iocb_cmd->ulpPU = 0; - } else - iocb_cmd->ulpPU = PARM_READ_CHECK; + } fcp_cmnd->fcpCntl3 = WRITE_DATA; phba->fc4OutputRequests++; } else { @@ -3984,7 +4000,7 @@ lpfc_scsi_prep_cmnd(struct lpfc_vport *vport, struct lpfc_scsi_buf *lpfc_cmd, * of the scsi_cmnd request_buffer */ piocbq->iocb.ulpContext = pnode->nlp_rpi; - if (phba->sli_rev == LPFC_SLI_REV4) + if (sli4) piocbq->iocb.ulpContext = phba->sli4_hba.rpi_ids[pnode->nlp_rpi]; if (pnode->nlp_fcp_info & NLP_FCP_2_DEVICE) @@ -4241,9 +4257,8 @@ void lpfc_poll_timeout(unsigned long ptr) * SCSI_MLQUEUE_HOST_BUSY - Block all devices served by this host temporarily. **/ static int -lpfc_queuecommand_lck(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *)) +lpfc_queuecommand(struct Scsi_Host *shost, struct scsi_cmnd *cmnd) { - struct Scsi_Host *shost = cmnd->device->host; struct lpfc_vport *vport = (struct lpfc_vport *) shost->hostdata; struct lpfc_hba *phba = vport->phba; struct lpfc_rport_data *rdata = cmnd->device->hostdata; @@ -4299,53 +4314,28 @@ lpfc_queuecommand_lck(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *)) lpfc_cmd->timeout = 0; lpfc_cmd->start_time = jiffies; cmnd->host_scribble = (unsigned char *)lpfc_cmd; - cmnd->scsi_done = done; if (scsi_get_prot_op(cmnd) != SCSI_PROT_NORMAL) { if (vport->phba->cfg_enable_bg) { - lpfc_printf_vlog(vport, KERN_WARNING, LOG_BG, - "9033 BLKGRD: rcvd protected cmd:%02x op=%s " - "guard=%s\n", cmnd->cmnd[0], - dif_op_str[scsi_get_prot_op(cmnd)], - dif_grd_str[scsi_host_get_guard(shost)]); - if (cmnd->cmnd[0] == READ_10) - lpfc_printf_vlog(vport, KERN_WARNING, LOG_BG, - "9035 BLKGRD: READ @ sector %llu, " - "cnt %u, rpt %d\n", - (unsigned long long)scsi_get_lba(cmnd), - blk_rq_sectors(cmnd->request), - (cmnd->cmnd[1]>>5)); - else if (cmnd->cmnd[0] == WRITE_10) - lpfc_printf_vlog(vport, KERN_WARNING, LOG_BG, - "9036 BLKGRD: WRITE @ sector %llu, " - "cnt %u, wpt %d\n", - (unsigned long long)scsi_get_lba(cmnd), - blk_rq_sectors(cmnd->request), - (cmnd->cmnd[1]>>5)); + lpfc_printf_vlog(vport, KERN_INFO, LOG_BG, + "9033 BLKGRD: rcvd %s cmd:x%x " + "sector x%llx cnt %u pt %x\n", + dif_op_str[scsi_get_prot_op(cmnd)], + cmnd->cmnd[0], + (unsigned long long)scsi_get_lba(cmnd), + blk_rq_sectors(cmnd->request), + (cmnd->cmnd[1]>>5)); } - err = lpfc_bg_scsi_prep_dma_buf(phba, lpfc_cmd); } else { if (vport->phba->cfg_enable_bg) { - lpfc_printf_vlog(vport, KERN_WARNING, LOG_BG, - "9038 BLKGRD: rcvd unprotected cmd:" - "%02x op=%s guard=%s\n", cmnd->cmnd[0], - dif_op_str[scsi_get_prot_op(cmnd)], - dif_grd_str[scsi_host_get_guard(shost)]); - if (cmnd->cmnd[0] == READ_10) - lpfc_printf_vlog(vport, KERN_WARNING, LOG_BG, - "9040 dbg: READ @ sector %llu, " - "cnt %u, rpt %d\n", - (unsigned long long)scsi_get_lba(cmnd), + lpfc_printf_vlog(vport, KERN_INFO, LOG_BG, + "9038 BLKGRD: rcvd PROT_NORMAL cmd: " + "x%x sector x%llx cnt %u pt %x\n", + cmnd->cmnd[0], + (unsigned long long)scsi_get_lba(cmnd), blk_rq_sectors(cmnd->request), - (cmnd->cmnd[1]>>5)); - else if (cmnd->cmnd[0] == WRITE_10) - lpfc_printf_vlog(vport, KERN_WARNING, LOG_BG, - "9041 dbg: WRITE @ sector %llu, " - "cnt %u, wpt %d\n", - (unsigned long long)scsi_get_lba(cmnd), - blk_rq_sectors(cmnd->request), - (cmnd->cmnd[1]>>5)); + (cmnd->cmnd[1]>>5)); } err = lpfc_scsi_prep_dma_buf(phba, lpfc_cmd); } @@ -4363,11 +4353,9 @@ lpfc_queuecommand_lck(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *)) goto out_host_busy_free_buf; } if (phba->cfg_poll & ENABLE_FCP_RING_POLLING) { - spin_unlock(shost->host_lock); lpfc_sli_handle_fast_ring_event(phba, &phba->sli.ring[LPFC_FCP_RING], HA_R0RE_REQ); - spin_lock(shost->host_lock); if (phba->cfg_poll & DISABLE_FCP_RING_INT) lpfc_poll_rearm_timer(phba); } @@ -4384,11 +4372,10 @@ lpfc_queuecommand_lck(struct scsi_cmnd *cmnd, void (*done) (struct scsi_cmnd *)) return SCSI_MLQUEUE_TARGET_BUSY; out_fail_command: - done(cmnd); + cmnd->scsi_done(cmnd); return 0; } -static DEF_SCSI_QCMD(lpfc_queuecommand) /** * lpfc_abort_handler - scsi_host_template eh_abort_handler entry point @@ -4414,7 +4401,7 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd) DECLARE_WAIT_QUEUE_HEAD_ONSTACK(waitq); status = fc_block_scsi_eh(cmnd); - if (status) + if (status != 0 && status != SUCCESS) return status; spin_lock_irq(&phba->hbalock); @@ -4428,7 +4415,7 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd) } lpfc_cmd = (struct lpfc_scsi_buf *)cmnd->host_scribble; - if (!lpfc_cmd) { + if (!lpfc_cmd || !lpfc_cmd->pCmd) { spin_unlock_irq(&phba->hbalock); lpfc_printf_vlog(vport, KERN_WARNING, LOG_FCP, "2873 SCSI Layer I/O Abort Request IO CMPL Status " @@ -4521,9 +4508,10 @@ lpfc_abort_handler(struct scsi_cmnd *cmnd) ret = FAILED; lpfc_printf_vlog(vport, KERN_ERR, LOG_FCP, "0748 abort handler timed out waiting " - "for abort to complete: ret %#x, ID %d, " - "LUN %d\n", - ret, cmnd->device->id, cmnd->device->lun); + "for abortng I/O (xri:x%x) to complete: " + "ret %#x, ID %d, LUN %d\n", + iocb->sli4_xritag, ret, + cmnd->device->id, cmnd->device->lun); } goto out; @@ -4769,7 +4757,7 @@ lpfc_device_reset_handler(struct scsi_cmnd *cmnd) } pnode = rdata->pnode; status = fc_block_scsi_eh(cmnd); - if (status) + if (status != 0 && status != SUCCESS) return status; status = lpfc_chk_tgt_mapped(vport, cmnd); @@ -4836,7 +4824,7 @@ lpfc_target_reset_handler(struct scsi_cmnd *cmnd) } pnode = rdata->pnode; status = fc_block_scsi_eh(cmnd); - if (status) + if (status != 0 && status != SUCCESS) return status; status = lpfc_chk_tgt_mapped(vport, cmnd); @@ -4904,7 +4892,7 @@ lpfc_bus_reset_handler(struct scsi_cmnd *cmnd) sizeof(scsi_event), (char *)&scsi_event, LPFC_NL_VENDOR_ID); status = fc_block_scsi_eh(cmnd); - if (status) + if (status != 0 && status != SUCCESS) return status; /* diff --git a/drivers/scsi/lpfc/lpfc_sli.c b/drivers/scsi/lpfc/lpfc_sli.c index 0e7e144..219bf53 100644 --- a/drivers/scsi/lpfc/lpfc_sli.c +++ b/drivers/scsi/lpfc/lpfc_sli.c @@ -69,6 +69,8 @@ static int lpfc_sli4_fp_handle_wcqe(struct lpfc_hba *, struct lpfc_queue *, struct lpfc_cqe *); static int lpfc_sli4_post_els_sgl_list(struct lpfc_hba *, struct list_head *, int); +static void lpfc_sli4_hba_handle_eqe(struct lpfc_hba *, struct lpfc_eqe *, + uint32_t); static IOCB_t * lpfc_get_iocb_from_iocbq(struct lpfc_iocbq *iocbq) @@ -94,6 +96,7 @@ lpfc_sli4_wq_put(struct lpfc_queue *q, union lpfc_wqe *wqe) union lpfc_wqe *temp_wqe; struct lpfc_register doorbell; uint32_t host_index; + uint32_t idx; /* sanity check on queue memory */ if (unlikely(!q)) @@ -101,8 +104,12 @@ lpfc_sli4_wq_put(struct lpfc_queue *q, union lpfc_wqe *wqe) temp_wqe = q->qe[q->host_index].wqe; /* If the host has not yet processed the next entry then we are done */ - if (((q->host_index + 1) % q->entry_count) == q->hba_index) + idx = ((q->host_index + 1) % q->entry_count); + if (idx == q->hba_index) { + q->WQ_overflow++; return -ENOMEM; + } + q->WQ_posted++; /* set consumption flag every once in a while */ if (!((q->host_index + 1) % q->entry_repost)) bf_set(wqe_wqec, &wqe->generic.wqe_com, 1); @@ -112,7 +119,8 @@ lpfc_sli4_wq_put(struct lpfc_queue *q, union lpfc_wqe *wqe) /* Update the host index before invoking device */ host_index = q->host_index; - q->host_index = ((q->host_index + 1) % q->entry_count); + + q->host_index = idx; /* Ring Doorbell */ doorbell.word0 = 0; @@ -120,7 +128,6 @@ lpfc_sli4_wq_put(struct lpfc_queue *q, union lpfc_wqe *wqe) bf_set(lpfc_wq_doorbell_index, &doorbell, host_index); bf_set(lpfc_wq_doorbell_id, &doorbell, q->queue_id); writel(doorbell.word0, q->phba->sli4_hba.WQDBregaddr); - readl(q->phba->sli4_hba.WQDBregaddr); /* Flush */ return 0; } @@ -194,7 +201,6 @@ lpfc_sli4_mq_put(struct lpfc_queue *q, struct lpfc_mqe *mqe) bf_set(lpfc_mq_doorbell_num_posted, &doorbell, 1); bf_set(lpfc_mq_doorbell_id, &doorbell, q->queue_id); writel(doorbell.word0, q->phba->sli4_hba.MQDBregaddr); - readl(q->phba->sli4_hba.MQDBregaddr); /* Flush */ return 0; } @@ -234,6 +240,7 @@ static struct lpfc_eqe * lpfc_sli4_eq_get(struct lpfc_queue *q) { struct lpfc_eqe *eqe; + uint32_t idx; /* sanity check on queue memory */ if (unlikely(!q)) @@ -244,14 +251,34 @@ lpfc_sli4_eq_get(struct lpfc_queue *q) if (!bf_get_le32(lpfc_eqe_valid, eqe)) return NULL; /* If the host has not yet processed the next entry then we are done */ - if (((q->hba_index + 1) % q->entry_count) == q->host_index) + idx = ((q->hba_index + 1) % q->entry_count); + if (idx == q->host_index) return NULL; - q->hba_index = ((q->hba_index + 1) % q->entry_count); + q->hba_index = idx; return eqe; } /** + * lpfc_sli4_eq_clr_intr - Turn off interrupts from this EQ + * @q: The Event Queue to disable interrupts + * + **/ +static inline void +lpfc_sli4_eq_clr_intr(struct lpfc_queue *q) +{ + struct lpfc_register doorbell; + + doorbell.word0 = 0; + bf_set(lpfc_eqcq_doorbell_eqci, &doorbell, 1); + bf_set(lpfc_eqcq_doorbell_qt, &doorbell, LPFC_QUEUE_TYPE_EVENT); + bf_set(lpfc_eqcq_doorbell_eqid_hi, &doorbell, + (q->queue_id >> LPFC_EQID_HI_FIELD_SHIFT)); + bf_set(lpfc_eqcq_doorbell_eqid_lo, &doorbell, q->queue_id); + writel(doorbell.word0, q->phba->sli4_hba.EQCQDBregaddr); +} + +/** * lpfc_sli4_eq_release - Indicates the host has finished processing an EQ * @q: The Event Queue that the host has completed processing for. * @arm: Indicates whether the host wants to arms this CQ. @@ -318,6 +345,7 @@ static struct lpfc_cqe * lpfc_sli4_cq_get(struct lpfc_queue *q) { struct lpfc_cqe *cqe; + uint32_t idx; /* sanity check on queue memory */ if (unlikely(!q)) @@ -327,11 +355,12 @@ lpfc_sli4_cq_get(struct lpfc_queue *q) if (!bf_get_le32(lpfc_cqe_valid, q->qe[q->hba_index].cqe)) return NULL; /* If the host has not yet processed the next entry then we are done */ - if (((q->hba_index + 1) % q->entry_count) == q->host_index) + idx = ((q->hba_index + 1) % q->entry_count); + if (idx == q->host_index) return NULL; cqe = q->qe[q->hba_index].cqe; - q->hba_index = ((q->hba_index + 1) % q->entry_count); + q->hba_index = idx; return cqe; } @@ -472,8 +501,8 @@ lpfc_sli4_rq_release(struct lpfc_queue *hq, struct lpfc_queue *dq) static inline IOCB_t * lpfc_cmd_iocb(struct lpfc_hba *phba, struct lpfc_sli_ring *pring) { - return (IOCB_t *) (((char *) pring->cmdringaddr) + - pring->cmdidx * phba->iocb_cmd_size); + return (IOCB_t *) (((char *) pring->sli.sli3.cmdringaddr) + + pring->sli.sli3.cmdidx * phba->iocb_cmd_size); } /** @@ -489,8 +518,8 @@ lpfc_cmd_iocb(struct lpfc_hba *phba, struct lpfc_sli_ring *pring) static inline IOCB_t * lpfc_resp_iocb(struct lpfc_hba *phba, struct lpfc_sli_ring *pring) { - return (IOCB_t *) (((char *) pring->rspringaddr) + - pring->rspidx * phba->iocb_rsp_size); + return (IOCB_t *) (((char *) pring->sli.sli3.rspringaddr) + + pring->sli.sli3.rspidx * phba->iocb_rsp_size); } /** @@ -1320,21 +1349,23 @@ static IOCB_t * lpfc_sli_next_iocb_slot (struct lpfc_hba *phba, struct lpfc_sli_ring *pring) { struct lpfc_pgp *pgp = &phba->port_gp[pring->ringno]; - uint32_t max_cmd_idx = pring->numCiocb; - if ((pring->next_cmdidx == pring->cmdidx) && - (++pring->next_cmdidx >= max_cmd_idx)) - pring->next_cmdidx = 0; + uint32_t max_cmd_idx = pring->sli.sli3.numCiocb; + if ((pring->sli.sli3.next_cmdidx == pring->sli.sli3.cmdidx) && + (++pring->sli.sli3.next_cmdidx >= max_cmd_idx)) + pring->sli.sli3.next_cmdidx = 0; - if (unlikely(pring->local_getidx == pring->next_cmdidx)) { + if (unlikely(pring->sli.sli3.local_getidx == + pring->sli.sli3.next_cmdidx)) { - pring->local_getidx = le32_to_cpu(pgp->cmdGetInx); + pring->sli.sli3.local_getidx = le32_to_cpu(pgp->cmdGetInx); - if (unlikely(pring->local_getidx >= max_cmd_idx)) { + if (unlikely(pring->sli.sli3.local_getidx >= max_cmd_idx)) { lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "0315 Ring %d issue: portCmdGet %d " "is bigger than cmd ring %d\n", pring->ringno, - pring->local_getidx, max_cmd_idx); + pring->sli.sli3.local_getidx, + max_cmd_idx); phba->link_state = LPFC_HBA_ERROR; /* @@ -1349,7 +1380,7 @@ lpfc_sli_next_iocb_slot (struct lpfc_hba *phba, struct lpfc_sli_ring *pring) return NULL; } - if (pring->local_getidx == pring->next_cmdidx) + if (pring->sli.sli3.local_getidx == pring->sli.sli3.next_cmdidx) return NULL; } @@ -1484,8 +1515,8 @@ lpfc_sli_submit_iocb(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, * Let the HBA know what IOCB slot will be the next one the * driver will put a command into. */ - pring->cmdidx = pring->next_cmdidx; - writel(pring->cmdidx, &phba->host_gp[pring->ringno].cmdPutInx); + pring->sli.sli3.cmdidx = pring->sli.sli3.next_cmdidx; + writel(pring->sli.sli3.cmdidx, &phba->host_gp[pring->ringno].cmdPutInx); } /** @@ -2056,6 +2087,7 @@ lpfc_sli_chk_mbx_command(uint8_t mbxCommand) case MBX_READ_EVENT_LOG: case MBX_SECURITY_MGMT: case MBX_AUTH_PORT: + case MBX_ACCESS_VDATA: ret = mbxCommand; break; default: @@ -2786,7 +2818,7 @@ lpfc_sli_rsp_pointers_error(struct lpfc_hba *phba, struct lpfc_sli_ring *pring) "0312 Ring %d handler: portRspPut %d " "is bigger than rsp ring %d\n", pring->ringno, le32_to_cpu(pgp->rspPutInx), - pring->numRiocb); + pring->sli.sli3.numRiocb); phba->link_state = LPFC_HBA_ERROR; @@ -2815,10 +2847,26 @@ lpfc_sli_rsp_pointers_error(struct lpfc_hba *phba, struct lpfc_sli_ring *pring) void lpfc_poll_eratt(unsigned long ptr) { struct lpfc_hba *phba; - uint32_t eratt = 0; + uint32_t eratt = 0, rem; + uint64_t sli_intr, cnt; phba = (struct lpfc_hba *)ptr; + /* Here we will also keep track of interrupts per sec of the hba */ + sli_intr = phba->sli.slistat.sli_intr; + + if (phba->sli.slistat.sli_prev_intr > sli_intr) + cnt = (((uint64_t)(-1) - phba->sli.slistat.sli_prev_intr) + + sli_intr); + else + cnt = (sli_intr - phba->sli.slistat.sli_prev_intr); + + /* 64-bit integer division not supporte on 32-bit x86 - use do_div */ + rem = do_div(cnt, LPFC_ERATT_POLL_INTERVAL); + phba->sli.slistat.sli_ips = cnt; + + phba->sli.slistat.sli_prev_intr = sli_intr; + /* Check chip HA register for error event */ eratt = lpfc_sli_check_eratt(phba); @@ -2873,7 +2921,7 @@ lpfc_sli_handle_fast_ring_event(struct lpfc_hba *phba, * The next available response entry should never exceed the maximum * entries. If it does, treat it as an adapter hardware error. */ - portRspMax = pring->numRiocb; + portRspMax = pring->sli.sli3.numRiocb; portRspPut = le32_to_cpu(pgp->rspPutInx); if (unlikely(portRspPut >= portRspMax)) { lpfc_sli_rsp_pointers_error(phba, pring); @@ -2887,7 +2935,7 @@ lpfc_sli_handle_fast_ring_event(struct lpfc_hba *phba, phba->fcp_ring_in_use = 1; rmb(); - while (pring->rspidx != portRspPut) { + while (pring->sli.sli3.rspidx != portRspPut) { /* * Fetch an entry off the ring and copy it into a local data * structure. The copy involves a byte-swap since the @@ -2896,8 +2944,8 @@ lpfc_sli_handle_fast_ring_event(struct lpfc_hba *phba, entry = lpfc_resp_iocb(phba, pring); phba->last_completion_time = jiffies; - if (++pring->rspidx >= portRspMax) - pring->rspidx = 0; + if (++pring->sli.sli3.rspidx >= portRspMax) + pring->sli.sli3.rspidx = 0; lpfc_sli_pcimem_bcopy((uint32_t *) entry, (uint32_t *) &rspiocbq.iocb, @@ -2915,7 +2963,8 @@ lpfc_sli_handle_fast_ring_event(struct lpfc_hba *phba, * queuedepths of the SCSI device. */ if ((irsp->ulpStatus == IOSTAT_LOCAL_REJECT) && - (irsp->un.ulpWord[4] == IOERR_NO_RESOURCES)) { + ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_NO_RESOURCES)) { spin_unlock_irqrestore(&phba->hbalock, iflag); phba->lpfc_rampdown_queue_depth(phba); spin_lock_irqsave(&phba->hbalock, iflag); @@ -2998,9 +3047,10 @@ lpfc_sli_handle_fast_ring_event(struct lpfc_hba *phba, * been updated, sync the pgp->rspPutInx and fetch the new port * response put pointer. */ - writel(pring->rspidx, &phba->host_gp[pring->ringno].rspGetInx); + writel(pring->sli.sli3.rspidx, + &phba->host_gp[pring->ringno].rspGetInx); - if (pring->rspidx == portRspPut) + if (pring->sli.sli3.rspidx == portRspPut) portRspPut = le32_to_cpu(pgp->rspPutInx); } @@ -3015,7 +3065,7 @@ lpfc_sli_handle_fast_ring_event(struct lpfc_hba *phba, pring->stats.iocb_cmd_empty++; /* Force update of the local copy of cmdGetInx */ - pring->local_getidx = le32_to_cpu(pgp->cmdGetInx); + pring->sli.sli3.local_getidx = le32_to_cpu(pgp->cmdGetInx); lpfc_sli_resume_iocb(phba, pring); if ((pring->lpfc_sli_cmd_available)) @@ -3086,7 +3136,8 @@ lpfc_sli_sp_handle_rspiocb(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, * queuedepths of the SCSI device. */ if ((irsp->ulpStatus == IOSTAT_LOCAL_REJECT) && - (irsp->un.ulpWord[4] == IOERR_NO_RESOURCES)) { + ((irsp->un.ulpWord[4] & IOERR_PARAM_MASK) == + IOERR_NO_RESOURCES)) { spin_unlock_irqrestore(&phba->hbalock, iflag); phba->lpfc_rampdown_queue_depth(phba); spin_lock_irqsave(&phba->hbalock, iflag); @@ -3247,7 +3298,7 @@ lpfc_sli_handle_slow_ring_event_s3(struct lpfc_hba *phba, * The next available response entry should never exceed the maximum * entries. If it does, treat it as an adapter hardware error. */ - portRspMax = pring->numRiocb; + portRspMax = pring->sli.sli3.numRiocb; portRspPut = le32_to_cpu(pgp->rspPutInx); if (portRspPut >= portRspMax) { /* @@ -3269,7 +3320,7 @@ lpfc_sli_handle_slow_ring_event_s3(struct lpfc_hba *phba, } rmb(); - while (pring->rspidx != portRspPut) { + while (pring->sli.sli3.rspidx != portRspPut) { /* * Build a completion list and call the appropriate handler. * The process is to get the next available response iocb, get @@ -3297,8 +3348,8 @@ lpfc_sli_handle_slow_ring_event_s3(struct lpfc_hba *phba, phba->iocb_rsp_size); irsp = &rspiocbp->iocb; - if (++pring->rspidx >= portRspMax) - pring->rspidx = 0; + if (++pring->sli.sli3.rspidx >= portRspMax) + pring->sli.sli3.rspidx = 0; if (pring->ringno == LPFC_ELS_RING) { lpfc_debugfs_slow_ring_trc(phba, @@ -3308,7 +3359,8 @@ lpfc_sli_handle_slow_ring_event_s3(struct lpfc_hba *phba, *(((uint32_t *) irsp) + 7)); } - writel(pring->rspidx, &phba->host_gp[pring->ringno].rspGetInx); + writel(pring->sli.sli3.rspidx, + &phba->host_gp[pring->ringno].rspGetInx); spin_unlock_irqrestore(&phba->hbalock, iflag); /* Handle the response IOCB */ @@ -3320,10 +3372,10 @@ lpfc_sli_handle_slow_ring_event_s3(struct lpfc_hba *phba, * the pgp->rspPutInx in the MAILBOX_tand fetch the new port * response put pointer. */ - if (pring->rspidx == portRspPut) { + if (pring->sli.sli3.rspidx == portRspPut) { portRspPut = le32_to_cpu(pgp->rspPutInx); } - } /* while (pring->rspidx != portRspPut) */ + } /* while (pring->sli.sli3.rspidx != portRspPut) */ if ((rspiocbp != NULL) && (mask & HA_R0RE_REQ)) { /* At least one response entry has been freed */ @@ -3338,7 +3390,7 @@ lpfc_sli_handle_slow_ring_event_s3(struct lpfc_hba *phba, pring->stats.iocb_cmd_empty++; /* Force update of the local copy of cmdGetInx */ - pring->local_getidx = le32_to_cpu(pgp->cmdGetInx); + pring->sli.sli3.local_getidx = le32_to_cpu(pgp->cmdGetInx); lpfc_sli_resume_iocb(phba, pring); if ((pring->lpfc_sli_cmd_available)) @@ -3859,10 +3911,10 @@ lpfc_sli_brdreset(struct lpfc_hba *phba) for (i = 0; i < psli->num_rings; i++) { pring = &psli->ring[i]; pring->flag = 0; - pring->rspidx = 0; - pring->next_cmdidx = 0; - pring->local_getidx = 0; - pring->cmdidx = 0; + pring->sli.sli3.rspidx = 0; + pring->sli.sli3.next_cmdidx = 0; + pring->sli.sli3.local_getidx = 0; + pring->sli.sli3.cmdidx = 0; pring->missbufcnt = 0; } @@ -4893,16 +4945,15 @@ lpfc_sli4_arm_cqeq_intr(struct lpfc_hba *phba) lpfc_sli4_cq_release(phba->sli4_hba.els_cq, LPFC_QUEUE_REARM); fcp_eqidx = 0; if (phba->sli4_hba.fcp_cq) { - do + do { lpfc_sli4_cq_release(phba->sli4_hba.fcp_cq[fcp_eqidx], LPFC_QUEUE_REARM); - while (++fcp_eqidx < phba->cfg_fcp_eq_count); + } while (++fcp_eqidx < phba->cfg_fcp_io_channel); } - lpfc_sli4_eq_release(phba->sli4_hba.sp_eq, LPFC_QUEUE_REARM); - if (phba->sli4_hba.fp_eq) { - for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_eq_count; + if (phba->sli4_hba.hba_eq) { + for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_io_channel; fcp_eqidx++) - lpfc_sli4_eq_release(phba->sli4_hba.fp_eq[fcp_eqidx], + lpfc_sli4_eq_release(phba->sli4_hba.hba_eq[fcp_eqidx], LPFC_QUEUE_REARM); } } @@ -7784,14 +7835,18 @@ lpfc_sli4_bpl2sgl(struct lpfc_hba *phba, struct lpfc_iocbq *piocbq, * * Return: index into SLI4 fast-path FCP queue index. **/ -static uint32_t +static inline uint32_t lpfc_sli4_scmd_to_wqidx_distr(struct lpfc_hba *phba) { - ++phba->fcp_qidx; - if (phba->fcp_qidx >= phba->cfg_fcp_wq_count) - phba->fcp_qidx = 0; + int i; + + if (phba->cfg_fcp_io_sched == LPFC_FCP_SCHED_BY_CPU) + i = smp_processor_id(); + else + i = atomic_add_return(1, &phba->fcp_qidx); - return phba->fcp_qidx; + i = (i % phba->cfg_fcp_io_channel); + return i; } /** @@ -8311,16 +8366,6 @@ __lpfc_sli_issue_iocb_s4(struct lpfc_hba *phba, uint32_t ring_number, if ((piocb->iocb_flag & LPFC_IO_FCP) || (piocb->iocb_flag & LPFC_USE_FCPWQIDX)) { - /* - * For FCP command IOCB, get a new WQ index to distribute - * WQE across the WQsr. On the other hand, for abort IOCB, - * it carries the same WQ index to the original command - * IOCB. - */ - if (piocb->iocb_flag & LPFC_IO_FCP) - piocb->fcp_wqidx = lpfc_sli4_scmd_to_wqidx_distr(phba); - if (unlikely(!phba->sli4_hba.fcp_wq)) - return IOCB_ERROR; if (lpfc_sli4_wq_put(phba->sli4_hba.fcp_wq[piocb->fcp_wqidx], &wqe)) return IOCB_ERROR; @@ -8401,13 +8446,68 @@ int lpfc_sli_issue_iocb(struct lpfc_hba *phba, uint32_t ring_number, struct lpfc_iocbq *piocb, uint32_t flag) { + struct lpfc_fcp_eq_hdl *fcp_eq_hdl; + struct lpfc_sli_ring *pring; + struct lpfc_queue *fpeq; + struct lpfc_eqe *eqe; unsigned long iflags; - int rc; + int rc, idx; - spin_lock_irqsave(&phba->hbalock, iflags); - rc = __lpfc_sli_issue_iocb(phba, ring_number, piocb, flag); - spin_unlock_irqrestore(&phba->hbalock, iflags); + if (phba->sli_rev == LPFC_SLI_REV4) { + if (piocb->iocb_flag & LPFC_IO_FCP) { + if (unlikely(!phba->sli4_hba.fcp_wq)) + return IOCB_ERROR; + idx = lpfc_sli4_scmd_to_wqidx_distr(phba); + piocb->fcp_wqidx = idx; + ring_number = MAX_SLI3_CONFIGURED_RINGS + idx; + + pring = &phba->sli.ring[ring_number]; + spin_lock_irqsave(&pring->ring_lock, iflags); + rc = __lpfc_sli_issue_iocb(phba, ring_number, piocb, + flag); + spin_unlock_irqrestore(&pring->ring_lock, iflags); + + if (lpfc_fcp_look_ahead) { + fcp_eq_hdl = &phba->sli4_hba.fcp_eq_hdl[idx]; + + if (atomic_dec_and_test(&fcp_eq_hdl-> + fcp_eq_in_use)) { + /* Get associated EQ with this index */ + fpeq = phba->sli4_hba.hba_eq[idx]; + + /* Turn off interrupts from this EQ */ + lpfc_sli4_eq_clr_intr(fpeq); + + /* + * Process all the events on FCP EQ + */ + while ((eqe = lpfc_sli4_eq_get(fpeq))) { + lpfc_sli4_hba_handle_eqe(phba, + eqe, idx); + fpeq->EQ_processed++; + } + + /* Always clear and re-arm the EQ */ + lpfc_sli4_eq_release(fpeq, + LPFC_QUEUE_REARM); + } + atomic_inc(&fcp_eq_hdl->fcp_eq_in_use); + } + } else { + pring = &phba->sli.ring[ring_number]; + spin_lock_irqsave(&pring->ring_lock, iflags); + rc = __lpfc_sli_issue_iocb(phba, ring_number, piocb, + flag); + spin_unlock_irqrestore(&pring->ring_lock, iflags); + + } + } else { + /* For now, SLI2/3 will still use hbalock */ + spin_lock_irqsave(&phba->hbalock, iflags); + rc = __lpfc_sli_issue_iocb(phba, ring_number, piocb, flag); + spin_unlock_irqrestore(&phba->hbalock, iflags); + } return rc; } @@ -8434,18 +8534,18 @@ lpfc_extra_ring_setup( struct lpfc_hba *phba) /* Take some away from the FCP ring */ pring = &psli->ring[psli->fcp_ring]; - pring->numCiocb -= SLI2_IOCB_CMD_R1XTRA_ENTRIES; - pring->numRiocb -= SLI2_IOCB_RSP_R1XTRA_ENTRIES; - pring->numCiocb -= SLI2_IOCB_CMD_R3XTRA_ENTRIES; - pring->numRiocb -= SLI2_IOCB_RSP_R3XTRA_ENTRIES; + pring->sli.sli3.numCiocb -= SLI2_IOCB_CMD_R1XTRA_ENTRIES; + pring->sli.sli3.numRiocb -= SLI2_IOCB_RSP_R1XTRA_ENTRIES; + pring->sli.sli3.numCiocb -= SLI2_IOCB_CMD_R3XTRA_ENTRIES; + pring->sli.sli3.numRiocb -= SLI2_IOCB_RSP_R3XTRA_ENTRIES; /* and give them to the extra ring */ pring = &psli->ring[psli->extra_ring]; - pring->numCiocb += SLI2_IOCB_CMD_R1XTRA_ENTRIES; - pring->numRiocb += SLI2_IOCB_RSP_R1XTRA_ENTRIES; - pring->numCiocb += SLI2_IOCB_CMD_R3XTRA_ENTRIES; - pring->numRiocb += SLI2_IOCB_RSP_R3XTRA_ENTRIES; + pring->sli.sli3.numCiocb += SLI2_IOCB_CMD_R1XTRA_ENTRIES; + pring->sli.sli3.numRiocb += SLI2_IOCB_RSP_R1XTRA_ENTRIES; + pring->sli.sli3.numCiocb += SLI2_IOCB_CMD_R3XTRA_ENTRIES; + pring->sli.sli3.numRiocb += SLI2_IOCB_RSP_R3XTRA_ENTRIES; /* Setup default profile for this ring */ pring->iotag_max = 4096; @@ -8457,56 +8557,6 @@ lpfc_extra_ring_setup( struct lpfc_hba *phba) return 0; } -/* lpfc_sli_abts_recover_port - Recover a port that failed an ABTS. - * @vport: pointer to virtual port object. - * @ndlp: nodelist pointer for the impacted rport. - * - * The driver calls this routine in response to a XRI ABORT CQE - * event from the port. In this event, the driver is required to - * recover its login to the rport even though its login may be valid - * from the driver's perspective. The failed ABTS notice from the - * port indicates the rport is not responding. - */ -static void -lpfc_sli_abts_recover_port(struct lpfc_vport *vport, - struct lpfc_nodelist *ndlp) -{ - struct Scsi_Host *shost; - struct lpfc_hba *phba; - unsigned long flags = 0; - - shost = lpfc_shost_from_vport(vport); - phba = vport->phba; - if (ndlp->nlp_state != NLP_STE_MAPPED_NODE) { - lpfc_printf_log(phba, KERN_INFO, - LOG_SLI, "3093 No rport recovery needed. " - "rport in state 0x%x\n", - ndlp->nlp_state); - return; - } - lpfc_printf_log(phba, KERN_ERR, LOG_SLI, - "3094 Start rport recovery on shost id 0x%x " - "fc_id 0x%06x vpi 0x%x rpi 0x%x state 0x%x " - "flags 0x%x\n", - shost->host_no, ndlp->nlp_DID, - vport->vpi, ndlp->nlp_rpi, ndlp->nlp_state, - ndlp->nlp_flag); - /* - * The rport is not responding. Don't attempt ADISC recovery. - * Remove the FCP-2 flag to force a PLOGI. - */ - spin_lock_irqsave(shost->host_lock, flags); - ndlp->nlp_fcp_info &= ~NLP_FCP_2_DEVICE; - spin_unlock_irqrestore(shost->host_lock, flags); - lpfc_disc_state_machine(vport, ndlp, NULL, - NLP_EVT_DEVICE_RECOVERY); - lpfc_cancel_retry_delay_tmo(vport, ndlp); - spin_lock_irqsave(shost->host_lock, flags); - ndlp->nlp_flag |= NLP_NPR_2B_DISC; - spin_unlock_irqrestore(shost->host_lock, flags); - lpfc_disc_start(vport); -} - /* lpfc_sli_abts_err_handler - handle a failed ABTS request from an SLI3 port. * @phba: Pointer to HBA context object. * @iocbq: Pointer to iocb object. @@ -8594,7 +8644,7 @@ lpfc_sli4_abts_err_handler(struct lpfc_hba *phba, * LOCAL_REJECT and 0 for a failed ABTS exchange and later OCe and * LPe FW releases returned LOCAL_REJECT and SEQUENCE_TIMEOUT. */ - ext_status = axri->parameter & WCQE_PARAM_MASK; + ext_status = axri->parameter & IOERR_PARAM_MASK; if ((bf_get(lpfc_wcqe_xa_status, axri) == IOSTAT_LOCAL_REJECT) && ((ext_status == IOERR_SEQUENCE_TIMEOUT) || (ext_status == 0))) lpfc_sli_abts_recover_port(vport, ndlp); @@ -8692,7 +8742,9 @@ lpfc_sli_setup(struct lpfc_hba *phba) struct lpfc_sli *psli = &phba->sli; struct lpfc_sli_ring *pring; - psli->num_rings = MAX_CONFIGURED_RINGS; + psli->num_rings = MAX_SLI3_CONFIGURED_RINGS; + if (phba->sli_rev == LPFC_SLI_REV4) + psli->num_rings += phba->cfg_fcp_io_channel; psli->sli_flag = 0; psli->fcp_ring = LPFC_FCP_RING; psli->next_ring = LPFC_FCP_NEXT_RING; @@ -8707,16 +8759,20 @@ lpfc_sli_setup(struct lpfc_hba *phba) switch (i) { case LPFC_FCP_RING: /* ring 0 - FCP */ /* numCiocb and numRiocb are used in config_port */ - pring->numCiocb = SLI2_IOCB_CMD_R0_ENTRIES; - pring->numRiocb = SLI2_IOCB_RSP_R0_ENTRIES; - pring->numCiocb += SLI2_IOCB_CMD_R1XTRA_ENTRIES; - pring->numRiocb += SLI2_IOCB_RSP_R1XTRA_ENTRIES; - pring->numCiocb += SLI2_IOCB_CMD_R3XTRA_ENTRIES; - pring->numRiocb += SLI2_IOCB_RSP_R3XTRA_ENTRIES; - pring->sizeCiocb = (phba->sli_rev == 3) ? + pring->sli.sli3.numCiocb = SLI2_IOCB_CMD_R0_ENTRIES; + pring->sli.sli3.numRiocb = SLI2_IOCB_RSP_R0_ENTRIES; + pring->sli.sli3.numCiocb += + SLI2_IOCB_CMD_R1XTRA_ENTRIES; + pring->sli.sli3.numRiocb += + SLI2_IOCB_RSP_R1XTRA_ENTRIES; + pring->sli.sli3.numCiocb += + SLI2_IOCB_CMD_R3XTRA_ENTRIES; + pring->sli.sli3.numRiocb += + SLI2_IOCB_RSP_R3XTRA_ENTRIES; + pring->sli.sli3.sizeCiocb = (phba->sli_rev == 3) ? SLI3_IOCB_CMD_SIZE : SLI2_IOCB_CMD_SIZE; - pring->sizeRiocb = (phba->sli_rev == 3) ? + pring->sli.sli3.sizeRiocb = (phba->sli_rev == 3) ? SLI3_IOCB_RSP_SIZE : SLI2_IOCB_RSP_SIZE; pring->iotag_ctr = 0; @@ -8727,12 +8783,12 @@ lpfc_sli_setup(struct lpfc_hba *phba) break; case LPFC_EXTRA_RING: /* ring 1 - EXTRA */ /* numCiocb and numRiocb are used in config_port */ - pring->numCiocb = SLI2_IOCB_CMD_R1_ENTRIES; - pring->numRiocb = SLI2_IOCB_RSP_R1_ENTRIES; - pring->sizeCiocb = (phba->sli_rev == 3) ? + pring->sli.sli3.numCiocb = SLI2_IOCB_CMD_R1_ENTRIES; + pring->sli.sli3.numRiocb = SLI2_IOCB_RSP_R1_ENTRIES; + pring->sli.sli3.sizeCiocb = (phba->sli_rev == 3) ? SLI3_IOCB_CMD_SIZE : SLI2_IOCB_CMD_SIZE; - pring->sizeRiocb = (phba->sli_rev == 3) ? + pring->sli.sli3.sizeRiocb = (phba->sli_rev == 3) ? SLI3_IOCB_RSP_SIZE : SLI2_IOCB_RSP_SIZE; pring->iotag_max = phba->cfg_hba_queue_depth; @@ -8740,12 +8796,12 @@ lpfc_sli_setup(struct lpfc_hba *phba) break; case LPFC_ELS_RING: /* ring 2 - ELS / CT */ /* numCiocb and numRiocb are used in config_port */ - pring->numCiocb = SLI2_IOCB_CMD_R2_ENTRIES; - pring->numRiocb = SLI2_IOCB_RSP_R2_ENTRIES; - pring->sizeCiocb = (phba->sli_rev == 3) ? + pring->sli.sli3.numCiocb = SLI2_IOCB_CMD_R2_ENTRIES; + pring->sli.sli3.numRiocb = SLI2_IOCB_RSP_R2_ENTRIES; + pring->sli.sli3.sizeCiocb = (phba->sli_rev == 3) ? SLI3_IOCB_CMD_SIZE : SLI2_IOCB_CMD_SIZE; - pring->sizeRiocb = (phba->sli_rev == 3) ? + pring->sli.sli3.sizeRiocb = (phba->sli_rev == 3) ? SLI3_IOCB_RSP_SIZE : SLI2_IOCB_RSP_SIZE; pring->fast_iotag = 0; @@ -8786,8 +8842,9 @@ lpfc_sli_setup(struct lpfc_hba *phba) lpfc_sli4_ct_abort_unsol_event; break; } - totiocbsize += (pring->numCiocb * pring->sizeCiocb) + - (pring->numRiocb * pring->sizeRiocb); + totiocbsize += (pring->sli.sli3.numCiocb * + pring->sli.sli3.sizeCiocb) + + (pring->sli.sli3.numRiocb * pring->sli.sli3.sizeRiocb); } if (totiocbsize > MAX_SLIM_IOCB_SIZE) { /* Too many cmd / rsp ring entries in SLI2 SLIM */ @@ -8828,14 +8885,15 @@ lpfc_sli_queue_setup(struct lpfc_hba *phba) for (i = 0; i < psli->num_rings; i++) { pring = &psli->ring[i]; pring->ringno = i; - pring->next_cmdidx = 0; - pring->local_getidx = 0; - pring->cmdidx = 0; + pring->sli.sli3.next_cmdidx = 0; + pring->sli.sli3.local_getidx = 0; + pring->sli.sli3.cmdidx = 0; INIT_LIST_HEAD(&pring->txq); INIT_LIST_HEAD(&pring->txcmplq); INIT_LIST_HEAD(&pring->iocb_continueq); INIT_LIST_HEAD(&pring->iocb_continue_saveq); INIT_LIST_HEAD(&pring->postbufq); + spin_lock_init(&pring->ring_lock); } spin_unlock_irq(&phba->hbalock); return 1; @@ -9334,6 +9392,7 @@ lpfc_sli_abort_iotag_issue(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, IOCB_t *icmd = NULL; IOCB_t *iabt = NULL; int retval; + unsigned long iflags; /* * There are certain command types we don't want to abort. And we @@ -9386,7 +9445,17 @@ lpfc_sli_abort_iotag_issue(struct lpfc_hba *phba, struct lpfc_sli_ring *pring, iabt->un.acxri.abortIoTag, iabt->un.acxri.abortContextTag, abtsiocbp->iotag); - retval = __lpfc_sli_issue_iocb(phba, pring->ringno, abtsiocbp, 0); + + if (phba->sli_rev == LPFC_SLI_REV4) { + /* Note: both hbalock and ring_lock need to be set here */ + spin_lock_irqsave(&pring->ring_lock, iflags); + retval = __lpfc_sli_issue_iocb(phba, pring->ringno, + abtsiocbp, 0); + spin_unlock_irqrestore(&pring->ring_lock, iflags); + } else { + retval = __lpfc_sli_issue_iocb(phba, pring->ringno, + abtsiocbp, 0); + } if (retval) __lpfc_sli_release_iocbq(phba, abtsiocbp); @@ -10947,12 +11016,12 @@ lpfc_sli4_els_wcqe_to_rspiocbq(struct lpfc_hba *phba, unsigned long iflags; wcqe = &irspiocbq->cq_event.cqe.wcqe_cmpl; - spin_lock_irqsave(&phba->hbalock, iflags); + spin_lock_irqsave(&pring->ring_lock, iflags); pring->stats.iocb_event++; /* Look up the ELS command IOCB and create pseudo response IOCB */ cmdiocbq = lpfc_sli_iocbq_lookup_by_tag(phba, pring, bf_get(lpfc_wcqe_c_request_tag, wcqe)); - spin_unlock_irqrestore(&phba->hbalock, iflags); + spin_unlock_irqrestore(&pring->ring_lock, iflags); if (unlikely(!cmdiocbq)) { lpfc_printf_log(phba, KERN_WARNING, LOG_SLI, @@ -11154,6 +11223,7 @@ lpfc_sli4_sp_handle_mcqe(struct lpfc_hba *phba, struct lpfc_cqe *cqe) /** * lpfc_sli4_sp_handle_els_wcqe - Handle els work-queue completion event * @phba: Pointer to HBA context object. + * @cq: Pointer to associated CQ * @wcqe: Pointer to work-queue completion queue entry. * * This routine handles an ELS work-queue completion event. @@ -11161,12 +11231,12 @@ lpfc_sli4_sp_handle_mcqe(struct lpfc_hba *phba, struct lpfc_cqe *cqe) * Return: true if work posted to worker thread, otherwise false. **/ static bool -lpfc_sli4_sp_handle_els_wcqe(struct lpfc_hba *phba, +lpfc_sli4_sp_handle_els_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq, struct lpfc_wcqe_complete *wcqe) { struct lpfc_iocbq *irspiocbq; unsigned long iflags; - struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_FCP_RING]; + struct lpfc_sli_ring *pring = cq->pring; /* Get an irspiocbq for later ELS response processing use */ irspiocbq = lpfc_sli_get_iocbq(phba); @@ -11311,14 +11381,17 @@ lpfc_sli4_sp_handle_rcqe(struct lpfc_hba *phba, struct lpfc_rcqe *rcqe) case FC_STATUS_RQ_BUF_LEN_EXCEEDED: lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "2537 Receive Frame Truncated!!\n"); + hrq->RQ_buf_trunc++; case FC_STATUS_RQ_SUCCESS: lpfc_sli4_rq_release(hrq, drq); spin_lock_irqsave(&phba->hbalock, iflags); dma_buf = lpfc_sli_hbqbuf_get(&phba->hbqs[0].hbq_buffer_list); if (!dma_buf) { + hrq->RQ_no_buf_found++; spin_unlock_irqrestore(&phba->hbalock, iflags); goto out; } + hrq->RQ_rcv_buf++; memcpy(&dma_buf->cq_event.cqe.rcqe_cmpl, rcqe, sizeof(*rcqe)); /* save off the frame for the word thread to process */ list_add_tail(&dma_buf->cq_event.list, @@ -11330,6 +11403,7 @@ lpfc_sli4_sp_handle_rcqe(struct lpfc_hba *phba, struct lpfc_rcqe *rcqe) break; case FC_STATUS_INSUFF_BUF_NEED_BUF: case FC_STATUS_INSUFF_BUF_FRM_DISC: + hrq->RQ_no_posted_buf++; /* Post more buffers if possible */ spin_lock_irqsave(&phba->hbalock, iflags); phba->hba_flag |= HBA_POST_RECEIVE_BUFFER; @@ -11367,7 +11441,7 @@ lpfc_sli4_sp_handle_cqe(struct lpfc_hba *phba, struct lpfc_queue *cq, case CQE_CODE_COMPL_WQE: /* Process the WQ/RQ complete event */ phba->last_completion_time = jiffies; - workposted = lpfc_sli4_sp_handle_els_wcqe(phba, + workposted = lpfc_sli4_sp_handle_els_wcqe(phba, cq, (struct lpfc_wcqe_complete *)&cqevt); break; case CQE_CODE_RELEASE_WQE: @@ -11411,31 +11485,18 @@ lpfc_sli4_sp_handle_cqe(struct lpfc_hba *phba, struct lpfc_queue *cq, * **/ static void -lpfc_sli4_sp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe) +lpfc_sli4_sp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe, + struct lpfc_queue *speq) { - struct lpfc_queue *cq = NULL, *childq, *speq; + struct lpfc_queue *cq = NULL, *childq; struct lpfc_cqe *cqe; bool workposted = false; int ecount = 0; uint16_t cqid; - if (bf_get_le32(lpfc_eqe_major_code, eqe) != 0) { - lpfc_printf_log(phba, KERN_ERR, LOG_SLI, - "0359 Not a valid slow-path completion " - "event: majorcode=x%x, minorcode=x%x\n", - bf_get_le32(lpfc_eqe_major_code, eqe), - bf_get_le32(lpfc_eqe_minor_code, eqe)); - return; - } - /* Get the reference to the corresponding CQ */ cqid = bf_get_le32(lpfc_eqe_resource_id, eqe); - /* Search for completion queue pointer matching this cqid */ - speq = phba->sli4_hba.sp_eq; - /* sanity check on queue memory */ - if (unlikely(!speq)) - return; list_for_each_entry(childq, &speq->child_list, list) { if (childq->queue_id == cqid) { cq = childq; @@ -11457,6 +11518,7 @@ lpfc_sli4_sp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe) workposted |= lpfc_sli4_sp_handle_mcqe(phba, cqe); if (!(++ecount % cq->entry_repost)) lpfc_sli4_cq_release(cq, LPFC_QUEUE_NOARM); + cq->CQ_mbox++; } break; case LPFC_WCQ: @@ -11470,6 +11532,10 @@ lpfc_sli4_sp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe) if (!(++ecount % cq->entry_repost)) lpfc_sli4_cq_release(cq, LPFC_QUEUE_NOARM); } + + /* Track the max number of CQEs processed in 1 EQ */ + if (ecount > cq->CQ_max_cqe) + cq->CQ_max_cqe = ecount; break; default: lpfc_printf_log(phba, KERN_ERR, LOG_SLI, @@ -11494,34 +11560,33 @@ lpfc_sli4_sp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe) /** * lpfc_sli4_fp_handle_fcp_wcqe - Process fast-path work queue completion entry - * @eqe: Pointer to fast-path completion queue entry. + * @phba: Pointer to HBA context object. + * @cq: Pointer to associated CQ + * @wcqe: Pointer to work-queue completion queue entry. * * This routine process a fast-path work queue completion entry from fast-path * event queue for FCP command response completion. **/ static void -lpfc_sli4_fp_handle_fcp_wcqe(struct lpfc_hba *phba, +lpfc_sli4_fp_handle_fcp_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq, struct lpfc_wcqe_complete *wcqe) { - struct lpfc_sli_ring *pring = &phba->sli.ring[LPFC_FCP_RING]; + struct lpfc_sli_ring *pring = cq->pring; struct lpfc_iocbq *cmdiocbq; struct lpfc_iocbq irspiocbq; unsigned long iflags; - spin_lock_irqsave(&phba->hbalock, iflags); - pring->stats.iocb_event++; - spin_unlock_irqrestore(&phba->hbalock, iflags); - /* Check for response status */ if (unlikely(bf_get(lpfc_wcqe_c_status, wcqe))) { /* If resource errors reported from HBA, reduce queue * depth of the SCSI device. */ - if ((bf_get(lpfc_wcqe_c_status, wcqe) == - IOSTAT_LOCAL_REJECT) && - (wcqe->parameter == IOERR_NO_RESOURCES)) { + if (((bf_get(lpfc_wcqe_c_status, wcqe) == + IOSTAT_LOCAL_REJECT)) && + ((wcqe->parameter & IOERR_PARAM_MASK) == + IOERR_NO_RESOURCES)) phba->lpfc_rampdown_queue_depth(phba); - } + /* Log the error status */ lpfc_printf_log(phba, KERN_WARNING, LOG_SLI, "0373 FCP complete error: status=x%x, " @@ -11534,10 +11599,11 @@ lpfc_sli4_fp_handle_fcp_wcqe(struct lpfc_hba *phba, } /* Look up the FCP command IOCB and create pseudo response IOCB */ - spin_lock_irqsave(&phba->hbalock, iflags); + spin_lock_irqsave(&pring->ring_lock, iflags); + pring->stats.iocb_event++; cmdiocbq = lpfc_sli_iocbq_lookup_by_tag(phba, pring, bf_get(lpfc_wcqe_c_request_tag, wcqe)); - spin_unlock_irqrestore(&phba->hbalock, iflags); + spin_unlock_irqrestore(&pring->ring_lock, iflags); if (unlikely(!cmdiocbq)) { lpfc_printf_log(phba, KERN_WARNING, LOG_SLI, "0374 FCP complete with no corresponding " @@ -11621,17 +11687,20 @@ lpfc_sli4_fp_handle_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq, /* Check and process for different type of WCQE and dispatch */ switch (bf_get(lpfc_wcqe_c_code, &wcqe)) { case CQE_CODE_COMPL_WQE: + cq->CQ_wq++; /* Process the WQ complete event */ phba->last_completion_time = jiffies; - lpfc_sli4_fp_handle_fcp_wcqe(phba, + lpfc_sli4_fp_handle_fcp_wcqe(phba, cq, (struct lpfc_wcqe_complete *)&wcqe); break; case CQE_CODE_RELEASE_WQE: + cq->CQ_release_wqe++; /* Process the WQ release event */ lpfc_sli4_fp_handle_rel_wcqe(phba, cq, (struct lpfc_wcqe_release *)&wcqe); break; case CQE_CODE_XRI_ABORTED: + cq->CQ_xri_aborted++; /* Process the WQ XRI abort event */ phba->last_completion_time = jiffies; workposted = lpfc_sli4_sp_handle_abort_xri_wcqe(phba, cq, @@ -11647,7 +11716,7 @@ lpfc_sli4_fp_handle_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq, } /** - * lpfc_sli4_fp_handle_eqe - Process a fast-path event queue entry + * lpfc_sli4_hba_handle_eqe - Process a fast-path event queue entry * @phba: Pointer to HBA context object. * @eqe: Pointer to fast-path event queue entry. * @@ -11659,8 +11728,8 @@ lpfc_sli4_fp_handle_wcqe(struct lpfc_hba *phba, struct lpfc_queue *cq, * completion queue, and then return. **/ static void -lpfc_sli4_fp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe, - uint32_t fcp_cqidx) +lpfc_sli4_hba_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe, + uint32_t qidx) { struct lpfc_queue *cq; struct lpfc_cqe *cqe; @@ -11670,30 +11739,38 @@ lpfc_sli4_fp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe, if (unlikely(bf_get_le32(lpfc_eqe_major_code, eqe) != 0)) { lpfc_printf_log(phba, KERN_ERR, LOG_SLI, - "0366 Not a valid fast-path completion " + "0366 Not a valid completion " "event: majorcode=x%x, minorcode=x%x\n", bf_get_le32(lpfc_eqe_major_code, eqe), bf_get_le32(lpfc_eqe_minor_code, eqe)); return; } + /* Get the reference to the corresponding CQ */ + cqid = bf_get_le32(lpfc_eqe_resource_id, eqe); + + /* Check if this is a Slow path event */ + if (unlikely(cqid != phba->sli4_hba.fcp_cq_map[qidx])) { + lpfc_sli4_sp_handle_eqe(phba, eqe, + phba->sli4_hba.hba_eq[qidx]); + return; + } + if (unlikely(!phba->sli4_hba.fcp_cq)) { lpfc_printf_log(phba, KERN_WARNING, LOG_SLI, "3146 Fast-path completion queues " "does not exist\n"); return; } - cq = phba->sli4_hba.fcp_cq[fcp_cqidx]; + cq = phba->sli4_hba.fcp_cq[qidx]; if (unlikely(!cq)) { if (phba->sli.sli_flag & LPFC_SLI_ACTIVE) lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "0367 Fast-path completion queue " - "(%d) does not exist\n", fcp_cqidx); + "(%d) does not exist\n", qidx); return; } - /* Get the reference to the corresponding CQ */ - cqid = bf_get_le32(lpfc_eqe_resource_id, eqe); if (unlikely(cqid != cq->queue_id)) { lpfc_printf_log(phba, KERN_ERR, LOG_SLI, "0368 Miss-matched fast-path completion " @@ -11709,6 +11786,10 @@ lpfc_sli4_fp_handle_eqe(struct lpfc_hba *phba, struct lpfc_eqe *eqe, lpfc_sli4_cq_release(cq, LPFC_QUEUE_NOARM); } + /* Track the max number of CQEs processed in 1 EQ */ + if (ecount > cq->CQ_max_cqe) + cq->CQ_max_cqe = ecount; + /* Catch the no cq entry condition */ if (unlikely(ecount == 0)) lpfc_printf_log(phba, KERN_ERR, LOG_SLI, @@ -11737,86 +11818,7 @@ lpfc_sli4_eq_flush(struct lpfc_hba *phba, struct lpfc_queue *eq) } /** - * lpfc_sli4_sp_intr_handler - Slow-path interrupt handler to SLI-4 device - * @irq: Interrupt number. - * @dev_id: The device context pointer. - * - * This function is directly called from the PCI layer as an interrupt - * service routine when device with SLI-4 interface spec is enabled with - * MSI-X multi-message interrupt mode and there are slow-path events in - * the HBA. However, when the device is enabled with either MSI or Pin-IRQ - * interrupt mode, this function is called as part of the device-level - * interrupt handler. When the PCI slot is in error recovery or the HBA is - * undergoing initialization, the interrupt handler will not process the - * interrupt. The link attention and ELS ring attention events are handled - * by the worker thread. The interrupt handler signals the worker thread - * and returns for these events. This function is called without any lock - * held. It gets the hbalock to access and update SLI data structures. - * - * This function returns IRQ_HANDLED when interrupt is handled else it - * returns IRQ_NONE. - **/ -irqreturn_t -lpfc_sli4_sp_intr_handler(int irq, void *dev_id) -{ - struct lpfc_hba *phba; - struct lpfc_queue *speq; - struct lpfc_eqe *eqe; - unsigned long iflag; - int ecount = 0; - - /* - * Get the driver's phba structure from the dev_id - */ - phba = (struct lpfc_hba *)dev_id; - - if (unlikely(!phba)) - return IRQ_NONE; - - /* Get to the EQ struct associated with this vector */ - speq = phba->sli4_hba.sp_eq; - if (unlikely(!speq)) - return IRQ_NONE; - - /* Check device state for handling interrupt */ - if (unlikely(lpfc_intr_state_check(phba))) { - /* Check again for link_state with lock held */ - spin_lock_irqsave(&phba->hbalock, iflag); - if (phba->link_state < LPFC_LINK_DOWN) - /* Flush, clear interrupt, and rearm the EQ */ - lpfc_sli4_eq_flush(phba, speq); - spin_unlock_irqrestore(&phba->hbalock, iflag); - return IRQ_NONE; - } - - /* - * Process all the event on FCP slow-path EQ - */ - while ((eqe = lpfc_sli4_eq_get(speq))) { - lpfc_sli4_sp_handle_eqe(phba, eqe); - if (!(++ecount % speq->entry_repost)) - lpfc_sli4_eq_release(speq, LPFC_QUEUE_NOARM); - } - - /* Always clear and re-arm the slow-path EQ */ - lpfc_sli4_eq_release(speq, LPFC_QUEUE_REARM); - - /* Catch the no cq entry condition */ - if (unlikely(ecount == 0)) { - if (phba->intr_type == MSIX) - /* MSI-X treated interrupt served as no EQ share INT */ - lpfc_printf_log(phba, KERN_WARNING, LOG_SLI, - "0357 MSI-X interrupt with no EQE\n"); - else - /* Non MSI-X treated on interrupt as EQ share INT */ - return IRQ_NONE; - } - - return IRQ_HANDLED; -} /* lpfc_sli4_sp_intr_handler */ - -/** - * lpfc_sli4_fp_intr_handler - Fast-path interrupt handler to SLI-4 device + * lpfc_sli4_hba_intr_handler - HBA interrupt handler to SLI-4 device * @irq: Interrupt number. * @dev_id: The device context pointer. * @@ -11833,11 +11835,16 @@ lpfc_sli4_sp_intr_handler(int irq, void *dev_id) * the FCP EQ to FCP CQ are one-to-one map such that the FCP EQ index is * equal to that of FCP CQ index. * + * The link attention and ELS ring attention events are handled + * by the worker thread. The interrupt handler signals the worker thread + * and returns for these events. This function is called without any lock + * held. It gets the hbalock to access and update SLI data structures. + * * This function returns IRQ_HANDLED when interrupt is handled else it * returns IRQ_NONE. **/ irqreturn_t -lpfc_sli4_fp_intr_handler(int irq, void *dev_id) +lpfc_sli4_hba_intr_handler(int irq, void *dev_id) { struct lpfc_hba *phba; struct lpfc_fcp_eq_hdl *fcp_eq_hdl; @@ -11854,22 +11861,34 @@ lpfc_sli4_fp_intr_handler(int irq, void *dev_id) if (unlikely(!phba)) return IRQ_NONE; - if (unlikely(!phba->sli4_hba.fp_eq)) + if (unlikely(!phba->sli4_hba.hba_eq)) return IRQ_NONE; /* Get to the EQ struct associated with this vector */ - fpeq = phba->sli4_hba.fp_eq[fcp_eqidx]; + fpeq = phba->sli4_hba.hba_eq[fcp_eqidx]; if (unlikely(!fpeq)) return IRQ_NONE; + if (lpfc_fcp_look_ahead) { + if (atomic_dec_and_test(&fcp_eq_hdl->fcp_eq_in_use)) + lpfc_sli4_eq_clr_intr(fpeq); + else { + atomic_inc(&fcp_eq_hdl->fcp_eq_in_use); + return IRQ_NONE; + } + } + /* Check device state for handling interrupt */ if (unlikely(lpfc_intr_state_check(phba))) { + fpeq->EQ_badstate++; /* Check again for link_state with lock held */ spin_lock_irqsave(&phba->hbalock, iflag); if (phba->link_state < LPFC_LINK_DOWN) /* Flush, clear interrupt, and rearm the EQ */ lpfc_sli4_eq_flush(phba, fpeq); spin_unlock_irqrestore(&phba->hbalock, iflag); + if (lpfc_fcp_look_ahead) + atomic_inc(&fcp_eq_hdl->fcp_eq_in_use); return IRQ_NONE; } @@ -11877,15 +11896,27 @@ lpfc_sli4_fp_intr_handler(int irq, void *dev_id) * Process all the event on FCP fast-path EQ */ while ((eqe = lpfc_sli4_eq_get(fpeq))) { - lpfc_sli4_fp_handle_eqe(phba, eqe, fcp_eqidx); + lpfc_sli4_hba_handle_eqe(phba, eqe, fcp_eqidx); if (!(++ecount % fpeq->entry_repost)) lpfc_sli4_eq_release(fpeq, LPFC_QUEUE_NOARM); + fpeq->EQ_processed++; } + /* Track the max number of EQEs processed in 1 intr */ + if (ecount > fpeq->EQ_max_eqe) + fpeq->EQ_max_eqe = ecount; + /* Always clear and re-arm the fast-path EQ */ lpfc_sli4_eq_release(fpeq, LPFC_QUEUE_REARM); if (unlikely(ecount == 0)) { + fpeq->EQ_no_entry++; + + if (lpfc_fcp_look_ahead) { + atomic_inc(&fcp_eq_hdl->fcp_eq_in_use); + return IRQ_NONE; + } + if (phba->intr_type == MSIX) /* MSI-X treated interrupt served as no EQ share INT */ lpfc_printf_log(phba, KERN_WARNING, LOG_SLI, @@ -11895,6 +11926,8 @@ lpfc_sli4_fp_intr_handler(int irq, void *dev_id) return IRQ_NONE; } + if (lpfc_fcp_look_ahead) + atomic_inc(&fcp_eq_hdl->fcp_eq_in_use); return IRQ_HANDLED; } /* lpfc_sli4_fp_intr_handler */ @@ -11919,8 +11952,8 @@ irqreturn_t lpfc_sli4_intr_handler(int irq, void *dev_id) { struct lpfc_hba *phba; - irqreturn_t sp_irq_rc, fp_irq_rc; - bool fp_handled = false; + irqreturn_t hba_irq_rc; + bool hba_handled = false; uint32_t fcp_eqidx; /* Get the driver's phba structure from the dev_id */ @@ -11930,21 +11963,16 @@ lpfc_sli4_intr_handler(int irq, void *dev_id) return IRQ_NONE; /* - * Invokes slow-path host attention interrupt handling as appropriate. - */ - sp_irq_rc = lpfc_sli4_sp_intr_handler(irq, dev_id); - - /* * Invoke fast-path host attention interrupt handling as appropriate. */ - for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_eq_count; fcp_eqidx++) { - fp_irq_rc = lpfc_sli4_fp_intr_handler(irq, + for (fcp_eqidx = 0; fcp_eqidx < phba->cfg_fcp_io_channel; fcp_eqidx++) { + hba_irq_rc = lpfc_sli4_hba_intr_handler(irq, &phba->sli4_hba.fcp_eq_hdl[fcp_eqidx]); - if (fp_irq_rc == IRQ_HANDLED) - fp_handled |= true; + if (hba_irq_rc == IRQ_HANDLED) + hba_handled |= true; } - return (fp_handled == true) ? IRQ_HANDLED : sp_irq_rc; + return (hba_handled == true) ? IRQ_HANDLED : IRQ_NONE; } /* lpfc_sli4_intr_handler */ /** @@ -12075,7 +12103,7 @@ lpfc_modify_fcp_eq_delay(struct lpfc_hba *phba, uint16_t startq) union lpfc_sli4_cfg_shdr *shdr; uint16_t dmult; - if (startq >= phba->cfg_fcp_eq_count) + if (startq >= phba->cfg_fcp_io_channel) return 0; mbox = mempool_alloc(phba->mbox_mem_pool, GFP_KERNEL); @@ -12089,12 +12117,13 @@ lpfc_modify_fcp_eq_delay(struct lpfc_hba *phba, uint16_t startq) eq_delay = &mbox->u.mqe.un.eq_delay; /* Calculate delay multiper from maximum interrupt per second */ - dmult = LPFC_DMULT_CONST/phba->cfg_fcp_imax - 1; + dmult = phba->cfg_fcp_imax / phba->cfg_fcp_io_channel; + dmult = LPFC_DMULT_CONST/dmult - 1; cnt = 0; - for (fcp_eqidx = startq; fcp_eqidx < phba->cfg_fcp_eq_count; + for (fcp_eqidx = startq; fcp_eqidx < phba->cfg_fcp_io_channel; fcp_eqidx++) { - eq = phba->sli4_hba.fp_eq[fcp_eqidx]; + eq = phba->sli4_hba.hba_eq[fcp_eqidx]; if (!eq) continue; eq_delay->u.request.eq[cnt].eq_id = eq->queue_id; diff --git a/drivers/scsi/lpfc/lpfc_sli.h b/drivers/scsi/lpfc/lpfc_sli.h index 2626f58c..2f48d00 100644 --- a/drivers/scsi/lpfc/lpfc_sli.h +++ b/drivers/scsi/lpfc/lpfc_sli.h @@ -131,7 +131,9 @@ typedef struct lpfcMboxq { #define LPFC_MAX_RING_MASK 5 /* max num of rctl/type masks allowed per ring */ -#define LPFC_MAX_RING 4 /* max num of SLI rings used by driver */ +#define LPFC_SLI3_MAX_RING 4 /* Max num of SLI3 rings used by driver. + For SLI4, an additional ring for each + FCP WQ will be allocated. */ struct lpfc_sli_ring; @@ -158,6 +160,24 @@ struct lpfc_sli_ring_stat { uint64_t iocb_rsp_full; /* IOCB rsp ring full */ }; +struct lpfc_sli3_ring { + uint32_t local_getidx; /* last available cmd index (from cmdGetInx) */ + uint32_t next_cmdidx; /* next_cmd index */ + uint32_t rspidx; /* current index in response ring */ + uint32_t cmdidx; /* current index in command ring */ + uint16_t numCiocb; /* number of command iocb's per ring */ + uint16_t numRiocb; /* number of rsp iocb's per ring */ + uint16_t sizeCiocb; /* Size of command iocb's in this ring */ + uint16_t sizeRiocb; /* Size of response iocb's in this ring */ + uint32_t *cmdringaddr; /* virtual address for cmd rings */ + uint32_t *rspringaddr; /* virtual address for rsp rings */ +}; + +struct lpfc_sli4_ring { + struct lpfc_queue *wqp; /* Pointer to associated WQ */ +}; + + /* Structure used to hold SLI ring information */ struct lpfc_sli_ring { uint16_t flag; /* ring flags */ @@ -166,16 +186,10 @@ struct lpfc_sli_ring { #define LPFC_STOP_IOCB_EVENT 0x020 /* Stop processing IOCB cmds event */ uint16_t abtsiotag; /* tracks next iotag to use for ABTS */ - uint32_t local_getidx; /* last available cmd index (from cmdGetInx) */ - uint32_t next_cmdidx; /* next_cmd index */ - uint32_t rspidx; /* current index in response ring */ - uint32_t cmdidx; /* current index in command ring */ uint8_t rsvd; uint8_t ringno; /* ring number */ - uint16_t numCiocb; /* number of command iocb's per ring */ - uint16_t numRiocb; /* number of rsp iocb's per ring */ - uint16_t sizeCiocb; /* Size of command iocb's in this ring */ - uint16_t sizeRiocb; /* Size of response iocb's in this ring */ + + spinlock_t ring_lock; /* lock for issuing commands */ uint32_t fast_iotag; /* max fastlookup based iotag */ uint32_t iotag_ctr; /* keeps track of the next iotag to use */ @@ -186,8 +200,6 @@ struct lpfc_sli_ring { struct list_head txcmplq; uint16_t txcmplq_cnt; /* current length of queue */ uint16_t txcmplq_max; /* max length */ - uint32_t *cmdringaddr; /* virtual address for cmd rings */ - uint32_t *rspringaddr; /* virtual address for rsp rings */ uint32_t missbufcnt; /* keep track of buffers to post */ struct list_head postbufq; uint16_t postbufq_cnt; /* current length of queue */ @@ -207,6 +219,10 @@ struct lpfc_sli_ring { /* cmd ring available */ void (*lpfc_sli_cmd_available) (struct lpfc_hba *, struct lpfc_sli_ring *); + union { + struct lpfc_sli3_ring sli3; + struct lpfc_sli4_ring sli4; + } sli; }; /* Structure used for configuring rings to a specific profile or rctl / type */ @@ -239,6 +255,8 @@ struct lpfc_sli_stat { uint64_t mbox_stat_err; /* Mbox cmds completed status error */ uint64_t mbox_cmd; /* Mailbox commands issued */ uint64_t sli_intr; /* Count of Host Attention interrupts */ + uint64_t sli_prev_intr; /* Previous cnt of Host Attention interrupts */ + uint64_t sli_ips; /* Host Attention interrupts per sec */ uint32_t err_attn_event; /* Error Attn event counters */ uint32_t link_event; /* Link event counters */ uint32_t mbox_event; /* Mailbox event counters */ @@ -270,7 +288,7 @@ struct lpfc_sli { #define LPFC_MENLO_MAINT 0x1000 /* need for menl fw download */ #define LPFC_SLI_ASYNC_MBX_BLK 0x2000 /* Async mailbox is blocked */ - struct lpfc_sli_ring ring[LPFC_MAX_RING]; + struct lpfc_sli_ring *ring; int fcp_ring; /* ring used for FCP initiator commands */ int next_ring; diff --git a/drivers/scsi/lpfc/lpfc_sli4.h b/drivers/scsi/lpfc/lpfc_sli4.h index ec75611..bd4bc43 100644 --- a/drivers/scsi/lpfc/lpfc_sli4.h +++ b/drivers/scsi/lpfc/lpfc_sli4.h @@ -34,18 +34,10 @@ /* Number of SGL entries can be posted in a 4KB nonembedded mbox command */ #define LPFC_NEMBED_MBOX_SGL_CNT 254 -/* Multi-queue arrangement for fast-path FCP work queues */ -#define LPFC_FN_EQN_MAX 8 -#define LPFC_SP_EQN_DEF 1 -#define LPFC_FP_EQN_DEF 4 -#define LPFC_FP_EQN_MIN 1 -#define LPFC_FP_EQN_MAX (LPFC_FN_EQN_MAX - LPFC_SP_EQN_DEF) - -#define LPFC_FN_WQN_MAX 32 -#define LPFC_SP_WQN_DEF 1 -#define LPFC_FP_WQN_DEF 4 -#define LPFC_FP_WQN_MIN 1 -#define LPFC_FP_WQN_MAX (LPFC_FN_WQN_MAX - LPFC_SP_WQN_DEF) +/* Multi-queue arrangement for FCP EQ/CQ/WQ tuples */ +#define LPFC_FCP_IO_CHAN_DEF 4 +#define LPFC_FCP_IO_CHAN_MIN 1 +#define LPFC_FCP_IO_CHAN_MAX 8 /* * Provide the default FCF Record attributes used by the driver @@ -141,6 +133,37 @@ struct lpfc_queue { uint32_t page_count; /* Number of pages allocated for this queue */ uint32_t host_index; /* The host's index for putting or getting */ uint32_t hba_index; /* The last known hba index for get or put */ + + struct lpfc_sli_ring *pring; /* ptr to io ring associated with q */ + + /* For q stats */ + uint32_t q_cnt_1; + uint32_t q_cnt_2; + uint32_t q_cnt_3; + uint64_t q_cnt_4; +/* defines for EQ stats */ +#define EQ_max_eqe q_cnt_1 +#define EQ_no_entry q_cnt_2 +#define EQ_badstate q_cnt_3 +#define EQ_processed q_cnt_4 + +/* defines for CQ stats */ +#define CQ_mbox q_cnt_1 +#define CQ_max_cqe q_cnt_1 +#define CQ_release_wqe q_cnt_2 +#define CQ_xri_aborted q_cnt_3 +#define CQ_wq q_cnt_4 + +/* defines for WQ stats */ +#define WQ_overflow q_cnt_1 +#define WQ_posted q_cnt_4 + +/* defines for RQ stats */ +#define RQ_no_posted_buf q_cnt_1 +#define RQ_no_buf_found q_cnt_2 +#define RQ_buf_trunc q_cnt_3 +#define RQ_rcv_buf q_cnt_4 + union sli4_qe qe[1]; /* array to index entries (must be last) */ }; @@ -350,6 +373,7 @@ struct lpfc_hba; struct lpfc_fcp_eq_hdl { uint32_t idx; struct lpfc_hba *phba; + atomic_t fcp_eq_in_use; }; /* Port Capabilities for SLI4 Parameters */ @@ -407,6 +431,8 @@ struct lpfc_sli4_lnk_info { uint8_t lnk_no; }; +#define LPFC_SLI4_HANDLER_NAME_SZ 16 + /* SLI4 HBA data structure entries */ struct lpfc_sli4_hba { void __iomem *conf_regs_memmap_p; /* Kernel memory mapped address for @@ -463,20 +489,23 @@ struct lpfc_sli4_hba { struct lpfc_register sli_intf; struct lpfc_pc_sli4_params pc_sli4_params; struct msix_entry *msix_entries; + uint8_t handler_name[LPFC_FCP_IO_CHAN_MAX][LPFC_SLI4_HANDLER_NAME_SZ]; uint32_t cfg_eqn; uint32_t msix_vec_nr; struct lpfc_fcp_eq_hdl *fcp_eq_hdl; /* FCP per-WQ handle */ + /* Pointers to the constructed SLI4 queues */ - struct lpfc_queue **fp_eq; /* Fast-path event queue */ - struct lpfc_queue *sp_eq; /* Slow-path event queue */ + struct lpfc_queue **hba_eq;/* Event queues for HBA */ + struct lpfc_queue **fcp_cq;/* Fast-path FCP compl queue */ struct lpfc_queue **fcp_wq;/* Fast-path FCP work queue */ + uint16_t *fcp_cq_map; + + struct lpfc_queue *mbx_cq; /* Slow-path mailbox complete queue */ + struct lpfc_queue *els_cq; /* Slow-path ELS response complete queue */ struct lpfc_queue *mbx_wq; /* Slow-path MBOX work queue */ struct lpfc_queue *els_wq; /* Slow-path ELS work queue */ struct lpfc_queue *hdr_rq; /* Slow-path Header Receive queue */ struct lpfc_queue *dat_rq; /* Slow-path Data Receive queue */ - struct lpfc_queue **fcp_cq;/* Fast-path FCP compl queue */ - struct lpfc_queue *mbx_cq; /* Slow-path mailbox complete queue */ - struct lpfc_queue *els_cq; /* Slow-path ELS response complete queue */ /* Setup information for various queue parameters */ int eq_esize; diff --git a/drivers/scsi/lpfc/lpfc_version.h b/drivers/scsi/lpfc/lpfc_version.h index 4704e5b..04265a1 100644 --- a/drivers/scsi/lpfc/lpfc_version.h +++ b/drivers/scsi/lpfc/lpfc_version.h @@ -18,11 +18,16 @@ * included with this package. * *******************************************************************/ -#define LPFC_DRIVER_VERSION "8.3.32" +#define LPFC_DRIVER_VERSION "8.3.34" #define LPFC_DRIVER_NAME "lpfc" + +/* Used for SLI 2/3 */ #define LPFC_SP_DRIVER_HANDLER_NAME "lpfc:sp" #define LPFC_FP_DRIVER_HANDLER_NAME "lpfc:fp" +/* Used for SLI4 */ +#define LPFC_DRIVER_HANDLER_NAME "lpfc:" + #define LPFC_MODULE_DESC "Emulex LightPulse Fibre Channel SCSI driver " \ LPFC_DRIVER_VERSION #define LPFC_COPYRIGHT "Copyright(c) 2004-2009 Emulex. All rights reserved." diff --git a/drivers/scsi/megaraid/megaraid_sas.h b/drivers/scsi/megaraid/megaraid_sas.h index e8f8926..fcb005f 100644 --- a/drivers/scsi/megaraid/megaraid_sas.h +++ b/drivers/scsi/megaraid/megaraid_sas.h @@ -33,9 +33,9 @@ /* * MegaRAID SAS Driver meta data */ -#define MEGASAS_VERSION "00.00.06.15-rc1" -#define MEGASAS_RELDATE "Mar. 19, 2012" -#define MEGASAS_EXT_VERSION "Mon. Mar. 19 17:00:00 PDT 2012" +#define MEGASAS_VERSION "00.00.06.18-rc1" +#define MEGASAS_RELDATE "Jun. 17, 2012" +#define MEGASAS_EXT_VERSION "Tue. Jun. 17 17:00:00 PDT 2012" /* * Device IDs @@ -747,6 +747,7 @@ struct megasas_ctrl_info { #define MEGASAS_RESET_NOTICE_INTERVAL 5 #define MEGASAS_IOCTL_CMD 0 #define MEGASAS_DEFAULT_CMD_TIMEOUT 90 +#define MEGASAS_THROTTLE_QUEUE_DEPTH 16 /* * FW reports the maximum of number of commands that it can accept (maximum @@ -1364,6 +1365,7 @@ struct megasas_instance { unsigned long bar; long reset_flags; struct mutex reset_mutex; + int throttlequeuedepth; }; enum { diff --git a/drivers/scsi/megaraid/megaraid_sas_base.c b/drivers/scsi/megaraid/megaraid_sas_base.c index ed38454..0393ec4 100644 --- a/drivers/scsi/megaraid/megaraid_sas_base.c +++ b/drivers/scsi/megaraid/megaraid_sas_base.c @@ -18,7 +18,7 @@ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA * * FILE: megaraid_sas_base.c - * Version : v00.00.06.15-rc1 + * Version : v00.00.06.18-rc1 * * Authors: LSI Corporation * Sreenivas Bagalkote @@ -71,6 +71,16 @@ static int msix_disable; module_param(msix_disable, int, S_IRUGO); MODULE_PARM_DESC(msix_disable, "Disable MSI-X interrupt handling. Default: 0"); +static int throttlequeuedepth = MEGASAS_THROTTLE_QUEUE_DEPTH; +module_param(throttlequeuedepth, int, S_IRUGO); +MODULE_PARM_DESC(throttlequeuedepth, + "Adapter queue depth when throttled due to I/O timeout. Default: 16"); + +int resetwaittime = MEGASAS_RESET_WAIT_TIME; +module_param(resetwaittime, int, S_IRUGO); +MODULE_PARM_DESC(resetwaittime, "Wait time in seconds after I/O timeout " + "before resetting adapter. Default: 180"); + MODULE_LICENSE("GPL"); MODULE_VERSION(MEGASAS_VERSION); MODULE_AUTHOR("megaraidlinux@lsi.com"); @@ -1595,8 +1605,9 @@ megasas_check_and_restore_queue_depth(struct megasas_instance *instance) { unsigned long flags; if (instance->flag & MEGASAS_FW_BUSY - && time_after(jiffies, instance->last_time + 5 * HZ) - && atomic_read(&instance->fw_outstanding) < 17) { + && time_after(jiffies, instance->last_time + 5 * HZ) + && atomic_read(&instance->fw_outstanding) < + instance->throttlequeuedepth + 1) { spin_lock_irqsave(instance->host->host_lock, flags); instance->flag &= ~MEGASAS_FW_BUSY; @@ -1772,7 +1783,7 @@ static int megasas_wait_for_outstanding(struct megasas_instance *instance) return SUCCESS; } - for (i = 0; i < wait_time; i++) { + for (i = 0; i < resetwaittime; i++) { int outstanding = atomic_read(&instance->fw_outstanding); @@ -1914,7 +1925,7 @@ blk_eh_timer_return megasas_reset_timer(struct scsi_cmnd *scmd) /* FW is busy, throttle IO */ spin_lock_irqsave(instance->host->host_lock, flags); - instance->host->can_queue = 16; + instance->host->can_queue = instance->throttlequeuedepth; instance->last_time = jiffies; instance->flag |= MEGASAS_FW_BUSY; @@ -3577,6 +3588,24 @@ static int megasas_init_fw(struct megasas_instance *instance) kfree(ctrl_info); + /* Check for valid throttlequeuedepth module parameter */ + if (instance->pdev->device == PCI_DEVICE_ID_LSI_SAS0073SKINNY || + instance->pdev->device == PCI_DEVICE_ID_LSI_SAS0071SKINNY) { + if (throttlequeuedepth > (instance->max_fw_cmds - + MEGASAS_SKINNY_INT_CMDS)) + instance->throttlequeuedepth = + MEGASAS_THROTTLE_QUEUE_DEPTH; + else + instance->throttlequeuedepth = throttlequeuedepth; + } else { + if (throttlequeuedepth > (instance->max_fw_cmds - + MEGASAS_INT_CMDS)) + instance->throttlequeuedepth = + MEGASAS_THROTTLE_QUEUE_DEPTH; + else + instance->throttlequeuedepth = throttlequeuedepth; + } + /* * Setup tasklet for cmd completion */ diff --git a/drivers/scsi/megaraid/megaraid_sas_fusion.c b/drivers/scsi/megaraid/megaraid_sas_fusion.c index a610cf1..ddf094e 100644 --- a/drivers/scsi/megaraid/megaraid_sas_fusion.c +++ b/drivers/scsi/megaraid/megaraid_sas_fusion.c @@ -94,6 +94,7 @@ int megasas_transition_to_ready(struct megasas_instance *instance, int ocr); void megaraid_sas_kill_hba(struct megasas_instance *instance); extern u32 megasas_dbg_lvl; +extern int resetwaittime; /** * megasas_enable_intr_fusion - Enables interrupts @@ -461,8 +462,8 @@ megasas_alloc_cmds_fusion(struct megasas_instance *instance) * Allocate the dynamic array first and then allocate individual * commands. */ - fusion->cmd_list = kmalloc(sizeof(struct megasas_cmd_fusion *) - *max_cmd, GFP_KERNEL); + fusion->cmd_list = kzalloc(sizeof(struct megasas_cmd_fusion *) + * max_cmd, GFP_KERNEL); if (!fusion->cmd_list) { printk(KERN_DEBUG "megasas: out of memory. Could not alloc " @@ -470,9 +471,6 @@ megasas_alloc_cmds_fusion(struct megasas_instance *instance) goto fail_cmd_list; } - memset(fusion->cmd_list, 0, sizeof(struct megasas_cmd_fusion *) - *max_cmd); - max_cmd = instance->max_fw_cmds; for (i = 0; i < max_cmd; i++) { fusion->cmd_list[i] = kmalloc(sizeof(struct megasas_cmd_fusion), @@ -2063,9 +2061,9 @@ megasas_check_reset_fusion(struct megasas_instance *instance, int megasas_wait_for_outstanding_fusion(struct megasas_instance *instance) { int i, outstanding, retval = 0; - u32 fw_state, wait_time = MEGASAS_RESET_WAIT_TIME; + u32 fw_state; - for (i = 0; i < wait_time; i++) { + for (i = 0; i < resetwaittime; i++) { /* Check if firmware is in fault state */ fw_state = instance->instancet->read_fw_status_reg( instance->reg_set) & MFI_STATE_MASK; diff --git a/drivers/scsi/mpt2sas/Kconfig b/drivers/scsi/mpt2sas/Kconfig index bbb7e4b..39f08dd 100644 --- a/drivers/scsi/mpt2sas/Kconfig +++ b/drivers/scsi/mpt2sas/Kconfig @@ -2,7 +2,7 @@ # Kernel configuration file for the MPT2SAS # # This code is based on drivers/scsi/mpt2sas/Kconfig -# Copyright (C) 2007-2010 LSI Corporation +# Copyright (C) 2007-2012 LSI Corporation # (mailto:DL-MPTFusionLinux@lsi.com) # This program is free software; you can redistribute it and/or diff --git a/drivers/scsi/mpt2sas/mpi/mpi2.h b/drivers/scsi/mpt2sas/mpi/mpi2.h index a80f322..e960f96 100644 --- a/drivers/scsi/mpt2sas/mpi/mpi2.h +++ b/drivers/scsi/mpt2sas/mpi/mpi2.h @@ -1,5 +1,5 @@ /* - * Copyright (c) 2000-2011 LSI Corporation. + * Copyright (c) 2000-2012 LSI Corporation. * * * Name: mpi2.h @@ -8,7 +8,7 @@ * scatter/gather formats. * Creation Date: June 21, 2006 * - * mpi2.h Version: 02.00.23 + * mpi2.h Version: 02.00.25 * * Version History * --------------- @@ -72,6 +72,9 @@ * 05-25-11 02.00.21 Bumped MPI2_HEADER_VERSION_UNIT. * 08-24-11 02.00.22 Bumped MPI2_HEADER_VERSION_UNIT. * 11-18-11 02.00.23 Bumped MPI2_HEADER_VERSION_UNIT. + * 02-06-12 02.00.24 Bumped MPI2_HEADER_VERSION_UNIT. + * 03-29-12 02.00.25 Bumped MPI2_HEADER_VERSION_UNIT. + * Added Hard Reset delay timings. * -------------------------------------------------------------------------- */ @@ -97,7 +100,7 @@ #define MPI2_VERSION_02_00 (0x0200) /* versioning for this MPI header set */ -#define MPI2_HEADER_VERSION_UNIT (0x17) +#define MPI2_HEADER_VERSION_UNIT (0x19) #define MPI2_HEADER_VERSION_DEV (0x00) #define MPI2_HEADER_VERSION_UNIT_MASK (0xFF00) #define MPI2_HEADER_VERSION_UNIT_SHIFT (8) @@ -275,6 +278,11 @@ typedef volatile struct _MPI2_SYSTEM_INTERFACE_REGS #define MPI2_REQUEST_DESCRIPTOR_POST_HIGH_OFFSET (0x000000C4) +/* Hard Reset delay timings */ +#define MPI2_HARD_RESET_PCIE_FIRST_READ_DELAY_MICRO_SEC (50000) +#define MPI2_HARD_RESET_PCIE_RESET_READ_WINDOW_MICRO_SEC (255000) +#define MPI2_HARD_RESET_PCIE_SECOND_READ_DELAY_MICRO_SEC (256000) + /***************************************************************************** * * Message Descriptors diff --git a/drivers/scsi/mpt2sas/mpi/mpi2_init.h b/drivers/scsi/mpt2sas/mpi/mpi2_init.h index de90162..38c5da3 100644 --- a/drivers/scsi/mpt2sas/mpi/mpi2_init.h +++ b/drivers/scsi/mpt2sas/mpi/mpi2_init.h @@ -1,12 +1,12 @@ /* - * Copyright (c) 2000-2010 LSI Corporation. + * Copyright (c) 2000-2012 LSI Corporation. * * * Name: mpi2_init.h * Title: MPI SCSI initiator mode messages and structures * Creation Date: June 23, 2006 * - * mpi2_init.h Version: 02.00.11 + * mpi2_init.h Version: 02.00.13 * * Version History * --------------- @@ -34,6 +34,8 @@ * 02-10-10 02.00.09 Removed unused structure that had "#if 0" around it. * 05-12-10 02.00.10 Added optional vendor-unique region to SCSI IO Request. * 11-10-10 02.00.11 Added MPI2_SCSIIO_NUM_SGLOFFSETS define. + * 02-06-12 02.00.13 Added alternate defines for Task Priority / Command + * Priority to match SAM-4. * -------------------------------------------------------------------------- */ @@ -194,6 +196,9 @@ typedef struct _MPI2_SCSI_IO_REQUEST #define MPI2_SCSIIO_CONTROL_TASKPRI_MASK (0x00007800) #define MPI2_SCSIIO_CONTROL_TASKPRI_SHIFT (11) +/* alternate name for the previous field; called Command Priority in SAM-4 */ +#define MPI2_SCSIIO_CONTROL_CMDPRI_MASK (0x00007800) +#define MPI2_SCSIIO_CONTROL_CMDPRI_SHIFT (11) #define MPI2_SCSIIO_CONTROL_TASKATTRIBUTE_MASK (0x00000700) #define MPI2_SCSIIO_CONTROL_SIMPLEQ (0x00000000) diff --git a/drivers/scsi/mpt2sas/mpi/mpi2_ioc.h b/drivers/scsi/mpt2sas/mpi/mpi2_ioc.h index 9a925c0..b0d4760 100644 --- a/drivers/scsi/mpt2sas/mpi/mpi2_ioc.h +++ b/drivers/scsi/mpt2sas/mpi/mpi2_ioc.h @@ -1,12 +1,12 @@ /* - * Copyright (c) 2000-2011 LSI Corporation. + * Copyright (c) 2000-2012 LSI Corporation. * * * Name: mpi2_ioc.h * Title: MPI IOC, Port, Event, FW Download, and FW Upload messages * Creation Date: October 11, 2006 * - * mpi2_ioc.h Version: 02.00.19 + * mpi2_ioc.h Version: 02.00.21 * * Version History * --------------- @@ -117,6 +117,7 @@ * 08-24-11 02.00.19 Added PhysicalPort field to * MPI2_EVENT_DATA_SAS_DEVICE_STATUS_CHANGE structure. * Marked MPI2_PM_CONTROL_FEATURE_PCIE_LINK as obsolete. + * 03-29-12 02.00.21 Added a product specific range to event values. * -------------------------------------------------------------------------- */ @@ -492,7 +493,8 @@ typedef struct _MPI2_EVENT_NOTIFICATION_REPLY #define MPI2_EVENT_SAS_NOTIFY_PRIMITIVE (0x0026) #define MPI2_EVENT_TEMP_THRESHOLD (0x0027) #define MPI2_EVENT_HOST_MESSAGE (0x0028) - +#define MPI2_EVENT_MIN_PRODUCT_SPECIFIC (0x006E) +#define MPI2_EVENT_MAX_PRODUCT_SPECIFIC (0x007F) /* Log Entry Added Event data */ diff --git a/drivers/scsi/mpt2sas/mpi/mpi2_raid.h b/drivers/scsi/mpt2sas/mpi/mpi2_raid.h index 0601612..2b38af2 100644 --- a/drivers/scsi/mpt2sas/mpi/mpi2_raid.h +++ b/drivers/scsi/mpt2sas/mpi/mpi2_raid.h @@ -1,12 +1,12 @@ /* - * Copyright (c) 2000-2010 LSI Corporation. + * Copyright (c) 2000-2012 LSI Corporation. * * * Name: mpi2_raid.h * Title: MPI Integrated RAID messages and structures * Creation Date: April 26, 2007 * - * mpi2_raid.h Version: 02.00.06 + * mpi2_raid.h Version: 02.00.08 * * Version History * --------------- @@ -26,7 +26,7 @@ * 08-24-10 02.00.06 Added MPI2_RAID_ACTION_COMPATIBILITY_CHECK along with * related structures and defines. * Added product-specific range to RAID Action values. - + * 02-06-12 02.00.08 Added MPI2_RAID_ACTION_PHYSDISK_HIDDEN. * -------------------------------------------------------------------------- */ @@ -181,6 +181,7 @@ typedef struct _MPI2_RAID_ACTION_REQUEST #define MPI2_RAID_ACTION_START_RAID_FUNCTION (0x21) #define MPI2_RAID_ACTION_STOP_RAID_FUNCTION (0x22) #define MPI2_RAID_ACTION_COMPATIBILITY_CHECK (0x23) +#define MPI2_RAID_ACTION_PHYSDISK_HIDDEN (0x24) #define MPI2_RAID_ACTION_MIN_PRODUCT_SPECIFIC (0x80) #define MPI2_RAID_ACTION_MAX_PRODUCT_SPECIFIC (0xFF) diff --git a/drivers/scsi/mpt2sas/mpt2sas_base.c b/drivers/scsi/mpt2sas/mpt2sas_base.c index 9d5a56c..ffd85c5 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_base.c +++ b/drivers/scsi/mpt2sas/mpt2sas_base.c @@ -3,7 +3,7 @@ * for access to MPT (Message Passing Technology) firmware. * * This code is based on drivers/scsi/mpt2sas/mpt2_base.c - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or @@ -1978,9 +1978,9 @@ _base_display_intel_branding(struct MPT2SAS_ADAPTER *ioc) printk(MPT2SAS_INFO_FMT "%s\n", ioc->name, MPT2SAS_INTEL_RMS2LL040_BRANDING); break; - case MPT2SAS_INTEL_RAMSDALE_SSDID: + case MPT2SAS_INTEL_SSD910_SSDID: printk(MPT2SAS_INFO_FMT "%s\n", ioc->name, - MPT2SAS_INTEL_RAMSDALE_BRANDING); + MPT2SAS_INTEL_SSD910_BRANDING); break; default: break; diff --git a/drivers/scsi/mpt2sas/mpt2sas_base.h b/drivers/scsi/mpt2sas/mpt2sas_base.h index b3a1a30..543d8d6 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_base.h +++ b/drivers/scsi/mpt2sas/mpt2sas_base.h @@ -3,7 +3,7 @@ * for access to MPT (Message Passing Technology) firmware. * * This code is based on drivers/scsi/mpt2sas/mpt2_base.h - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or @@ -69,8 +69,8 @@ #define MPT2SAS_DRIVER_NAME "mpt2sas" #define MPT2SAS_AUTHOR "LSI Corporation <DL-MPTFusionLinux@lsi.com>" #define MPT2SAS_DESCRIPTION "LSI MPT Fusion SAS 2.0 Device Driver" -#define MPT2SAS_DRIVER_VERSION "13.100.00.00" -#define MPT2SAS_MAJOR_VERSION 13 +#define MPT2SAS_DRIVER_VERSION "14.100.00.00" +#define MPT2SAS_MAJOR_VERSION 14 #define MPT2SAS_MINOR_VERSION 100 #define MPT2SAS_BUILD_VERSION 00 #define MPT2SAS_RELEASE_VERSION 00 @@ -171,8 +171,8 @@ "Intel Integrated RAID Module RMS2LL040" #define MPT2SAS_INTEL_RS25GB008_BRANDING \ "Intel(R) RAID Controller RS25GB008" -#define MPT2SAS_INTEL_RAMSDALE_BRANDING \ - "Intel 720 Series SSD" +#define MPT2SAS_INTEL_SSD910_BRANDING \ + "Intel(R) SSD 910 Series" /* * Intel HBA SSDIDs */ @@ -183,7 +183,7 @@ #define MPT2SAS_INTEL_RMS2LL080_SSDID 0x350E #define MPT2SAS_INTEL_RMS2LL040_SSDID 0x350F #define MPT2SAS_INTEL_RS25GB008_SSDID 0x3000 -#define MPT2SAS_INTEL_RAMSDALE_SSDID 0x3700 +#define MPT2SAS_INTEL_SSD910_SSDID 0x3700 /* * HP HBA branding @@ -1096,6 +1096,8 @@ int mpt2sas_config_get_iounit_pg1(struct MPT2SAS_ADAPTER *ioc, Mpi2ConfigReply_t *mpi_reply, Mpi2IOUnitPage1_t *config_page); int mpt2sas_config_set_iounit_pg1(struct MPT2SAS_ADAPTER *ioc, Mpi2ConfigReply_t *mpi_reply, Mpi2IOUnitPage1_t *config_page); +int mpt2sas_config_get_iounit_pg3(struct MPT2SAS_ADAPTER *ioc, + Mpi2ConfigReply_t *mpi_reply, Mpi2IOUnitPage3_t *config_page, u16 sz); int mpt2sas_config_get_sas_iounit_pg1(struct MPT2SAS_ADAPTER *ioc, Mpi2ConfigReply_t *mpi_reply, Mpi2SasIOUnitPage1_t *config_page, u16 sz); int mpt2sas_config_set_sas_iounit_pg1(struct MPT2SAS_ADAPTER *ioc, diff --git a/drivers/scsi/mpt2sas/mpt2sas_config.c b/drivers/scsi/mpt2sas/mpt2sas_config.c index 2b4d376..8637780 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_config.c +++ b/drivers/scsi/mpt2sas/mpt2sas_config.c @@ -2,7 +2,7 @@ * This module provides common API for accessing firmware configuration pages * * This code is based on drivers/scsi/mpt2sas/mpt2_base.c - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or @@ -683,6 +683,42 @@ mpt2sas_config_set_iounit_pg1(struct MPT2SAS_ADAPTER *ioc, } /** + * mpt2sas_config_get_iounit_pg3 - obtain iounit page 3 + * @ioc: per adapter object + * @mpi_reply: reply mf payload returned from firmware + * @config_page: contents of the config page + * @sz: size of buffer passed in config_page + * Context: sleep. + * + * Returns 0 for success, non-zero for failure. + */ +int +mpt2sas_config_get_iounit_pg3(struct MPT2SAS_ADAPTER *ioc, + Mpi2ConfigReply_t *mpi_reply, Mpi2IOUnitPage3_t *config_page, u16 sz) +{ + Mpi2ConfigRequest_t mpi_request; + int r; + + memset(&mpi_request, 0, sizeof(Mpi2ConfigRequest_t)); + mpi_request.Function = MPI2_FUNCTION_CONFIG; + mpi_request.Action = MPI2_CONFIG_ACTION_PAGE_HEADER; + mpi_request.Header.PageType = MPI2_CONFIG_PAGETYPE_IO_UNIT; + mpi_request.Header.PageNumber = 3; + mpi_request.Header.PageVersion = MPI2_IOUNITPAGE3_PAGEVERSION; + mpt2sas_base_build_zero_len_sge(ioc, &mpi_request.PageBufferSGE); + r = _config_request(ioc, &mpi_request, mpi_reply, + MPT2_CONFIG_PAGE_DEFAULT_TIMEOUT, NULL, 0); + if (r) + goto out; + + mpi_request.Action = MPI2_CONFIG_ACTION_PAGE_READ_CURRENT; + r = _config_request(ioc, &mpi_request, mpi_reply, + MPT2_CONFIG_PAGE_DEFAULT_TIMEOUT, config_page, sz); + out: + return r; +} + +/** * mpt2sas_config_get_ioc_pg8 - obtain ioc page 8 * @ioc: per adapter object * @mpi_reply: reply mf payload returned from firmware diff --git a/drivers/scsi/mpt2sas/mpt2sas_ctl.c b/drivers/scsi/mpt2sas/mpt2sas_ctl.c index 49bdd2d..08685c4 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_ctl.c +++ b/drivers/scsi/mpt2sas/mpt2sas_ctl.c @@ -3,7 +3,7 @@ * controllers * * This code is based on drivers/scsi/mpt2sas/mpt2_ctl.c - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or @@ -2181,10 +2181,12 @@ _ctl_ioctl_main(struct file *file, unsigned int cmd, void __user *arg, return -EAGAIN; state = (file->f_flags & O_NONBLOCK) ? NON_BLOCKING : BLOCKING; - if (state == NON_BLOCKING && !mutex_trylock(&ioc->ctl_cmds.mutex)) - return -EAGAIN; - else if (mutex_lock_interruptible(&ioc->ctl_cmds.mutex)) + if (state == NON_BLOCKING) { + if (!mutex_trylock(&ioc->ctl_cmds.mutex)) + return -EAGAIN; + } else if (mutex_lock_interruptible(&ioc->ctl_cmds.mutex)) { return -ERESTARTSYS; + } switch (cmd) { case MPT2IOCINFO: @@ -2690,6 +2692,75 @@ _ctl_ioc_reply_queue_count_show(struct device *cdev, static DEVICE_ATTR(reply_queue_count, S_IRUGO, _ctl_ioc_reply_queue_count_show, NULL); +/** + * _ctl_BRM_status_show - Backup Rail Monitor Status + * @cdev - pointer to embedded class device + * @buf - the buffer returned + * + * This is number of reply queues + * + * A sysfs 'read-only' shost attribute. + */ +static ssize_t +_ctl_BRM_status_show(struct device *cdev, struct device_attribute *attr, + char *buf) +{ + struct Scsi_Host *shost = class_to_shost(cdev); + struct MPT2SAS_ADAPTER *ioc = shost_priv(shost); + Mpi2IOUnitPage3_t *io_unit_pg3 = NULL; + Mpi2ConfigReply_t mpi_reply; + u16 backup_rail_monitor_status = 0; + u16 ioc_status; + int sz; + ssize_t rc = 0; + + if (!ioc->is_warpdrive) { + printk(MPT2SAS_ERR_FMT "%s: BRM attribute is only for"\ + "warpdrive\n", ioc->name, __func__); + goto out; + } + + /* allocate upto GPIOVal 36 entries */ + sz = offsetof(Mpi2IOUnitPage3_t, GPIOVal) + (sizeof(u16) * 36); + io_unit_pg3 = kzalloc(sz, GFP_KERNEL); + if (!io_unit_pg3) { + printk(MPT2SAS_ERR_FMT "%s: failed allocating memory"\ + "for iounit_pg3: (%d) bytes\n", ioc->name, __func__, sz); + goto out; + } + + if (mpt2sas_config_get_iounit_pg3(ioc, &mpi_reply, io_unit_pg3, sz) != + 0) { + printk(MPT2SAS_ERR_FMT + "%s: failed reading iounit_pg3\n", ioc->name, + __func__); + goto out; + } + + ioc_status = le16_to_cpu(mpi_reply.IOCStatus) & MPI2_IOCSTATUS_MASK; + if (ioc_status != MPI2_IOCSTATUS_SUCCESS) { + printk(MPT2SAS_ERR_FMT "%s: iounit_pg3 failed with"\ + "ioc_status(0x%04x)\n", ioc->name, __func__, ioc_status); + goto out; + } + + if (io_unit_pg3->GPIOCount < 25) { + printk(MPT2SAS_ERR_FMT "%s: iounit_pg3->GPIOCount less than"\ + "25 entries, detected (%d) entries\n", ioc->name, __func__, + io_unit_pg3->GPIOCount); + goto out; + } + + /* BRM status is in bit zero of GPIOVal[24] */ + backup_rail_monitor_status = le16_to_cpu(io_unit_pg3->GPIOVal[24]); + rc = snprintf(buf, PAGE_SIZE, "%d\n", (backup_rail_monitor_status & 1)); + + out: + kfree(io_unit_pg3); + return rc; +} +static DEVICE_ATTR(BRM_status, S_IRUGO, _ctl_BRM_status_show, NULL); + struct DIAG_BUFFER_START { __le32 Size; __le32 DiagVersion; @@ -2901,6 +2972,7 @@ struct device_attribute *mpt2sas_host_attrs[] = { &dev_attr_host_trace_buffer, &dev_attr_host_trace_buffer_enable, &dev_attr_reply_queue_count, + &dev_attr_BRM_status, NULL, }; diff --git a/drivers/scsi/mpt2sas/mpt2sas_ctl.h b/drivers/scsi/mpt2sas/mpt2sas_ctl.h index 11ff1d5..b5eb0d1 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_ctl.h +++ b/drivers/scsi/mpt2sas/mpt2sas_ctl.h @@ -3,7 +3,7 @@ * controllers * * This code is based on drivers/scsi/mpt2sas/mpt2_ctl.h - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or diff --git a/drivers/scsi/mpt2sas/mpt2sas_debug.h b/drivers/scsi/mpt2sas/mpt2sas_debug.h index 9731f8e..69cc7d0 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_debug.h +++ b/drivers/scsi/mpt2sas/mpt2sas_debug.h @@ -2,7 +2,7 @@ * Logging Support for MPT (Message Passing Technology) based controllers * * This code is based on drivers/scsi/mpt2sas/mpt2_debug.c - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or diff --git a/drivers/scsi/mpt2sas/mpt2sas_scsih.c b/drivers/scsi/mpt2sas/mpt2sas_scsih.c index 1ccae45..af4e6c4 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_scsih.c +++ b/drivers/scsi/mpt2sas/mpt2sas_scsih.c @@ -2,7 +2,7 @@ * Scsi Host Layer for MPT (Message Passing Technology) based controllers * * This code is based on drivers/scsi/mpt2sas/mpt2_scsih.c - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or @@ -119,6 +119,15 @@ module_param(diag_buffer_enable, int, 0); MODULE_PARM_DESC(diag_buffer_enable, " post diag buffers " "(TRACE=1/SNAPSHOT=2/EXTENDED=4/default=0)"); +static int disable_discovery = -1; +module_param(disable_discovery, int, 0); +MODULE_PARM_DESC(disable_discovery, " disable discovery "); + +/* permit overriding the host protection capabilities mask (EEDP/T10 PI) */ +static int prot_mask = 0; +module_param(prot_mask, int, 0); +MODULE_PARM_DESC(prot_mask, " host protection capabilities mask, def=7 "); + /** * struct sense_info - common structure for obtaining sense keys * @skey: sense key @@ -3768,8 +3777,6 @@ static void _scsih_eedp_error_handling(struct scsi_cmnd *scmd, u16 ioc_status) { u8 ascq; - u8 sk; - u8 host_byte; switch (ioc_status) { case MPI2_IOCSTATUS_EEDP_GUARD_ERROR: @@ -3786,16 +3793,8 @@ _scsih_eedp_error_handling(struct scsi_cmnd *scmd, u16 ioc_status) break; } - if (scmd->sc_data_direction == DMA_TO_DEVICE) { - sk = ILLEGAL_REQUEST; - host_byte = DID_ABORT; - } else { - sk = ABORTED_COMMAND; - host_byte = DID_OK; - } - - scsi_build_sense_buffer(0, scmd->sense_buffer, sk, 0x10, ascq); - scmd->result = DRIVER_SENSE << 24 | (host_byte << 16) | + scsi_build_sense_buffer(0, scmd->sense_buffer, ILLEGAL_REQUEST, 0x10, ascq); + scmd->result = DRIVER_SENSE << 24 | (DID_ABORT << 16) | SAM_STAT_CHECK_CONDITION; } @@ -5973,8 +5972,14 @@ _scsih_sas_discovery_event(struct MPT2SAS_ADAPTER *ioc, #endif if (event_data->ReasonCode == MPI2_EVENT_SAS_DISC_RC_STARTED && - !ioc->sas_hba.num_phys) + !ioc->sas_hba.num_phys) { + if (disable_discovery > 0 && ioc->shost_recovery) { + /* Wait for the reset to complete */ + while (ioc->shost_recovery) + ssleep(1); + } _scsih_sas_host_add(ioc); + } } /** @@ -7254,7 +7259,8 @@ mpt2sas_scsih_reset_handler(struct MPT2SAS_ADAPTER *ioc, int reset_phase) _scsih_search_responding_sas_devices(ioc); _scsih_search_responding_raid_devices(ioc); _scsih_search_responding_expanders(ioc); - if (!ioc->is_driver_loading) { + if ((!ioc->is_driver_loading) && !(disable_discovery > 0 && + !ioc->sas_hba.num_phys)) { _scsih_prep_device_scan(ioc); _scsih_search_responding_sas_devices(ioc); _scsih_search_responding_raid_devices(ioc); @@ -7929,6 +7935,9 @@ _scsih_scan_start(struct Scsi_Host *shost) if (diag_buffer_enable != -1 && diag_buffer_enable != 0) mpt2sas_enable_diag_buffer(ioc, diag_buffer_enable); + if (disable_discovery > 0) + return; + ioc->start_scan = 1; rc = mpt2sas_port_enable(ioc); @@ -7950,6 +7959,12 @@ _scsih_scan_finished(struct Scsi_Host *shost, unsigned long time) { struct MPT2SAS_ADAPTER *ioc = shost_priv(shost); + if (disable_discovery > 0) { + ioc->is_driver_loading = 0; + ioc->wait_for_discovery_to_complete = 0; + return 1; + } + if (time >= (300 * HZ)) { ioc->base_cmds.status = MPT2_CMD_NOT_USED; printk(MPT2SAS_INFO_FMT "port enable: FAILED with timeout " @@ -8055,8 +8070,8 @@ _scsih_probe(struct pci_dev *pdev, const struct pci_device_id *id) if (max_sectors != 0xFFFF) { if (max_sectors < 64) { shost->max_sectors = 64; - printk(MPT2SAS_WARN_FMT "Invalid value %d passed " - "for max_sectors, range is 64 to 8192. Assigning " + printk(MPT2SAS_WARN_FMT "Invalid value %d passed "\ + "for max_sectors, range is 64 to 32767. Assigning "\ "value of 64.\n", ioc->name, max_sectors); } else if (max_sectors > 32767) { shost->max_sectors = 32767; @@ -8078,8 +8093,14 @@ _scsih_probe(struct pci_dev *pdev, const struct pci_device_id *id) goto out_add_shost_fail; } - scsi_host_set_prot(shost, SHOST_DIF_TYPE1_PROTECTION - | SHOST_DIF_TYPE2_PROTECTION | SHOST_DIF_TYPE3_PROTECTION); + /* register EEDP capabilities with SCSI layer */ + if (prot_mask) + scsi_host_set_prot(shost, prot_mask); + else + scsi_host_set_prot(shost, SHOST_DIF_TYPE1_PROTECTION + | SHOST_DIF_TYPE2_PROTECTION + | SHOST_DIF_TYPE3_PROTECTION); + scsi_host_set_guard(shost, SHOST_DIX_GUARD_CRC); /* event thread */ diff --git a/drivers/scsi/mpt2sas/mpt2sas_transport.c b/drivers/scsi/mpt2sas/mpt2sas_transport.c index c6cf20f..8c2ffbe 100644 --- a/drivers/scsi/mpt2sas/mpt2sas_transport.c +++ b/drivers/scsi/mpt2sas/mpt2sas_transport.c @@ -2,7 +2,7 @@ * SAS Transport Layer for MPT (Message Passing Technology) based controllers * * This code is based on drivers/scsi/mpt2sas/mpt2_transport.c - * Copyright (C) 2007-2010 LSI Corporation + * Copyright (C) 2007-2012 LSI Corporation * (mailto:DL-MPTFusionLinux@lsi.com) * * This program is free software; you can redistribute it and/or diff --git a/drivers/scsi/mvsas/mv_sas.c b/drivers/scsi/mvsas/mv_sas.c index 4539d59..a3776d6 100644 --- a/drivers/scsi/mvsas/mv_sas.c +++ b/drivers/scsi/mvsas/mv_sas.c @@ -1629,7 +1629,7 @@ int mvs_abort_task(struct sas_task *task) mv_dprintk("mvs_abort_task() mvi=%p task=%p " "slot=%p slot_idx=x%x\n", mvi, task, slot, slot_idx); - mvs_tmf_timedout((unsigned long)task); + task->task_state_flags |= SAS_TASK_STATE_ABORTED; mvs_slot_task_free(mvi, task, slot, slot_idx); rc = TMF_RESP_FUNC_COMPLETE; goto out; diff --git a/drivers/scsi/qla2xxx/qla_attr.c b/drivers/scsi/qla2xxx/qla_attr.c index 5ab9530..1c28215 100644 --- a/drivers/scsi/qla2xxx/qla_attr.c +++ b/drivers/scsi/qla2xxx/qla_attr.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -26,7 +26,7 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj, struct qla_hw_data *ha = vha->hw; int rval = 0; - if (ha->fw_dump_reading == 0) + if (!(ha->fw_dump_reading || ha->mctp_dump_reading)) return 0; if (IS_QLA82XX(ha)) { @@ -39,9 +39,14 @@ qla2x00_sysfs_read_fw_dump(struct file *filp, struct kobject *kobj, rval = memory_read_from_buffer(buf, count, &off, ha->md_dump, ha->md_dump_size); return rval; - } else + } else if (ha->mctp_dumped && ha->mctp_dump_reading) + return memory_read_from_buffer(buf, count, &off, ha->mctp_dump, + MCTP_DUMP_SIZE); + else if (ha->fw_dump_reading) return memory_read_from_buffer(buf, count, &off, ha->fw_dump, ha->fw_dump_len); + else + return 0; } static ssize_t @@ -107,6 +112,22 @@ qla2x00_sysfs_write_fw_dump(struct file *filp, struct kobject *kobj, if (IS_QLA82XX(ha)) set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); break; + case 6: + if (!ha->mctp_dump_reading) + break; + ql_log(ql_log_info, vha, 0x70c1, + "MCTP dump cleared on (%ld).\n", vha->host_no); + ha->mctp_dump_reading = 0; + ha->mctp_dumped = 0; + break; + case 7: + if (ha->mctp_dumped && !ha->mctp_dump_reading) { + ha->mctp_dump_reading = 1; + ql_log(ql_log_info, vha, 0x70c2, + "Raw mctp dump ready for read on (%ld).\n", + vha->host_no); + } + break; } return count; } @@ -564,6 +585,7 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj, struct qla_hw_data *ha = vha->hw; struct scsi_qla_host *base_vha = pci_get_drvdata(ha->pdev); int type; + uint32_t idc_control; if (off != 0) return -EINVAL; @@ -587,22 +609,36 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj, scsi_unblock_requests(vha->host); break; case 0x2025d: - if (!IS_QLA81XX(ha) || !IS_QLA8031(ha)) + if (!IS_QLA81XX(ha) && !IS_QLA83XX(ha)) return -EPERM; ql_log(ql_log_info, vha, 0x706f, "Issuing MPI reset.\n"); - /* Make sure FC side is not in reset */ - qla2x00_wait_for_hba_online(vha); - - /* Issue MPI reset */ - scsi_block_requests(vha->host); - if (qla81xx_restart_mpi_firmware(vha) != QLA_SUCCESS) - ql_log(ql_log_warn, vha, 0x7070, - "MPI reset failed.\n"); - scsi_unblock_requests(vha->host); - break; + if (IS_QLA83XX(ha)) { + uint32_t idc_control; + + qla83xx_idc_lock(vha, 0); + __qla83xx_get_idc_control(vha, &idc_control); + idc_control |= QLA83XX_IDC_GRACEFUL_RESET; + __qla83xx_set_idc_control(vha, idc_control); + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, + QLA8XXX_DEV_NEED_RESET); + qla83xx_idc_audit(vha, IDC_AUDIT_TIMESTAMP); + qla83xx_idc_unlock(vha, 0); + break; + } else { + /* Make sure FC side is not in reset */ + qla2x00_wait_for_hba_online(vha); + + /* Issue MPI reset */ + scsi_block_requests(vha->host); + if (qla81xx_restart_mpi_firmware(vha) != QLA_SUCCESS) + ql_log(ql_log_warn, vha, 0x7070, + "MPI reset failed.\n"); + scsi_unblock_requests(vha->host); + break; + } case 0x2025e: if (!IS_QLA82XX(ha) || vha != base_vha) { ql_log(ql_log_info, vha, 0x7071, @@ -616,6 +652,29 @@ qla2x00_sysfs_write_reset(struct file *filp, struct kobject *kobj, qla2xxx_wake_dpc(vha); qla2x00_wait_for_fcoe_ctx_reset(vha); break; + case 0x2025f: + if (!IS_QLA8031(ha)) + return -EPERM; + ql_log(ql_log_info, vha, 0x70bc, + "Disabling Reset by IDC control\n"); + qla83xx_idc_lock(vha, 0); + __qla83xx_get_idc_control(vha, &idc_control); + idc_control |= QLA83XX_IDC_RESET_DISABLED; + __qla83xx_set_idc_control(vha, idc_control); + qla83xx_idc_unlock(vha, 0); + break; + case 0x20260: + if (!IS_QLA8031(ha)) + return -EPERM; + ql_log(ql_log_info, vha, 0x70bd, + "Enabling Reset by IDC control\n"); + qla83xx_idc_lock(vha, 0); + __qla83xx_get_idc_control(vha, &idc_control); + idc_control &= ~QLA83XX_IDC_RESET_DISABLED; + __qla83xx_set_idc_control(vha, idc_control); + qla83xx_idc_unlock(vha, 0); + break; + } return count; } @@ -1251,6 +1310,49 @@ qla2x00_fw_state_show(struct device *dev, struct device_attribute *attr, state[1], state[2], state[3], state[4]); } +static ssize_t +qla2x00_diag_requests_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + scsi_qla_host_t *vha = shost_priv(class_to_shost(dev)); + + if (!IS_BIDI_CAPABLE(vha->hw)) + return snprintf(buf, PAGE_SIZE, "\n"); + + return snprintf(buf, PAGE_SIZE, "%llu\n", vha->bidi_stats.io_count); +} + +static ssize_t +qla2x00_diag_megabytes_show(struct device *dev, + struct device_attribute *attr, char *buf) +{ + scsi_qla_host_t *vha = shost_priv(class_to_shost(dev)); + + if (!IS_BIDI_CAPABLE(vha->hw)) + return snprintf(buf, PAGE_SIZE, "\n"); + + return snprintf(buf, PAGE_SIZE, "%llu\n", + vha->bidi_stats.transfer_bytes >> 20); +} + +static ssize_t +qla2x00_fw_dump_size_show(struct device *dev, struct device_attribute *attr, + char *buf) +{ + scsi_qla_host_t *vha = shost_priv(class_to_shost(dev)); + struct qla_hw_data *ha = vha->hw; + uint32_t size; + + if (!ha->fw_dumped) + size = 0; + else if (IS_QLA82XX(ha)) + size = ha->md_template_size + ha->md_dump_size; + else + size = ha->fw_dump_len; + + return snprintf(buf, PAGE_SIZE, "%d\n", size); +} + static DEVICE_ATTR(driver_version, S_IRUGO, qla2x00_drvr_version_show, NULL); static DEVICE_ATTR(fw_version, S_IRUGO, qla2x00_fw_version_show, NULL); static DEVICE_ATTR(serial_num, S_IRUGO, qla2x00_serial_num_show, NULL); @@ -1289,6 +1391,9 @@ static DEVICE_ATTR(vn_port_mac_address, S_IRUGO, static DEVICE_ATTR(fabric_param, S_IRUGO, qla2x00_fabric_param_show, NULL); static DEVICE_ATTR(fw_state, S_IRUGO, qla2x00_fw_state_show, NULL); static DEVICE_ATTR(thermal_temp, S_IRUGO, qla2x00_thermal_temp_show, NULL); +static DEVICE_ATTR(diag_requests, S_IRUGO, qla2x00_diag_requests_show, NULL); +static DEVICE_ATTR(diag_megabytes, S_IRUGO, qla2x00_diag_megabytes_show, NULL); +static DEVICE_ATTR(fw_dump_size, S_IRUGO, qla2x00_fw_dump_size_show, NULL); struct device_attribute *qla2x00_host_attrs[] = { &dev_attr_driver_version, @@ -1318,6 +1423,9 @@ struct device_attribute *qla2x00_host_attrs[] = { &dev_attr_fw_state, &dev_attr_optrom_gold_fw_version, &dev_attr_thermal_temp, + &dev_attr_diag_requests, + &dev_attr_diag_megabytes, + &dev_attr_fw_dump_size, NULL, }; @@ -1704,7 +1812,7 @@ qla24xx_vport_create(struct fc_vport *fc_vport, bool disable) if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) { - int prot = 0; + int prot = 0, guard; vha->flags.difdix_supported = 1; ql_dbg(ql_dbg_user, vha, 0x7082, "Registered for DIF/DIX type 1 and 3 protection.\n"); @@ -1717,7 +1825,14 @@ qla24xx_vport_create(struct fc_vport *fc_vport, bool disable) | SHOST_DIX_TYPE1_PROTECTION | SHOST_DIX_TYPE2_PROTECTION | SHOST_DIX_TYPE3_PROTECTION); - scsi_host_set_guard(vha->host, SHOST_DIX_GUARD_CRC); + + guard = SHOST_DIX_GUARD_CRC; + + if (IS_PI_IPGUARD_CAPABLE(ha) && + (ql2xenabledif > 1 || IS_PI_DIFB_DIX0_CAPABLE(ha))) + guard |= SHOST_DIX_GUARD_IP; + + scsi_host_set_guard(vha->host, guard); } else vha->flags.difdix_supported = 0; } diff --git a/drivers/scsi/qla2xxx/qla_bsg.c b/drivers/scsi/qla2xxx/qla_bsg.c index c688838..2f9bddd 100644 --- a/drivers/scsi/qla2xxx/qla_bsg.c +++ b/drivers/scsi/qla2xxx/qla_bsg.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -530,13 +530,13 @@ done_unmap_sg: done: return rval; } - -/* Set the port configuration to enable the - * internal loopback on ISP81XX +/* + * Set the port configuration to enable the internal or external loopback + * depending on the loopback mode. */ static inline int -qla81xx_set_internal_loopback(scsi_qla_host_t *vha, uint16_t *config, - uint16_t *new_config) +qla81xx_set_loopback_mode(scsi_qla_host_t *vha, uint16_t *config, + uint16_t *new_config, uint16_t mode) { int ret = 0; int rval = 0; @@ -545,8 +545,14 @@ qla81xx_set_internal_loopback(scsi_qla_host_t *vha, uint16_t *config, if (!IS_QLA81XX(ha) && !IS_QLA8031(ha)) goto done_set_internal; - new_config[0] = config[0] | (ENABLE_INTERNAL_LOOPBACK << 1); - memcpy(&new_config[1], &config[1], sizeof(uint16_t) * 3) ; + if (mode == INTERNAL_LOOPBACK) + new_config[0] = config[0] | (ENABLE_INTERNAL_LOOPBACK << 1); + else if (mode == EXTERNAL_LOOPBACK) + new_config[0] = config[0] | (ENABLE_EXTERNAL_LOOPBACK << 1); + ql_dbg(ql_dbg_user, vha, 0x70be, + "new_config[0]=%02x\n", (new_config[0] & INTERNAL_LOOPBACK_MASK)); + + memcpy(&new_config[1], &config[1], sizeof(uint16_t) * 3); ha->notify_dcbx_comp = 1; ret = qla81xx_set_port_config(vha, new_config); @@ -562,9 +568,17 @@ qla81xx_set_internal_loopback(scsi_qla_host_t *vha, uint16_t *config, if (!wait_for_completion_timeout(&ha->dcbx_comp, (20 * HZ))) { ql_dbg(ql_dbg_user, vha, 0x7022, "State change notification not received.\n"); - } else - ql_dbg(ql_dbg_user, vha, 0x7023, - "State change received.\n"); + rval = -EINVAL; + } else { + if (ha->flags.idc_compl_status) { + ql_dbg(ql_dbg_user, vha, 0x70c3, + "Bad status in IDC Completion AEN\n"); + rval = -EINVAL; + ha->flags.idc_compl_status = 0; + } else + ql_dbg(ql_dbg_user, vha, 0x7023, + "State change received.\n"); + } ha->notify_dcbx_comp = 0; @@ -572,11 +586,9 @@ done_set_internal: return rval; } -/* Set the port configuration to disable the - * internal loopback on ISP81XX - */ +/* Disable loopback mode */ static inline int -qla81xx_reset_internal_loopback(scsi_qla_host_t *vha, uint16_t *config, +qla81xx_reset_loopback_mode(scsi_qla_host_t *vha, uint16_t *config, int wait) { int ret = 0; @@ -589,8 +601,12 @@ qla81xx_reset_internal_loopback(scsi_qla_host_t *vha, uint16_t *config, memset(new_config, 0 , sizeof(new_config)); if ((config[0] & INTERNAL_LOOPBACK_MASK) >> 1 == - ENABLE_INTERNAL_LOOPBACK) { + ENABLE_INTERNAL_LOOPBACK || + (config[0] & INTERNAL_LOOPBACK_MASK) >> 1 == + ENABLE_EXTERNAL_LOOPBACK) { new_config[0] = config[0] & ~INTERNAL_LOOPBACK_MASK; + ql_dbg(ql_dbg_user, vha, 0x70bf, "new_config[0]=%02x\n", + (new_config[0] & INTERNAL_LOOPBACK_MASK)); memcpy(&new_config[1], &config[1], sizeof(uint16_t) * 3) ; ha->notify_dcbx_comp = wait; @@ -707,7 +723,8 @@ qla2x00_process_loopback(struct fc_bsg_job *bsg_job) elreq.options = bsg_job->request->rqst_data.h_vendor.vendor_cmd[1]; - if ((ha->current_topology == ISP_CFG_F || + if (atomic_read(&vha->loop_state) == LOOP_READY && + (ha->current_topology == ISP_CFG_F || ((IS_QLA81XX(ha) || IS_QLA8031(ha)) && le32_to_cpu(*(uint32_t *)req_data) == ELS_OPCODE_BYTE && req_data_len == MAX_ELS_FRAME_PAYLOAD)) && @@ -729,30 +746,24 @@ qla2x00_process_loopback(struct fc_bsg_job *bsg_job) goto done_free_dma_req; } - if (elreq.options != EXTERNAL_LOOPBACK) { - ql_dbg(ql_dbg_user, vha, 0x7020, - "Internal: current port config = %x\n", - config[0]); - if (qla81xx_set_internal_loopback(vha, config, - new_config)) { - ql_log(ql_log_warn, vha, 0x7024, - "Internal loopback failed.\n"); - bsg_job->reply->result = - (DID_ERROR << 16); - rval = -EPERM; - goto done_free_dma_req; - } - } else { - /* For external loopback to work - * ensure internal loopback is disabled - */ - if (qla81xx_reset_internal_loopback(vha, - config, 1)) { - bsg_job->reply->result = - (DID_ERROR << 16); - rval = -EPERM; - goto done_free_dma_req; - } + ql_dbg(ql_dbg_user, vha, 0x70c0, + "elreq.options=%04x\n", elreq.options); + + if (elreq.options == EXTERNAL_LOOPBACK) + if (IS_QLA8031(ha)) + rval = qla81xx_set_loopback_mode(vha, + config, new_config, elreq.options); + else + rval = qla81xx_reset_loopback_mode(vha, + config, 1); + else + rval = qla81xx_set_loopback_mode(vha, config, + new_config, elreq.options); + + if (rval) { + bsg_job->reply->result = (DID_ERROR << 16); + rval = -EPERM; + goto done_free_dma_req; } type = "FC_BSG_HST_VENDOR_LOOPBACK"; @@ -766,7 +777,7 @@ qla2x00_process_loopback(struct fc_bsg_job *bsg_job) /* Revert back to original port config * Also clear internal loopback */ - qla81xx_reset_internal_loopback(vha, + qla81xx_reset_loopback_mode(vha, new_config, 0); } @@ -1364,7 +1375,7 @@ qla2x00_read_optrom(struct fc_bsg_job *bsg_job) struct qla_hw_data *ha = vha->hw; int rval = 0; - if (ha->flags.isp82xx_reset_hdlr_active) + if (ha->flags.nic_core_reset_hdlr_active) return -EBUSY; rval = qla2x00_optrom_setup(bsg_job, vha, 0); @@ -1560,6 +1571,276 @@ done: } static int +qla2x00_write_i2c(struct fc_bsg_job *bsg_job) +{ + struct Scsi_Host *host = bsg_job->shost; + scsi_qla_host_t *vha = shost_priv(host); + struct qla_hw_data *ha = vha->hw; + int rval = 0; + uint8_t bsg[DMA_POOL_SIZE]; + struct qla_i2c_access *i2c = (void *)bsg; + dma_addr_t sfp_dma; + uint8_t *sfp = dma_pool_alloc(ha->s_dma_pool, GFP_KERNEL, &sfp_dma); + if (!sfp) { + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = + EXT_STATUS_NO_MEMORY; + goto done; + } + + sg_copy_to_buffer(bsg_job->request_payload.sg_list, + bsg_job->request_payload.sg_cnt, i2c, sizeof(*i2c)); + + memcpy(sfp, i2c->buffer, i2c->length); + rval = qla2x00_write_sfp(vha, sfp_dma, sfp, + i2c->device, i2c->offset, i2c->length, i2c->option); + + if (rval) { + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = + EXT_STATUS_MAILBOX; + goto dealloc; + } + + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = 0; + +dealloc: + dma_pool_free(ha->s_dma_pool, sfp, sfp_dma); + +done: + bsg_job->reply_len = sizeof(struct fc_bsg_reply); + bsg_job->reply->result = DID_OK << 16; + bsg_job->job_done(bsg_job); + + return 0; +} + +static int +qla2x00_read_i2c(struct fc_bsg_job *bsg_job) +{ + struct Scsi_Host *host = bsg_job->shost; + scsi_qla_host_t *vha = shost_priv(host); + struct qla_hw_data *ha = vha->hw; + int rval = 0; + uint8_t bsg[DMA_POOL_SIZE]; + struct qla_i2c_access *i2c = (void *)bsg; + dma_addr_t sfp_dma; + uint8_t *sfp = dma_pool_alloc(ha->s_dma_pool, GFP_KERNEL, &sfp_dma); + if (!sfp) { + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = + EXT_STATUS_NO_MEMORY; + goto done; + } + + sg_copy_to_buffer(bsg_job->request_payload.sg_list, + bsg_job->request_payload.sg_cnt, i2c, sizeof(*i2c)); + + rval = qla2x00_read_sfp(vha, sfp_dma, sfp, + i2c->device, i2c->offset, i2c->length, i2c->option); + + if (rval) { + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = + EXT_STATUS_MAILBOX; + goto dealloc; + } + + memcpy(i2c->buffer, sfp, i2c->length); + sg_copy_from_buffer(bsg_job->reply_payload.sg_list, + bsg_job->reply_payload.sg_cnt, i2c, sizeof(*i2c)); + + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = 0; + +dealloc: + dma_pool_free(ha->s_dma_pool, sfp, sfp_dma); + +done: + bsg_job->reply_len = sizeof(struct fc_bsg_reply); + bsg_job->reply->reply_payload_rcv_len = sizeof(*i2c); + bsg_job->reply->result = DID_OK << 16; + bsg_job->job_done(bsg_job); + + return 0; +} + +static int +qla24xx_process_bidir_cmd(struct fc_bsg_job *bsg_job) +{ + struct Scsi_Host *host = bsg_job->shost; + scsi_qla_host_t *vha = shost_priv(host); + struct qla_hw_data *ha = vha->hw; + uint16_t thread_id; + uint32_t rval = EXT_STATUS_OK; + uint16_t req_sg_cnt = 0; + uint16_t rsp_sg_cnt = 0; + uint16_t nextlid = 0; + uint32_t tot_dsds; + srb_t *sp = NULL; + uint32_t req_data_len = 0; + uint32_t rsp_data_len = 0; + + /* Check the type of the adapter */ + if (!IS_BIDI_CAPABLE(ha)) { + ql_log(ql_log_warn, vha, 0x70a0, + "This adapter is not supported\n"); + rval = EXT_STATUS_NOT_SUPPORTED; + goto done; + } + + if (test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags) || + test_bit(ABORT_ISP_ACTIVE, &vha->dpc_flags) || + test_bit(ISP_ABORT_RETRY, &vha->dpc_flags)) { + rval = EXT_STATUS_BUSY; + goto done; + } + + /* Check if host is online */ + if (!vha->flags.online) { + ql_log(ql_log_warn, vha, 0x70a1, + "Host is not online\n"); + rval = EXT_STATUS_DEVICE_OFFLINE; + goto done; + } + + /* Check if cable is plugged in or not */ + if (vha->device_flags & DFLG_NO_CABLE) { + ql_log(ql_log_warn, vha, 0x70a2, + "Cable is unplugged...\n"); + rval = EXT_STATUS_INVALID_CFG; + goto done; + } + + /* Check if the switch is connected or not */ + if (ha->current_topology != ISP_CFG_F) { + ql_log(ql_log_warn, vha, 0x70a3, + "Host is not connected to the switch\n"); + rval = EXT_STATUS_INVALID_CFG; + goto done; + } + + /* Check if operating mode is P2P */ + if (ha->operating_mode != P2P) { + ql_log(ql_log_warn, vha, 0x70a4, + "Host is operating mode is not P2p\n"); + rval = EXT_STATUS_INVALID_CFG; + goto done; + } + + thread_id = bsg_job->request->rqst_data.h_vendor.vendor_cmd[1]; + + mutex_lock(&ha->selflogin_lock); + if (vha->self_login_loop_id == 0) { + /* Initialize all required fields of fcport */ + vha->bidir_fcport.vha = vha; + vha->bidir_fcport.d_id.b.al_pa = vha->d_id.b.al_pa; + vha->bidir_fcport.d_id.b.area = vha->d_id.b.area; + vha->bidir_fcport.d_id.b.domain = vha->d_id.b.domain; + vha->bidir_fcport.loop_id = vha->loop_id; + + if (qla2x00_fabric_login(vha, &(vha->bidir_fcport), &nextlid)) { + ql_log(ql_log_warn, vha, 0x70a7, + "Failed to login port %06X for bidirectional IOCB\n", + vha->bidir_fcport.d_id.b24); + mutex_unlock(&ha->selflogin_lock); + rval = EXT_STATUS_MAILBOX; + goto done; + } + vha->self_login_loop_id = nextlid - 1; + + } + /* Assign the self login loop id to fcport */ + mutex_unlock(&ha->selflogin_lock); + + vha->bidir_fcport.loop_id = vha->self_login_loop_id; + + req_sg_cnt = dma_map_sg(&ha->pdev->dev, + bsg_job->request_payload.sg_list, + bsg_job->request_payload.sg_cnt, + DMA_TO_DEVICE); + + if (!req_sg_cnt) { + rval = EXT_STATUS_NO_MEMORY; + goto done; + } + + rsp_sg_cnt = dma_map_sg(&ha->pdev->dev, + bsg_job->reply_payload.sg_list, bsg_job->reply_payload.sg_cnt, + DMA_FROM_DEVICE); + + if (!rsp_sg_cnt) { + rval = EXT_STATUS_NO_MEMORY; + goto done_unmap_req_sg; + } + + if ((req_sg_cnt != bsg_job->request_payload.sg_cnt) || + (rsp_sg_cnt != bsg_job->reply_payload.sg_cnt)) { + ql_dbg(ql_dbg_user, vha, 0x70a9, + "Dma mapping resulted in different sg counts " + "[request_sg_cnt: %x dma_request_sg_cnt: %x reply_sg_cnt: " + "%x dma_reply_sg_cnt: %x]\n", + bsg_job->request_payload.sg_cnt, req_sg_cnt, + bsg_job->reply_payload.sg_cnt, rsp_sg_cnt); + rval = EXT_STATUS_NO_MEMORY; + goto done_unmap_sg; + } + + if (req_data_len != rsp_data_len) { + rval = EXT_STATUS_BUSY; + ql_log(ql_log_warn, vha, 0x70aa, + "req_data_len != rsp_data_len\n"); + goto done_unmap_sg; + } + + req_data_len = bsg_job->request_payload.payload_len; + rsp_data_len = bsg_job->reply_payload.payload_len; + + + /* Alloc SRB structure */ + sp = qla2x00_get_sp(vha, &(vha->bidir_fcport), GFP_KERNEL); + if (!sp) { + ql_dbg(ql_dbg_user, vha, 0x70ac, + "Alloc SRB structure failed\n"); + rval = EXT_STATUS_NO_MEMORY; + goto done_unmap_sg; + } + + /*Populate srb->ctx with bidir ctx*/ + sp->u.bsg_job = bsg_job; + sp->free = qla2x00_bsg_sp_free; + sp->type = SRB_BIDI_CMD; + sp->done = qla2x00_bsg_job_done; + + /* Add the read and write sg count */ + tot_dsds = rsp_sg_cnt + req_sg_cnt; + + rval = qla2x00_start_bidir(sp, vha, tot_dsds); + if (rval != EXT_STATUS_OK) + goto done_free_srb; + /* the bsg request will be completed in the interrupt handler */ + return rval; + +done_free_srb: + mempool_free(sp, ha->srb_mempool); +done_unmap_sg: + dma_unmap_sg(&ha->pdev->dev, + bsg_job->reply_payload.sg_list, + bsg_job->reply_payload.sg_cnt, DMA_FROM_DEVICE); +done_unmap_req_sg: + dma_unmap_sg(&ha->pdev->dev, + bsg_job->request_payload.sg_list, + bsg_job->request_payload.sg_cnt, DMA_TO_DEVICE); +done: + + /* Return an error vendor specific response + * and complete the bsg request + */ + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = rval; + bsg_job->reply_len = sizeof(struct fc_bsg_reply); + bsg_job->reply->reply_payload_rcv_len = 0; + bsg_job->reply->result = (DID_OK) << 16; + bsg_job->job_done(bsg_job); + /* Always retrun success, vendor rsp carries correct status */ + return 0; +} + +static int qla2x00_process_vendor_specific(struct fc_bsg_job *bsg_job) { switch (bsg_job->request->rqst_data.h_vendor.vendor_cmd[0]) { @@ -1596,6 +1877,15 @@ qla2x00_process_vendor_specific(struct fc_bsg_job *bsg_job) case QL_VND_WRITE_FRU_STATUS: return qla2x00_write_fru_status(bsg_job); + case QL_VND_WRITE_I2C: + return qla2x00_write_i2c(bsg_job); + + case QL_VND_READ_I2C: + return qla2x00_read_i2c(bsg_job); + + case QL_VND_DIAG_IO_CMD: + return qla24xx_process_bidir_cmd(bsg_job); + default: bsg_job->reply->result = (DID_ERROR << 16); bsg_job->job_done(bsg_job); diff --git a/drivers/scsi/qla2xxx/qla_bsg.h b/drivers/scsi/qla2xxx/qla_bsg.h index 70caa63..37b8b7b 100644 --- a/drivers/scsi/qla2xxx/qla_bsg.h +++ b/drivers/scsi/qla2xxx/qla_bsg.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -19,21 +19,41 @@ #define QL_VND_SET_FRU_VERSION 0x0B #define QL_VND_READ_FRU_STATUS 0x0C #define QL_VND_WRITE_FRU_STATUS 0x0D +#define QL_VND_DIAG_IO_CMD 0x0A +#define QL_VND_WRITE_I2C 0x10 +#define QL_VND_READ_I2C 0x11 /* BSG Vendor specific subcode returns */ #define EXT_STATUS_OK 0 #define EXT_STATUS_ERR 1 +#define EXT_STATUS_BUSY 2 #define EXT_STATUS_INVALID_PARAM 6 +#define EXT_STATUS_DATA_OVERRUN 7 +#define EXT_STATUS_DATA_UNDERRUN 8 #define EXT_STATUS_MAILBOX 11 #define EXT_STATUS_NO_MEMORY 17 +#define EXT_STATUS_DEVICE_OFFLINE 22 + +/* + * To support bidirectional iocb + * BSG Vendor specific returns + */ +#define EXT_STATUS_NOT_SUPPORTED 27 +#define EXT_STATUS_INVALID_CFG 28 +#define EXT_STATUS_DMA_ERR 29 +#define EXT_STATUS_TIMEOUT 30 +#define EXT_STATUS_THREAD_FAILED 31 +#define EXT_STATUS_DATA_CMP_FAILED 32 /* BSG definations for interpreting CommandSent field */ #define INT_DEF_LB_LOOPBACK_CMD 0 #define INT_DEF_LB_ECHO_CMD 1 /* Loopback related definations */ +#define INTERNAL_LOOPBACK 0xF1 #define EXTERNAL_LOOPBACK 0xF2 #define ENABLE_INTERNAL_LOOPBACK 0x02 +#define ENABLE_EXTERNAL_LOOPBACK 0x04 #define INTERNAL_LOOPBACK_MASK 0x000E #define MAX_ELS_FRAME_PAYLOAD 252 #define ELS_OPCODE_BYTE 0x10 @@ -183,4 +203,12 @@ struct qla_status_reg { uint8_t reserved[7]; } __packed; +struct qla_i2c_access { + uint16_t device; + uint16_t offset; + uint16_t option; + uint16_t length; + uint8_t buffer[0x40]; +} __packed; + #endif diff --git a/drivers/scsi/qla2xxx/qla_dbg.c b/drivers/scsi/qla2xxx/qla_dbg.c index fdee561..44efe3c 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.c +++ b/drivers/scsi/qla2xxx/qla_dbg.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -11,26 +11,31 @@ * ---------------------------------------------------------------------- * | Level | Last Value Used | Holes | * ---------------------------------------------------------------------- - * | Module Init and Probe | 0x0122 | 0x4b,0xba,0xfa | - * | Mailbox commands | 0x1140 | 0x111a-0x111b | + * | Module Init and Probe | 0x0124 | 0x4b,0xba,0xfa | + * | Mailbox commands | 0x114f | 0x111a-0x111b | * | | | 0x112c-0x112e | * | | | 0x113a | - * | Device Discovery | 0x2086 | 0x2020-0x2022 | - * | Queue Command and IO tracing | 0x3030 | 0x3006,0x3008 | + * | Device Discovery | 0x2087 | 0x2020-0x2022, | + * | | | 0x2016 | + * | Queue Command and IO tracing | 0x3030 | 0x3006-0x300b | + * | | | 0x3027-0x3028 | * | | | 0x302d-0x302e | - * | DPC Thread | 0x401c | 0x4002,0x4013 | - * | Async Events | 0x505f | 0x502b-0x502f | + * | DPC Thread | 0x401d | 0x4002,0x4013 | + * | Async Events | 0x5071 | 0x502b-0x502f | * | | | 0x5047,0x5052 | * | Timer Routines | 0x6011 | | - * | User Space Interactions | 0x709f | 0x7018,0x702e, | + * | User Space Interactions | 0x70c3 | 0x7018,0x702e, | * | | | 0x7039,0x7045, | * | | | 0x7073-0x7075, | - * | | | 0x708c | + * | | | 0x708c, | + * | | | 0x70a5,0x70a6, | + * | | | 0x70a8,0x70ab, | + * | | | 0x70ad-0x70ae | * | Task Management | 0x803c | 0x8025-0x8026 | * | | | 0x800b,0x8039 | * | AER/EEH | 0x9011 | | * | Virtual Port | 0xa007 | | - * | ISP82XX Specific | 0xb054 | 0xb024 | + * | ISP82XX Specific | 0xb084 | 0xb002,0xb024 | * | MultiQ | 0xc00c | | * | Misc | 0xd010 | | * | Target Mode | 0xe06f | | @@ -2357,7 +2362,7 @@ ql_dbg(uint32_t level, scsi_qla_host_t *vha, int32_t id, const char *fmt, ...) /* * This function is for formatting and logging debug information. - * It is to be used when vha is not available and pci is availble, + * It is to be used when vha is not available and pci is available, * i.e., before host allocation. It formats the message and logs it * to the messages file. * parameters: @@ -2452,7 +2457,7 @@ ql_log(uint32_t level, scsi_qla_host_t *vha, int32_t id, const char *fmt, ...) /* * This function is for formatting and logging log messages. - * It is to be used when vha is not available and pci is availble, + * It is to be used when vha is not available and pci is available, * i.e., before host allocation. It formats the message and logs * it to the messages file. All the messages are logged irrespective * of the value of ql2xextended_error_logging. diff --git a/drivers/scsi/qla2xxx/qla_dbg.h b/drivers/scsi/qla2xxx/qla_dbg.h index f278df8..8f911c0 100644 --- a/drivers/scsi/qla2xxx/qla_dbg.h +++ b/drivers/scsi/qla2xxx/qla_dbg.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla2xxx/qla_def.h b/drivers/scsi/qla2xxx/qla_def.h index 39007f5..a9725bf 100644 --- a/drivers/scsi/qla2xxx/qla_def.h +++ b/drivers/scsi/qla2xxx/qla_def.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -115,6 +115,82 @@ #define WRT_REG_DWORD(addr, data) writel(data,addr) /* + * ISP83XX specific remote register addresses + */ +#define QLA83XX_LED_PORT0 0x00201320 +#define QLA83XX_LED_PORT1 0x00201328 +#define QLA83XX_IDC_DEV_STATE 0x22102384 +#define QLA83XX_IDC_MAJOR_VERSION 0x22102380 +#define QLA83XX_IDC_MINOR_VERSION 0x22102398 +#define QLA83XX_IDC_DRV_PRESENCE 0x22102388 +#define QLA83XX_IDC_DRIVER_ACK 0x2210238c +#define QLA83XX_IDC_CONTROL 0x22102390 +#define QLA83XX_IDC_AUDIT 0x22102394 +#define QLA83XX_IDC_LOCK_RECOVERY 0x2210239c +#define QLA83XX_DRIVER_LOCKID 0x22102104 +#define QLA83XX_DRIVER_LOCK 0x8111c028 +#define QLA83XX_DRIVER_UNLOCK 0x8111c02c +#define QLA83XX_FLASH_LOCKID 0x22102100 +#define QLA83XX_FLASH_LOCK 0x8111c010 +#define QLA83XX_FLASH_UNLOCK 0x8111c014 +#define QLA83XX_DEV_PARTINFO1 0x221023e0 +#define QLA83XX_DEV_PARTINFO2 0x221023e4 +#define QLA83XX_FW_HEARTBEAT 0x221020b0 +#define QLA83XX_PEG_HALT_STATUS1 0x221020a8 +#define QLA83XX_PEG_HALT_STATUS2 0x221020ac + +/* 83XX: Macros defining 8200 AEN Reason codes */ +#define IDC_DEVICE_STATE_CHANGE BIT_0 +#define IDC_PEG_HALT_STATUS_CHANGE BIT_1 +#define IDC_NIC_FW_REPORTED_FAILURE BIT_2 +#define IDC_HEARTBEAT_FAILURE BIT_3 + +/* 83XX: Macros defining 8200 AEN Error-levels */ +#define ERR_LEVEL_NON_FATAL 0x1 +#define ERR_LEVEL_RECOVERABLE_FATAL 0x2 +#define ERR_LEVEL_UNRECOVERABLE_FATAL 0x4 + +/* 83XX: Macros for IDC Version */ +#define QLA83XX_SUPP_IDC_MAJOR_VERSION 0x01 +#define QLA83XX_SUPP_IDC_MINOR_VERSION 0x0 + +/* 83XX: Macros for scheduling dpc tasks */ +#define QLA83XX_NIC_CORE_RESET 0x1 +#define QLA83XX_IDC_STATE_HANDLER 0x2 +#define QLA83XX_NIC_CORE_UNRECOVERABLE 0x3 + +/* 83XX: Macros for defining IDC-Control bits */ +#define QLA83XX_IDC_RESET_DISABLED BIT_0 +#define QLA83XX_IDC_GRACEFUL_RESET BIT_1 + +/* 83XX: Macros for different timeouts */ +#define QLA83XX_IDC_INITIALIZATION_TIMEOUT 30 +#define QLA83XX_IDC_RESET_ACK_TIMEOUT 10 +#define QLA83XX_MAX_LOCK_RECOVERY_WAIT (2 * HZ) + +/* 83XX: Macros for defining class in DEV-Partition Info register */ +#define QLA83XX_CLASS_TYPE_NONE 0x0 +#define QLA83XX_CLASS_TYPE_NIC 0x1 +#define QLA83XX_CLASS_TYPE_FCOE 0x2 +#define QLA83XX_CLASS_TYPE_ISCSI 0x3 + +/* 83XX: Macros for IDC Lock-Recovery stages */ +#define IDC_LOCK_RECOVERY_STAGE1 0x1 /* Stage1: Intent for + * lock-recovery + */ +#define IDC_LOCK_RECOVERY_STAGE2 0x2 /* Stage2: Perform lock-recovery */ + +/* 83XX: Macros for IDC Audit type */ +#define IDC_AUDIT_TIMESTAMP 0x0 /* IDC-AUDIT: Record timestamp of + * dev-state change to NEED-RESET + * or NEED-QUIESCENT + */ +#define IDC_AUDIT_COMPLETION 0x1 /* IDC-AUDIT: Record duration of + * reset-recovery completion is + * second + */ + +/* * The ISP2312 v2 chip cannot access the FLASH/GPIO registers via MMIO in an * 133Mhz slot. */ @@ -129,6 +205,7 @@ #define MAX_FIBRE_DEVICES_2400 2048 #define MAX_FIBRE_DEVICES_LOOP 128 #define MAX_FIBRE_DEVICES_MAX MAX_FIBRE_DEVICES_2400 +#define LOOPID_MAP_SIZE (ha->max_fibre_devices) #define MAX_FIBRE_LUNS 0xFFFF #define MAX_HOST_COUNT 16 @@ -259,6 +336,7 @@ struct srb_iocb { #define SRB_ADISC_CMD 6 #define SRB_TM_CMD 7 #define SRB_SCSI_CMD 8 +#define SRB_BIDI_CMD 9 typedef struct srb { atomic_t ref_count; @@ -594,6 +672,20 @@ typedef struct { #define MBA_DISCARD_RND_FRAME 0x8048 /* discard RND frame due to error. */ #define MBA_REJECTED_FCP_CMD 0x8049 /* rejected FCP_CMD. */ +/* 83XX FCoE specific */ +#define MBA_IDC_AEN 0x8200 /* FCoE: NIC Core state change AEN */ + +/* Interrupt type codes */ +#define INTR_ROM_MB_SUCCESS 0x1 +#define INTR_ROM_MB_FAILED 0x2 +#define INTR_MB_SUCCESS 0x10 +#define INTR_MB_FAILED 0x11 +#define INTR_ASYNC_EVENT 0x12 +#define INTR_RSP_QUE_UPDATE 0x13 +#define INTR_RSP_QUE_UPDATE_83XX 0x14 +#define INTR_ATIO_QUE_UPDATE 0x1C +#define INTR_ATIO_RSP_QUE_UPDATE 0x1D + /* ISP mailbox loopback echo diagnostic error code */ #define MBS_LB_RESET 0x17 /* @@ -718,6 +810,7 @@ typedef struct { #define MBC_SEND_RNFT_ELS 0x5e /* Send RNFT ELS request */ #define MBC_GET_LINK_PRIV_STATS 0x6d /* Get link & private data. */ #define MBC_SET_VENDOR_ID 0x76 /* Set Vendor ID. */ +#define MBC_PORT_RESET 0x120 /* Port Reset */ #define MBC_SET_PORT_CONFIG 0x122 /* Set port configuration */ #define MBC_GET_PORT_CONFIG 0x123 /* Get port configuration */ @@ -1375,9 +1468,10 @@ typedef struct { } cont_a64_entry_t; #define PO_MODE_DIF_INSERT 0 -#define PO_MODE_DIF_REMOVE BIT_0 -#define PO_MODE_DIF_PASS BIT_1 -#define PO_MODE_DIF_REPLACE (BIT_0 + BIT_1) +#define PO_MODE_DIF_REMOVE 1 +#define PO_MODE_DIF_PASS 2 +#define PO_MODE_DIF_REPLACE 3 +#define PO_MODE_DIF_TCP_CKSUM 6 #define PO_ENABLE_DIF_BUNDLING BIT_8 #define PO_ENABLE_INCR_GUARD_SEED BIT_3 #define PO_DISABLE_INCR_REF_TAG BIT_5 @@ -1509,6 +1603,13 @@ typedef struct { #define CS_RETRY 0x82 /* Driver defined */ #define CS_LOOP_DOWN_ABORT 0x83 /* Driver defined */ +#define CS_BIDIR_RD_OVERRUN 0x700 +#define CS_BIDIR_RD_WR_OVERRUN 0x707 +#define CS_BIDIR_RD_OVERRUN_WR_UNDERRUN 0x715 +#define CS_BIDIR_RD_UNDERRUN 0x1500 +#define CS_BIDIR_RD_UNDERRUN_WR_OVERRUN 0x1507 +#define CS_BIDIR_RD_WR_UNDERRUN 0x1515 +#define CS_BIDIR_DMA 0x200 /* * Status entry status flags */ @@ -2373,6 +2474,11 @@ struct qla_statistics { uint64_t output_bytes; }; +struct bidi_statistics { + unsigned long long io_count; + unsigned long long transfer_bytes; +}; + /* Multi queue support */ #define MBC_INITIALIZE_MULTIQ 0x1f #define QLA_QUE_PAGE 0X1000 @@ -2509,14 +2615,16 @@ struct qla_hw_data { uint32_t disable_msix_handshake :1; uint32_t fcp_prio_enabled :1; uint32_t isp82xx_fw_hung:1; + uint32_t nic_core_hung:1; uint32_t quiesce_owner:1; uint32_t thermal_supported:1; - uint32_t isp82xx_reset_hdlr_active:1; - uint32_t isp82xx_reset_owner:1; + uint32_t nic_core_reset_hdlr_active:1; + uint32_t nic_core_reset_owner:1; uint32_t isp82xx_no_md_cap:1; uint32_t host_shutting_down:1; - /* 30 bits */ + uint32_t idc_compl_status:1; + /* 32 bits */ } flags; /* This spinlock is used to protect "io transactions", you must @@ -2670,6 +2778,16 @@ struct qla_hw_data { #define HAS_EXTENDED_IDS(ha) ((ha)->device_type & DT_EXTENDED_IDS) #define IS_CT6_SUPPORTED(ha) ((ha)->device_type & DT_CT6_SUPPORTED) #define IS_MQUE_CAPABLE(ha) ((ha)->mqenable || IS_QLA83XX(ha)) +#define IS_BIDI_CAPABLE(ha) ((IS_QLA25XX(ha) || IS_QLA2031(ha))) +/* Bit 21 of fw_attributes decides the MCTP capabilities */ +#define IS_MCTP_CAPABLE(ha) (IS_QLA2031(ha) && \ + ((ha)->fw_attributes_ext[0] & BIT_0)) +#define IS_PI_UNINIT_CAPABLE(ha) (IS_QLA83XX(ha)) +#define IS_PI_IPGUARD_CAPABLE(ha) (IS_QLA83XX(ha)) +#define IS_PI_DIFB_DIX0_CAPABLE(ha) (0) +#define IS_PI_SPLIT_DET_CAPABLE_HBA(ha) (IS_QLA83XX(ha)) +#define IS_PI_SPLIT_DET_CAPABLE(ha) (IS_PI_SPLIT_DET_CAPABLE_HBA(ha) && \ + (((ha)->fw_attributes_h << 16 | (ha)->fw_attributes) & BIT_22)) /* HBA serial number */ uint8_t serial0; @@ -2753,6 +2871,7 @@ struct qla_hw_data { struct completion mbx_intr_comp; /* Used for completion notification */ struct completion dcbx_comp; /* For set port config notification */ int notify_dcbx_comp; + struct mutex selflogin_lock; /* Basic firmware related information. */ uint16_t fw_major_version; @@ -2784,7 +2903,12 @@ struct qla_hw_data { int fw_dump_reading; dma_addr_t eft_dma; void *eft; - +/* Current size of mctp dump is 0x086064 bytes */ +#define MCTP_DUMP_SIZE 0x086064 + dma_addr_t mctp_dump_dma; + void *mctp_dump; + int mctp_dumped; + int mctp_dump_reading; uint32_t chain_offset; struct dentry *dfs_dir; struct dentry *dfs_fce; @@ -2896,8 +3020,8 @@ struct qla_hw_data { unsigned long mn_win_crb; unsigned long ms_win_crb; int qdr_sn_window; - uint32_t nx_dev_init_timeout; - uint32_t nx_reset_timeout; + uint32_t fcoe_dev_init_timeout; + uint32_t fcoe_reset_timeout; rwlock_t hw_lock; uint16_t portnum; /* port number */ int link_width; @@ -2918,6 +3042,20 @@ struct qla_hw_data { void *md_dump; uint32_t md_dump_size; + void *loop_id_map; + + /* QLA83XX IDC specific fields */ + uint32_t idc_audit_ts; + + /* DPC low-priority workqueue */ + struct workqueue_struct *dpc_lp_wq; + struct work_struct idc_aen; + /* DPC high-priority workqueue */ + struct workqueue_struct *dpc_hp_wq; + struct work_struct nic_core_reset; + struct work_struct idc_state_handler; + struct work_struct nic_core_unrecoverable; + struct qlt_hw_data tgt; }; @@ -2985,6 +3123,13 @@ typedef struct scsi_qla_host { /* ISP configuration data. */ uint16_t loop_id; /* Host adapter loop id */ + uint16_t self_login_loop_id; /* host adapter loop id + * get it on self login + */ + fc_port_t bidir_fcport; /* fcport used for bidir cmnds + * no need of allocating it for + * each command + */ port_id_t d_id; /* Host adapter port id */ uint8_t marker_needed; @@ -3038,6 +3183,7 @@ typedef struct scsi_qla_host { int seconds_since_last_heartbeat; struct fc_host_statistics fc_host_stat; struct qla_statistics qla_stats; + struct bidi_statistics bidi_stats; atomic_t vref_count; } scsi_qla_host_t; diff --git a/drivers/scsi/qla2xxx/qla_dfs.c b/drivers/scsi/qla2xxx/qla_dfs.c index 499c74e..706c4f7 100644 --- a/drivers/scsi/qla2xxx/qla_dfs.c +++ b/drivers/scsi/qla2xxx/qla_dfs.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla2xxx/qla_fw.h b/drivers/scsi/qla2xxx/qla_fw.h index 6d7d775..59524aa 100644 --- a/drivers/scsi/qla2xxx/qla_fw.h +++ b/drivers/scsi/qla2xxx/qla_fw.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -381,6 +381,44 @@ struct init_cb_24xx { /* * ISP queue - command entry structure definition. */ +#define COMMAND_BIDIRECTIONAL 0x75 +struct cmd_bidir { + uint8_t entry_type; /* Entry type. */ + uint8_t entry_count; /* Entry count. */ + uint8_t sys_define; /* System defined */ + uint8_t entry_status; /* Entry status. */ + + uint32_t handle; /* System handle. */ + + uint16_t nport_handle; /* N_PORT hanlde. */ + + uint16_t timeout; /* Commnad timeout. */ + + uint16_t wr_dseg_count; /* Write Data segment count. */ + uint16_t rd_dseg_count; /* Read Data segment count. */ + + struct scsi_lun lun; /* FCP LUN (BE). */ + + uint16_t control_flags; /* Control flags. */ +#define BD_WRAP_BACK BIT_3 +#define BD_READ_DATA BIT_1 +#define BD_WRITE_DATA BIT_0 + + uint16_t fcp_cmnd_dseg_len; /* Data segment length. */ + uint32_t fcp_cmnd_dseg_address[2]; /* Data segment address. */ + + uint16_t reserved[2]; /* Reserved */ + + uint32_t rd_byte_count; /* Total Byte count Read. */ + uint32_t wr_byte_count; /* Total Byte count write. */ + + uint8_t port_id[3]; /* PortID of destination port.*/ + uint8_t vp_index; + + uint32_t fcp_data_dseg_address[2]; /* Data segment address. */ + uint16_t fcp_data_dseg_len; /* Data segment length. */ +}; + #define COMMAND_TYPE_6 0x48 /* Command Type 6 entry */ struct cmd_type_6 { uint8_t entry_type; /* Entry type. */ @@ -1130,7 +1168,7 @@ struct mid_db_entry_24xx { /* * Virtual Port Control IOCB */ -#define VP_CTRL_IOCB_TYPE 0x30 /* Vitual Port Control entry. */ +#define VP_CTRL_IOCB_TYPE 0x30 /* Virtual Port Control entry. */ struct vp_ctrl_entry_24xx { uint8_t entry_type; /* Entry type. */ uint8_t entry_count; /* Entry count. */ @@ -1166,7 +1204,7 @@ struct vp_ctrl_entry_24xx { /* * Modify Virtual Port Configuration IOCB */ -#define VP_CONFIG_IOCB_TYPE 0x31 /* Vitual Port Config entry. */ +#define VP_CONFIG_IOCB_TYPE 0x31 /* Virtual Port Config entry. */ struct vp_config_entry_24xx { uint8_t entry_type; /* Entry type. */ uint8_t entry_count; /* Entry count. */ @@ -1502,7 +1540,10 @@ struct access_chip_rsp_84xx { /* * ISP83xx mailbox commands */ -#define MBC_WRITE_REMOTE_REG 0x0001 /* Write remote register */ +#define MBC_WRITE_REMOTE_REG 0x0001 /* Write remote register */ +#define MBC_READ_REMOTE_REG 0x0009 /* Read remote register */ +#define MBC_RESTART_NIC_FIRMWARE 0x003d /* Restart NIC firmware */ +#define MBC_SET_ACCESS_CONTROL 0x003e /* Access control command */ /* Flash access control option field bit definitions */ #define FAC_OPT_FORCE_SEMAPHORE BIT_15 diff --git a/drivers/scsi/qla2xxx/qla_gbl.h b/drivers/scsi/qla2xxx/qla_gbl.h index 9eacd2d..6acb397 100644 --- a/drivers/scsi/qla2xxx/qla_gbl.h +++ b/drivers/scsi/qla2xxx/qla_gbl.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -48,7 +48,7 @@ extern void qla2x00_update_fcports(scsi_qla_host_t *); extern int qla2x00_abort_isp(scsi_qla_host_t *); extern void qla2x00_abort_isp_cleanup(scsi_qla_host_t *); -extern void qla82xx_quiescent_state_cleanup(scsi_qla_host_t *); +extern void qla2x00_quiesce_io(scsi_qla_host_t *); extern void qla2x00_update_fcport(scsi_qla_host_t *, fc_port_t *); @@ -76,6 +76,14 @@ extern int qla24xx_update_fcport_fcp_prio(scsi_qla_host_t *, fc_port_t *); extern fc_port_t * qla2x00_alloc_fcport(scsi_qla_host_t *, gfp_t ); + +extern int __qla83xx_set_idc_control(scsi_qla_host_t *, uint32_t); +extern int __qla83xx_get_idc_control(scsi_qla_host_t *, uint32_t *); +extern void qla83xx_idc_audit(scsi_qla_host_t *, int); +extern int qla83xx_nic_core_reset(scsi_qla_host_t *); +extern void qla83xx_reset_ownership(scsi_qla_host_t *); +extern int qla2xxx_mctp_dump(scsi_qla_host_t *); + /* * Global Data in qla_os.c source file. */ @@ -133,6 +141,20 @@ extern void qla2x00_relogin(struct scsi_qla_host *); extern void qla2x00_do_work(struct scsi_qla_host *); extern void qla2x00_free_fcports(struct scsi_qla_host *); +extern void qla83xx_schedule_work(scsi_qla_host_t *, int); +extern void qla83xx_service_idc_aen(struct work_struct *); +extern void qla83xx_nic_core_unrecoverable_work(struct work_struct *); +extern void qla83xx_idc_state_handler_work(struct work_struct *); +extern void qla83xx_nic_core_reset_work(struct work_struct *); + +extern void qla83xx_idc_lock(scsi_qla_host_t *, uint16_t); +extern void qla83xx_idc_unlock(scsi_qla_host_t *, uint16_t); +extern int qla83xx_idc_state_handler(scsi_qla_host_t *); +extern int qla83xx_set_drv_presence(scsi_qla_host_t *vha); +extern int __qla83xx_set_drv_presence(scsi_qla_host_t *vha); +extern int qla83xx_clear_drv_presence(scsi_qla_host_t *vha); +extern int __qla83xx_clear_drv_presence(scsi_qla_host_t *vha); + /* * Global Functions in qla_mid.c source file. */ @@ -188,6 +210,8 @@ extern int qla2x00_start_sp(srb_t *); extern uint16_t qla24xx_calc_iocbs(scsi_qla_host_t *, uint16_t); extern void qla24xx_build_scsi_iocbs(srb_t *, struct cmd_type_7 *, uint16_t); extern int qla24xx_dif_start_scsi(srb_t *); +extern int qla2x00_start_bidir(srb_t *, struct scsi_qla_host *, uint32_t); +extern unsigned long qla2x00_get_async_timeout(struct scsi_qla_host *); extern void *qla2x00_alloc_iocbs(scsi_qla_host_t *, srb_t *); extern int qla2x00_issue_marker(scsi_qla_host_t *, int); @@ -376,6 +400,9 @@ qla81xx_set_port_config(scsi_qla_host_t *, uint16_t *); extern int qla2x00_port_logout(scsi_qla_host_t *, struct fc_port *); +extern int +qla2x00_dump_mctp_data(scsi_qla_host_t *, dma_addr_t, uint32_t, uint32_t); + /* * Global Function Prototypes in qla_isr.c source file. */ @@ -419,7 +446,11 @@ extern void qla24xx_beacon_blink(struct scsi_qla_host *); extern void qla83xx_beacon_blink(struct scsi_qla_host *); extern int qla82xx_beacon_on(struct scsi_qla_host *); extern int qla82xx_beacon_off(struct scsi_qla_host *); -extern int qla83xx_write_remote_reg(struct scsi_qla_host *, uint32_t, uint32_t); +extern int qla83xx_wr_reg(scsi_qla_host_t *, uint32_t, uint32_t); +extern int qla83xx_rd_reg(scsi_qla_host_t *, uint32_t, uint32_t *); +extern int qla83xx_restart_nic_firmware(scsi_qla_host_t *); +extern int qla83xx_access_control(scsi_qla_host_t *, uint16_t, uint32_t, + uint32_t, uint16_t *); extern uint8_t *qla2x00_read_optrom_data(struct scsi_qla_host *, uint8_t *, uint32_t, uint32_t); @@ -527,7 +558,6 @@ extern void qla24xx_wrt_rsp_reg(struct qla_hw_data *, uint16_t, uint16_t); /* PCI related functions */ extern int qla82xx_pci_config(struct scsi_qla_host *); extern int qla82xx_pci_mem_read_2M(struct qla_hw_data *, u64, void *, int); -extern char *qla82xx_pci_info_str(struct scsi_qla_host *, char *); extern int qla82xx_pci_region_offset(struct pci_dev *, int); extern int qla82xx_iospace_config(struct qla_hw_data *); @@ -580,6 +610,7 @@ extern uint32_t qla82xx_wait_for_state_change(scsi_qla_host_t *, uint32_t); extern int qla82xx_idc_lock(struct qla_hw_data *); extern void qla82xx_idc_unlock(struct qla_hw_data *); extern int qla82xx_device_state_handler(scsi_qla_host_t *); +extern void qla8xxx_dev_failed_handler(scsi_qla_host_t *); extern void qla82xx_clear_qsnt_ready(scsi_qla_host_t *); extern void qla2x00_set_model_info(scsi_qla_host_t *, uint8_t *, diff --git a/drivers/scsi/qla2xxx/qla_gs.c b/drivers/scsi/qla2xxx/qla_gs.c index 05260d2..f4e4bd7 100644 --- a/drivers/scsi/qla2xxx/qla_gs.c +++ b/drivers/scsi/qla2xxx/qla_gs.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -1131,7 +1131,7 @@ qla2x00_mgmt_svr_login(scsi_qla_host_t *vha) return ret; rval = ha->isp_ops->fabric_login(vha, vha->mgmt_svr_loop_id, 0xff, 0xff, - 0xfa, mb, BIT_1|BIT_0); + 0xfa, mb, BIT_1); if (rval != QLA_SUCCESS || mb[0] != MBS_COMMAND_COMPLETE) { if (rval == QLA_MEMORY_ALLOC_FAILED) ql_dbg(ql_dbg_disc, vha, 0x2085, diff --git a/drivers/scsi/qla2xxx/qla_init.c b/drivers/scsi/qla2xxx/qla_init.c index a44653b..799a58b 100644 --- a/drivers/scsi/qla2xxx/qla_init.c +++ b/drivers/scsi/qla2xxx/qla_init.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -77,7 +77,7 @@ qla2x00_sp_free(void *data, void *ptr) /* Asynchronous Login/Logout Routines -------------------------------------- */ -static inline unsigned long +unsigned long qla2x00_get_async_timeout(struct scsi_qla_host *vha) { unsigned long tmo; @@ -429,6 +429,79 @@ qla2x00_async_adisc_done(struct scsi_qla_host *vha, fc_port_t *fcport, /* QLogic ISP2x00 Hardware Support Functions. */ /****************************************************************************/ +int +qla83xx_nic_core_fw_load(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + struct qla_hw_data *ha = vha->hw; + uint32_t idc_major_ver, idc_minor_ver; + uint16_t config[4]; + + qla83xx_idc_lock(vha, 0); + + /* SV: TODO: Assign initialization timeout from + * flash-info / other param + */ + ha->fcoe_dev_init_timeout = QLA83XX_IDC_INITIALIZATION_TIMEOUT; + ha->fcoe_reset_timeout = QLA83XX_IDC_RESET_ACK_TIMEOUT; + + /* Set our fcoe function presence */ + if (__qla83xx_set_drv_presence(vha) != QLA_SUCCESS) { + ql_dbg(ql_dbg_p3p, vha, 0xb077, + "Error while setting DRV-Presence.\n"); + rval = QLA_FUNCTION_FAILED; + goto exit; + } + + /* Decide the reset ownership */ + qla83xx_reset_ownership(vha); + + /* + * On first protocol driver load: + * Init-Owner: Set IDC-Major-Version and Clear IDC-Lock-Recovery + * register. + * Others: Check compatibility with current IDC Major version. + */ + qla83xx_rd_reg(vha, QLA83XX_IDC_MAJOR_VERSION, &idc_major_ver); + if (ha->flags.nic_core_reset_owner) { + /* Set IDC Major version */ + idc_major_ver = QLA83XX_SUPP_IDC_MAJOR_VERSION; + qla83xx_wr_reg(vha, QLA83XX_IDC_MAJOR_VERSION, idc_major_ver); + + /* Clearing IDC-Lock-Recovery register */ + qla83xx_wr_reg(vha, QLA83XX_IDC_LOCK_RECOVERY, 0); + } else if (idc_major_ver != QLA83XX_SUPP_IDC_MAJOR_VERSION) { + /* + * Clear further IDC participation if we are not compatible with + * the current IDC Major Version. + */ + ql_log(ql_log_warn, vha, 0xb07d, + "Failing load, idc_major_ver=%d, expected_major_ver=%d.\n", + idc_major_ver, QLA83XX_SUPP_IDC_MAJOR_VERSION); + __qla83xx_clear_drv_presence(vha); + rval = QLA_FUNCTION_FAILED; + goto exit; + } + /* Each function sets its supported Minor version. */ + qla83xx_rd_reg(vha, QLA83XX_IDC_MINOR_VERSION, &idc_minor_ver); + idc_minor_ver |= (QLA83XX_SUPP_IDC_MINOR_VERSION << (ha->portnum * 2)); + qla83xx_wr_reg(vha, QLA83XX_IDC_MINOR_VERSION, idc_minor_ver); + + if (ha->flags.nic_core_reset_owner) { + memset(config, 0, sizeof(config)); + if (!qla81xx_get_port_config(vha, config)) + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, + QLA8XXX_DEV_READY); + } + + rval = qla83xx_idc_state_handler(vha); + +exit: + qla83xx_idc_unlock(vha, 0); + + return rval; +} + /* * qla2x00_initialize_adapter * Initialize board. @@ -537,6 +610,14 @@ qla2x00_initialize_adapter(scsi_qla_host_t *vha) } } + /* Load the NIC Core f/w if we are the first protocol driver. */ + if (IS_QLA8031(ha)) { + rval = qla83xx_nic_core_fw_load(vha); + if (rval) + ql_log(ql_log_warn, vha, 0x0124, + "Error in initializing NIC Core f/w.\n"); + } + if (IS_QLA24XX_TYPE(ha) || IS_QLA25XX(ha)) qla24xx_read_fcp_prio_cfg(vha); @@ -686,7 +767,7 @@ qla24xx_pci_config(scsi_qla_host_t *vha) /* PCIe -- adjust Maximum Read Request Size (2048). */ if (pci_is_pcie(ha->pdev)) - pcie_set_readrq(ha->pdev, 2048); + pcie_set_readrq(ha->pdev, 4096); pci_disable_rom(ha->pdev); @@ -722,7 +803,7 @@ qla25xx_pci_config(scsi_qla_host_t *vha) /* PCIe -- adjust Maximum Read Request Size (2048). */ if (pci_is_pcie(ha->pdev)) - pcie_set_readrq(ha->pdev, 2048); + pcie_set_readrq(ha->pdev, 4096); pci_disable_rom(ha->pdev); @@ -1480,7 +1561,8 @@ enable_82xx_npiv: "ISP Firmware failed checksum.\n"); goto failed; } - } + } else + goto failed; if (!IS_FWI2_CAPABLE(ha) && !IS_QLA2100(ha) && !IS_QLA2200(ha)) { /* Enable proper parity. */ @@ -1825,7 +1907,7 @@ qla2x00_init_rings(scsi_qla_host_t *vha) ql_dbg(ql_dbg_init, vha, 0x00d1, "Issue init firmware.\n"); if (ha->flags.npiv_supported) { - if (ha->operating_mode == LOOP) + if (ha->operating_mode == LOOP && !IS_CNA_CAPABLE(ha)) ha->max_npiv_vports = MIN_MULTI_ID_FABRIC - 1; mid_init_cb->count = cpu_to_le16(ha->max_npiv_vports); } @@ -2682,11 +2764,6 @@ qla2x00_configure_local_loop(scsi_qla_host_t *vha) new_fcport = NULL; entries = MAX_FIBRE_DEVICES_LOOP; - ql_dbg(ql_dbg_disc, vha, 0x2016, - "Getting FCAL position map.\n"); - if (ql2xextended_error_logging & ql_dbg_disc) - qla2x00_get_fcal_position_map(vha, NULL); - /* Get list of logged in devices. */ memset(ha->gid_list, 0, qla2x00_gid_list_size(ha)); rval = qla2x00_get_id_list(vha, ha->gid_list, ha->gid_list_dma, @@ -2753,6 +2830,8 @@ qla2x00_configure_local_loop(scsi_qla_host_t *vha) if (loop_id > LAST_LOCAL_LOOP_ID) continue; + memset(new_fcport, 0, sizeof(fc_port_t)); + /* Fill in member data. */ new_fcport->d_id.b.domain = domain; new_fcport->d_id.b.area = area; @@ -3285,7 +3364,7 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha, */ if ((fcport->flags & FCF_FABRIC_DEVICE) == 0) { fcport->d_id.b24 = new_fcport->d_id.b24; - fcport->loop_id = FC_NO_LOOP_ID; + qla2x00_clear_loop_id(fcport); fcport->flags |= (FCF_FABRIC_DEVICE | FCF_LOGIN_NEEDED); break; @@ -3306,7 +3385,7 @@ qla2x00_find_all_fabric_devs(scsi_qla_host_t *vha, ha->isp_ops->fabric_logout(vha, fcport->loop_id, fcport->d_id.b.domain, fcport->d_id.b.area, fcport->d_id.b.al_pa); - fcport->loop_id = FC_NO_LOOP_ID; + qla2x00_clear_loop_id(fcport); } break; @@ -3352,71 +3431,32 @@ int qla2x00_find_new_loop_id(scsi_qla_host_t *vha, fc_port_t *dev) { int rval; - int found; - fc_port_t *fcport; - uint16_t first_loop_id; struct qla_hw_data *ha = vha->hw; - struct scsi_qla_host *vp; - struct scsi_qla_host *tvp; unsigned long flags = 0; rval = QLA_SUCCESS; - /* Save starting loop ID. */ - first_loop_id = dev->loop_id; - - for (;;) { - /* Skip loop ID if already used by adapter. */ - if (dev->loop_id == vha->loop_id) - dev->loop_id++; - - /* Skip reserved loop IDs. */ - while (qla2x00_is_reserved_id(vha, dev->loop_id)) - dev->loop_id++; - - /* Reset loop ID if passed the end. */ - if (dev->loop_id > ha->max_loop_id) { - /* first loop ID. */ - dev->loop_id = ha->min_external_loopid; - } - - /* Check for loop ID being already in use. */ - found = 0; - fcport = NULL; - - spin_lock_irqsave(&ha->vport_slock, flags); - list_for_each_entry_safe(vp, tvp, &ha->vp_list, list) { - list_for_each_entry(fcport, &vp->vp_fcports, list) { - if (fcport->loop_id == dev->loop_id && - fcport != dev) { - /* ID possibly in use */ - found++; - break; - } - } - if (found) - break; - } - spin_unlock_irqrestore(&ha->vport_slock, flags); + spin_lock_irqsave(&ha->vport_slock, flags); - /* If not in use then it is free to use. */ - if (!found) { - ql_dbg(ql_dbg_disc, dev->vha, 0x2086, - "Assigning new loopid=%x, portid=%x.\n", - dev->loop_id, dev->d_id.b24); - break; - } + dev->loop_id = find_first_zero_bit(ha->loop_id_map, + LOOPID_MAP_SIZE); + if (dev->loop_id >= LOOPID_MAP_SIZE || + qla2x00_is_reserved_id(vha, dev->loop_id)) { + dev->loop_id = FC_NO_LOOP_ID; + rval = QLA_FUNCTION_FAILED; + } else + set_bit(dev->loop_id, ha->loop_id_map); - /* ID in use. Try next value. */ - dev->loop_id++; + spin_unlock_irqrestore(&ha->vport_slock, flags); - /* If wrap around. No free ID to use. */ - if (dev->loop_id == first_loop_id) { - dev->loop_id = FC_NO_LOOP_ID; - rval = QLA_FUNCTION_FAILED; - break; - } - } + if (rval == QLA_SUCCESS) + ql_dbg(ql_dbg_disc, dev->vha, 0x2086, + "Assigning new loopid=%x, portid=%x.\n", + dev->loop_id, dev->d_id.b24); + else + ql_log(ql_log_warn, dev->vha, 0x2087, + "No loop_id's available, portid=%x.\n", + dev->d_id.b24); return (rval); } @@ -3616,7 +3656,7 @@ qla2x00_fabric_login(scsi_qla_host_t *vha, fc_port_t *fcport, ha->isp_ops->fabric_logout(vha, fcport->loop_id, fcport->d_id.b.domain, fcport->d_id.b.area, fcport->d_id.b.al_pa); - fcport->loop_id = FC_NO_LOOP_ID; + qla2x00_clear_loop_id(fcport); fcport->login_retry = 0; rval = 3; @@ -3775,8 +3815,363 @@ qla2x00_update_fcports(scsi_qla_host_t *base_vha) spin_unlock_irqrestore(&ha->vport_slock, flags); } +/* Assumes idc_lock always held on entry */ +void +qla83xx_reset_ownership(scsi_qla_host_t *vha) +{ + struct qla_hw_data *ha = vha->hw; + uint32_t drv_presence, drv_presence_mask; + uint32_t dev_part_info1, dev_part_info2, class_type; + uint32_t class_type_mask = 0x3; + uint16_t fcoe_other_function = 0xffff, i; + + qla83xx_rd_reg(vha, QLA83XX_IDC_DRV_PRESENCE, &drv_presence); + + qla83xx_rd_reg(vha, QLA83XX_DEV_PARTINFO1, &dev_part_info1); + qla83xx_rd_reg(vha, QLA83XX_DEV_PARTINFO2, &dev_part_info2); + for (i = 0; i < 8; i++) { + class_type = ((dev_part_info1 >> (i * 4)) & class_type_mask); + if ((class_type == QLA83XX_CLASS_TYPE_FCOE) && + (i != ha->portnum)) { + fcoe_other_function = i; + break; + } + } + if (fcoe_other_function == 0xffff) { + for (i = 0; i < 8; i++) { + class_type = ((dev_part_info2 >> (i * 4)) & + class_type_mask); + if ((class_type == QLA83XX_CLASS_TYPE_FCOE) && + ((i + 8) != ha->portnum)) { + fcoe_other_function = i + 8; + break; + } + } + } + /* + * Prepare drv-presence mask based on fcoe functions present. + * However consider only valid physical fcoe function numbers (0-15). + */ + drv_presence_mask = ~((1 << (ha->portnum)) | + ((fcoe_other_function == 0xffff) ? + 0 : (1 << (fcoe_other_function)))); + + /* We are the reset owner iff: + * - No other protocol drivers present. + * - This is the lowest among fcoe functions. */ + if (!(drv_presence & drv_presence_mask) && + (ha->portnum < fcoe_other_function)) { + ql_dbg(ql_dbg_p3p, vha, 0xb07f, + "This host is Reset owner.\n"); + ha->flags.nic_core_reset_owner = 1; + } +} + +int +__qla83xx_set_drv_ack(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + struct qla_hw_data *ha = vha->hw; + uint32_t drv_ack; + + rval = qla83xx_rd_reg(vha, QLA83XX_IDC_DRIVER_ACK, &drv_ack); + if (rval == QLA_SUCCESS) { + drv_ack |= (1 << ha->portnum); + rval = qla83xx_wr_reg(vha, QLA83XX_IDC_DRIVER_ACK, drv_ack); + } + + return rval; +} + +int +qla83xx_set_drv_ack(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + + qla83xx_idc_lock(vha, 0); + rval = __qla83xx_set_drv_ack(vha); + qla83xx_idc_unlock(vha, 0); + + return rval; +} + +int +__qla83xx_clear_drv_ack(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + struct qla_hw_data *ha = vha->hw; + uint32_t drv_ack; + + rval = qla83xx_rd_reg(vha, QLA83XX_IDC_DRIVER_ACK, &drv_ack); + if (rval == QLA_SUCCESS) { + drv_ack &= ~(1 << ha->portnum); + rval = qla83xx_wr_reg(vha, QLA83XX_IDC_DRIVER_ACK, drv_ack); + } + + return rval; +} + +int +qla83xx_clear_drv_ack(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + + qla83xx_idc_lock(vha, 0); + rval = __qla83xx_clear_drv_ack(vha); + qla83xx_idc_unlock(vha, 0); + + return rval; +} + +const char * +qla83xx_dev_state_to_string(uint32_t dev_state) +{ + switch (dev_state) { + case QLA8XXX_DEV_COLD: + return "COLD/RE-INIT"; + case QLA8XXX_DEV_INITIALIZING: + return "INITIALIZING"; + case QLA8XXX_DEV_READY: + return "READY"; + case QLA8XXX_DEV_NEED_RESET: + return "NEED RESET"; + case QLA8XXX_DEV_NEED_QUIESCENT: + return "NEED QUIESCENT"; + case QLA8XXX_DEV_FAILED: + return "FAILED"; + case QLA8XXX_DEV_QUIESCENT: + return "QUIESCENT"; + default: + return "Unknown"; + } +} + +/* Assumes idc-lock always held on entry */ +void +qla83xx_idc_audit(scsi_qla_host_t *vha, int audit_type) +{ + struct qla_hw_data *ha = vha->hw; + uint32_t idc_audit_reg = 0, duration_secs = 0; + + switch (audit_type) { + case IDC_AUDIT_TIMESTAMP: + ha->idc_audit_ts = (jiffies_to_msecs(jiffies) / 1000); + idc_audit_reg = (ha->portnum) | + (IDC_AUDIT_TIMESTAMP << 7) | (ha->idc_audit_ts << 8); + qla83xx_wr_reg(vha, QLA83XX_IDC_AUDIT, idc_audit_reg); + break; + + case IDC_AUDIT_COMPLETION: + duration_secs = ((jiffies_to_msecs(jiffies) - + jiffies_to_msecs(ha->idc_audit_ts)) / 1000); + idc_audit_reg = (ha->portnum) | + (IDC_AUDIT_COMPLETION << 7) | (duration_secs << 8); + qla83xx_wr_reg(vha, QLA83XX_IDC_AUDIT, idc_audit_reg); + break; + + default: + ql_log(ql_log_warn, vha, 0xb078, + "Invalid audit type specified.\n"); + break; + } +} + +/* Assumes idc_lock always held on entry */ +int +qla83xx_initiating_reset(scsi_qla_host_t *vha) +{ + struct qla_hw_data *ha = vha->hw; + uint32_t idc_control, dev_state; + + __qla83xx_get_idc_control(vha, &idc_control); + if ((idc_control & QLA83XX_IDC_RESET_DISABLED)) { + ql_log(ql_log_info, vha, 0xb080, + "NIC Core reset has been disabled. idc-control=0x%x\n", + idc_control); + return QLA_FUNCTION_FAILED; + } + + /* Set NEED-RESET iff in READY state and we are the reset-owner */ + qla83xx_rd_reg(vha, QLA83XX_IDC_DEV_STATE, &dev_state); + if (ha->flags.nic_core_reset_owner && dev_state == QLA8XXX_DEV_READY) { + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, + QLA8XXX_DEV_NEED_RESET); + ql_log(ql_log_info, vha, 0xb056, "HW State: NEED RESET.\n"); + qla83xx_idc_audit(vha, IDC_AUDIT_TIMESTAMP); + } else { + const char *state = qla83xx_dev_state_to_string(dev_state); + ql_log(ql_log_info, vha, 0xb057, "HW State: %s.\n", state); + + /* SV: XXX: Is timeout required here? */ + /* Wait for IDC state change READY -> NEED_RESET */ + while (dev_state == QLA8XXX_DEV_READY) { + qla83xx_idc_unlock(vha, 0); + msleep(200); + qla83xx_idc_lock(vha, 0); + qla83xx_rd_reg(vha, QLA83XX_IDC_DEV_STATE, &dev_state); + } + } + + /* Send IDC ack by writing to drv-ack register */ + __qla83xx_set_drv_ack(vha); + + return QLA_SUCCESS; +} + +int +__qla83xx_set_idc_control(scsi_qla_host_t *vha, uint32_t idc_control) +{ + return qla83xx_wr_reg(vha, QLA83XX_IDC_CONTROL, idc_control); +} + +int +qla83xx_set_idc_control(scsi_qla_host_t *vha, uint32_t idc_control) +{ + int rval = QLA_SUCCESS; + + qla83xx_idc_lock(vha, 0); + rval = __qla83xx_set_idc_control(vha, idc_control); + qla83xx_idc_unlock(vha, 0); + + return rval; +} + +int +__qla83xx_get_idc_control(scsi_qla_host_t *vha, uint32_t *idc_control) +{ + return qla83xx_rd_reg(vha, QLA83XX_IDC_CONTROL, idc_control); +} + +int +qla83xx_get_idc_control(scsi_qla_host_t *vha, uint32_t *idc_control) +{ + int rval = QLA_SUCCESS; + + qla83xx_idc_lock(vha, 0); + rval = __qla83xx_get_idc_control(vha, idc_control); + qla83xx_idc_unlock(vha, 0); + + return rval; +} + +int +qla83xx_check_driver_presence(scsi_qla_host_t *vha) +{ + uint32_t drv_presence = 0; + struct qla_hw_data *ha = vha->hw; + + qla83xx_rd_reg(vha, QLA83XX_IDC_DRV_PRESENCE, &drv_presence); + if (drv_presence & (1 << ha->portnum)) + return QLA_SUCCESS; + else + return QLA_TEST_FAILED; +} + +int +qla83xx_nic_core_reset(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + struct qla_hw_data *ha = vha->hw; + + ql_dbg(ql_dbg_p3p, vha, 0xb058, + "Entered %s().\n", __func__); + + if (vha->device_flags & DFLG_DEV_FAILED) { + ql_log(ql_log_warn, vha, 0xb059, + "Device in unrecoverable FAILED state.\n"); + return QLA_FUNCTION_FAILED; + } + + qla83xx_idc_lock(vha, 0); + + if (qla83xx_check_driver_presence(vha) != QLA_SUCCESS) { + ql_log(ql_log_warn, vha, 0xb05a, + "Function=0x%x has been removed from IDC participation.\n", + ha->portnum); + rval = QLA_FUNCTION_FAILED; + goto exit; + } + + qla83xx_reset_ownership(vha); + + rval = qla83xx_initiating_reset(vha); + + /* + * Perform reset if we are the reset-owner, + * else wait till IDC state changes to READY/FAILED. + */ + if (rval == QLA_SUCCESS) { + rval = qla83xx_idc_state_handler(vha); + + if (rval == QLA_SUCCESS) + ha->flags.nic_core_hung = 0; + __qla83xx_clear_drv_ack(vha); + } + +exit: + qla83xx_idc_unlock(vha, 0); + + ql_dbg(ql_dbg_p3p, vha, 0xb05b, "Exiting %s.\n", __func__); + + return rval; +} + +int +qla2xxx_mctp_dump(scsi_qla_host_t *vha) +{ + struct qla_hw_data *ha = vha->hw; + int rval = QLA_FUNCTION_FAILED; + + if (!IS_MCTP_CAPABLE(ha)) { + /* This message can be removed from the final version */ + ql_log(ql_log_info, vha, 0x506d, + "This board is not MCTP capable\n"); + return rval; + } + + if (!ha->mctp_dump) { + ha->mctp_dump = dma_alloc_coherent(&ha->pdev->dev, + MCTP_DUMP_SIZE, &ha->mctp_dump_dma, GFP_KERNEL); + + if (!ha->mctp_dump) { + ql_log(ql_log_warn, vha, 0x506e, + "Failed to allocate memory for mctp dump\n"); + return rval; + } + } + +#define MCTP_DUMP_STR_ADDR 0x00000000 + rval = qla2x00_dump_mctp_data(vha, ha->mctp_dump_dma, + MCTP_DUMP_STR_ADDR, MCTP_DUMP_SIZE/4); + if (rval != QLA_SUCCESS) { + ql_log(ql_log_warn, vha, 0x506f, + "Failed to capture mctp dump\n"); + } else { + ql_log(ql_log_info, vha, 0x5070, + "Mctp dump capture for host (%ld/%p).\n", + vha->host_no, ha->mctp_dump); + ha->mctp_dumped = 1; + } + + if (!ha->flags.nic_core_reset_hdlr_active && !ha->portnum) { + ha->flags.nic_core_reset_hdlr_active = 1; + rval = qla83xx_restart_nic_firmware(vha); + if (rval) + /* NIC Core reset failed. */ + ql_log(ql_log_warn, vha, 0x5071, + "Failed to restart nic firmware\n"); + else + ql_dbg(ql_dbg_p3p, vha, 0xb084, + "Restarted NIC firmware successfully.\n"); + ha->flags.nic_core_reset_hdlr_active = 0; + } + + return rval; + +} + /* -* qla82xx_quiescent_state_cleanup +* qla2x00_quiesce_io * Description: This function will block the new I/Os * Its not aborting any I/Os as context * is not destroyed during quiescence @@ -3784,20 +4179,20 @@ qla2x00_update_fcports(scsi_qla_host_t *base_vha) * return : void */ void -qla82xx_quiescent_state_cleanup(scsi_qla_host_t *vha) +qla2x00_quiesce_io(scsi_qla_host_t *vha) { struct qla_hw_data *ha = vha->hw; struct scsi_qla_host *vp; - ql_dbg(ql_dbg_p3p, vha, 0xb002, - "Performing ISP error recovery - ha=%p.\n", ha); + ql_dbg(ql_dbg_dpc, vha, 0x401d, + "Quiescing I/O - ha=%p.\n", ha); atomic_set(&ha->loop_down_timer, LOOP_DOWN_TIME); if (atomic_read(&vha->loop_state) != LOOP_DOWN) { atomic_set(&vha->loop_state, LOOP_DOWN); qla2x00_mark_all_devices_lost(vha, 0); list_for_each_entry(vp, &ha->vp_list, list) - qla2x00_mark_all_devices_lost(vha, 0); + qla2x00_mark_all_devices_lost(vp, 0); } else { if (!atomic_read(&vha->loop_down_timer)) atomic_set(&vha->loop_down_timer, @@ -3913,6 +4308,14 @@ qla2x00_abort_isp(scsi_qla_host_t *vha) if (vha->flags.online) { qla2x00_abort_isp_cleanup(vha); + if (IS_QLA8031(ha)) { + ql_dbg(ql_dbg_p3p, vha, 0xb05c, + "Clearing fcoe driver presence.\n"); + if (qla83xx_clear_drv_presence(vha) != QLA_SUCCESS) + ql_dbg(ql_dbg_p3p, vha, 0xb073, + "Error while clearing DRV-Presence.\n"); + } + if (unlikely(pci_channel_offline(ha->pdev) && ha->flags.pci_channel_io_perm_failure)) { clear_bit(ISP_ABORT_RETRY, &vha->dpc_flags); @@ -4021,6 +4424,13 @@ qla2x00_abort_isp(scsi_qla_host_t *vha) } spin_unlock_irqrestore(&ha->vport_slock, flags); + if (IS_QLA8031(ha)) { + ql_dbg(ql_dbg_p3p, vha, 0xb05d, + "Setting back fcoe driver presence.\n"); + if (qla83xx_set_drv_presence(vha) != QLA_SUCCESS) + ql_dbg(ql_dbg_p3p, vha, 0xb074, + "Error while setting DRV-Presence.\n"); + } } else { ql_log(ql_log_warn, vha, 0x8023, "%s **** FAILED ****.\n", __func__); @@ -5088,6 +5498,9 @@ qla81xx_nvram_config(scsi_qla_host_t *vha) rval = 1; } + if (IS_T10_PI_CAPABLE(ha)) + nv->frame_payload_size &= ~7; + /* Reset Initialization control block */ memset(icb, 0, ha->init_cb_size); diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h index 6e45764..c0462c0 100644 --- a/drivers/scsi/qla2xxx/qla_inline.h +++ b/drivers/scsi/qla2xxx/qla_inline.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -57,6 +57,20 @@ host_to_fcp_swap(uint8_t *fcp, uint32_t bsize) return fcp; } +static inline void +qla2x00_set_reserved_loop_ids(struct qla_hw_data *ha) +{ + int i; + + if (IS_FWI2_CAPABLE(ha)) + return; + + for (i = 0; i < SNS_FIRST_LOOP_ID; i++) + set_bit(i, ha->loop_id_map); + set_bit(MANAGEMENT_SERVER, ha->loop_id_map); + set_bit(BROADCAST, ha->loop_id_map); +} + static inline int qla2x00_is_reserved_id(scsi_qla_host_t *vha, uint16_t loop_id) { @@ -69,6 +83,18 @@ qla2x00_is_reserved_id(scsi_qla_host_t *vha, uint16_t loop_id) } static inline void +qla2x00_clear_loop_id(fc_port_t *fcport) { + struct qla_hw_data *ha = fcport->vha->hw; + + if (fcport->loop_id == FC_NO_LOOP_ID || + qla2x00_is_reserved_id(fcport->vha, fcport->loop_id)) + return; + + clear_bit(fcport->loop_id, ha->loop_id_map); + fcport->loop_id = FC_NO_LOOP_ID; +} + +static inline void qla2x00_clean_dsd_pool(struct qla_hw_data *ha, srb_t *sp) { struct dsd_dma *dsd_ptr, *tdsd_ptr; diff --git a/drivers/scsi/qla2xxx/qla_iocb.c b/drivers/scsi/qla2xxx/qla_iocb.c index 70dbf53..03b7526 100644 --- a/drivers/scsi/qla2xxx/qla_iocb.c +++ b/drivers/scsi/qla2xxx/qla_iocb.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -147,13 +147,6 @@ qla24xx_configure_prot_mode(srb_t *sp, uint16_t *fw_prot_opts) struct scsi_cmnd *cmd = GET_CMD_SP(sp); uint8_t guard = scsi_host_get_guard(cmd->device->host); - /* We only support T10 DIF right now */ - if (guard != SHOST_DIX_GUARD_CRC) { - ql_dbg(ql_dbg_io, sp->fcport->vha, 0x3007, - "Unsupported guard: %d for cmd=%p.\n", guard, cmd); - return 0; - } - /* We always use DIFF Bundling for best performance */ *fw_prot_opts = 0; @@ -172,10 +165,11 @@ qla24xx_configure_prot_mode(srb_t *sp, uint16_t *fw_prot_opts) *fw_prot_opts |= PO_MODE_DIF_REMOVE; break; case SCSI_PROT_READ_PASS: - *fw_prot_opts |= PO_MODE_DIF_PASS; - break; case SCSI_PROT_WRITE_PASS: - *fw_prot_opts |= PO_MODE_DIF_PASS; + if (guard & SHOST_DIX_GUARD_IP) + *fw_prot_opts |= PO_MODE_DIF_TCP_CKSUM; + else + *fw_prot_opts |= PO_MODE_DIF_PASS; break; default: /* Normal Request */ *fw_prot_opts |= PO_MODE_DIF_PASS; @@ -821,7 +815,6 @@ qla24xx_set_t10dif_tags(srb_t *sp, struct fw_dif_context *pkt, unsigned int protcnt) { struct scsi_cmnd *cmd = GET_CMD_SP(sp); - scsi_qla_host_t *vha = shost_priv(cmd->device->host); switch (scsi_get_prot_type(cmd)) { case SCSI_PROT_DIF_TYPE0: @@ -891,12 +884,6 @@ qla24xx_set_t10dif_tags(srb_t *sp, struct fw_dif_context *pkt, pkt->ref_tag_mask[3] = 0xff; break; } - - ql_dbg(ql_dbg_io, vha, 0x3009, - "Setting protection Tags: (BIG) ref tag = 0x%x, app tag = 0x%x, " - "prot SG count %d, cmd lba 0x%x, prot_type=%u cmd=%p.\n", - pkt->ref_tag, pkt->app_tag, protcnt, (int)scsi_get_lba(cmd), - scsi_get_prot_type(cmd), cmd); } struct qla2_sgx { @@ -1068,9 +1055,6 @@ qla24xx_walk_and_build_sglist(struct qla_hw_data *ha, srb_t *sp, uint32_t *dsd, int i; uint16_t used_dsds = tot_dsds; struct scsi_cmnd *cmd = GET_CMD_SP(sp); - scsi_qla_host_t *vha = shost_priv(cmd->device->host); - - uint8_t *cp; scsi_for_each_sg(cmd, sg, tot_dsds, i) { dma_addr_t sle_dma; @@ -1113,19 +1097,12 @@ qla24xx_walk_and_build_sglist(struct qla_hw_data *ha, srb_t *sp, uint32_t *dsd, cur_dsd = (uint32_t *)next_dsd; } sle_dma = sg_dma_address(sg); - ql_dbg(ql_dbg_io, vha, 0x300a, - "sg entry %d - addr=0x%x 0x%x, " "len=%d for cmd=%p.\n", - i, LSD(sle_dma), MSD(sle_dma), sg_dma_len(sg), cmd); + *cur_dsd++ = cpu_to_le32(LSD(sle_dma)); *cur_dsd++ = cpu_to_le32(MSD(sle_dma)); *cur_dsd++ = cpu_to_le32(sg_dma_len(sg)); avail_dsds--; - if (scsi_get_prot_op(cmd) == SCSI_PROT_WRITE_PASS) { - cp = page_address(sg_page(sg)) + sg->offset; - ql_dbg(ql_dbg_io, vha, 0x300b, - "User data buffer=%p for cmd=%p.\n", cp, cmd); - } } /* Null termination */ *cur_dsd++ = 0; @@ -1148,8 +1125,6 @@ qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *ha, srb_t *sp, struct scsi_cmnd *cmd; uint32_t *cur_dsd = dsd; uint16_t used_dsds = tot_dsds; - scsi_qla_host_t *vha = pci_get_drvdata(ha->pdev); - uint8_t *cp; cmd = GET_CMD_SP(sp); scsi_for_each_prot_sg(cmd, sg, tot_dsds, i) { @@ -1193,23 +1168,11 @@ qla24xx_walk_and_build_prot_sglist(struct qla_hw_data *ha, srb_t *sp, cur_dsd = (uint32_t *)next_dsd; } sle_dma = sg_dma_address(sg); - if (scsi_get_prot_op(cmd) == SCSI_PROT_WRITE_PASS) { - ql_dbg(ql_dbg_io, vha, 0x3027, - "%s(): %p, sg_entry %d - " - "addr=0x%x0x%x, len=%d.\n", - __func__, cur_dsd, i, - LSD(sle_dma), MSD(sle_dma), sg_dma_len(sg)); - } + *cur_dsd++ = cpu_to_le32(LSD(sle_dma)); *cur_dsd++ = cpu_to_le32(MSD(sle_dma)); *cur_dsd++ = cpu_to_le32(sg_dma_len(sg)); - if (scsi_get_prot_op(cmd) == SCSI_PROT_WRITE_PASS) { - cp = page_address(sg_page(sg)) + sg->offset; - ql_dbg(ql_dbg_io, vha, 0x3028, - "%s(): Protection Data buffer = %p.\n", __func__, - cp); - } avail_dsds--; } /* Null termination */ @@ -1386,6 +1349,16 @@ qla24xx_build_scsi_crc_2_iocbs(srb_t *sp, struct cmd_type_crc_2 *cmd_pkt, if (!qla2x00_hba_err_chk_enabled(sp)) fw_prot_opts |= 0x10; /* Disable Guard tag checking */ + /* HBA error checking enabled */ + else if (IS_PI_UNINIT_CAPABLE(ha)) { + if ((scsi_get_prot_type(GET_CMD_SP(sp)) == SCSI_PROT_DIF_TYPE1) + || (scsi_get_prot_type(GET_CMD_SP(sp)) == + SCSI_PROT_DIF_TYPE2)) + fw_prot_opts |= BIT_10; + else if (scsi_get_prot_type(GET_CMD_SP(sp)) == + SCSI_PROT_DIF_TYPE3) + fw_prot_opts |= BIT_11; + } if (!bundling) { cur_dsd = (uint32_t *) &crc_ctx_pkt->u.nobundling.data_address; @@ -1858,7 +1831,7 @@ qla2x00_alloc_iocbs(scsi_qla_host_t *vha, srb_t *sp) } if (index == MAX_OUTSTANDING_COMMANDS) { ql_log(ql_log_warn, vha, 0x700b, - "No room on oustanding cmd array.\n"); + "No room on outstanding cmd array.\n"); goto queuing_error; } @@ -2665,3 +2638,201 @@ done: spin_unlock_irqrestore(&ha->hardware_lock, flags); return rval; } + +static void +qla25xx_build_bidir_iocb(srb_t *sp, struct scsi_qla_host *vha, + struct cmd_bidir *cmd_pkt, uint32_t tot_dsds) +{ + uint16_t avail_dsds; + uint32_t *cur_dsd; + uint32_t req_data_len = 0; + uint32_t rsp_data_len = 0; + struct scatterlist *sg; + int index; + int entry_count = 1; + struct fc_bsg_job *bsg_job = sp->u.bsg_job; + + /*Update entry type to indicate bidir command */ + *((uint32_t *)(&cmd_pkt->entry_type)) = + __constant_cpu_to_le32(COMMAND_BIDIRECTIONAL); + + /* Set the transfer direction, in this set both flags + * Also set the BD_WRAP_BACK flag, firmware will take care + * assigning DID=SID for outgoing pkts. + */ + cmd_pkt->wr_dseg_count = cpu_to_le16(bsg_job->request_payload.sg_cnt); + cmd_pkt->rd_dseg_count = cpu_to_le16(bsg_job->reply_payload.sg_cnt); + cmd_pkt->control_flags = + __constant_cpu_to_le16(BD_WRITE_DATA | BD_READ_DATA | + BD_WRAP_BACK); + + req_data_len = rsp_data_len = bsg_job->request_payload.payload_len; + cmd_pkt->wr_byte_count = cpu_to_le32(req_data_len); + cmd_pkt->rd_byte_count = cpu_to_le32(rsp_data_len); + cmd_pkt->timeout = cpu_to_le16(qla2x00_get_async_timeout(vha) + 2); + + vha->bidi_stats.transfer_bytes += req_data_len; + vha->bidi_stats.io_count++; + + /* Only one dsd is available for bidirectional IOCB, remaining dsds + * are bundled in continuation iocb + */ + avail_dsds = 1; + cur_dsd = (uint32_t *)&cmd_pkt->fcp_data_dseg_address; + + index = 0; + + for_each_sg(bsg_job->request_payload.sg_list, sg, + bsg_job->request_payload.sg_cnt, index) { + dma_addr_t sle_dma; + cont_a64_entry_t *cont_pkt; + + /* Allocate additional continuation packets */ + if (avail_dsds == 0) { + /* Continuation type 1 IOCB can accomodate + * 5 DSDS + */ + cont_pkt = qla2x00_prep_cont_type1_iocb(vha, vha->req); + cur_dsd = (uint32_t *) cont_pkt->dseg_0_address; + avail_dsds = 5; + entry_count++; + } + sle_dma = sg_dma_address(sg); + *cur_dsd++ = cpu_to_le32(LSD(sle_dma)); + *cur_dsd++ = cpu_to_le32(MSD(sle_dma)); + *cur_dsd++ = cpu_to_le32(sg_dma_len(sg)); + avail_dsds--; + } + /* For read request DSD will always goes to continuation IOCB + * and follow the write DSD. If there is room on the current IOCB + * then it is added to that IOCB else new continuation IOCB is + * allocated. + */ + for_each_sg(bsg_job->reply_payload.sg_list, sg, + bsg_job->reply_payload.sg_cnt, index) { + dma_addr_t sle_dma; + cont_a64_entry_t *cont_pkt; + + /* Allocate additional continuation packets */ + if (avail_dsds == 0) { + /* Continuation type 1 IOCB can accomodate + * 5 DSDS + */ + cont_pkt = qla2x00_prep_cont_type1_iocb(vha, vha->req); + cur_dsd = (uint32_t *) cont_pkt->dseg_0_address; + avail_dsds = 5; + entry_count++; + } + sle_dma = sg_dma_address(sg); + *cur_dsd++ = cpu_to_le32(LSD(sle_dma)); + *cur_dsd++ = cpu_to_le32(MSD(sle_dma)); + *cur_dsd++ = cpu_to_le32(sg_dma_len(sg)); + avail_dsds--; + } + /* This value should be same as number of IOCB required for this cmd */ + cmd_pkt->entry_count = entry_count; +} + +int +qla2x00_start_bidir(srb_t *sp, struct scsi_qla_host *vha, uint32_t tot_dsds) +{ + + struct qla_hw_data *ha = vha->hw; + unsigned long flags; + uint32_t handle; + uint32_t index; + uint16_t req_cnt; + uint16_t cnt; + uint32_t *clr_ptr; + struct cmd_bidir *cmd_pkt = NULL; + struct rsp_que *rsp; + struct req_que *req; + int rval = EXT_STATUS_OK; + device_reg_t __iomem *reg = ISP_QUE_REG(ha, vha->req->id); + + rval = QLA_SUCCESS; + + rsp = ha->rsp_q_map[0]; + req = vha->req; + + /* Send marker if required */ + if (vha->marker_needed != 0) { + if (qla2x00_marker(vha, req, + rsp, 0, 0, MK_SYNC_ALL) != QLA_SUCCESS) + return EXT_STATUS_MAILBOX; + vha->marker_needed = 0; + } + + /* Acquire ring specific lock */ + spin_lock_irqsave(&ha->hardware_lock, flags); + + /* Check for room in outstanding command list. */ + handle = req->current_outstanding_cmd; + for (index = 1; index < MAX_OUTSTANDING_COMMANDS; index++) { + handle++; + if (handle == MAX_OUTSTANDING_COMMANDS) + handle = 1; + if (!req->outstanding_cmds[handle]) + break; + } + + if (index == MAX_OUTSTANDING_COMMANDS) { + rval = EXT_STATUS_BUSY; + goto queuing_error; + } + + /* Calculate number of IOCB required */ + req_cnt = qla24xx_calc_iocbs(vha, tot_dsds); + + /* Check for room on request queue. */ + if (req->cnt < req_cnt + 2) { + if (ha->mqenable) + cnt = RD_REG_DWORD(®->isp25mq.req_q_out); + else if (IS_QLA82XX(ha)) + cnt = RD_REG_DWORD(®->isp82.req_q_out); + else if (IS_FWI2_CAPABLE(ha)) + cnt = RD_REG_DWORD(®->isp24.req_q_out); + else + cnt = qla2x00_debounce_register( + ISP_REQ_Q_OUT(ha, ®->isp)); + + if (req->ring_index < cnt) + req->cnt = cnt - req->ring_index; + else + req->cnt = req->length - + (req->ring_index - cnt); + } + if (req->cnt < req_cnt + 2) { + rval = EXT_STATUS_BUSY; + goto queuing_error; + } + + cmd_pkt = (struct cmd_bidir *)req->ring_ptr; + cmd_pkt->handle = MAKE_HANDLE(req->id, handle); + + /* Zero out remaining portion of packet. */ + /* tagged queuing modifier -- default is TSK_SIMPLE (0).*/ + clr_ptr = (uint32_t *)cmd_pkt + 2; + memset(clr_ptr, 0, REQUEST_ENTRY_SIZE - 8); + + /* Set NPORT-ID (of vha)*/ + cmd_pkt->nport_handle = cpu_to_le16(vha->self_login_loop_id); + cmd_pkt->port_id[0] = vha->d_id.b.al_pa; + cmd_pkt->port_id[1] = vha->d_id.b.area; + cmd_pkt->port_id[2] = vha->d_id.b.domain; + + qla25xx_build_bidir_iocb(sp, vha, cmd_pkt, tot_dsds); + cmd_pkt->entry_status = (uint8_t) rsp->id; + /* Build command packet. */ + req->current_outstanding_cmd = handle; + req->outstanding_cmds[handle] = sp; + sp->handle = handle; + req->cnt -= req_cnt; + + /* Send the command to the firmware */ + wmb(); + qla2x00_start_iocbs(vha, req); +queuing_error: + spin_unlock_irqrestore(&ha->hardware_lock, flags); + return rval; +} diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c index 6f67a9d..5733811 100644 --- a/drivers/scsi/qla2xxx/qla_isr.c +++ b/drivers/scsi/qla2xxx/qla_isr.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -294,6 +294,11 @@ qla81xx_idc_event(scsi_qla_host_t *vha, uint16_t aen, uint16_t descr) "%04x %04x %04x %04x %04x %04x %04x.\n", event[aen & 0xff], mb[0], mb[1], mb[2], mb[3], mb[4], mb[5], mb[6]); + if ((aen == MBA_IDC_COMPLETE && mb[1] >> 15)) { + vha->hw->flags.idc_compl_status = 1; + if (vha->hw->notify_dcbx_comp) + complete(&vha->hw->dcbx_comp); + } /* Acknowledgement needed? [Notify && non-zero timeout]. */ timeout = (descr >> 8) & 0xf; @@ -332,6 +337,166 @@ qla2x00_get_link_speed_str(struct qla_hw_data *ha) return link_speed; } +void +qla83xx_handle_8200_aen(scsi_qla_host_t *vha, uint16_t *mb) +{ + struct qla_hw_data *ha = vha->hw; + + /* + * 8200 AEN Interpretation: + * mb[0] = AEN code + * mb[1] = AEN Reason code + * mb[2] = LSW of Peg-Halt Status-1 Register + * mb[6] = MSW of Peg-Halt Status-1 Register + * mb[3] = LSW of Peg-Halt Status-2 register + * mb[7] = MSW of Peg-Halt Status-2 register + * mb[4] = IDC Device-State Register value + * mb[5] = IDC Driver-Presence Register value + */ + ql_dbg(ql_dbg_async, vha, 0x506b, "AEN Code: mb[0] = 0x%x AEN reason: " + "mb[1] = 0x%x PH-status1: mb[2] = 0x%x PH-status1: mb[6] = 0x%x.\n", + mb[0], mb[1], mb[2], mb[6]); + ql_dbg(ql_dbg_async, vha, 0x506c, "PH-status2: mb[3] = 0x%x " + "PH-status2: mb[7] = 0x%x Device-State: mb[4] = 0x%x " + "Drv-Presence: mb[5] = 0x%x.\n", mb[3], mb[7], mb[4], mb[5]); + + if (mb[1] & (IDC_PEG_HALT_STATUS_CHANGE | IDC_NIC_FW_REPORTED_FAILURE | + IDC_HEARTBEAT_FAILURE)) { + ha->flags.nic_core_hung = 1; + ql_log(ql_log_warn, vha, 0x5060, + "83XX: F/W Error Reported: Check if reset required.\n"); + + if (mb[1] & IDC_PEG_HALT_STATUS_CHANGE) { + uint32_t protocol_engine_id, fw_err_code, err_level; + + /* + * IDC_PEG_HALT_STATUS_CHANGE interpretation: + * - PEG-Halt Status-1 Register: + * (LSW = mb[2], MSW = mb[6]) + * Bits 0-7 = protocol-engine ID + * Bits 8-28 = f/w error code + * Bits 29-31 = Error-level + * Error-level 0x1 = Non-Fatal error + * Error-level 0x2 = Recoverable Fatal error + * Error-level 0x4 = UnRecoverable Fatal error + * - PEG-Halt Status-2 Register: + * (LSW = mb[3], MSW = mb[7]) + */ + protocol_engine_id = (mb[2] & 0xff); + fw_err_code = (((mb[2] & 0xff00) >> 8) | + ((mb[6] & 0x1fff) << 8)); + err_level = ((mb[6] & 0xe000) >> 13); + ql_log(ql_log_warn, vha, 0x5061, "PegHalt Status-1 " + "Register: protocol_engine_id=0x%x " + "fw_err_code=0x%x err_level=0x%x.\n", + protocol_engine_id, fw_err_code, err_level); + ql_log(ql_log_warn, vha, 0x5062, "PegHalt Status-2 " + "Register: 0x%x%x.\n", mb[7], mb[3]); + if (err_level == ERR_LEVEL_NON_FATAL) { + ql_log(ql_log_warn, vha, 0x5063, + "Not a fatal error, f/w has recovered " + "iteself.\n"); + } else if (err_level == ERR_LEVEL_RECOVERABLE_FATAL) { + ql_log(ql_log_fatal, vha, 0x5064, + "Recoverable Fatal error: Chip reset " + "required.\n"); + qla83xx_schedule_work(vha, + QLA83XX_NIC_CORE_RESET); + } else if (err_level == ERR_LEVEL_UNRECOVERABLE_FATAL) { + ql_log(ql_log_fatal, vha, 0x5065, + "Unrecoverable Fatal error: Set FAILED " + "state, reboot required.\n"); + qla83xx_schedule_work(vha, + QLA83XX_NIC_CORE_UNRECOVERABLE); + } + } + + if (mb[1] & IDC_NIC_FW_REPORTED_FAILURE) { + uint16_t peg_fw_state, nw_interface_link_up; + uint16_t nw_interface_signal_detect, sfp_status; + uint16_t htbt_counter, htbt_monitor_enable; + uint16_t sfp_additonal_info, sfp_multirate; + uint16_t sfp_tx_fault, link_speed, dcbx_status; + + /* + * IDC_NIC_FW_REPORTED_FAILURE interpretation: + * - PEG-to-FC Status Register: + * (LSW = mb[2], MSW = mb[6]) + * Bits 0-7 = Peg-Firmware state + * Bit 8 = N/W Interface Link-up + * Bit 9 = N/W Interface signal detected + * Bits 10-11 = SFP Status + * SFP Status 0x0 = SFP+ transceiver not expected + * SFP Status 0x1 = SFP+ transceiver not present + * SFP Status 0x2 = SFP+ transceiver invalid + * SFP Status 0x3 = SFP+ transceiver present and + * valid + * Bits 12-14 = Heartbeat Counter + * Bit 15 = Heartbeat Monitor Enable + * Bits 16-17 = SFP Additional Info + * SFP info 0x0 = Unregocnized transceiver for + * Ethernet + * SFP info 0x1 = SFP+ brand validation failed + * SFP info 0x2 = SFP+ speed validation failed + * SFP info 0x3 = SFP+ access error + * Bit 18 = SFP Multirate + * Bit 19 = SFP Tx Fault + * Bits 20-22 = Link Speed + * Bits 23-27 = Reserved + * Bits 28-30 = DCBX Status + * DCBX Status 0x0 = DCBX Disabled + * DCBX Status 0x1 = DCBX Enabled + * DCBX Status 0x2 = DCBX Exchange error + * Bit 31 = Reserved + */ + peg_fw_state = (mb[2] & 0x00ff); + nw_interface_link_up = ((mb[2] & 0x0100) >> 8); + nw_interface_signal_detect = ((mb[2] & 0x0200) >> 9); + sfp_status = ((mb[2] & 0x0c00) >> 10); + htbt_counter = ((mb[2] & 0x7000) >> 12); + htbt_monitor_enable = ((mb[2] & 0x8000) >> 15); + sfp_additonal_info = (mb[6] & 0x0003); + sfp_multirate = ((mb[6] & 0x0004) >> 2); + sfp_tx_fault = ((mb[6] & 0x0008) >> 3); + link_speed = ((mb[6] & 0x0070) >> 4); + dcbx_status = ((mb[6] & 0x7000) >> 12); + + ql_log(ql_log_warn, vha, 0x5066, + "Peg-to-Fc Status Register:\n" + "peg_fw_state=0x%x, nw_interface_link_up=0x%x, " + "nw_interface_signal_detect=0x%x" + "\nsfp_statis=0x%x.\n ", peg_fw_state, + nw_interface_link_up, nw_interface_signal_detect, + sfp_status); + ql_log(ql_log_warn, vha, 0x5067, + "htbt_counter=0x%x, htbt_monitor_enable=0x%x, " + "sfp_additonal_info=0x%x, sfp_multirate=0x%x.\n ", + htbt_counter, htbt_monitor_enable, + sfp_additonal_info, sfp_multirate); + ql_log(ql_log_warn, vha, 0x5068, + "sfp_tx_fault=0x%x, link_state=0x%x, " + "dcbx_status=0x%x.\n", sfp_tx_fault, link_speed, + dcbx_status); + + qla83xx_schedule_work(vha, QLA83XX_NIC_CORE_RESET); + } + + if (mb[1] & IDC_HEARTBEAT_FAILURE) { + ql_log(ql_log_warn, vha, 0x5069, + "Heartbeat Failure encountered, chip reset " + "required.\n"); + + qla83xx_schedule_work(vha, QLA83XX_NIC_CORE_RESET); + } + } + + if (mb[1] & IDC_DEVICE_STATE_CHANGE) { + ql_log(ql_log_info, vha, 0x506a, + "IDC Device-State changed = 0x%x.\n", mb[4]); + qla83xx_schedule_work(vha, MBA_IDC_AEN); + } +} + /** * qla2x00_async_event() - Process aynchronous events. * @ha: SCSI driver HA context @@ -681,8 +846,7 @@ skip_rio: * it. Otherwise ignore it and Wait for RSCN to come in. */ atomic_set(&vha->loop_down_timer, 0); - if (atomic_read(&vha->loop_state) != LOOP_DOWN && - atomic_read(&vha->loop_state) != LOOP_DEAD) { + if (mb[1] != 0xffff || (mb[2] != 0x6 && mb[2] != 0x4)) { ql_dbg(ql_dbg_async, vha, 0x5011, "Asynchronous PORT UPDATE ignored %04x/%04x/%04x.\n", mb[1], mb[2], mb[3]); @@ -822,11 +986,28 @@ skip_rio: "FCF Configuration Error -- %04x %04x %04x.\n", mb[1], mb[2], mb[3]); break; - case MBA_IDC_COMPLETE: case MBA_IDC_NOTIFY: + /* See if we need to quiesce any I/O */ + if (IS_QLA8031(vha->hw)) + if ((mb[2] & 0x7fff) == MBC_PORT_RESET || + (mb[2] & 0x7fff) == MBC_SET_PORT_CONFIG) { + set_bit(ISP_QUIESCE_NEEDED, &vha->dpc_flags); + qla2xxx_wake_dpc(vha); + } + case MBA_IDC_COMPLETE: case MBA_IDC_TIME_EXT: - qla81xx_idc_event(vha, mb[0], mb[1]); + if (IS_QLA81XX(vha->hw) || IS_QLA8031(vha->hw)) + qla81xx_idc_event(vha, mb[0], mb[1]); break; + + case MBA_IDC_AEN: + mb[4] = RD_REG_WORD(®24->mailbox4); + mb[5] = RD_REG_WORD(®24->mailbox5); + mb[6] = RD_REG_WORD(®24->mailbox6); + mb[7] = RD_REG_WORD(®24->mailbox7); + qla83xx_handle_8200_aen(vha, mb); + break; + default: ql_dbg(ql_dbg_async, vha, 0x5057, "Unknown AEN:%04x %04x %04x %04x\n", @@ -1414,7 +1595,7 @@ qla2x00_handle_sense(srb_t *sp, uint8_t *sense_data, uint32_t par_sense_len, struct scsi_dif_tuple { __be16 guard; /* Checksum */ - __be16 app_tag; /* APPL identifer */ + __be16 app_tag; /* APPL identifier */ __be32 ref_tag; /* Target LBA or indirect LBA */ }; @@ -1546,6 +1727,149 @@ qla2x00_handle_dif_error(srb_t *sp, struct sts_entry_24xx *sts24) return 1; } +static void +qla25xx_process_bidir_status_iocb(scsi_qla_host_t *vha, void *pkt, + struct req_que *req, uint32_t index) +{ + struct qla_hw_data *ha = vha->hw; + srb_t *sp; + uint16_t comp_status; + uint16_t scsi_status; + uint16_t thread_id; + uint32_t rval = EXT_STATUS_OK; + struct fc_bsg_job *bsg_job = NULL; + sts_entry_t *sts; + struct sts_entry_24xx *sts24; + sts = (sts_entry_t *) pkt; + sts24 = (struct sts_entry_24xx *) pkt; + + /* Validate handle. */ + if (index >= MAX_OUTSTANDING_COMMANDS) { + ql_log(ql_log_warn, vha, 0x70af, + "Invalid SCSI completion handle 0x%x.\n", index); + set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); + return; + } + + sp = req->outstanding_cmds[index]; + if (sp) { + /* Free outstanding command slot. */ + req->outstanding_cmds[index] = NULL; + bsg_job = sp->u.bsg_job; + } else { + ql_log(ql_log_warn, vha, 0x70b0, + "Req:%d: Invalid ISP SCSI completion handle(0x%x)\n", + req->id, index); + + set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); + return; + } + + if (IS_FWI2_CAPABLE(ha)) { + comp_status = le16_to_cpu(sts24->comp_status); + scsi_status = le16_to_cpu(sts24->scsi_status) & SS_MASK; + } else { + comp_status = le16_to_cpu(sts->comp_status); + scsi_status = le16_to_cpu(sts->scsi_status) & SS_MASK; + } + + thread_id = bsg_job->request->rqst_data.h_vendor.vendor_cmd[1]; + switch (comp_status) { + case CS_COMPLETE: + if (scsi_status == 0) { + bsg_job->reply->reply_payload_rcv_len = + bsg_job->reply_payload.payload_len; + rval = EXT_STATUS_OK; + } + goto done; + + case CS_DATA_OVERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b1, + "Command completed with date overrun thread_id=%d\n", + thread_id); + rval = EXT_STATUS_DATA_OVERRUN; + break; + + case CS_DATA_UNDERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b2, + "Command completed with date underrun thread_id=%d\n", + thread_id); + rval = EXT_STATUS_DATA_UNDERRUN; + break; + case CS_BIDIR_RD_OVERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b3, + "Command completed with read data overrun thread_id=%d\n", + thread_id); + rval = EXT_STATUS_DATA_OVERRUN; + break; + + case CS_BIDIR_RD_WR_OVERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b4, + "Command completed with read and write data overrun " + "thread_id=%d\n", thread_id); + rval = EXT_STATUS_DATA_OVERRUN; + break; + + case CS_BIDIR_RD_OVERRUN_WR_UNDERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b5, + "Command completed with read data over and write data " + "underrun thread_id=%d\n", thread_id); + rval = EXT_STATUS_DATA_OVERRUN; + break; + + case CS_BIDIR_RD_UNDERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b6, + "Command completed with read data data underrun " + "thread_id=%d\n", thread_id); + rval = EXT_STATUS_DATA_UNDERRUN; + break; + + case CS_BIDIR_RD_UNDERRUN_WR_OVERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b7, + "Command completed with read data under and write data " + "overrun thread_id=%d\n", thread_id); + rval = EXT_STATUS_DATA_UNDERRUN; + break; + + case CS_BIDIR_RD_WR_UNDERRUN: + ql_dbg(ql_dbg_user, vha, 0x70b8, + "Command completed with read and write data underrun " + "thread_id=%d\n", thread_id); + rval = EXT_STATUS_DATA_UNDERRUN; + break; + + case CS_BIDIR_DMA: + ql_dbg(ql_dbg_user, vha, 0x70b9, + "Command completed with data DMA error thread_id=%d\n", + thread_id); + rval = EXT_STATUS_DMA_ERR; + break; + + case CS_TIMEOUT: + ql_dbg(ql_dbg_user, vha, 0x70ba, + "Command completed with timeout thread_id=%d\n", + thread_id); + rval = EXT_STATUS_TIMEOUT; + break; + default: + ql_dbg(ql_dbg_user, vha, 0x70bb, + "Command completed with completion status=0x%x " + "thread_id=%d\n", comp_status, thread_id); + rval = EXT_STATUS_ERR; + break; + } + bsg_job->reply->reply_payload_rcv_len = 0; + +done: + /* Return the vendor specific reply to API */ + bsg_job->reply->reply_data.vendor_reply.vendor_rsp[0] = rval; + bsg_job->reply_len = sizeof(struct fc_bsg_reply); + /* Always return DID_OK, bsg will send the vendor specific response + * in this case only */ + sp->done(vha, sp, (DID_OK << 6)); + +} + /** * qla2x00_status_entry() - Process a Status IOCB entry. * @ha: SCSI driver HA context @@ -1573,12 +1897,14 @@ qla2x00_status_entry(scsi_qla_host_t *vha, struct rsp_que *rsp, void *pkt) struct req_que *req; int logit = 1; int res = 0; + uint16_t state_flags = 0; sts = (sts_entry_t *) pkt; sts24 = (struct sts_entry_24xx *) pkt; if (IS_FWI2_CAPABLE(ha)) { comp_status = le16_to_cpu(sts24->comp_status); scsi_status = le16_to_cpu(sts24->scsi_status) & SS_MASK; + state_flags = le16_to_cpu(sts24->state_flags); } else { comp_status = le16_to_cpu(sts->comp_status); scsi_status = le16_to_cpu(sts->scsi_status) & SS_MASK; @@ -1587,17 +1913,9 @@ qla2x00_status_entry(scsi_qla_host_t *vha, struct rsp_que *rsp, void *pkt) que = MSW(sts->handle); req = ha->req_q_map[que]; - /* Fast path completion. */ - if (comp_status == CS_COMPLETE && scsi_status == 0) { - qla2x00_process_completed_request(vha, req, handle); - - return; - } - /* Validate handle. */ if (handle < MAX_OUTSTANDING_COMMANDS) { sp = req->outstanding_cmds[handle]; - req->outstanding_cmds[handle] = NULL; } else sp = NULL; @@ -1612,6 +1930,20 @@ qla2x00_status_entry(scsi_qla_host_t *vha, struct rsp_que *rsp, void *pkt) qla2xxx_wake_dpc(vha); return; } + + if (unlikely((state_flags & BIT_1) && (sp->type == SRB_BIDI_CMD))) { + qla25xx_process_bidir_status_iocb(vha, pkt, req, handle); + return; + } + + /* Fast path completion. */ + if (comp_status == CS_COMPLETE && scsi_status == 0) { + qla2x00_process_completed_request(vha, req, handle); + + return; + } + + req->outstanding_cmds[handle] = NULL; cp = GET_CMD_SP(sp); if (cp == NULL) { ql_dbg(ql_dbg_io, vha, 0x3018, @@ -1830,7 +2162,21 @@ check_scsi_status: case CS_DIF_ERROR: logit = qla2x00_handle_dif_error(sp, sts24); + res = cp->result; break; + + case CS_TRANSPORT: + res = DID_ERROR << 16; + + if (!IS_PI_SPLIT_DET_CAPABLE(ha)) + break; + + if (state_flags & BIT_4) + scmd_printk(KERN_WARNING, cp, + "Unsupported device '%s' found.\n", + cp->device->vendor); + break; + default: res = DID_ERROR << 16; break; @@ -2150,7 +2496,7 @@ qla24xx_intr_handler(int irq, void *dev_id) unsigned long iter; uint32_t stat; uint32_t hccr; - uint16_t mb[4]; + uint16_t mb[8]; struct rsp_que *rsp; unsigned long flags; @@ -2191,29 +2537,29 @@ qla24xx_intr_handler(int irq, void *dev_id) break; switch (stat & 0xff) { - case 0x1: - case 0x2: - case 0x10: - case 0x11: + case INTR_ROM_MB_SUCCESS: + case INTR_ROM_MB_FAILED: + case INTR_MB_SUCCESS: + case INTR_MB_FAILED: qla24xx_mbx_completion(vha, MSW(stat)); status |= MBX_INTERRUPT; break; - case 0x12: + case INTR_ASYNC_EVENT: mb[0] = MSW(stat); mb[1] = RD_REG_WORD(®->mailbox1); mb[2] = RD_REG_WORD(®->mailbox2); mb[3] = RD_REG_WORD(®->mailbox3); qla2x00_async_event(vha, rsp, mb); break; - case 0x13: - case 0x14: + case INTR_RSP_QUE_UPDATE: + case INTR_RSP_QUE_UPDATE_83XX: qla24xx_process_response_queue(vha, rsp); break; - case 0x1C: /* ATIO queue updated */ + case INTR_ATIO_QUE_UPDATE: qlt_24xx_process_atio_queue(vha); break; - case 0x1D: /* ATIO and response queues updated */ + case INTR_ATIO_RSP_QUE_UPDATE: qlt_24xx_process_atio_queue(vha); qla24xx_process_response_queue(vha, rsp); break; @@ -2224,6 +2570,8 @@ qla24xx_intr_handler(int irq, void *dev_id) } WRT_REG_DWORD(®->hccr, HCCRX_CLR_RISC_INT); RD_REG_DWORD_RELAXED(®->hccr); + if (unlikely(IS_QLA83XX(ha) && (ha->pdev->revision == 1))) + ndelay(3500); } spin_unlock_irqrestore(&ha->hardware_lock, flags); @@ -2306,7 +2654,7 @@ qla24xx_msix_default(int irq, void *dev_id) int status; uint32_t stat; uint32_t hccr; - uint16_t mb[4]; + uint16_t mb[8]; unsigned long flags; rsp = (struct rsp_que *) dev_id; @@ -2342,29 +2690,29 @@ qla24xx_msix_default(int irq, void *dev_id) break; switch (stat & 0xff) { - case 0x1: - case 0x2: - case 0x10: - case 0x11: + case INTR_ROM_MB_SUCCESS: + case INTR_ROM_MB_FAILED: + case INTR_MB_SUCCESS: + case INTR_MB_FAILED: qla24xx_mbx_completion(vha, MSW(stat)); status |= MBX_INTERRUPT; break; - case 0x12: + case INTR_ASYNC_EVENT: mb[0] = MSW(stat); mb[1] = RD_REG_WORD(®->mailbox1); mb[2] = RD_REG_WORD(®->mailbox2); mb[3] = RD_REG_WORD(®->mailbox3); qla2x00_async_event(vha, rsp, mb); break; - case 0x13: - case 0x14: + case INTR_RSP_QUE_UPDATE: + case INTR_RSP_QUE_UPDATE_83XX: qla24xx_process_response_queue(vha, rsp); break; - case 0x1C: /* ATIO queue updated */ + case INTR_ATIO_QUE_UPDATE: qlt_24xx_process_atio_queue(vha); break; - case 0x1D: /* ATIO and response queues updated */ + case INTR_ATIO_RSP_QUE_UPDATE: qlt_24xx_process_atio_queue(vha); qla24xx_process_response_queue(vha, rsp); break; @@ -2570,7 +2918,7 @@ qla2x00_request_irqs(struct qla_hw_data *ha, struct rsp_que *rsp) skip_msix: if (!IS_QLA24XX(ha) && !IS_QLA2532(ha) && !IS_QLA8432(ha) && - !IS_QLA8001(ha)) + !IS_QLA8001(ha) && !IS_QLA82XX(ha)) goto skip_msi; ret = pci_enable_msi(ha->pdev); @@ -2581,6 +2929,11 @@ skip_msix: } else ql_log(ql_log_warn, vha, 0x0039, "MSI-X; Falling back-to INTa mode -- %d.\n", ret); + + /* Skip INTx on ISP82xx. */ + if (!ha->flags.msi_enabled && IS_QLA82XX(ha)) + return QLA_FUNCTION_FAILED; + skip_msi: ret = request_irq(ha->pdev->irq, ha->isp_ops->intr_handler, @@ -2595,21 +2948,9 @@ skip_msi: clear_risc_ints: - /* - * FIXME: Noted that 8014s were being dropped during NK testing. - * Timing deltas during MSI-X/INTa transitions? - */ - if (IS_QLA81XX(ha) || IS_QLA82XX(ha) || IS_QLA83XX(ha)) - goto fail; spin_lock_irq(&ha->hardware_lock); - if (IS_FWI2_CAPABLE(ha)) { - WRT_REG_DWORD(®->isp24.hccr, HCCRX_CLR_HOST_INT); - WRT_REG_DWORD(®->isp24.hccr, HCCRX_CLR_RISC_INT); - } else { + if (!IS_FWI2_CAPABLE(ha)) WRT_REG_WORD(®->isp.semaphore, 0); - WRT_REG_WORD(®->isp.hccr, HCCR_CLR_RISC_INT); - WRT_REG_WORD(®->isp.hccr, HCCR_CLR_HOST_INT); - } spin_unlock_irq(&ha->hardware_lock); fail: diff --git a/drivers/scsi/qla2xxx/qla_mbx.c b/drivers/scsi/qla2xxx/qla_mbx.c index d5ce92c..18c509f 100644 --- a/drivers/scsi/qla2xxx/qla_mbx.c +++ b/drivers/scsi/qla2xxx/qla_mbx.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -75,7 +75,7 @@ qla2x00_mailbox_command(scsi_qla_host_t *vha, mbx_cmd_t *mcp) return QLA_FUNCTION_TIMEOUT; } - if (ha->flags.isp82xx_fw_hung) { + if (IS_QLA82XX(ha) && ha->flags.isp82xx_fw_hung) { /* Setting Link-Down error */ mcp->mb[0] = MBS_LINK_DOWN_ERROR; ql_log(ql_log_warn, vha, 0x1004, @@ -232,7 +232,7 @@ qla2x00_mailbox_command(scsi_qla_host_t *vha, mbx_cmd_t *mcp) ha->flags.mbox_int = 0; clear_bit(MBX_INTERRUPT, &ha->mbx_cmd_flags); - if (ha->flags.isp82xx_fw_hung) { + if ((IS_QLA82XX(ha) && ha->flags.isp82xx_fw_hung)) { ha->flags.mbox_busy = 0; /* Setting Link-Down error */ mcp->mb[0] = MBS_LINK_DOWN_ERROR; @@ -369,7 +369,7 @@ premature_exit: mbx_done: if (rval) { - ql_dbg(ql_dbg_mbx, base_vha, 0x1020, + ql_log(ql_log_warn, base_vha, 0x1020, "**** Failed mbx[0]=%x, mb[1]=%x, mb[2]=%x, mb[3]=%x, cmd=%x ****.\n", mcp->mb[0], mcp->mb[1], mcp->mb[2], mcp->mb[3], command); } else { @@ -533,7 +533,7 @@ qla2x00_get_fw_version(scsi_qla_host_t *vha) mcp->in_mb = MBX_6|MBX_5|MBX_4|MBX_3|MBX_2|MBX_1|MBX_0; if (IS_QLA81XX(vha->hw) || IS_QLA8031(ha)) mcp->in_mb |= MBX_13|MBX_12|MBX_11|MBX_10|MBX_9|MBX_8; - if (IS_QLA83XX(vha->hw)) + if (IS_FWI2_CAPABLE(ha)) mcp->in_mb |= MBX_17|MBX_16|MBX_15; mcp->flags = 0; mcp->tov = MBX_TOV_SECONDS; @@ -559,18 +559,16 @@ qla2x00_get_fw_version(scsi_qla_host_t *vha) ha->phy_version[1] = mcp->mb[9] >> 8; ha->phy_version[2] = mcp->mb[9] & 0xff; } - if (IS_QLA83XX(ha)) { - if (mcp->mb[6] & BIT_15) { - ha->fw_attributes_h = mcp->mb[15]; - ha->fw_attributes_ext[0] = mcp->mb[16]; - ha->fw_attributes_ext[1] = mcp->mb[17]; - ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x1139, - "%s: FW_attributes Upper: 0x%x, Lower: 0x%x.\n", - __func__, mcp->mb[15], mcp->mb[6]); - } else - ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x112f, - "%s: FwAttributes [Upper] invalid, MB6:%04x\n", - __func__, mcp->mb[6]); + if (IS_FWI2_CAPABLE(ha)) { + ha->fw_attributes_h = mcp->mb[15]; + ha->fw_attributes_ext[0] = mcp->mb[16]; + ha->fw_attributes_ext[1] = mcp->mb[17]; + ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x1139, + "%s: FW_attributes Upper: 0x%x, Lower: 0x%x.\n", + __func__, mcp->mb[15], mcp->mb[6]); + ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x112f, + "%s: Ext_FwAttributes Upper: 0x%x, Lower: 0x%x.\n", + __func__, mcp->mb[17], mcp->mb[16]); } failed: @@ -3408,7 +3406,6 @@ qla2x00_dump_ram(scsi_qla_host_t *vha, dma_addr_t req_dma, uint32_t addr, return rval; } - /* 84XX Support **************************************************************/ struct cs84xx_mgmt_cmd { @@ -4428,7 +4425,8 @@ qla2x00_get_thermal_temp(scsi_qla_host_t *vha, uint16_t *temp, uint16_t *frac) "Entered %s.\n", __func__); /* Integer part */ - rval = qla2x00_read_sfp(vha, 0, &byte, 0x98, 0x01, 1, BIT_13|BIT_0); + rval = qla2x00_read_sfp(vha, 0, &byte, 0x98, 0x01, 1, + BIT_13|BIT_12|BIT_0); if (rval != QLA_SUCCESS) { ql_dbg(ql_dbg_mbx, vha, 0x10c9, "Failed=%x.\n", rval); ha->flags.thermal_supported = 0; @@ -4437,7 +4435,8 @@ qla2x00_get_thermal_temp(scsi_qla_host_t *vha, uint16_t *temp, uint16_t *frac) *temp = byte; /* Fraction part */ - rval = qla2x00_read_sfp(vha, 0, &byte, 0x98, 0x10, 1, BIT_13|BIT_0); + rval = qla2x00_read_sfp(vha, 0, &byte, 0x98, 0x10, 1, + BIT_13|BIT_12|BIT_0); if (rval != QLA_SUCCESS) { ql_dbg(ql_dbg_mbx, vha, 0x1019, "Failed=%x.\n", rval); ha->flags.thermal_supported = 0; @@ -4741,7 +4740,7 @@ qla82xx_mbx_beacon_ctl(scsi_qla_host_t *vha, int enable) } int -qla83xx_write_remote_reg(scsi_qla_host_t *vha, uint32_t reg, uint32_t data) +qla83xx_wr_reg(scsi_qla_host_t *vha, uint32_t reg, uint32_t data) { int rval; struct qla_hw_data *ha = vha->hw; @@ -4814,3 +4813,186 @@ qla2x00_port_logout(scsi_qla_host_t *vha, struct fc_port *fcport) return rval; } +int +qla83xx_rd_reg(scsi_qla_host_t *vha, uint32_t reg, uint32_t *data) +{ + int rval; + mbx_cmd_t mc; + mbx_cmd_t *mcp = &mc; + struct qla_hw_data *ha = vha->hw; + unsigned long retry_max_time = jiffies + (2 * HZ); + + if (!IS_QLA83XX(ha)) + return QLA_FUNCTION_FAILED; + + ql_dbg(ql_dbg_mbx, vha, 0x114b, "Entered %s.\n", __func__); + +retry_rd_reg: + mcp->mb[0] = MBC_READ_REMOTE_REG; + mcp->mb[1] = LSW(reg); + mcp->mb[2] = MSW(reg); + mcp->out_mb = MBX_2|MBX_1|MBX_0; + mcp->in_mb = MBX_4|MBX_3|MBX_1|MBX_0; + mcp->tov = MBX_TOV_SECONDS; + mcp->flags = 0; + rval = qla2x00_mailbox_command(vha, mcp); + + if (rval != QLA_SUCCESS) { + ql_dbg(ql_dbg_mbx, vha, 0x114c, + "Failed=%x mb[0]=%x mb[1]=%x.\n", + rval, mcp->mb[0], mcp->mb[1]); + } else { + *data = (mcp->mb[3] | (mcp->mb[4] << 16)); + if (*data == QLA8XXX_BAD_VALUE) { + /* + * During soft-reset CAMRAM register reads might + * return 0xbad0bad0. So retry for MAX of 2 sec + * while reading camram registers. + */ + if (time_after(jiffies, retry_max_time)) { + ql_dbg(ql_dbg_mbx, vha, 0x1141, + "Failure to read CAMRAM register. " + "data=0x%x.\n", *data); + return QLA_FUNCTION_FAILED; + } + msleep(100); + goto retry_rd_reg; + } + ql_dbg(ql_dbg_mbx, vha, 0x1142, "Done %s.\n", __func__); + } + + return rval; +} + +int +qla83xx_restart_nic_firmware(scsi_qla_host_t *vha) +{ + int rval; + mbx_cmd_t mc; + mbx_cmd_t *mcp = &mc; + struct qla_hw_data *ha = vha->hw; + + if (!IS_QLA83XX(ha)) + return QLA_FUNCTION_FAILED; + + ql_dbg(ql_dbg_mbx, vha, 0x1143, "Entered %s.\n", __func__); + + mcp->mb[0] = MBC_RESTART_NIC_FIRMWARE; + mcp->out_mb = MBX_0; + mcp->in_mb = MBX_1|MBX_0; + mcp->tov = MBX_TOV_SECONDS; + mcp->flags = 0; + rval = qla2x00_mailbox_command(vha, mcp); + + if (rval != QLA_SUCCESS) { + ql_dbg(ql_dbg_mbx, vha, 0x1144, + "Failed=%x mb[0]=%x mb[1]=%x.\n", + rval, mcp->mb[0], mcp->mb[1]); + ha->isp_ops->fw_dump(vha, 0); + } else { + ql_dbg(ql_dbg_mbx, vha, 0x1145, "Done %s.\n", __func__); + } + + return rval; +} + +int +qla83xx_access_control(scsi_qla_host_t *vha, uint16_t options, + uint32_t start_addr, uint32_t end_addr, uint16_t *sector_size) +{ + int rval; + mbx_cmd_t mc; + mbx_cmd_t *mcp = &mc; + uint8_t subcode = (uint8_t)options; + struct qla_hw_data *ha = vha->hw; + + if (!IS_QLA8031(ha)) + return QLA_FUNCTION_FAILED; + + ql_dbg(ql_dbg_mbx, vha, 0x1146, "Entered %s.\n", __func__); + + mcp->mb[0] = MBC_SET_ACCESS_CONTROL; + mcp->mb[1] = options; + mcp->out_mb = MBX_1|MBX_0; + if (subcode & BIT_2) { + mcp->mb[2] = LSW(start_addr); + mcp->mb[3] = MSW(start_addr); + mcp->mb[4] = LSW(end_addr); + mcp->mb[5] = MSW(end_addr); + mcp->out_mb |= MBX_5|MBX_4|MBX_3|MBX_2; + } + mcp->in_mb = MBX_2|MBX_1|MBX_0; + if (!(subcode & (BIT_2 | BIT_5))) + mcp->in_mb |= MBX_4|MBX_3; + mcp->tov = MBX_TOV_SECONDS; + mcp->flags = 0; + rval = qla2x00_mailbox_command(vha, mcp); + + if (rval != QLA_SUCCESS) { + ql_dbg(ql_dbg_mbx, vha, 0x1147, + "Failed=%x mb[0]=%x mb[1]=%x mb[2]=%x mb[3]=%x mb[4]=%x.\n", + rval, mcp->mb[0], mcp->mb[1], mcp->mb[2], mcp->mb[3], + mcp->mb[4]); + ha->isp_ops->fw_dump(vha, 0); + } else { + if (subcode & BIT_5) + *sector_size = mcp->mb[1]; + else if (subcode & (BIT_6 | BIT_7)) { + ql_dbg(ql_dbg_mbx, vha, 0x1148, + "Driver-lock id=%x%x", mcp->mb[4], mcp->mb[3]); + } else if (subcode & (BIT_3 | BIT_4)) { + ql_dbg(ql_dbg_mbx, vha, 0x1149, + "Flash-lock id=%x%x", mcp->mb[4], mcp->mb[3]); + } + ql_dbg(ql_dbg_mbx, vha, 0x114a, "Done %s.\n", __func__); + } + + return rval; +} + +int +qla2x00_dump_mctp_data(scsi_qla_host_t *vha, dma_addr_t req_dma, uint32_t addr, + uint32_t size) +{ + int rval; + mbx_cmd_t mc; + mbx_cmd_t *mcp = &mc; + + if (!IS_MCTP_CAPABLE(vha->hw)) + return QLA_FUNCTION_FAILED; + + ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x114f, + "Entered %s.\n", __func__); + + mcp->mb[0] = MBC_DUMP_RISC_RAM_EXTENDED; + mcp->mb[1] = LSW(addr); + mcp->mb[2] = MSW(req_dma); + mcp->mb[3] = LSW(req_dma); + mcp->mb[4] = MSW(size); + mcp->mb[5] = LSW(size); + mcp->mb[6] = MSW(MSD(req_dma)); + mcp->mb[7] = LSW(MSD(req_dma)); + mcp->mb[8] = MSW(addr); + /* Setting RAM ID to valid */ + mcp->mb[10] |= BIT_7; + /* For MCTP RAM ID is 0x40 */ + mcp->mb[10] |= 0x40; + + mcp->out_mb |= MBX_10|MBX_8|MBX_7|MBX_6|MBX_5|MBX_4|MBX_3|MBX_2|MBX_1| + MBX_0; + + mcp->in_mb = MBX_0; + mcp->tov = MBX_TOV_SECONDS; + mcp->flags = 0; + rval = qla2x00_mailbox_command(vha, mcp); + + if (rval != QLA_SUCCESS) { + ql_dbg(ql_dbg_mbx, vha, 0x114e, + "Failed=%x mb[0]=%x.\n", rval, mcp->mb[0]); + } else { + ql_dbg(ql_dbg_mbx + ql_dbg_verbose, vha, 0x114d, + "Done %s.\n", __func__); + } + + return rval; +} diff --git a/drivers/scsi/qla2xxx/qla_mid.c b/drivers/scsi/qla2xxx/qla_mid.c index 3e8b324..bd4708a 100644 --- a/drivers/scsi/qla2xxx/qla_mid.c +++ b/drivers/scsi/qla2xxx/qla_mid.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -476,7 +476,6 @@ qla24xx_create_vhost(struct fc_vport *fc_vport) vha->req = base_vha->req; host->can_queue = base_vha->req->length + 128; - host->this_id = 255; host->cmd_per_lun = 3; if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) host->max_cmd_len = 32; @@ -643,7 +642,7 @@ qla25xx_create_req_que(struct qla_hw_data *ha, uint16_t options, &req->dma, GFP_KERNEL); if (req->ring == NULL) { ql_log(ql_log_fatal, base_vha, 0x00da, - "Failed to allocte memory for request_ring.\n"); + "Failed to allocate memory for request_ring.\n"); goto que_failed; } diff --git a/drivers/scsi/qla2xxx/qla_nx.c b/drivers/scsi/qla2xxx/qla_nx.c index 7cfdf2b..14cd361 100644 --- a/drivers/scsi/qla2xxx/qla_nx.c +++ b/drivers/scsi/qla2xxx/qla_nx.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -1612,23 +1612,6 @@ qla82xx_get_fw_offs(struct qla_hw_data *ha) } /* PCI related functions */ -char * -qla82xx_pci_info_str(struct scsi_qla_host *vha, char *str) -{ - struct qla_hw_data *ha = vha->hw; - char lwstr[6]; - uint16_t lnk; - - pcie_capability_read_word(ha->pdev, PCI_EXP_LNKSTA, &lnk); - ha->link_width = (lnk >> 4) & 0x3f; - - strcpy(str, "PCIe ("); - strcat(str, "2.5Gb/s "); - snprintf(lwstr, sizeof(lwstr), "x%d)", ha->link_width); - strcat(str, lwstr); - return str; -} - int qla82xx_pci_region_offset(struct pci_dev *pdev, int region) { unsigned long val = 0; @@ -2320,6 +2303,29 @@ void qla82xx_init_flags(struct qla_hw_data *ha) } inline void +qla82xx_set_idc_version(scsi_qla_host_t *vha) +{ + int idc_ver; + uint32_t drv_active; + struct qla_hw_data *ha = vha->hw; + + drv_active = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); + if (drv_active == (QLA82XX_DRV_ACTIVE << (ha->portnum * 4))) { + qla82xx_wr_32(ha, QLA82XX_CRB_DRV_IDC_VERSION, + QLA82XX_IDC_VERSION); + ql_log(ql_log_info, vha, 0xb082, + "IDC version updated to %d\n", QLA82XX_IDC_VERSION); + } else { + idc_ver = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_IDC_VERSION); + if (idc_ver != QLA82XX_IDC_VERSION) + ql_log(ql_log_info, vha, 0xb083, + "qla2xxx driver IDC version %d is not compatible " + "with IDC version %d of the other drivers\n", + QLA82XX_IDC_VERSION, idc_ver); + } +} + +inline void qla82xx_set_drv_active(scsi_qla_host_t *vha) { uint32_t drv_active; @@ -2353,7 +2359,7 @@ qla82xx_need_reset(struct qla_hw_data *ha) uint32_t drv_state; int rval; - if (ha->flags.isp82xx_reset_owner) + if (ha->flags.nic_core_reset_owner) return 1; else { drv_state = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_STATE); @@ -2860,7 +2866,7 @@ qla82xx_device_bootstrap(scsi_qla_host_t *vha) timeout = msleep_interruptible(200); if (timeout) { qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + QLA8XXX_DEV_FAILED); return QLA_FUNCTION_FAILED; } @@ -2891,10 +2897,7 @@ dev_initialize: /* set to DEV_INITIALIZING */ ql_log(ql_log_info, vha, 0x009e, "HW State: INITIALIZING.\n"); - qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_INITIALIZING); - - /* Driver that sets device state to initializating sets IDC version */ - qla82xx_wr_32(ha, QLA82XX_CRB_DRV_IDC_VERSION, QLA82XX_IDC_VERSION); + qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA8XXX_DEV_INITIALIZING); qla82xx_idc_unlock(ha); rval = qla82xx_start_firmware(vha); @@ -2904,14 +2907,14 @@ dev_initialize: ql_log(ql_log_fatal, vha, 0x00ad, "HW State: FAILED.\n"); qla82xx_clear_drv_active(ha); - qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_FAILED); + qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA8XXX_DEV_FAILED); return rval; } dev_ready: ql_log(ql_log_info, vha, 0x00ae, "HW State: READY.\n"); - qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_READY); + qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA8XXX_DEV_READY); return QLA_SUCCESS; } @@ -2935,7 +2938,7 @@ qla82xx_need_qsnt_handler(scsi_qla_host_t *vha) if (vha->flags.online) { /*Block any further I/O and wait for pending cmnds to complete*/ - qla82xx_quiescent_state_cleanup(vha); + qla2x00_quiesce_io(vha); } /* Set the quiescence ready bit */ @@ -2960,7 +2963,7 @@ qla82xx_need_qsnt_handler(scsi_qla_host_t *vha) "DRV_STATE:%d.\n", QLA2XXX_DRIVER_NAME, drv_active, drv_state); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_READY); + QLA8XXX_DEV_READY); ql_log(ql_log_info, vha, 0xb025, "HW State: DEV_READY.\n"); qla82xx_idc_unlock(ha); @@ -2981,10 +2984,10 @@ qla82xx_need_qsnt_handler(scsi_qla_host_t *vha) } dev_state = qla82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); /* everyone acked so set the state to DEV_QUIESCENCE */ - if (dev_state == QLA82XX_DEV_NEED_QUIESCENT) { + if (dev_state == QLA8XXX_DEV_NEED_QUIESCENT) { ql_log(ql_log_info, vha, 0xb026, "HW State: DEV_QUIESCENT.\n"); - qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_QUIESCENT); + qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA8XXX_DEV_QUIESCENT); } } @@ -3014,8 +3017,8 @@ qla82xx_wait_for_state_change(scsi_qla_host_t *vha, uint32_t curr_state) return dev_state; } -static void -qla82xx_dev_failed_handler(scsi_qla_host_t *vha) +void +qla8xxx_dev_failed_handler(scsi_qla_host_t *vha) { struct qla_hw_data *ha = vha->hw; @@ -3023,9 +3026,10 @@ qla82xx_dev_failed_handler(scsi_qla_host_t *vha) ql_log(ql_log_fatal, vha, 0x00b8, "Disabling the board.\n"); - qla82xx_idc_lock(ha); - qla82xx_clear_drv_active(ha); - qla82xx_idc_unlock(ha); + if (IS_QLA82XX(ha)) { + qla82xx_clear_drv_active(ha); + qla82xx_idc_unlock(ha); + } /* Set DEV_FAILED flag to disable timer */ vha->device_flags |= DFLG_DEV_FAILED; @@ -3064,7 +3068,7 @@ qla82xx_need_reset_handler(scsi_qla_host_t *vha) } drv_active = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); - if (!ha->flags.isp82xx_reset_owner) { + if (!ha->flags.nic_core_reset_owner) { ql_dbg(ql_dbg_p3p, vha, 0xb028, "reset_acknowledged by 0x%x\n", ha->portnum); qla82xx_set_rst_ready(ha); @@ -3076,7 +3080,7 @@ qla82xx_need_reset_handler(scsi_qla_host_t *vha) } /* wait for 10 seconds for reset ack from all functions */ - reset_timeout = jiffies + (ha->nx_reset_timeout * HZ); + reset_timeout = jiffies + (ha->fcoe_reset_timeout * HZ); drv_state = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_STATE); drv_active = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); @@ -3088,7 +3092,7 @@ qla82xx_need_reset_handler(scsi_qla_host_t *vha) drv_state, drv_active, dev_state, active_mask); while (drv_state != drv_active && - dev_state != QLA82XX_DEV_INITIALIZING) { + dev_state != QLA8XXX_DEV_INITIALIZING) { if (time_after_eq(jiffies, reset_timeout)) { ql_log(ql_log_warn, vha, 0x00b5, "Reset timeout.\n"); @@ -3099,7 +3103,7 @@ qla82xx_need_reset_handler(scsi_qla_host_t *vha) qla82xx_idc_lock(ha); drv_state = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_STATE); drv_active = qla82xx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); - if (ha->flags.isp82xx_reset_owner) + if (ha->flags.nic_core_reset_owner) drv_active &= active_mask; dev_state = qla82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); } @@ -3115,11 +3119,11 @@ qla82xx_need_reset_handler(scsi_qla_host_t *vha) dev_state < MAX_STATES ? qdev_state(dev_state) : "Unknown"); /* Force to DEV_COLD unless someone else is starting a reset */ - if (dev_state != QLA82XX_DEV_INITIALIZING && - dev_state != QLA82XX_DEV_COLD) { + if (dev_state != QLA8XXX_DEV_INITIALIZING && + dev_state != QLA8XXX_DEV_COLD) { ql_log(ql_log_info, vha, 0x00b7, "HW State: COLD/RE-INIT.\n"); - qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_COLD); + qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA8XXX_DEV_COLD); qla82xx_set_rst_ready(ha); if (ql2xmdenable) { if (qla82xx_md_collect(vha)) @@ -3226,8 +3230,10 @@ qla82xx_device_state_handler(scsi_qla_host_t *vha) int loopcount = 0; qla82xx_idc_lock(ha); - if (!vha->flags.init_done) + if (!vha->flags.init_done) { qla82xx_set_drv_active(vha); + qla82xx_set_idc_version(vha); + } dev_state = qla82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); old_dev_state = dev_state; @@ -3237,7 +3243,7 @@ qla82xx_device_state_handler(scsi_qla_host_t *vha) dev_state < MAX_STATES ? qdev_state(dev_state) : "Unknown"); /* wait for 30 seconds for device to go ready */ - dev_init_timeout = jiffies + (ha->nx_dev_init_timeout * HZ); + dev_init_timeout = jiffies + (ha->fcoe_dev_init_timeout * HZ); while (1) { @@ -3261,18 +3267,18 @@ qla82xx_device_state_handler(scsi_qla_host_t *vha) } switch (dev_state) { - case QLA82XX_DEV_READY: - ha->flags.isp82xx_reset_owner = 0; - goto exit; - case QLA82XX_DEV_COLD: + case QLA8XXX_DEV_READY: + ha->flags.nic_core_reset_owner = 0; + goto rel_lock; + case QLA8XXX_DEV_COLD: rval = qla82xx_device_bootstrap(vha); break; - case QLA82XX_DEV_INITIALIZING: + case QLA8XXX_DEV_INITIALIZING: qla82xx_idc_unlock(ha); msleep(1000); qla82xx_idc_lock(ha); break; - case QLA82XX_DEV_NEED_RESET: + case QLA8XXX_DEV_NEED_RESET: if (!ql2xdontresethba) qla82xx_need_reset_handler(vha); else { @@ -3281,31 +3287,31 @@ qla82xx_device_state_handler(scsi_qla_host_t *vha) qla82xx_idc_lock(ha); } dev_init_timeout = jiffies + - (ha->nx_dev_init_timeout * HZ); + (ha->fcoe_dev_init_timeout * HZ); break; - case QLA82XX_DEV_NEED_QUIESCENT: + case QLA8XXX_DEV_NEED_QUIESCENT: qla82xx_need_qsnt_handler(vha); /* Reset timeout value after quiescence handler */ - dev_init_timeout = jiffies + (ha->nx_dev_init_timeout\ + dev_init_timeout = jiffies + (ha->fcoe_dev_init_timeout\ * HZ); break; - case QLA82XX_DEV_QUIESCENT: + case QLA8XXX_DEV_QUIESCENT: /* Owner will exit and other will wait for the state * to get changed */ if (ha->flags.quiesce_owner) - goto exit; + goto rel_lock; qla82xx_idc_unlock(ha); msleep(1000); qla82xx_idc_lock(ha); /* Reset timeout value after quiescence handler */ - dev_init_timeout = jiffies + (ha->nx_dev_init_timeout\ + dev_init_timeout = jiffies + (ha->fcoe_dev_init_timeout\ * HZ); break; - case QLA82XX_DEV_FAILED: - qla82xx_dev_failed_handler(vha); + case QLA8XXX_DEV_FAILED: + qla8xxx_dev_failed_handler(vha); rval = QLA_FUNCTION_FAILED; goto exit; default: @@ -3315,8 +3321,9 @@ qla82xx_device_state_handler(scsi_qla_host_t *vha) } loopcount++; } -exit: +rel_lock: qla82xx_idc_unlock(ha); +exit: return rval; } @@ -3364,22 +3371,30 @@ void qla82xx_watchdog(scsi_qla_host_t *vha) struct qla_hw_data *ha = vha->hw; /* don't poll if reset is going on */ - if (!ha->flags.isp82xx_reset_hdlr_active) { + if (!ha->flags.nic_core_reset_hdlr_active) { dev_state = qla82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); if (qla82xx_check_temp(vha)) { set_bit(ISP_UNRECOVERABLE, &vha->dpc_flags); ha->flags.isp82xx_fw_hung = 1; qla82xx_clear_pending_mbx(vha); - } else if (dev_state == QLA82XX_DEV_NEED_RESET && + } else if (dev_state == QLA8XXX_DEV_NEED_RESET && !test_bit(ISP_ABORT_NEEDED, &vha->dpc_flags)) { ql_log(ql_log_warn, vha, 0x6001, "Adapter reset needed.\n"); set_bit(ISP_ABORT_NEEDED, &vha->dpc_flags); - } else if (dev_state == QLA82XX_DEV_NEED_QUIESCENT && + } else if (dev_state == QLA8XXX_DEV_NEED_QUIESCENT && !test_bit(ISP_QUIESCE_NEEDED, &vha->dpc_flags)) { ql_log(ql_log_warn, vha, 0x6002, "Quiescent needed.\n"); set_bit(ISP_QUIESCE_NEEDED, &vha->dpc_flags); + } else if (dev_state == QLA8XXX_DEV_FAILED && + !test_bit(ISP_UNRECOVERABLE, &vha->dpc_flags) && + vha->flags.online == 1) { + ql_log(ql_log_warn, vha, 0xb055, + "Adapter state is failed. Offlining.\n"); + set_bit(ISP_UNRECOVERABLE, &vha->dpc_flags); + ha->flags.isp82xx_fw_hung = 1; + qla82xx_clear_pending_mbx(vha); } else { if (qla82xx_check_fw_alive(vha)) { ql_dbg(ql_dbg_timer, vha, 0x6011, @@ -3441,12 +3456,12 @@ qla82xx_set_reset_owner(scsi_qla_host_t *vha) uint32_t dev_state; dev_state = qla82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); - if (dev_state == QLA82XX_DEV_READY) { + if (dev_state == QLA8XXX_DEV_READY) { ql_log(ql_log_info, vha, 0xb02f, "HW State: NEED RESET\n"); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_NEED_RESET); - ha->flags.isp82xx_reset_owner = 1; + QLA8XXX_DEV_NEED_RESET); + ha->flags.nic_core_reset_owner = 1; ql_dbg(ql_dbg_p3p, vha, 0xb030, "reset_owner is 0x%x\n", ha->portnum); } else @@ -3477,7 +3492,7 @@ qla82xx_abort_isp(scsi_qla_host_t *vha) "Device in failed state, exiting.\n"); return QLA_SUCCESS; } - ha->flags.isp82xx_reset_hdlr_active = 1; + ha->flags.nic_core_reset_hdlr_active = 1; qla82xx_idc_lock(ha); qla82xx_set_reset_owner(vha); @@ -3491,7 +3506,7 @@ qla82xx_abort_isp(scsi_qla_host_t *vha) if (rval == QLA_SUCCESS) { ha->flags.isp82xx_fw_hung = 0; - ha->flags.isp82xx_reset_hdlr_active = 0; + ha->flags.nic_core_reset_hdlr_active = 0; qla82xx_restart_isp(vha); } @@ -4026,7 +4041,7 @@ qla82xx_minidump_process_rdmem(scsi_qla_host_t *vha, if (r_addr & 0xf) { ql_log(ql_log_warn, vha, 0xb033, - "Read addr 0x%x not 16 bytes alligned\n", r_addr); + "Read addr 0x%x not 16 bytes aligned\n", r_addr); return rval; } diff --git a/drivers/scsi/qla2xxx/qla_nx.h b/drivers/scsi/qla2xxx/qla_nx.h index 6eb210e..6c953e8 100644 --- a/drivers/scsi/qla2xxx/qla_nx.h +++ b/drivers/scsi/qla2xxx/qla_nx.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -542,14 +542,15 @@ #define QLA82XX_CRB_DRV_IDC_VERSION (QLA82XX_CAM_RAM(0x174)) /* Every driver should use these Device State */ -#define QLA82XX_DEV_COLD 1 -#define QLA82XX_DEV_INITIALIZING 2 -#define QLA82XX_DEV_READY 3 -#define QLA82XX_DEV_NEED_RESET 4 -#define QLA82XX_DEV_NEED_QUIESCENT 5 -#define QLA82XX_DEV_FAILED 6 -#define QLA82XX_DEV_QUIESCENT 7 +#define QLA8XXX_DEV_COLD 1 +#define QLA8XXX_DEV_INITIALIZING 2 +#define QLA8XXX_DEV_READY 3 +#define QLA8XXX_DEV_NEED_RESET 4 +#define QLA8XXX_DEV_NEED_QUIESCENT 5 +#define QLA8XXX_DEV_FAILED 6 +#define QLA8XXX_DEV_QUIESCENT 7 #define MAX_STATES 8 /* Increment if new state added */ +#define QLA8XXX_BAD_VALUE 0xbad0bad0 #define QLA82XX_IDC_VERSION 1 #define QLA82XX_ROM_DEV_INIT_TIMEOUT 30 diff --git a/drivers/scsi/qla2xxx/qla_os.c b/drivers/scsi/qla2xxx/qla_os.c index d305262..d501bf5 100644 --- a/drivers/scsi/qla2xxx/qla_os.c +++ b/drivers/scsi/qla2xxx/qla_os.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -113,11 +113,11 @@ MODULE_PARM_DESC(ql2xfdmienable, static int ql2xmaxqdepth = MAX_Q_DEPTH; module_param(ql2xmaxqdepth, int, S_IRUGO|S_IWUSR); MODULE_PARM_DESC(ql2xmaxqdepth, - "Maximum queue depth to report for target devices."); + "Maximum queue depth to set for each LUN. " + "Default is 32."); -/* Do not change the value of this after module load */ -int ql2xenabledif = 0; -module_param(ql2xenabledif, int, S_IRUGO|S_IWUSR); +int ql2xenabledif = 2; +module_param(ql2xenabledif, int, S_IRUGO); MODULE_PARM_DESC(ql2xenabledif, " Enable T10-CRC-DIF " " Default is 0 - No DIF Support. 1 - Enable it" @@ -1078,7 +1078,7 @@ __qla2xxx_eh_generic_reset(char *name, enum nexus_wait_type type, if (qla2x00_eh_wait_for_pending_commands(vha, cmd->device->id, cmd->device->lun, type) != QLA_SUCCESS) { ql_log(ql_log_warn, vha, 0x800d, - "wait for peding cmds failed for cmd=%p.\n", cmd); + "wait for pending cmds failed for cmd=%p.\n", cmd); goto eh_reset_failed; } @@ -1177,7 +1177,7 @@ qla2xxx_eh_bus_reset(struct scsi_cmnd *cmd) eh_bus_reset_done: ql_log(ql_log_warn, vha, 0x802b, "BUS RESET %s nexus=%ld:%d:%d.\n", - (ret == FAILED) ? "FAILED" : "SUCCEDED", vha->host_no, id, lun); + (ret == FAILED) ? "FAILED" : "SUCCEEDED", vha->host_no, id, lun); return ret; } @@ -1357,6 +1357,9 @@ qla2xxx_slave_configure(struct scsi_device *sdev) scsi_qla_host_t *vha = shost_priv(sdev->host); struct req_que *req = vha->req; + if (IS_T10_PI_CAPABLE(vha->hw)) + blk_queue_update_dma_alignment(sdev->request_queue, 0x7); + if (sdev->tagged_supported) scsi_activate_tcq(sdev, req->max_q_depth); else @@ -1919,7 +1922,7 @@ static struct isp_operations qla82xx_isp_ops = { .nvram_config = qla81xx_nvram_config, .update_fw_options = qla24xx_update_fw_options, .load_risc = qla82xx_load_risc, - .pci_info_str = qla82xx_pci_info_str, + .pci_info_str = qla24xx_pci_info_str, .fw_version_str = qla24xx_fw_version_str, .intr_handler = qla82xx_intr_handler, .enable_intrs = qla82xx_enable_intrs, @@ -2149,7 +2152,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) scsi_qla_host_t *base_vha = NULL; struct qla_hw_data *ha; char pci_info[30]; - char fw_str[30]; + char fw_str[30], wq_name[30]; struct scsi_host_template *sht; int bars, mem_only = 0; uint16_t req_length = 0, rsp_length = 0; @@ -2203,12 +2206,14 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) ha->mem_only = mem_only; spin_lock_init(&ha->hardware_lock); spin_lock_init(&ha->vport_slock); + mutex_init(&ha->selflogin_lock); /* Set ISP-type information. */ qla2x00_set_isp_flags(ha); /* Set EEH reset type to fundamental if required by hba */ - if (IS_QLA24XX(ha) || IS_QLA25XX(ha) || IS_QLA81XX(ha)) + if (IS_QLA24XX(ha) || IS_QLA25XX(ha) || IS_QLA81XX(ha) || + IS_QLA83XX(ha)) pdev->needs_freset = 1; ha->prev_topology = 0; @@ -2318,6 +2323,7 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) ha->nvram_conf_off = FARX_ACCESS_NVRAM_CONF; ha->nvram_data_off = FARX_ACCESS_NVRAM_DATA; } else if (IS_QLA83XX(ha)) { + ha->portnum = PCI_FUNC(ha->pdev->devfn); ha->max_fibre_devices = MAX_FIBRE_DEVICES_2400; ha->mbx_count = MAILBOX_REGISTER_COUNT; req_length = REQUEST_ENTRY_CNT_24XX; @@ -2416,7 +2422,6 @@ qla2x00_probe_one(struct pci_dev *pdev, const struct pci_device_id *id) host->can_queue, base_vha->req, base_vha->mgmt_svr_loop_id, host->sg_tablesize); host->max_id = ha->max_fibre_devices; - host->this_id = 255; host->cmd_per_lun = 3; host->unique_id = host->host_no; if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) @@ -2499,7 +2504,7 @@ que_init: if (IS_QLA82XX(ha)) { qla82xx_idc_lock(ha); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + QLA8XXX_DEV_FAILED); qla82xx_idc_unlock(ha); ql_log(ql_log_fatal, base_vha, 0x00d7, "HW State: FAILED.\n"); @@ -2542,6 +2547,20 @@ que_init: */ qla2xxx_wake_dpc(base_vha); + if (IS_QLA8031(ha) || IS_MCTP_CAPABLE(ha)) { + sprintf(wq_name, "qla2xxx_%lu_dpc_lp_wq", base_vha->host_no); + ha->dpc_lp_wq = create_singlethread_workqueue(wq_name); + INIT_WORK(&ha->idc_aen, qla83xx_service_idc_aen); + + sprintf(wq_name, "qla2xxx_%lu_dpc_hp_wq", base_vha->host_no); + ha->dpc_hp_wq = create_singlethread_workqueue(wq_name); + INIT_WORK(&ha->nic_core_reset, qla83xx_nic_core_reset_work); + INIT_WORK(&ha->idc_state_handler, + qla83xx_idc_state_handler_work); + INIT_WORK(&ha->nic_core_unrecoverable, + qla83xx_nic_core_unrecoverable_work); + } + skip_dpc: list_add_tail(&base_vha->list, &ha->vp_list); base_vha->host->irq = ha->pdev->irq; @@ -2557,7 +2576,7 @@ skip_dpc: if (IS_T10_PI_CAPABLE(ha) && ql2xenabledif) { if (ha->fw_attributes & BIT_4) { - int prot = 0; + int prot = 0, guard; base_vha->flags.difdix_supported = 1; ql_dbg(ql_dbg_init, base_vha, 0x00f1, "Registering for DIF/DIX type 1 and 3 protection.\n"); @@ -2570,7 +2589,14 @@ skip_dpc: | SHOST_DIX_TYPE1_PROTECTION | SHOST_DIX_TYPE2_PROTECTION | SHOST_DIX_TYPE3_PROTECTION); - scsi_host_set_guard(host, SHOST_DIX_GUARD_CRC); + + guard = SHOST_DIX_GUARD_CRC; + + if (IS_PI_IPGUARD_CAPABLE(ha) && + (ql2xenabledif > 1 || IS_PI_DIFB_DIX0_CAPABLE(ha))) + guard |= SHOST_DIX_GUARD_IP; + + scsi_host_set_guard(host, guard); } else base_vha->flags.difdix_supported = 0; } @@ -2750,6 +2776,14 @@ qla2x00_remove_one(struct pci_dev *pdev) } mutex_unlock(&ha->vport_lock); + if (IS_QLA8031(ha)) { + ql_dbg(ql_dbg_p3p, base_vha, 0xb07e, + "Clearing fcoe driver presence.\n"); + if (qla83xx_clear_drv_presence(base_vha) != QLA_SUCCESS) + ql_dbg(ql_dbg_p3p, base_vha, 0xb079, + "Error while clearing DRV-Presence.\n"); + } + set_bit(UNLOADING, &base_vha->dpc_flags); qla2x00_abort_all_cmds(base_vha, DID_NO_CONNECT << 16); @@ -2771,6 +2805,21 @@ qla2x00_remove_one(struct pci_dev *pdev) ha->wq = NULL; } + /* Cancel all work and destroy DPC workqueues */ + if (ha->dpc_lp_wq) { + cancel_work_sync(&ha->idc_aen); + destroy_workqueue(ha->dpc_lp_wq); + ha->dpc_lp_wq = NULL; + } + + if (ha->dpc_hp_wq) { + cancel_work_sync(&ha->nic_core_reset); + cancel_work_sync(&ha->idc_state_handler); + cancel_work_sync(&ha->nic_core_unrecoverable); + destroy_workqueue(ha->dpc_hp_wq); + ha->dpc_hp_wq = NULL; + } + /* Kill the kernel thread for this host */ if (ha->dpc_thread) { struct task_struct *t = ha->dpc_thread; @@ -2837,7 +2886,6 @@ qla2x00_free_device(scsi_qla_host_t *vha) qla2x00_stop_dpc_thread(vha); qla25xx_delete_queues(vha); - if (ha->flags.fce_enabled) qla2x00_disable_fce_trace(vha, NULL, NULL); @@ -2872,6 +2920,7 @@ void qla2x00_free_fcports(struct scsi_qla_host *vha) list_for_each_entry_safe(fcport, tfcport, &vha->vp_fcports, list) { list_del(&fcport->list); + qla2x00_clear_loop_id(fcport); kfree(fcport); fcport = NULL; } @@ -3169,6 +3218,18 @@ qla2x00_mem_alloc(struct qla_hw_data *ha, uint16_t req_len, uint16_t rsp_len, } INIT_LIST_HEAD(&ha->vp_list); + + /* Allocate memory for our loop_id bitmap */ + ha->loop_id_map = kzalloc(BITS_TO_LONGS(LOOPID_MAP_SIZE) * sizeof(long), + GFP_KERNEL); + if (!ha->loop_id_map) + goto fail_async_pd; + else { + qla2x00_set_reserved_loop_ids(ha); + ql_dbg_pci(ql_dbg_init, ha->pdev, 0x0123, + "loop_id_map=%p. \n", ha->loop_id_map); + } + return 1; fail_async_pd: @@ -3280,6 +3341,10 @@ qla2x00_mem_free(struct qla_hw_data *ha) { qla2x00_free_fw_dump(ha); + if (ha->mctp_dump) + dma_free_coherent(&ha->pdev->dev, MCTP_DUMP_SIZE, ha->mctp_dump, + ha->mctp_dump_dma); + if (ha->srb_mempool) mempool_destroy(ha->srb_mempool); @@ -3352,6 +3417,7 @@ qla2x00_mem_free(struct qla_hw_data *ha) kfree(ha->nvram); kfree(ha->npiv_info); kfree(ha->swl); + kfree(ha->loop_id_map); ha->srb_mempool = NULL; ha->ctx_mempool = NULL; @@ -3687,13 +3753,651 @@ void qla2x00_relogin(struct scsi_qla_host *vha) } if (fcport->login_retry == 0 && status != QLA_SUCCESS) - fcport->loop_id = FC_NO_LOOP_ID; + qla2x00_clear_loop_id(fcport); } if (test_bit(LOOP_RESYNC_NEEDED, &vha->dpc_flags)) break; } } +/* Schedule work on any of the dpc-workqueues */ +void +qla83xx_schedule_work(scsi_qla_host_t *base_vha, int work_code) +{ + struct qla_hw_data *ha = base_vha->hw; + + switch (work_code) { + case MBA_IDC_AEN: /* 0x8200 */ + if (ha->dpc_lp_wq) + queue_work(ha->dpc_lp_wq, &ha->idc_aen); + break; + + case QLA83XX_NIC_CORE_RESET: /* 0x1 */ + if (!ha->flags.nic_core_reset_hdlr_active) { + if (ha->dpc_hp_wq) + queue_work(ha->dpc_hp_wq, &ha->nic_core_reset); + } else + ql_dbg(ql_dbg_p3p, base_vha, 0xb05e, + "NIC Core reset is already active. Skip " + "scheduling it again.\n"); + break; + case QLA83XX_IDC_STATE_HANDLER: /* 0x2 */ + if (ha->dpc_hp_wq) + queue_work(ha->dpc_hp_wq, &ha->idc_state_handler); + break; + case QLA83XX_NIC_CORE_UNRECOVERABLE: /* 0x3 */ + if (ha->dpc_hp_wq) + queue_work(ha->dpc_hp_wq, &ha->nic_core_unrecoverable); + break; + default: + ql_log(ql_log_warn, base_vha, 0xb05f, + "Unknow work-code=0x%x.\n", work_code); + } + + return; +} + +/* Work: Perform NIC Core Unrecoverable state handling */ +void +qla83xx_nic_core_unrecoverable_work(struct work_struct *work) +{ + struct qla_hw_data *ha = + container_of(work, struct qla_hw_data, nic_core_unrecoverable); + scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev); + uint32_t dev_state = 0; + + qla83xx_idc_lock(base_vha, 0); + qla83xx_rd_reg(base_vha, QLA83XX_IDC_DEV_STATE, &dev_state); + qla83xx_reset_ownership(base_vha); + if (ha->flags.nic_core_reset_owner) { + ha->flags.nic_core_reset_owner = 0; + qla83xx_wr_reg(base_vha, QLA83XX_IDC_DEV_STATE, + QLA8XXX_DEV_FAILED); + ql_log(ql_log_info, base_vha, 0xb060, "HW State: FAILED.\n"); + qla83xx_schedule_work(base_vha, QLA83XX_IDC_STATE_HANDLER); + } + qla83xx_idc_unlock(base_vha, 0); +} + +/* Work: Execute IDC state handler */ +void +qla83xx_idc_state_handler_work(struct work_struct *work) +{ + struct qla_hw_data *ha = + container_of(work, struct qla_hw_data, idc_state_handler); + scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev); + uint32_t dev_state = 0; + + qla83xx_idc_lock(base_vha, 0); + qla83xx_rd_reg(base_vha, QLA83XX_IDC_DEV_STATE, &dev_state); + if (dev_state == QLA8XXX_DEV_FAILED || + dev_state == QLA8XXX_DEV_NEED_QUIESCENT) + qla83xx_idc_state_handler(base_vha); + qla83xx_idc_unlock(base_vha, 0); +} + +int +qla83xx_check_nic_core_fw_alive(scsi_qla_host_t *base_vha) +{ + int rval = QLA_SUCCESS; + unsigned long heart_beat_wait = jiffies + (1 * HZ); + uint32_t heart_beat_counter1, heart_beat_counter2; + + do { + if (time_after(jiffies, heart_beat_wait)) { + ql_dbg(ql_dbg_p3p, base_vha, 0xb07c, + "Nic Core f/w is not alive.\n"); + rval = QLA_FUNCTION_FAILED; + break; + } + + qla83xx_idc_lock(base_vha, 0); + qla83xx_rd_reg(base_vha, QLA83XX_FW_HEARTBEAT, + &heart_beat_counter1); + qla83xx_idc_unlock(base_vha, 0); + msleep(100); + qla83xx_idc_lock(base_vha, 0); + qla83xx_rd_reg(base_vha, QLA83XX_FW_HEARTBEAT, + &heart_beat_counter2); + qla83xx_idc_unlock(base_vha, 0); + } while (heart_beat_counter1 == heart_beat_counter2); + + return rval; +} + +/* Work: Perform NIC Core Reset handling */ +void +qla83xx_nic_core_reset_work(struct work_struct *work) +{ + struct qla_hw_data *ha = + container_of(work, struct qla_hw_data, nic_core_reset); + scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev); + uint32_t dev_state = 0; + + if (IS_QLA2031(ha)) { + if (qla2xxx_mctp_dump(base_vha) != QLA_SUCCESS) + ql_log(ql_log_warn, base_vha, 0xb081, + "Failed to dump mctp\n"); + return; + } + + if (!ha->flags.nic_core_reset_hdlr_active) { + if (qla83xx_check_nic_core_fw_alive(base_vha) == QLA_SUCCESS) { + qla83xx_idc_lock(base_vha, 0); + qla83xx_rd_reg(base_vha, QLA83XX_IDC_DEV_STATE, + &dev_state); + qla83xx_idc_unlock(base_vha, 0); + if (dev_state != QLA8XXX_DEV_NEED_RESET) { + ql_dbg(ql_dbg_p3p, base_vha, 0xb07a, + "Nic Core f/w is alive.\n"); + return; + } + } + + ha->flags.nic_core_reset_hdlr_active = 1; + if (qla83xx_nic_core_reset(base_vha)) { + /* NIC Core reset failed. */ + ql_dbg(ql_dbg_p3p, base_vha, 0xb061, + "NIC Core reset failed.\n"); + } + ha->flags.nic_core_reset_hdlr_active = 0; + } +} + +/* Work: Handle 8200 IDC aens */ +void +qla83xx_service_idc_aen(struct work_struct *work) +{ + struct qla_hw_data *ha = + container_of(work, struct qla_hw_data, idc_aen); + scsi_qla_host_t *base_vha = pci_get_drvdata(ha->pdev); + uint32_t dev_state, idc_control; + + qla83xx_idc_lock(base_vha, 0); + qla83xx_rd_reg(base_vha, QLA83XX_IDC_DEV_STATE, &dev_state); + qla83xx_rd_reg(base_vha, QLA83XX_IDC_CONTROL, &idc_control); + qla83xx_idc_unlock(base_vha, 0); + if (dev_state == QLA8XXX_DEV_NEED_RESET) { + if (idc_control & QLA83XX_IDC_GRACEFUL_RESET) { + ql_dbg(ql_dbg_p3p, base_vha, 0xb062, + "Application requested NIC Core Reset.\n"); + qla83xx_schedule_work(base_vha, QLA83XX_NIC_CORE_RESET); + } else if (qla83xx_check_nic_core_fw_alive(base_vha) == + QLA_SUCCESS) { + ql_dbg(ql_dbg_p3p, base_vha, 0xb07b, + "Other protocol driver requested NIC Core Reset.\n"); + qla83xx_schedule_work(base_vha, QLA83XX_NIC_CORE_RESET); + } + } else if (dev_state == QLA8XXX_DEV_FAILED || + dev_state == QLA8XXX_DEV_NEED_QUIESCENT) { + qla83xx_schedule_work(base_vha, QLA83XX_IDC_STATE_HANDLER); + } +} + +static void +qla83xx_wait_logic(void) +{ + int i; + + /* Yield CPU */ + if (!in_interrupt()) { + /* + * Wait about 200ms before retrying again. + * This controls the number of retries for single + * lock operation. + */ + msleep(100); + schedule(); + } else { + for (i = 0; i < 20; i++) + cpu_relax(); /* This a nop instr on i386 */ + } +} + +int +qla83xx_force_lock_recovery(scsi_qla_host_t *base_vha) +{ + int rval; + uint32_t data; + uint32_t idc_lck_rcvry_stage_mask = 0x3; + uint32_t idc_lck_rcvry_owner_mask = 0x3c; + struct qla_hw_data *ha = base_vha->hw; + + rval = qla83xx_rd_reg(base_vha, QLA83XX_IDC_LOCK_RECOVERY, &data); + if (rval) + return rval; + + if ((data & idc_lck_rcvry_stage_mask) > 0) { + return QLA_SUCCESS; + } else { + data = (IDC_LOCK_RECOVERY_STAGE1) | (ha->portnum << 2); + rval = qla83xx_wr_reg(base_vha, QLA83XX_IDC_LOCK_RECOVERY, + data); + if (rval) + return rval; + + msleep(200); + + rval = qla83xx_rd_reg(base_vha, QLA83XX_IDC_LOCK_RECOVERY, + &data); + if (rval) + return rval; + + if (((data & idc_lck_rcvry_owner_mask) >> 2) == ha->portnum) { + data &= (IDC_LOCK_RECOVERY_STAGE2 | + ~(idc_lck_rcvry_stage_mask)); + rval = qla83xx_wr_reg(base_vha, + QLA83XX_IDC_LOCK_RECOVERY, data); + if (rval) + return rval; + + /* Forcefully perform IDC UnLock */ + rval = qla83xx_rd_reg(base_vha, QLA83XX_DRIVER_UNLOCK, + &data); + if (rval) + return rval; + /* Clear lock-id by setting 0xff */ + rval = qla83xx_wr_reg(base_vha, QLA83XX_DRIVER_LOCKID, + 0xff); + if (rval) + return rval; + /* Clear lock-recovery by setting 0x0 */ + rval = qla83xx_wr_reg(base_vha, + QLA83XX_IDC_LOCK_RECOVERY, 0x0); + if (rval) + return rval; + } else + return QLA_SUCCESS; + } + + return rval; +} + +int +qla83xx_idc_lock_recovery(scsi_qla_host_t *base_vha) +{ + int rval = QLA_SUCCESS; + uint32_t o_drv_lockid, n_drv_lockid; + unsigned long lock_recovery_timeout; + + lock_recovery_timeout = jiffies + QLA83XX_MAX_LOCK_RECOVERY_WAIT; +retry_lockid: + rval = qla83xx_rd_reg(base_vha, QLA83XX_DRIVER_LOCKID, &o_drv_lockid); + if (rval) + goto exit; + + /* MAX wait time before forcing IDC Lock recovery = 2 secs */ + if (time_after_eq(jiffies, lock_recovery_timeout)) { + if (qla83xx_force_lock_recovery(base_vha) == QLA_SUCCESS) + return QLA_SUCCESS; + else + return QLA_FUNCTION_FAILED; + } + + rval = qla83xx_rd_reg(base_vha, QLA83XX_DRIVER_LOCKID, &n_drv_lockid); + if (rval) + goto exit; + + if (o_drv_lockid == n_drv_lockid) { + qla83xx_wait_logic(); + goto retry_lockid; + } else + return QLA_SUCCESS; + +exit: + return rval; +} + +void +qla83xx_idc_lock(scsi_qla_host_t *base_vha, uint16_t requester_id) +{ + uint16_t options = (requester_id << 15) | BIT_6; + uint32_t data; + struct qla_hw_data *ha = base_vha->hw; + + /* IDC-lock implementation using driver-lock/lock-id remote registers */ +retry_lock: + if (qla83xx_rd_reg(base_vha, QLA83XX_DRIVER_LOCK, &data) + == QLA_SUCCESS) { + if (data) { + /* Setting lock-id to our function-number */ + qla83xx_wr_reg(base_vha, QLA83XX_DRIVER_LOCKID, + ha->portnum); + } else { + ql_dbg(ql_dbg_p3p, base_vha, 0xb063, + "Failed to acquire IDC lock. retrying...\n"); + + /* Retry/Perform IDC-Lock recovery */ + if (qla83xx_idc_lock_recovery(base_vha) + == QLA_SUCCESS) { + qla83xx_wait_logic(); + goto retry_lock; + } else + ql_log(ql_log_warn, base_vha, 0xb075, + "IDC Lock recovery FAILED.\n"); + } + + } + + return; + + /* XXX: IDC-lock implementation using access-control mbx */ +retry_lock2: + if (qla83xx_access_control(base_vha, options, 0, 0, NULL)) { + ql_dbg(ql_dbg_p3p, base_vha, 0xb072, + "Failed to acquire IDC lock. retrying...\n"); + /* Retry/Perform IDC-Lock recovery */ + if (qla83xx_idc_lock_recovery(base_vha) == QLA_SUCCESS) { + qla83xx_wait_logic(); + goto retry_lock2; + } else + ql_log(ql_log_warn, base_vha, 0xb076, + "IDC Lock recovery FAILED.\n"); + } + + return; +} + +void +qla83xx_idc_unlock(scsi_qla_host_t *base_vha, uint16_t requester_id) +{ + uint16_t options = (requester_id << 15) | BIT_7, retry; + uint32_t data; + struct qla_hw_data *ha = base_vha->hw; + + /* IDC-unlock implementation using driver-unlock/lock-id + * remote registers + */ + retry = 0; +retry_unlock: + if (qla83xx_rd_reg(base_vha, QLA83XX_DRIVER_LOCKID, &data) + == QLA_SUCCESS) { + if (data == ha->portnum) { + qla83xx_rd_reg(base_vha, QLA83XX_DRIVER_UNLOCK, &data); + /* Clearing lock-id by setting 0xff */ + qla83xx_wr_reg(base_vha, QLA83XX_DRIVER_LOCKID, 0xff); + } else if (retry < 10) { + /* SV: XXX: IDC unlock retrying needed here? */ + + /* Retry for IDC-unlock */ + qla83xx_wait_logic(); + retry++; + ql_dbg(ql_dbg_p3p, base_vha, 0xb064, + "Failed to release IDC lock, retyring=%d\n", retry); + goto retry_unlock; + } + } else if (retry < 10) { + /* Retry for IDC-unlock */ + qla83xx_wait_logic(); + retry++; + ql_dbg(ql_dbg_p3p, base_vha, 0xb065, + "Failed to read drv-lockid, retyring=%d\n", retry); + goto retry_unlock; + } + + return; + + /* XXX: IDC-unlock implementation using access-control mbx */ + retry = 0; +retry_unlock2: + if (qla83xx_access_control(base_vha, options, 0, 0, NULL)) { + if (retry < 10) { + /* Retry for IDC-unlock */ + qla83xx_wait_logic(); + retry++; + ql_dbg(ql_dbg_p3p, base_vha, 0xb066, + "Failed to release IDC lock, retyring=%d\n", retry); + goto retry_unlock2; + } + } + + return; +} + +int +__qla83xx_set_drv_presence(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + struct qla_hw_data *ha = vha->hw; + uint32_t drv_presence; + + rval = qla83xx_rd_reg(vha, QLA83XX_IDC_DRV_PRESENCE, &drv_presence); + if (rval == QLA_SUCCESS) { + drv_presence |= (1 << ha->portnum); + rval = qla83xx_wr_reg(vha, QLA83XX_IDC_DRV_PRESENCE, + drv_presence); + } + + return rval; +} + +int +qla83xx_set_drv_presence(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + + qla83xx_idc_lock(vha, 0); + rval = __qla83xx_set_drv_presence(vha); + qla83xx_idc_unlock(vha, 0); + + return rval; +} + +int +__qla83xx_clear_drv_presence(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + struct qla_hw_data *ha = vha->hw; + uint32_t drv_presence; + + rval = qla83xx_rd_reg(vha, QLA83XX_IDC_DRV_PRESENCE, &drv_presence); + if (rval == QLA_SUCCESS) { + drv_presence &= ~(1 << ha->portnum); + rval = qla83xx_wr_reg(vha, QLA83XX_IDC_DRV_PRESENCE, + drv_presence); + } + + return rval; +} + +int +qla83xx_clear_drv_presence(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + + qla83xx_idc_lock(vha, 0); + rval = __qla83xx_clear_drv_presence(vha); + qla83xx_idc_unlock(vha, 0); + + return rval; +} + +void +qla83xx_need_reset_handler(scsi_qla_host_t *vha) +{ + struct qla_hw_data *ha = vha->hw; + uint32_t drv_ack, drv_presence; + unsigned long ack_timeout; + + /* Wait for IDC ACK from all functions (DRV-ACK == DRV-PRESENCE) */ + ack_timeout = jiffies + (ha->fcoe_reset_timeout * HZ); + while (1) { + qla83xx_rd_reg(vha, QLA83XX_IDC_DRIVER_ACK, &drv_ack); + qla83xx_rd_reg(vha, QLA83XX_IDC_DRV_PRESENCE, &drv_presence); + if (drv_ack == drv_presence) + break; + + if (time_after_eq(jiffies, ack_timeout)) { + ql_log(ql_log_warn, vha, 0xb067, + "RESET ACK TIMEOUT! drv_presence=0x%x " + "drv_ack=0x%x\n", drv_presence, drv_ack); + /* + * The function(s) which did not ack in time are forced + * to withdraw any further participation in the IDC + * reset. + */ + if (drv_ack != drv_presence) + qla83xx_wr_reg(vha, QLA83XX_IDC_DRV_PRESENCE, + drv_ack); + break; + } + + qla83xx_idc_unlock(vha, 0); + msleep(1000); + qla83xx_idc_lock(vha, 0); + } + + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, QLA8XXX_DEV_COLD); + ql_log(ql_log_info, vha, 0xb068, "HW State: COLD/RE-INIT.\n"); +} + +int +qla83xx_device_bootstrap(scsi_qla_host_t *vha) +{ + int rval = QLA_SUCCESS; + uint32_t idc_control; + + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, QLA8XXX_DEV_INITIALIZING); + ql_log(ql_log_info, vha, 0xb069, "HW State: INITIALIZING.\n"); + + /* Clearing IDC-Control Graceful-Reset Bit before resetting f/w */ + __qla83xx_get_idc_control(vha, &idc_control); + idc_control &= ~QLA83XX_IDC_GRACEFUL_RESET; + __qla83xx_set_idc_control(vha, 0); + + qla83xx_idc_unlock(vha, 0); + rval = qla83xx_restart_nic_firmware(vha); + qla83xx_idc_lock(vha, 0); + + if (rval != QLA_SUCCESS) { + ql_log(ql_log_fatal, vha, 0xb06a, + "Failed to restart NIC f/w.\n"); + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, QLA8XXX_DEV_FAILED); + ql_log(ql_log_info, vha, 0xb06b, "HW State: FAILED.\n"); + } else { + ql_dbg(ql_dbg_p3p, vha, 0xb06c, + "Success in restarting nic f/w.\n"); + qla83xx_wr_reg(vha, QLA83XX_IDC_DEV_STATE, QLA8XXX_DEV_READY); + ql_log(ql_log_info, vha, 0xb06d, "HW State: READY.\n"); + } + + return rval; +} + +/* Assumes idc_lock always held on entry */ +int +qla83xx_idc_state_handler(scsi_qla_host_t *base_vha) +{ + struct qla_hw_data *ha = base_vha->hw; + int rval = QLA_SUCCESS; + unsigned long dev_init_timeout; + uint32_t dev_state; + + /* Wait for MAX-INIT-TIMEOUT for the device to go ready */ + dev_init_timeout = jiffies + (ha->fcoe_dev_init_timeout * HZ); + + while (1) { + + if (time_after_eq(jiffies, dev_init_timeout)) { + ql_log(ql_log_warn, base_vha, 0xb06e, + "Initialization TIMEOUT!\n"); + /* Init timeout. Disable further NIC Core + * communication. + */ + qla83xx_wr_reg(base_vha, QLA83XX_IDC_DEV_STATE, + QLA8XXX_DEV_FAILED); + ql_log(ql_log_info, base_vha, 0xb06f, + "HW State: FAILED.\n"); + } + + qla83xx_rd_reg(base_vha, QLA83XX_IDC_DEV_STATE, &dev_state); + switch (dev_state) { + case QLA8XXX_DEV_READY: + if (ha->flags.nic_core_reset_owner) + qla83xx_idc_audit(base_vha, + IDC_AUDIT_COMPLETION); + ha->flags.nic_core_reset_owner = 0; + ql_dbg(ql_dbg_p3p, base_vha, 0xb070, + "Reset_owner reset by 0x%x.\n", + ha->portnum); + goto exit; + case QLA8XXX_DEV_COLD: + if (ha->flags.nic_core_reset_owner) + rval = qla83xx_device_bootstrap(base_vha); + else { + /* Wait for AEN to change device-state */ + qla83xx_idc_unlock(base_vha, 0); + msleep(1000); + qla83xx_idc_lock(base_vha, 0); + } + break; + case QLA8XXX_DEV_INITIALIZING: + /* Wait for AEN to change device-state */ + qla83xx_idc_unlock(base_vha, 0); + msleep(1000); + qla83xx_idc_lock(base_vha, 0); + break; + case QLA8XXX_DEV_NEED_RESET: + if (!ql2xdontresethba && ha->flags.nic_core_reset_owner) + qla83xx_need_reset_handler(base_vha); + else { + /* Wait for AEN to change device-state */ + qla83xx_idc_unlock(base_vha, 0); + msleep(1000); + qla83xx_idc_lock(base_vha, 0); + } + /* reset timeout value after need reset handler */ + dev_init_timeout = jiffies + + (ha->fcoe_dev_init_timeout * HZ); + break; + case QLA8XXX_DEV_NEED_QUIESCENT: + /* XXX: DEBUG for now */ + qla83xx_idc_unlock(base_vha, 0); + msleep(1000); + qla83xx_idc_lock(base_vha, 0); + break; + case QLA8XXX_DEV_QUIESCENT: + /* XXX: DEBUG for now */ + if (ha->flags.quiesce_owner) + goto exit; + + qla83xx_idc_unlock(base_vha, 0); + msleep(1000); + qla83xx_idc_lock(base_vha, 0); + dev_init_timeout = jiffies + + (ha->fcoe_dev_init_timeout * HZ); + break; + case QLA8XXX_DEV_FAILED: + if (ha->flags.nic_core_reset_owner) + qla83xx_idc_audit(base_vha, + IDC_AUDIT_COMPLETION); + ha->flags.nic_core_reset_owner = 0; + __qla83xx_clear_drv_presence(base_vha); + qla83xx_idc_unlock(base_vha, 0); + qla8xxx_dev_failed_handler(base_vha); + rval = QLA_FUNCTION_FAILED; + qla83xx_idc_lock(base_vha, 0); + goto exit; + case QLA8XXX_BAD_VALUE: + qla83xx_idc_unlock(base_vha, 0); + msleep(1000); + qla83xx_idc_lock(base_vha, 0); + break; + default: + ql_log(ql_log_warn, base_vha, 0xb071, + "Unknow Device State: %x.\n", dev_state); + qla83xx_idc_unlock(base_vha, 0); + qla8xxx_dev_failed_handler(base_vha); + rval = QLA_FUNCTION_FAILED; + qla83xx_idc_lock(base_vha, 0); + goto exit; + } + } + +exit: + return rval; +} + /************************************************************************** * qla2x00_do_dpc * This kernel thread is a task that is schedule by the interrupt handler @@ -3749,7 +4453,7 @@ qla2x00_do_dpc(void *data) &base_vha->dpc_flags)) { qla82xx_idc_lock(ha); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + QLA8XXX_DEV_FAILED); qla82xx_idc_unlock(ha); ql_log(ql_log_info, base_vha, 0x4004, "HW State: FAILED.\n"); @@ -3819,14 +4523,21 @@ qla2x00_do_dpc(void *data) if (test_bit(ISP_QUIESCE_NEEDED, &base_vha->dpc_flags)) { ql_dbg(ql_dbg_dpc, base_vha, 0x4009, "Quiescence mode scheduled.\n"); - qla82xx_device_state_handler(base_vha); - clear_bit(ISP_QUIESCE_NEEDED, &base_vha->dpc_flags); - if (!ha->flags.quiesce_owner) { - qla2x00_perform_loop_resync(base_vha); - - qla82xx_idc_lock(ha); - qla82xx_clear_qsnt_ready(base_vha); - qla82xx_idc_unlock(ha); + if (IS_QLA82XX(ha)) { + qla82xx_device_state_handler(base_vha); + clear_bit(ISP_QUIESCE_NEEDED, + &base_vha->dpc_flags); + if (!ha->flags.quiesce_owner) { + qla2x00_perform_loop_resync(base_vha); + + qla82xx_idc_lock(ha); + qla82xx_clear_qsnt_ready(base_vha); + qla82xx_idc_unlock(ha); + } + } else { + clear_bit(ISP_QUIESCE_NEEDED, + &base_vha->dpc_flags); + qla2x00_quiesce_io(base_vha); } ql_dbg(ql_dbg_dpc, base_vha, 0x400a, "Quiescence mode end.\n"); @@ -4326,7 +5037,7 @@ uint32_t qla82xx_error_recovery(scsi_qla_host_t *base_vha) qla82xx_idc_lock(ha); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_INITIALIZING); + QLA8XXX_DEV_INITIALIZING); qla82xx_wr_32(ha, QLA82XX_CRB_DRV_IDC_VERSION, QLA82XX_IDC_VERSION); @@ -4350,12 +5061,12 @@ uint32_t qla82xx_error_recovery(scsi_qla_host_t *base_vha) "HW State: FAILED.\n"); qla82xx_clear_drv_active(ha); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + QLA8XXX_DEV_FAILED); } else { ql_log(ql_log_info, base_vha, 0x900c, "HW State: READY.\n"); qla82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_READY); + QLA8XXX_DEV_READY); qla82xx_idc_unlock(ha); ha->flags.isp82xx_fw_hung = 0; rval = qla82xx_restart_isp(base_vha); @@ -4370,7 +5081,7 @@ uint32_t qla82xx_error_recovery(scsi_qla_host_t *base_vha) "This devfn is not reset owner = 0x%x.\n", ha->pdev->devfn); if ((qla82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE) == - QLA82XX_DEV_READY)) { + QLA8XXX_DEV_READY)) { ha->flags.isp82xx_fw_hung = 0; rval = qla82xx_restart_isp(base_vha); qla82xx_idc_lock(ha); @@ -4495,6 +5206,7 @@ static struct pci_device_id qla2xxx_pci_tbl[] = { { PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP2031) }, { PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP8001) }, { PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP8021) }, + { PCI_DEVICE(PCI_VENDOR_ID_QLOGIC, PCI_DEVICE_ID_QLOGIC_ISP8031) }, { 0 }, }; MODULE_DEVICE_TABLE(pci, qla2xxx_pci_tbl); diff --git a/drivers/scsi/qla2xxx/qla_settings.h b/drivers/scsi/qla2xxx/qla_settings.h index d70f030..892a81e 100644 --- a/drivers/scsi/qla2xxx/qla_settings.h +++ b/drivers/scsi/qla2xxx/qla_settings.h @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla2xxx/qla_sup.c b/drivers/scsi/qla2xxx/qla_sup.c index a683e76..32fdc2a 100644 --- a/drivers/scsi/qla2xxx/qla_sup.c +++ b/drivers/scsi/qla2xxx/qla_sup.c @@ -1,6 +1,6 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ @@ -966,16 +966,16 @@ qla2xxx_get_idc_param(scsi_qla_host_t *vha) QLA82XX_IDC_PARAM_ADDR , 8); if (*wptr == __constant_cpu_to_le32(0xffffffff)) { - ha->nx_dev_init_timeout = QLA82XX_ROM_DEV_INIT_TIMEOUT; - ha->nx_reset_timeout = QLA82XX_ROM_DRV_RESET_ACK_TIMEOUT; + ha->fcoe_dev_init_timeout = QLA82XX_ROM_DEV_INIT_TIMEOUT; + ha->fcoe_reset_timeout = QLA82XX_ROM_DRV_RESET_ACK_TIMEOUT; } else { - ha->nx_dev_init_timeout = le32_to_cpu(*wptr++); - ha->nx_reset_timeout = le32_to_cpu(*wptr); + ha->fcoe_dev_init_timeout = le32_to_cpu(*wptr++); + ha->fcoe_reset_timeout = le32_to_cpu(*wptr); } ql_dbg(ql_dbg_init, vha, 0x004e, - "nx_dev_init_timeout=%d " - "nx_reset_timeout=%d.\n", ha->nx_dev_init_timeout, - ha->nx_reset_timeout); + "fcoe_dev_init_timeout=%d " + "fcoe_reset_timeout=%d.\n", ha->fcoe_dev_init_timeout, + ha->fcoe_reset_timeout); return; } @@ -1017,7 +1017,7 @@ qla2xxx_flash_npiv_conf(scsi_qla_host_t *vha) !IS_CNA_CAPABLE(ha) && !IS_QLA2031(ha)) return; - if (ha->flags.isp82xx_reset_hdlr_active) + if (ha->flags.nic_core_reset_hdlr_active) return; ha->isp_ops->read_optrom(vha, (uint8_t *)&hdr, @@ -1662,6 +1662,23 @@ qla24xx_beacon_blink(struct scsi_qla_host *vha) spin_unlock_irqrestore(&ha->hardware_lock, flags); } +static uint32_t +qla83xx_select_led_port(struct qla_hw_data *ha) +{ + uint32_t led_select_value = 0; + + if (!IS_QLA83XX(ha)) + goto out; + + if (ha->flags.port0) + led_select_value = QLA83XX_LED_PORT0; + else + led_select_value = QLA83XX_LED_PORT1; + +out: + return led_select_value; +} + void qla83xx_beacon_blink(struct scsi_qla_host *vha) { @@ -1669,22 +1686,34 @@ qla83xx_beacon_blink(struct scsi_qla_host *vha) struct qla_hw_data *ha = vha->hw; uint16_t led_cfg[6]; uint16_t orig_led_cfg[6]; + uint32_t led_10_value, led_43_value; if (!IS_QLA83XX(ha) && !IS_QLA81XX(ha)) return; - if (IS_QLA2031(ha) && ha->beacon_blink_led) { - if (ha->flags.port0) - led_select_value = 0x00201320; - else - led_select_value = 0x00201328; + if (!ha->beacon_blink_led) + return; + + if (IS_QLA2031(ha)) { + led_select_value = qla83xx_select_led_port(ha); - qla83xx_write_remote_reg(vha, led_select_value, 0x40002000); - qla83xx_write_remote_reg(vha, led_select_value + 4, 0x40002000); + qla83xx_wr_reg(vha, led_select_value, 0x40002000); + qla83xx_wr_reg(vha, led_select_value + 4, 0x40002000); + msleep(1000); + qla83xx_wr_reg(vha, led_select_value, 0x40004000); + qla83xx_wr_reg(vha, led_select_value + 4, 0x40004000); + } else if (IS_QLA8031(ha)) { + led_select_value = qla83xx_select_led_port(ha); + + qla83xx_rd_reg(vha, led_select_value, &led_10_value); + qla83xx_rd_reg(vha, led_select_value + 0x10, &led_43_value); + qla83xx_wr_reg(vha, led_select_value, 0x01f44000); + msleep(500); + qla83xx_wr_reg(vha, led_select_value, 0x400001f4); msleep(1000); - qla83xx_write_remote_reg(vha, led_select_value, 0x40004000); - qla83xx_write_remote_reg(vha, led_select_value + 4, 0x40004000); - } else if ((IS_QLA8031(ha) || IS_QLA81XX(ha)) && ha->beacon_blink_led) { + qla83xx_wr_reg(vha, led_select_value, led_10_value); + qla83xx_wr_reg(vha, led_select_value + 0x10, led_43_value); + } else if (IS_QLA81XX(ha)) { int rval; /* Save Current */ diff --git a/drivers/scsi/qla2xxx/qla_version.h b/drivers/scsi/qla2xxx/qla_version.h index f5fdb16..cfe934e 100644 --- a/drivers/scsi/qla2xxx/qla_version.h +++ b/drivers/scsi/qla2xxx/qla_version.h @@ -1,15 +1,15 @@ /* * QLogic Fibre Channel HBA Driver - * Copyright (c) 2003-2011 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla2xxx for copyright and licensing details. */ /* * Driver version */ -#define QLA2XXX_VERSION "8.04.00.03-k" +#define QLA2XXX_VERSION "8.04.00.07-k" #define QLA_DRIVER_MAJOR_VER 8 #define QLA_DRIVER_MINOR_VER 4 #define QLA_DRIVER_PATCH_VER 0 -#define QLA_DRIVER_BETA_VER 3 +#define QLA_DRIVER_BETA_VER 0 diff --git a/drivers/scsi/qla4xxx/Kconfig b/drivers/scsi/qla4xxx/Kconfig index f1ad02e..e4dc7c7 100644 --- a/drivers/scsi/qla4xxx/Kconfig +++ b/drivers/scsi/qla4xxx/Kconfig @@ -4,5 +4,5 @@ config SCSI_QLA_ISCSI select SCSI_ISCSI_ATTRS select ISCSI_BOOT_SYSFS ---help--- - This driver supports the QLogic 40xx (ISP4XXX) and 8022 (ISP82XX) - iSCSI host adapter family. + This driver supports the QLogic 40xx (ISP4XXX), 8022 (ISP82XX) + and 8032 (ISP83XX) iSCSI host adapter family. diff --git a/drivers/scsi/qla4xxx/Makefile b/drivers/scsi/qla4xxx/Makefile index 5b44139..4230977 100644 --- a/drivers/scsi/qla4xxx/Makefile +++ b/drivers/scsi/qla4xxx/Makefile @@ -1,5 +1,5 @@ qla4xxx-y := ql4_os.o ql4_init.o ql4_mbx.o ql4_iocb.o ql4_isr.o \ - ql4_nx.o ql4_nvram.o ql4_dbg.o ql4_attr.o ql4_bsg.o + ql4_nx.o ql4_nvram.o ql4_dbg.o ql4_attr.o ql4_bsg.o ql4_83xx.o obj-$(CONFIG_SCSI_QLA_ISCSI) += qla4xxx.o diff --git a/drivers/scsi/qla4xxx/ql4_83xx.c b/drivers/scsi/qla4xxx/ql4_83xx.c new file mode 100644 index 0000000..6e9af20 --- /dev/null +++ b/drivers/scsi/qla4xxx/ql4_83xx.c @@ -0,0 +1,1611 @@ +/* + * QLogic iSCSI HBA Driver + * Copyright (c) 2003-2012 QLogic Corporation + * + * See LICENSE.qla4xxx for copyright and licensing details. + */ + +#include <linux/ratelimit.h> + +#include "ql4_def.h" +#include "ql4_version.h" +#include "ql4_glbl.h" +#include "ql4_dbg.h" +#include "ql4_inline.h" + +uint32_t qla4_83xx_rd_reg(struct scsi_qla_host *ha, ulong addr) +{ + return readl((void __iomem *)(ha->nx_pcibase + addr)); +} + +void qla4_83xx_wr_reg(struct scsi_qla_host *ha, ulong addr, uint32_t val) +{ + writel(val, (void __iomem *)(ha->nx_pcibase + addr)); +} + +static int qla4_83xx_set_win_base(struct scsi_qla_host *ha, uint32_t addr) +{ + uint32_t val; + int ret_val = QLA_SUCCESS; + + qla4_83xx_wr_reg(ha, QLA83XX_CRB_WIN_FUNC(ha->func_num), addr); + val = qla4_83xx_rd_reg(ha, QLA83XX_CRB_WIN_FUNC(ha->func_num)); + if (val != addr) { + ql4_printk(KERN_ERR, ha, "%s: Failed to set register window : addr written 0x%x, read 0x%x!\n", + __func__, addr, val); + ret_val = QLA_ERROR; + } + + return ret_val; +} + +int qla4_83xx_rd_reg_indirect(struct scsi_qla_host *ha, uint32_t addr, + uint32_t *data) +{ + int ret_val; + + ret_val = qla4_83xx_set_win_base(ha, addr); + + if (ret_val == QLA_SUCCESS) + *data = qla4_83xx_rd_reg(ha, QLA83XX_WILDCARD); + else + ql4_printk(KERN_ERR, ha, "%s: failed read of addr 0x%x!\n", + __func__, addr); + + return ret_val; +} + +int qla4_83xx_wr_reg_indirect(struct scsi_qla_host *ha, uint32_t addr, + uint32_t data) +{ + int ret_val; + + ret_val = qla4_83xx_set_win_base(ha, addr); + + if (ret_val == QLA_SUCCESS) + qla4_83xx_wr_reg(ha, QLA83XX_WILDCARD, data); + else + ql4_printk(KERN_ERR, ha, "%s: failed wrt to addr 0x%x, data 0x%x\n", + __func__, addr, data); + + return ret_val; +} + +static int qla4_83xx_flash_lock(struct scsi_qla_host *ha) +{ + int lock_owner; + int timeout = 0; + uint32_t lock_status = 0; + int ret_val = QLA_SUCCESS; + + while (lock_status == 0) { + lock_status = qla4_83xx_rd_reg(ha, QLA83XX_FLASH_LOCK); + if (lock_status) + break; + + if (++timeout >= QLA83XX_FLASH_LOCK_TIMEOUT / 20) { + lock_owner = qla4_83xx_rd_reg(ha, + QLA83XX_FLASH_LOCK_ID); + ql4_printk(KERN_ERR, ha, "%s: flash lock by func %d failed, held by func %d\n", + __func__, ha->func_num, lock_owner); + ret_val = QLA_ERROR; + break; + } + msleep(20); + } + + qla4_83xx_wr_reg(ha, QLA83XX_FLASH_LOCK_ID, ha->func_num); + return ret_val; +} + +static void qla4_83xx_flash_unlock(struct scsi_qla_host *ha) +{ + /* Reading FLASH_UNLOCK register unlocks the Flash */ + qla4_83xx_wr_reg(ha, QLA83XX_FLASH_LOCK_ID, 0xFF); + qla4_83xx_rd_reg(ha, QLA83XX_FLASH_UNLOCK); +} + +int qla4_83xx_flash_read_u32(struct scsi_qla_host *ha, uint32_t flash_addr, + uint8_t *p_data, int u32_word_count) +{ + int i; + uint32_t u32_word; + uint32_t addr = flash_addr; + int ret_val = QLA_SUCCESS; + + ret_val = qla4_83xx_flash_lock(ha); + if (ret_val == QLA_ERROR) + goto exit_lock_error; + + if (addr & 0x03) { + ql4_printk(KERN_ERR, ha, "%s: Illegal addr = 0x%x\n", + __func__, addr); + ret_val = QLA_ERROR; + goto exit_flash_read; + } + + for (i = 0; i < u32_word_count; i++) { + ret_val = qla4_83xx_wr_reg_indirect(ha, + QLA83XX_FLASH_DIRECT_WINDOW, + (addr & 0xFFFF0000)); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to write addr 0x%x to FLASH_DIRECT_WINDOW\n!", + __func__, addr); + goto exit_flash_read; + } + + ret_val = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_FLASH_DIRECT_DATA(addr), + &u32_word); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to read addr 0x%x!\n", + __func__, addr); + goto exit_flash_read; + } + + *(__le32 *)p_data = le32_to_cpu(u32_word); + p_data = p_data + 4; + addr = addr + 4; + } + +exit_flash_read: + qla4_83xx_flash_unlock(ha); + +exit_lock_error: + return ret_val; +} + +int qla4_83xx_lockless_flash_read_u32(struct scsi_qla_host *ha, + uint32_t flash_addr, uint8_t *p_data, + int u32_word_count) +{ + uint32_t i; + uint32_t u32_word; + uint32_t flash_offset; + uint32_t addr = flash_addr; + int ret_val = QLA_SUCCESS; + + flash_offset = addr & (QLA83XX_FLASH_SECTOR_SIZE - 1); + + if (addr & 0x3) { + ql4_printk(KERN_ERR, ha, "%s: Illegal addr = 0x%x\n", + __func__, addr); + ret_val = QLA_ERROR; + goto exit_lockless_read; + } + + ret_val = qla4_83xx_wr_reg_indirect(ha, QLA83XX_FLASH_DIRECT_WINDOW, + addr); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to write addr 0x%x to FLASH_DIRECT_WINDOW!\n", + __func__, addr); + goto exit_lockless_read; + } + + /* Check if data is spread across multiple sectors */ + if ((flash_offset + (u32_word_count * sizeof(uint32_t))) > + (QLA83XX_FLASH_SECTOR_SIZE - 1)) { + + /* Multi sector read */ + for (i = 0; i < u32_word_count; i++) { + ret_val = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_FLASH_DIRECT_DATA(addr), + &u32_word); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to read addr 0x%x!\n", + __func__, addr); + goto exit_lockless_read; + } + + *(__le32 *)p_data = le32_to_cpu(u32_word); + p_data = p_data + 4; + addr = addr + 4; + flash_offset = flash_offset + 4; + + if (flash_offset > (QLA83XX_FLASH_SECTOR_SIZE - 1)) { + /* This write is needed once for each sector */ + ret_val = qla4_83xx_wr_reg_indirect(ha, + QLA83XX_FLASH_DIRECT_WINDOW, + addr); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to write addr 0x%x to FLASH_DIRECT_WINDOW!\n", + __func__, addr); + goto exit_lockless_read; + } + flash_offset = 0; + } + } + } else { + /* Single sector read */ + for (i = 0; i < u32_word_count; i++) { + ret_val = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_FLASH_DIRECT_DATA(addr), + &u32_word); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to read addr 0x%x!\n", + __func__, addr); + goto exit_lockless_read; + } + + *(__le32 *)p_data = le32_to_cpu(u32_word); + p_data = p_data + 4; + addr = addr + 4; + } + } + +exit_lockless_read: + return ret_val; +} + +void qla4_83xx_rom_lock_recovery(struct scsi_qla_host *ha) +{ + if (qla4_83xx_flash_lock(ha)) + ql4_printk(KERN_INFO, ha, "%s: Resetting rom lock\n", __func__); + + /* + * We got the lock, or someone else is holding the lock + * since we are restting, forcefully unlock + */ + qla4_83xx_flash_unlock(ha); +} + +/** + * qla4_83xx_ms_mem_write_128b - Writes data to MS/off-chip memory + * @ha: Pointer to adapter structure + * @addr: Flash address to write to + * @data: Data to be written + * @count: word_count to be written + * + * Return: On success return QLA_SUCCESS + * On error return QLA_ERROR + **/ +static int qla4_83xx_ms_mem_write_128b(struct scsi_qla_host *ha, uint64_t addr, + uint32_t *data, uint32_t count) +{ + int i, j; + uint32_t agt_ctrl; + unsigned long flags; + int ret_val = QLA_SUCCESS; + + /* Only 128-bit aligned access */ + if (addr & 0xF) { + ret_val = QLA_ERROR; + goto exit_ms_mem_write; + } + + write_lock_irqsave(&ha->hw_lock, flags); + + /* Write address */ + ret_val = qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_HI, 0); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: write to AGT_ADDR_HI failed\n", + __func__); + goto exit_ms_mem_write_unlock; + } + + for (i = 0; i < count; i++, addr += 16) { + if (!((QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_QDR_NET, + QLA8XXX_ADDR_QDR_NET_MAX)) || + (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_DDR_NET, + QLA8XXX_ADDR_DDR_NET_MAX)))) { + ret_val = QLA_ERROR; + goto exit_ms_mem_write_unlock; + } + + ret_val = qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_LO, + addr); + /* Write data */ + ret_val |= qla4_83xx_wr_reg_indirect(ha, + MD_MIU_TEST_AGT_WRDATA_LO, + *data++); + ret_val |= qla4_83xx_wr_reg_indirect(ha, + MD_MIU_TEST_AGT_WRDATA_HI, + *data++); + ret_val |= qla4_83xx_wr_reg_indirect(ha, + MD_MIU_TEST_AGT_WRDATA_ULO, + *data++); + ret_val |= qla4_83xx_wr_reg_indirect(ha, + MD_MIU_TEST_AGT_WRDATA_UHI, + *data++); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: write to AGT_WRDATA failed\n", + __func__); + goto exit_ms_mem_write_unlock; + } + + /* Check write status */ + ret_val = qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL, + MIU_TA_CTL_WRITE_ENABLE); + ret_val |= qla4_83xx_wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL, + MIU_TA_CTL_WRITE_START); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: write to AGT_CTRL failed\n", + __func__); + goto exit_ms_mem_write_unlock; + } + + for (j = 0; j < MAX_CTL_CHECK; j++) { + ret_val = qla4_83xx_rd_reg_indirect(ha, + MD_MIU_TEST_AGT_CTRL, + &agt_ctrl); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: failed to read MD_MIU_TEST_AGT_CTRL\n", + __func__); + goto exit_ms_mem_write_unlock; + } + if ((agt_ctrl & MIU_TA_CTL_BUSY) == 0) + break; + } + + /* Status check failed */ + if (j >= MAX_CTL_CHECK) { + printk_ratelimited(KERN_ERR "%s: MS memory write failed!\n", + __func__); + ret_val = QLA_ERROR; + goto exit_ms_mem_write_unlock; + } + } + +exit_ms_mem_write_unlock: + write_unlock_irqrestore(&ha->hw_lock, flags); + +exit_ms_mem_write: + return ret_val; +} + +#define INTENT_TO_RECOVER 0x01 +#define PROCEED_TO_RECOVER 0x02 + +static int qla4_83xx_lock_recovery(struct scsi_qla_host *ha) +{ + + uint32_t lock = 0, lockid; + int ret_val = QLA_ERROR; + + lockid = ha->isp_ops->rd_reg_direct(ha, QLA83XX_DRV_LOCKRECOVERY); + + /* Check for other Recovery in progress, go wait */ + if ((lockid & 0x3) != 0) + goto exit_lock_recovery; + + /* Intent to Recover */ + ha->isp_ops->wr_reg_direct(ha, QLA83XX_DRV_LOCKRECOVERY, + (ha->func_num << 2) | INTENT_TO_RECOVER); + + msleep(200); + + /* Check Intent to Recover is advertised */ + lockid = ha->isp_ops->rd_reg_direct(ha, QLA83XX_DRV_LOCKRECOVERY); + if ((lockid & 0x3C) != (ha->func_num << 2)) + goto exit_lock_recovery; + + ql4_printk(KERN_INFO, ha, "%s: IDC Lock recovery initiated for func %d\n", + __func__, ha->func_num); + + /* Proceed to Recover */ + ha->isp_ops->wr_reg_direct(ha, QLA83XX_DRV_LOCKRECOVERY, + (ha->func_num << 2) | PROCEED_TO_RECOVER); + + /* Force Unlock */ + ha->isp_ops->wr_reg_direct(ha, QLA83XX_DRV_LOCK_ID, 0xFF); + ha->isp_ops->rd_reg_direct(ha, QLA83XX_DRV_UNLOCK); + + /* Clear bits 0-5 in IDC_RECOVERY register*/ + ha->isp_ops->wr_reg_direct(ha, QLA83XX_DRV_LOCKRECOVERY, 0); + + /* Get lock */ + lock = ha->isp_ops->rd_reg_direct(ha, QLA83XX_DRV_LOCK); + if (lock) { + lockid = ha->isp_ops->rd_reg_direct(ha, QLA83XX_DRV_LOCK_ID); + lockid = ((lockid + (1 << 8)) & ~0xFF) | ha->func_num; + ha->isp_ops->wr_reg_direct(ha, QLA83XX_DRV_LOCK_ID, lockid); + ret_val = QLA_SUCCESS; + } + +exit_lock_recovery: + return ret_val; +} + +#define QLA83XX_DRV_LOCK_MSLEEP 200 + +int qla4_83xx_drv_lock(struct scsi_qla_host *ha) +{ + int timeout = 0; + uint32_t status = 0; + int ret_val = QLA_SUCCESS; + uint32_t first_owner = 0; + uint32_t tmo_owner = 0; + uint32_t lock_id; + uint32_t func_num; + uint32_t lock_cnt; + + while (status == 0) { + status = qla4_83xx_rd_reg(ha, QLA83XX_DRV_LOCK); + if (status) { + /* Increment Counter (8-31) and update func_num (0-7) on + * getting a successful lock */ + lock_id = qla4_83xx_rd_reg(ha, QLA83XX_DRV_LOCK_ID); + lock_id = ((lock_id + (1 << 8)) & ~0xFF) | ha->func_num; + qla4_83xx_wr_reg(ha, QLA83XX_DRV_LOCK_ID, lock_id); + break; + } + + if (timeout == 0) + /* Save counter + ID of function holding the lock for + * first failure */ + first_owner = ha->isp_ops->rd_reg_direct(ha, + QLA83XX_DRV_LOCK_ID); + + if (++timeout >= + (QLA83XX_DRV_LOCK_TIMEOUT / QLA83XX_DRV_LOCK_MSLEEP)) { + tmo_owner = qla4_83xx_rd_reg(ha, QLA83XX_DRV_LOCK_ID); + func_num = tmo_owner & 0xFF; + lock_cnt = tmo_owner >> 8; + ql4_printk(KERN_INFO, ha, "%s: Lock by func %d failed after 2s, lock held by func %d, lock count %d, first_owner %d\n", + __func__, ha->func_num, func_num, lock_cnt, + (first_owner & 0xFF)); + + if (first_owner != tmo_owner) { + /* Some other driver got lock, OR same driver + * got lock again (counter value changed), when + * we were waiting for lock. + * Retry for another 2 sec */ + ql4_printk(KERN_INFO, ha, "%s: IDC lock failed for func %d\n", + __func__, ha->func_num); + timeout = 0; + } else { + /* Same driver holding lock > 2sec. + * Force Recovery */ + ret_val = qla4_83xx_lock_recovery(ha); + if (ret_val == QLA_SUCCESS) { + /* Recovered and got lock */ + ql4_printk(KERN_INFO, ha, "%s: IDC lock Recovery by %d successful\n", + __func__, ha->func_num); + break; + } + /* Recovery Failed, some other function + * has the lock, wait for 2secs and retry */ + ql4_printk(KERN_INFO, ha, "%s: IDC lock Recovery by %d failed, Retrying timout\n", + __func__, ha->func_num); + timeout = 0; + } + } + msleep(QLA83XX_DRV_LOCK_MSLEEP); + } + + return ret_val; +} + +void qla4_83xx_drv_unlock(struct scsi_qla_host *ha) +{ + int id; + + id = qla4_83xx_rd_reg(ha, QLA83XX_DRV_LOCK_ID); + + if ((id & 0xFF) != ha->func_num) { + ql4_printk(KERN_ERR, ha, "%s: IDC Unlock by %d failed, lock owner is %d\n", + __func__, ha->func_num, (id & 0xFF)); + return; + } + + /* Keep lock counter value, update the ha->func_num to 0xFF */ + qla4_83xx_wr_reg(ha, QLA83XX_DRV_LOCK_ID, (id | 0xFF)); + qla4_83xx_rd_reg(ha, QLA83XX_DRV_UNLOCK); +} + +void qla4_83xx_set_idc_dontreset(struct scsi_qla_host *ha) +{ + uint32_t idc_ctrl; + + idc_ctrl = qla4_83xx_rd_reg(ha, QLA83XX_IDC_DRV_CTRL); + idc_ctrl |= DONTRESET_BIT0; + qla4_83xx_wr_reg(ha, QLA83XX_IDC_DRV_CTRL, idc_ctrl); + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: idc_ctrl = %d\n", __func__, + idc_ctrl)); +} + +void qla4_83xx_clear_idc_dontreset(struct scsi_qla_host *ha) +{ + uint32_t idc_ctrl; + + idc_ctrl = qla4_83xx_rd_reg(ha, QLA83XX_IDC_DRV_CTRL); + idc_ctrl &= ~DONTRESET_BIT0; + qla4_83xx_wr_reg(ha, QLA83XX_IDC_DRV_CTRL, idc_ctrl); + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: idc_ctrl = %d\n", __func__, + idc_ctrl)); +} + +int qla4_83xx_idc_dontreset(struct scsi_qla_host *ha) +{ + uint32_t idc_ctrl; + + idc_ctrl = qla4_83xx_rd_reg(ha, QLA83XX_IDC_DRV_CTRL); + return idc_ctrl & DONTRESET_BIT0; +} + +/*-------------------------IDC State Machine ---------------------*/ + +enum { + UNKNOWN_CLASS = 0, + NIC_CLASS, + FCOE_CLASS, + ISCSI_CLASS +}; + +struct device_info { + int func_num; + int device_type; + int port_num; +}; + +static int qla4_83xx_can_perform_reset(struct scsi_qla_host *ha) +{ + uint32_t drv_active; + uint32_t dev_part, dev_part1, dev_part2; + int i; + struct device_info device_map[16]; + int func_nibble; + int nibble; + int nic_present = 0; + int iscsi_present = 0; + int iscsi_func_low = 0; + + /* Use the dev_partition register to determine the PCI function number + * and then check drv_active register to see which driver is loaded */ + dev_part1 = qla4_83xx_rd_reg(ha, + ha->reg_tbl[QLA8XXX_CRB_DEV_PART_INFO]); + dev_part2 = qla4_83xx_rd_reg(ha, QLA83XX_CRB_DEV_PART_INFO2); + drv_active = qla4_83xx_rd_reg(ha, ha->reg_tbl[QLA8XXX_CRB_DRV_ACTIVE]); + + /* Each function has 4 bits in dev_partition Info register, + * Lower 2 bits - device type, Upper 2 bits - physical port number */ + dev_part = dev_part1; + for (i = nibble = 0; i <= 15; i++, nibble++) { + func_nibble = dev_part & (0xF << (nibble * 4)); + func_nibble >>= (nibble * 4); + device_map[i].func_num = i; + device_map[i].device_type = func_nibble & 0x3; + device_map[i].port_num = func_nibble & 0xC; + + if (device_map[i].device_type == NIC_CLASS) { + if (drv_active & (1 << device_map[i].func_num)) { + nic_present++; + break; + } + } else if (device_map[i].device_type == ISCSI_CLASS) { + if (drv_active & (1 << device_map[i].func_num)) { + if (!iscsi_present || + (iscsi_present && + (iscsi_func_low > device_map[i].func_num))) + iscsi_func_low = device_map[i].func_num; + + iscsi_present++; + } + } + + /* For function_num[8..15] get info from dev_part2 register */ + if (nibble == 7) { + nibble = 0; + dev_part = dev_part2; + } + } + + /* NIC, iSCSI and FCOE are the Reset owners based on order, NIC gets + * precedence over iSCSI and FCOE and iSCSI over FCOE, based on drivers + * present. */ + if (!nic_present && (ha->func_num == iscsi_func_low)) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: can reset - NIC not present and lower iSCSI function is %d\n", + __func__, ha->func_num)); + return 1; + } + + return 0; +} + +/** + * qla4_83xx_need_reset_handler - Code to start reset sequence + * @ha: pointer to adapter structure + * + * Note: IDC lock must be held upon entry + **/ +void qla4_83xx_need_reset_handler(struct scsi_qla_host *ha) +{ + uint32_t dev_state, drv_state, drv_active; + unsigned long reset_timeout, dev_init_timeout; + + ql4_printk(KERN_INFO, ha, "%s: Performing ISP error recovery\n", + __func__); + + if (!test_bit(AF_8XXX_RST_OWNER, &ha->flags)) { + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: reset acknowledged\n", + __func__)); + qla4_8xxx_set_rst_ready(ha); + + /* Non-reset owners ACK Reset and wait for device INIT state + * as part of Reset Recovery by Reset Owner */ + dev_init_timeout = jiffies + (ha->nx_dev_init_timeout * HZ); + + do { + if (time_after_eq(jiffies, dev_init_timeout)) { + ql4_printk(KERN_INFO, ha, "%s: Non Reset owner dev init timeout\n", + __func__); + break; + } + + ha->isp_ops->idc_unlock(ha); + msleep(1000); + ha->isp_ops->idc_lock(ha); + + dev_state = qla4_8xxx_rd_direct(ha, + QLA8XXX_CRB_DEV_STATE); + } while (dev_state == QLA8XXX_DEV_NEED_RESET); + } else { + qla4_8xxx_set_rst_ready(ha); + reset_timeout = jiffies + (ha->nx_reset_timeout * HZ); + drv_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_STATE); + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + + ql4_printk(KERN_INFO, ha, "%s: drv_state = 0x%x, drv_active = 0x%x\n", + __func__, drv_state, drv_active); + + while (drv_state != drv_active) { + if (time_after_eq(jiffies, reset_timeout)) { + ql4_printk(KERN_INFO, ha, "%s: %s: RESET TIMEOUT! drv_state: 0x%08x, drv_active: 0x%08x\n", + __func__, DRIVER_NAME, drv_state, + drv_active); + break; + } + + ha->isp_ops->idc_unlock(ha); + msleep(1000); + ha->isp_ops->idc_lock(ha); + + drv_state = qla4_8xxx_rd_direct(ha, + QLA8XXX_CRB_DRV_STATE); + drv_active = qla4_8xxx_rd_direct(ha, + QLA8XXX_CRB_DRV_ACTIVE); + } + + if (drv_state != drv_active) { + ql4_printk(KERN_INFO, ha, "%s: Reset_owner turning off drv_active of non-acking function 0x%x\n", + __func__, (drv_active ^ drv_state)); + drv_active = drv_active & drv_state; + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_ACTIVE, + drv_active); + } + + clear_bit(AF_8XXX_RST_OWNER, &ha->flags); + /* Start Reset Recovery */ + qla4_8xxx_device_bootstrap(ha); + } +} + +void qla4_83xx_get_idc_param(struct scsi_qla_host *ha) +{ + uint32_t idc_params, ret_val; + + ret_val = qla4_83xx_flash_read_u32(ha, QLA83XX_IDC_PARAM_ADDR, + (uint8_t *)&idc_params, 1); + if (ret_val == QLA_SUCCESS) { + ha->nx_dev_init_timeout = idc_params & 0xFFFF; + ha->nx_reset_timeout = (idc_params >> 16) & 0xFFFF; + } else { + ha->nx_dev_init_timeout = ROM_DEV_INIT_TIMEOUT; + ha->nx_reset_timeout = ROM_DRV_RESET_ACK_TIMEOUT; + } + + DEBUG2(ql4_printk(KERN_DEBUG, ha, + "%s: ha->nx_dev_init_timeout = %d, ha->nx_reset_timeout = %d\n", + __func__, ha->nx_dev_init_timeout, + ha->nx_reset_timeout)); +} + +/*-------------------------Reset Sequence Functions-----------------------*/ + +static void qla4_83xx_dump_reset_seq_hdr(struct scsi_qla_host *ha) +{ + uint8_t *phdr; + + if (!ha->reset_tmplt.buff) { + ql4_printk(KERN_ERR, ha, "%s: Error: Invalid reset_seq_template\n", + __func__); + return; + } + + phdr = ha->reset_tmplt.buff; + + DEBUG2(ql4_printk(KERN_INFO, ha, + "Reset Template: 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", + *phdr, *(phdr+1), *(phdr+2), *(phdr+3), *(phdr+4), + *(phdr+5), *(phdr+6), *(phdr+7), *(phdr + 8), + *(phdr+9), *(phdr+10), *(phdr+11), *(phdr+12), + *(phdr+13), *(phdr+14), *(phdr+15))); +} + +static int qla4_83xx_copy_bootloader(struct scsi_qla_host *ha) +{ + uint8_t *p_cache; + uint32_t src, count, size; + uint64_t dest; + int ret_val = QLA_SUCCESS; + + src = QLA83XX_BOOTLOADER_FLASH_ADDR; + dest = qla4_83xx_rd_reg(ha, QLA83XX_BOOTLOADER_ADDR); + size = qla4_83xx_rd_reg(ha, QLA83XX_BOOTLOADER_SIZE); + + /* 128 bit alignment check */ + if (size & 0xF) + size = (size + 16) & ~0xF; + + /* 16 byte count */ + count = size/16; + + p_cache = vmalloc(size); + if (p_cache == NULL) { + ql4_printk(KERN_ERR, ha, "%s: Failed to allocate memory for boot loader cache\n", + __func__); + ret_val = QLA_ERROR; + goto exit_copy_bootloader; + } + + ret_val = qla4_83xx_lockless_flash_read_u32(ha, src, p_cache, + size / sizeof(uint32_t)); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: Error reading firmware from flash\n", + __func__); + goto exit_copy_error; + } + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: Read firmware from flash\n", + __func__)); + + /* 128 bit/16 byte write to MS memory */ + ret_val = qla4_83xx_ms_mem_write_128b(ha, dest, (uint32_t *)p_cache, + count); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: Error writing firmware to MS\n", + __func__); + goto exit_copy_error; + } + + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: Wrote firmware size %d to MS\n", + __func__, size)); + +exit_copy_error: + vfree(p_cache); + +exit_copy_bootloader: + return ret_val; +} + +static int qla4_83xx_check_cmd_peg_status(struct scsi_qla_host *ha) +{ + uint32_t val, ret_val = QLA_ERROR; + int retries = CRB_CMDPEG_CHECK_RETRY_COUNT; + + do { + val = qla4_83xx_rd_reg(ha, QLA83XX_CMDPEG_STATE); + if (val == PHAN_INITIALIZE_COMPLETE) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Command Peg initialization complete. State=0x%x\n", + __func__, val)); + ret_val = QLA_SUCCESS; + break; + } + msleep(CRB_CMDPEG_CHECK_DELAY); + } while (--retries); + + return ret_val; +} + +/** + * qla4_83xx_poll_reg - Poll the given CRB addr for duration msecs till + * value read ANDed with test_mask is equal to test_result. + * + * @ha : Pointer to adapter structure + * @addr : CRB register address + * @duration : Poll for total of "duration" msecs + * @test_mask : Mask value read with "test_mask" + * @test_result : Compare (value&test_mask) with test_result. + **/ +static int qla4_83xx_poll_reg(struct scsi_qla_host *ha, uint32_t addr, + int duration, uint32_t test_mask, + uint32_t test_result) +{ + uint32_t value; + uint8_t retries; + int ret_val = QLA_SUCCESS; + + ret_val = qla4_83xx_rd_reg_indirect(ha, addr, &value); + if (ret_val == QLA_ERROR) + goto exit_poll_reg; + + retries = duration / 10; + do { + if ((value & test_mask) != test_result) { + msleep(duration / 10); + ret_val = qla4_83xx_rd_reg_indirect(ha, addr, &value); + if (ret_val == QLA_ERROR) + goto exit_poll_reg; + + ret_val = QLA_ERROR; + } else { + ret_val = QLA_SUCCESS; + break; + } + } while (retries--); + +exit_poll_reg: + if (ret_val == QLA_ERROR) { + ha->reset_tmplt.seq_error++; + ql4_printk(KERN_ERR, ha, "%s: Poll Failed: 0x%08x 0x%08x 0x%08x\n", + __func__, value, test_mask, test_result); + } + + return ret_val; +} + +static int qla4_83xx_reset_seq_checksum_test(struct scsi_qla_host *ha) +{ + uint32_t sum = 0; + uint16_t *buff = (uint16_t *)ha->reset_tmplt.buff; + int u16_count = ha->reset_tmplt.hdr->size / sizeof(uint16_t); + int ret_val; + + while (u16_count-- > 0) + sum += *buff++; + + while (sum >> 16) + sum = (sum & 0xFFFF) + (sum >> 16); + + /* checksum of 0 indicates a valid template */ + if (~sum) { + ret_val = QLA_SUCCESS; + } else { + ql4_printk(KERN_ERR, ha, "%s: Reset seq checksum failed\n", + __func__); + ret_val = QLA_ERROR; + } + + return ret_val; +} + +/** + * qla4_83xx_read_reset_template - Read Reset Template from Flash + * @ha: Pointer to adapter structure + **/ +void qla4_83xx_read_reset_template(struct scsi_qla_host *ha) +{ + uint8_t *p_buff; + uint32_t addr, tmplt_hdr_def_size, tmplt_hdr_size; + uint32_t ret_val; + + ha->reset_tmplt.seq_error = 0; + ha->reset_tmplt.buff = vmalloc(QLA83XX_RESTART_TEMPLATE_SIZE); + if (ha->reset_tmplt.buff == NULL) { + ql4_printk(KERN_ERR, ha, "%s: Failed to allocate reset template resources\n", + __func__); + goto exit_read_reset_template; + } + + p_buff = ha->reset_tmplt.buff; + addr = QLA83XX_RESET_TEMPLATE_ADDR; + + tmplt_hdr_def_size = sizeof(struct qla4_83xx_reset_template_hdr) / + sizeof(uint32_t); + + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Read template hdr size %d from Flash\n", + __func__, tmplt_hdr_def_size)); + + /* Copy template header from flash */ + ret_val = qla4_83xx_flash_read_u32(ha, addr, p_buff, + tmplt_hdr_def_size); + if (ret_val != QLA_SUCCESS) { + ql4_printk(KERN_ERR, ha, "%s: Failed to read reset template\n", + __func__); + goto exit_read_template_error; + } + + ha->reset_tmplt.hdr = + (struct qla4_83xx_reset_template_hdr *)ha->reset_tmplt.buff; + + /* Validate the template header size and signature */ + tmplt_hdr_size = ha->reset_tmplt.hdr->hdr_size/sizeof(uint32_t); + if ((tmplt_hdr_size != tmplt_hdr_def_size) || + (ha->reset_tmplt.hdr->signature != RESET_TMPLT_HDR_SIGNATURE)) { + ql4_printk(KERN_ERR, ha, "%s: Template Header size %d is invalid, tmplt_hdr_def_size %d\n", + __func__, tmplt_hdr_size, tmplt_hdr_def_size); + goto exit_read_template_error; + } + + addr = QLA83XX_RESET_TEMPLATE_ADDR + ha->reset_tmplt.hdr->hdr_size; + p_buff = ha->reset_tmplt.buff + ha->reset_tmplt.hdr->hdr_size; + tmplt_hdr_def_size = (ha->reset_tmplt.hdr->size - + ha->reset_tmplt.hdr->hdr_size) / sizeof(uint32_t); + + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Read rest of the template size %d\n", + __func__, ha->reset_tmplt.hdr->size)); + + /* Copy rest of the template */ + ret_val = qla4_83xx_flash_read_u32(ha, addr, p_buff, + tmplt_hdr_def_size); + if (ret_val != QLA_SUCCESS) { + ql4_printk(KERN_ERR, ha, "%s: Failed to read reset tempelate\n", + __func__); + goto exit_read_template_error; + } + + /* Integrity check */ + if (qla4_83xx_reset_seq_checksum_test(ha)) { + ql4_printk(KERN_ERR, ha, "%s: Reset Seq checksum failed!\n", + __func__); + goto exit_read_template_error; + } + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Reset Seq checksum passed, Get stop, start and init seq offsets\n", + __func__)); + + /* Get STOP, START, INIT sequence offsets */ + ha->reset_tmplt.init_offset = ha->reset_tmplt.buff + + ha->reset_tmplt.hdr->init_seq_offset; + ha->reset_tmplt.start_offset = ha->reset_tmplt.buff + + ha->reset_tmplt.hdr->start_seq_offset; + ha->reset_tmplt.stop_offset = ha->reset_tmplt.buff + + ha->reset_tmplt.hdr->hdr_size; + qla4_83xx_dump_reset_seq_hdr(ha); + + goto exit_read_reset_template; + +exit_read_template_error: + vfree(ha->reset_tmplt.buff); + +exit_read_reset_template: + return; +} + +/** + * qla4_83xx_read_write_crb_reg - Read from raddr and write value to waddr. + * + * @ha : Pointer to adapter structure + * @raddr : CRB address to read from + * @waddr : CRB address to write to + **/ +static void qla4_83xx_read_write_crb_reg(struct scsi_qla_host *ha, + uint32_t raddr, uint32_t waddr) +{ + uint32_t value; + + qla4_83xx_rd_reg_indirect(ha, raddr, &value); + qla4_83xx_wr_reg_indirect(ha, waddr, value); +} + +/** + * qla4_83xx_rmw_crb_reg - Read Modify Write crb register + * + * This function read value from raddr, AND with test_mask, + * Shift Left,Right/OR/XOR with values RMW header and write value to waddr. + * + * @ha : Pointer to adapter structure + * @raddr : CRB address to read from + * @waddr : CRB address to write to + * @p_rmw_hdr : header with shift/or/xor values. + **/ +static void qla4_83xx_rmw_crb_reg(struct scsi_qla_host *ha, uint32_t raddr, + uint32_t waddr, + struct qla4_83xx_rmw *p_rmw_hdr) +{ + uint32_t value; + + if (p_rmw_hdr->index_a) + value = ha->reset_tmplt.array[p_rmw_hdr->index_a]; + else + qla4_83xx_rd_reg_indirect(ha, raddr, &value); + + value &= p_rmw_hdr->test_mask; + value <<= p_rmw_hdr->shl; + value >>= p_rmw_hdr->shr; + value |= p_rmw_hdr->or_value; + value ^= p_rmw_hdr->xor_value; + + qla4_83xx_wr_reg_indirect(ha, waddr, value); + + return; +} + +static void qla4_83xx_write_list(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + struct qla4_83xx_entry *p_entry; + uint32_t i; + + p_entry = (struct qla4_83xx_entry *) + ((char *)p_hdr + sizeof(struct qla4_83xx_reset_entry_hdr)); + + for (i = 0; i < p_hdr->count; i++, p_entry++) { + qla4_83xx_wr_reg_indirect(ha, p_entry->arg1, p_entry->arg2); + if (p_hdr->delay) + udelay((uint32_t)(p_hdr->delay)); + } +} + +static void qla4_83xx_read_write_list(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + struct qla4_83xx_entry *p_entry; + uint32_t i; + + p_entry = (struct qla4_83xx_entry *) + ((char *)p_hdr + sizeof(struct qla4_83xx_reset_entry_hdr)); + + for (i = 0; i < p_hdr->count; i++, p_entry++) { + qla4_83xx_read_write_crb_reg(ha, p_entry->arg1, p_entry->arg2); + if (p_hdr->delay) + udelay((uint32_t)(p_hdr->delay)); + } +} + +static void qla4_83xx_poll_list(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + long delay; + struct qla4_83xx_entry *p_entry; + struct qla4_83xx_poll *p_poll; + uint32_t i; + uint32_t value; + + p_poll = (struct qla4_83xx_poll *) + ((char *)p_hdr + sizeof(struct qla4_83xx_reset_entry_hdr)); + + /* Entries start after 8 byte qla4_83xx_poll, poll header contains + * the test_mask, test_value. */ + p_entry = (struct qla4_83xx_entry *)((char *)p_poll + + sizeof(struct qla4_83xx_poll)); + + delay = (long)p_hdr->delay; + if (!delay) { + for (i = 0; i < p_hdr->count; i++, p_entry++) { + qla4_83xx_poll_reg(ha, p_entry->arg1, delay, + p_poll->test_mask, + p_poll->test_value); + } + } else { + for (i = 0; i < p_hdr->count; i++, p_entry++) { + if (qla4_83xx_poll_reg(ha, p_entry->arg1, delay, + p_poll->test_mask, + p_poll->test_value)) { + qla4_83xx_rd_reg_indirect(ha, p_entry->arg1, + &value); + qla4_83xx_rd_reg_indirect(ha, p_entry->arg2, + &value); + } + } + } +} + +static void qla4_83xx_poll_write_list(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + long delay; + struct qla4_83xx_quad_entry *p_entry; + struct qla4_83xx_poll *p_poll; + uint32_t i; + + p_poll = (struct qla4_83xx_poll *) + ((char *)p_hdr + sizeof(struct qla4_83xx_reset_entry_hdr)); + p_entry = (struct qla4_83xx_quad_entry *) + ((char *)p_poll + sizeof(struct qla4_83xx_poll)); + delay = (long)p_hdr->delay; + + for (i = 0; i < p_hdr->count; i++, p_entry++) { + qla4_83xx_wr_reg_indirect(ha, p_entry->dr_addr, + p_entry->dr_value); + qla4_83xx_wr_reg_indirect(ha, p_entry->ar_addr, + p_entry->ar_value); + if (delay) { + if (qla4_83xx_poll_reg(ha, p_entry->ar_addr, delay, + p_poll->test_mask, + p_poll->test_value)) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Timeout Error: poll list, item_num %d, entry_num %d\n", + __func__, i, + ha->reset_tmplt.seq_index)); + } + } + } +} + +static void qla4_83xx_read_modify_write(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + struct qla4_83xx_entry *p_entry; + struct qla4_83xx_rmw *p_rmw_hdr; + uint32_t i; + + p_rmw_hdr = (struct qla4_83xx_rmw *) + ((char *)p_hdr + sizeof(struct qla4_83xx_reset_entry_hdr)); + p_entry = (struct qla4_83xx_entry *) + ((char *)p_rmw_hdr + sizeof(struct qla4_83xx_rmw)); + + for (i = 0; i < p_hdr->count; i++, p_entry++) { + qla4_83xx_rmw_crb_reg(ha, p_entry->arg1, p_entry->arg2, + p_rmw_hdr); + if (p_hdr->delay) + udelay((uint32_t)(p_hdr->delay)); + } +} + +static void qla4_83xx_pause(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + if (p_hdr->delay) + mdelay((uint32_t)((long)p_hdr->delay)); +} + +static void qla4_83xx_poll_read_list(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + long delay; + int index; + struct qla4_83xx_quad_entry *p_entry; + struct qla4_83xx_poll *p_poll; + uint32_t i; + uint32_t value; + + p_poll = (struct qla4_83xx_poll *) + ((char *)p_hdr + sizeof(struct qla4_83xx_reset_entry_hdr)); + p_entry = (struct qla4_83xx_quad_entry *) + ((char *)p_poll + sizeof(struct qla4_83xx_poll)); + delay = (long)p_hdr->delay; + + for (i = 0; i < p_hdr->count; i++, p_entry++) { + qla4_83xx_wr_reg_indirect(ha, p_entry->ar_addr, + p_entry->ar_value); + if (delay) { + if (qla4_83xx_poll_reg(ha, p_entry->ar_addr, delay, + p_poll->test_mask, + p_poll->test_value)) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Timeout Error: poll list, Item_num %d, entry_num %d\n", + __func__, i, + ha->reset_tmplt.seq_index)); + } else { + index = ha->reset_tmplt.array_index; + qla4_83xx_rd_reg_indirect(ha, p_entry->dr_addr, + &value); + ha->reset_tmplt.array[index++] = value; + + if (index == QLA83XX_MAX_RESET_SEQ_ENTRIES) + ha->reset_tmplt.array_index = 1; + } + } + } +} + +static void qla4_83xx_seq_end(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + ha->reset_tmplt.seq_end = 1; +} + +static void qla4_83xx_template_end(struct scsi_qla_host *ha, + struct qla4_83xx_reset_entry_hdr *p_hdr) +{ + ha->reset_tmplt.template_end = 1; + + if (ha->reset_tmplt.seq_error == 0) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Reset sequence completed SUCCESSFULLY.\n", + __func__)); + } else { + ql4_printk(KERN_ERR, ha, "%s: Reset sequence completed with some timeout errors.\n", + __func__); + } +} + +/** + * qla4_83xx_process_reset_template - Process reset template. + * + * Process all entries in reset template till entry with SEQ_END opcode, + * which indicates end of the reset template processing. Each entry has a + * Reset Entry header, entry opcode/command, with size of the entry, number + * of entries in sub-sequence and delay in microsecs or timeout in millisecs. + * + * @ha : Pointer to adapter structure + * @p_buff : Common reset entry header. + **/ +static void qla4_83xx_process_reset_template(struct scsi_qla_host *ha, + char *p_buff) +{ + int index, entries; + struct qla4_83xx_reset_entry_hdr *p_hdr; + char *p_entry = p_buff; + + ha->reset_tmplt.seq_end = 0; + ha->reset_tmplt.template_end = 0; + entries = ha->reset_tmplt.hdr->entries; + index = ha->reset_tmplt.seq_index; + + for (; (!ha->reset_tmplt.seq_end) && (index < entries); index++) { + + p_hdr = (struct qla4_83xx_reset_entry_hdr *)p_entry; + switch (p_hdr->cmd) { + case OPCODE_NOP: + break; + case OPCODE_WRITE_LIST: + qla4_83xx_write_list(ha, p_hdr); + break; + case OPCODE_READ_WRITE_LIST: + qla4_83xx_read_write_list(ha, p_hdr); + break; + case OPCODE_POLL_LIST: + qla4_83xx_poll_list(ha, p_hdr); + break; + case OPCODE_POLL_WRITE_LIST: + qla4_83xx_poll_write_list(ha, p_hdr); + break; + case OPCODE_READ_MODIFY_WRITE: + qla4_83xx_read_modify_write(ha, p_hdr); + break; + case OPCODE_SEQ_PAUSE: + qla4_83xx_pause(ha, p_hdr); + break; + case OPCODE_SEQ_END: + qla4_83xx_seq_end(ha, p_hdr); + break; + case OPCODE_TMPL_END: + qla4_83xx_template_end(ha, p_hdr); + break; + case OPCODE_POLL_READ_LIST: + qla4_83xx_poll_read_list(ha, p_hdr); + break; + default: + ql4_printk(KERN_ERR, ha, "%s: Unknown command ==> 0x%04x on entry = %d\n", + __func__, p_hdr->cmd, index); + break; + } + + /* Set pointer to next entry in the sequence. */ + p_entry += p_hdr->size; + } + + ha->reset_tmplt.seq_index = index; +} + +static void qla4_83xx_process_stop_seq(struct scsi_qla_host *ha) +{ + ha->reset_tmplt.seq_index = 0; + qla4_83xx_process_reset_template(ha, ha->reset_tmplt.stop_offset); + + if (ha->reset_tmplt.seq_end != 1) + ql4_printk(KERN_ERR, ha, "%s: Abrupt STOP Sub-Sequence end.\n", + __func__); +} + +static void qla4_83xx_process_start_seq(struct scsi_qla_host *ha) +{ + qla4_83xx_process_reset_template(ha, ha->reset_tmplt.start_offset); + + if (ha->reset_tmplt.template_end != 1) + ql4_printk(KERN_ERR, ha, "%s: Abrupt START Sub-Sequence end.\n", + __func__); +} + +static void qla4_83xx_process_init_seq(struct scsi_qla_host *ha) +{ + qla4_83xx_process_reset_template(ha, ha->reset_tmplt.init_offset); + + if (ha->reset_tmplt.seq_end != 1) + ql4_printk(KERN_ERR, ha, "%s: Abrupt INIT Sub-Sequence end.\n", + __func__); +} + +static int qla4_83xx_restart(struct scsi_qla_host *ha) +{ + int ret_val = QLA_SUCCESS; + + qla4_83xx_process_stop_seq(ha); + + /* Collect minidump*/ + if (!test_and_clear_bit(AF_83XX_NO_FW_DUMP, &ha->flags)) + qla4_8xxx_get_minidump(ha); + + qla4_83xx_process_init_seq(ha); + + if (qla4_83xx_copy_bootloader(ha)) { + ql4_printk(KERN_ERR, ha, "%s: Copy bootloader, firmware restart failed!\n", + __func__); + ret_val = QLA_ERROR; + goto exit_restart; + } + + qla4_83xx_wr_reg(ha, QLA83XX_FW_IMAGE_VALID, QLA83XX_BOOT_FROM_FLASH); + qla4_83xx_process_start_seq(ha); + +exit_restart: + return ret_val; +} + +int qla4_83xx_start_firmware(struct scsi_qla_host *ha) +{ + int ret_val = QLA_SUCCESS; + + ret_val = qla4_83xx_restart(ha); + if (ret_val == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: Restart error\n", __func__); + goto exit_start_fw; + } else { + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: Restart done\n", + __func__)); + } + + ret_val = qla4_83xx_check_cmd_peg_status(ha); + if (ret_val == QLA_ERROR) + ql4_printk(KERN_ERR, ha, "%s: Peg not initialized\n", + __func__); + +exit_start_fw: + return ret_val; +} + +/*----------------------Interrupt Related functions ---------------------*/ + +void qla4_83xx_disable_intrs(struct scsi_qla_host *ha) +{ + uint32_t mb_int, ret; + + if (test_and_clear_bit(AF_INTERRUPTS_ON, &ha->flags)) + qla4_8xxx_mbx_intr_disable(ha); + + ret = readl(&ha->qla4_83xx_reg->mbox_int); + mb_int = ret & ~INT_ENABLE_FW_MB; + writel(mb_int, &ha->qla4_83xx_reg->mbox_int); + writel(1, &ha->qla4_83xx_reg->leg_int_mask); +} + +void qla4_83xx_enable_intrs(struct scsi_qla_host *ha) +{ + uint32_t mb_int; + + qla4_8xxx_mbx_intr_enable(ha); + mb_int = INT_ENABLE_FW_MB; + writel(mb_int, &ha->qla4_83xx_reg->mbox_int); + writel(0, &ha->qla4_83xx_reg->leg_int_mask); + + set_bit(AF_INTERRUPTS_ON, &ha->flags); +} + +void qla4_83xx_queue_mbox_cmd(struct scsi_qla_host *ha, uint32_t *mbx_cmd, + int incount) +{ + int i; + + /* Load all mailbox registers, except mailbox 0. */ + for (i = 1; i < incount; i++) + writel(mbx_cmd[i], &ha->qla4_83xx_reg->mailbox_in[i]); + + writel(mbx_cmd[0], &ha->qla4_83xx_reg->mailbox_in[0]); + + /* Set Host Interrupt register to 1, to tell the firmware that + * a mailbox command is pending. Firmware after reading the + * mailbox command, clears the host interrupt register */ + writel(HINT_MBX_INT_PENDING, &ha->qla4_83xx_reg->host_intr); +} + +void qla4_83xx_process_mbox_intr(struct scsi_qla_host *ha, int outcount) +{ + int intr_status; + + intr_status = readl(&ha->qla4_83xx_reg->risc_intr); + if (intr_status) { + ha->mbox_status_count = outcount; + ha->isp_ops->interrupt_service_routine(ha, intr_status); + } +} + +/** + * qla4_83xx_isp_reset - Resets ISP and aborts all outstanding commands. + * @ha: pointer to host adapter structure. + **/ +int qla4_83xx_isp_reset(struct scsi_qla_host *ha) +{ + int rval; + uint32_t dev_state; + + ha->isp_ops->idc_lock(ha); + dev_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE); + + if (ql4xdontresethba) + qla4_83xx_set_idc_dontreset(ha); + + if (dev_state == QLA8XXX_DEV_READY) { + /* If IDC_CTRL DONTRESETHBA_BIT0 is set dont do reset + * recovery */ + if (qla4_83xx_idc_dontreset(ha) == DONTRESET_BIT0) { + ql4_printk(KERN_ERR, ha, "%s: Reset recovery disabled\n", + __func__); + rval = QLA_ERROR; + goto exit_isp_reset; + } + + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: HW State: NEED RESET\n", + __func__)); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_NEED_RESET); + + } else { + /* If device_state is NEED_RESET, go ahead with + * Reset,irrespective of ql4xdontresethba. This is to allow a + * non-reset-owner to force a reset. Non-reset-owner sets + * the IDC_CTRL BIT0 to prevent Reset-owner from doing a Reset + * and then forces a Reset by setting device_state to + * NEED_RESET. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: HW state already set to NEED_RESET\n", + __func__)); + } + + /* For ISP8324, Reset owner is NIC, iSCSI or FCOE based on priority + * and which drivers are present. Unlike ISP8022, the function setting + * NEED_RESET, may not be the Reset owner. */ + if (qla4_83xx_can_perform_reset(ha)) + set_bit(AF_8XXX_RST_OWNER, &ha->flags); + + ha->isp_ops->idc_unlock(ha); + rval = qla4_8xxx_device_state_handler(ha); + + ha->isp_ops->idc_lock(ha); + qla4_8xxx_clear_rst_ready(ha); +exit_isp_reset: + ha->isp_ops->idc_unlock(ha); + + if (rval == QLA_SUCCESS) + clear_bit(AF_FW_RECOVERY, &ha->flags); + + return rval; +} + +static void qla4_83xx_dump_pause_control_regs(struct scsi_qla_host *ha) +{ + u32 val = 0, val1 = 0; + int i, status = QLA_SUCCESS; + + status = qla4_83xx_rd_reg_indirect(ha, QLA83XX_SRE_SHIM_CONTROL, &val); + DEBUG2(ql4_printk(KERN_INFO, ha, "SRE-Shim Ctrl:0x%x\n", val)); + + /* Port 0 Rx Buffer Pause Threshold Registers. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "Port 0 Rx Buffer Pause Threshold Registers[TC7..TC0]:")); + for (i = 0; i < 8; i++) { + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT0_RXB_PAUSE_THRS + (i * 0x4), &val); + DEBUG2(pr_info("0x%x ", val)); + } + + DEBUG2(pr_info("\n")); + + /* Port 1 Rx Buffer Pause Threshold Registers. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "Port 1 Rx Buffer Pause Threshold Registers[TC7..TC0]:")); + for (i = 0; i < 8; i++) { + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT1_RXB_PAUSE_THRS + (i * 0x4), &val); + DEBUG2(pr_info("0x%x ", val)); + } + + DEBUG2(pr_info("\n")); + + /* Port 0 RxB Traffic Class Max Cell Registers. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "Port 0 RxB Traffic Class Max Cell Registers[3..0]:")); + for (i = 0; i < 4; i++) { + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT0_RXB_TC_MAX_CELL + (i * 0x4), &val); + DEBUG2(pr_info("0x%x ", val)); + } + + DEBUG2(pr_info("\n")); + + /* Port 1 RxB Traffic Class Max Cell Registers. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "Port 1 RxB Traffic Class Max Cell Registers[3..0]:")); + for (i = 0; i < 4; i++) { + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT1_RXB_TC_MAX_CELL + (i * 0x4), &val); + DEBUG2(pr_info("0x%x ", val)); + } + + DEBUG2(pr_info("\n")); + + /* Port 0 RxB Rx Traffic Class Stats. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "Port 0 RxB Rx Traffic Class Stats [TC7..TC0]")); + for (i = 7; i >= 0; i--) { + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT0_RXB_TC_STATS, + &val); + val &= ~(0x7 << 29); /* Reset bits 29 to 31 */ + qla4_83xx_wr_reg_indirect(ha, QLA83XX_PORT0_RXB_TC_STATS, + (val | (i << 29))); + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT0_RXB_TC_STATS, + &val); + DEBUG2(pr_info("0x%x ", val)); + } + + DEBUG2(pr_info("\n")); + + /* Port 1 RxB Rx Traffic Class Stats. */ + DEBUG2(ql4_printk(KERN_INFO, ha, + "Port 1 RxB Rx Traffic Class Stats [TC7..TC0]")); + for (i = 7; i >= 0; i--) { + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT1_RXB_TC_STATS, + &val); + val &= ~(0x7 << 29); /* Reset bits 29 to 31 */ + qla4_83xx_wr_reg_indirect(ha, QLA83XX_PORT1_RXB_TC_STATS, + (val | (i << 29))); + status = qla4_83xx_rd_reg_indirect(ha, + QLA83XX_PORT1_RXB_TC_STATS, + &val); + DEBUG2(pr_info("0x%x ", val)); + } + + DEBUG2(pr_info("\n")); + + status = qla4_83xx_rd_reg_indirect(ha, QLA83XX_PORT2_IFB_PAUSE_THRS, + &val); + status = qla4_83xx_rd_reg_indirect(ha, QLA83XX_PORT3_IFB_PAUSE_THRS, + &val1); + + DEBUG2(ql4_printk(KERN_INFO, ha, + "IFB-Pause Thresholds: Port 2:0x%x, Port 3:0x%x\n", + val, val1)); +} + +static void __qla4_83xx_disable_pause(struct scsi_qla_host *ha) +{ + int i; + + /* set SRE-Shim Control Register */ + qla4_83xx_wr_reg_indirect(ha, QLA83XX_SRE_SHIM_CONTROL, + QLA83XX_SET_PAUSE_VAL); + + for (i = 0; i < 8; i++) { + /* Port 0 Rx Buffer Pause Threshold Registers. */ + qla4_83xx_wr_reg_indirect(ha, + QLA83XX_PORT0_RXB_PAUSE_THRS + (i * 0x4), + QLA83XX_SET_PAUSE_VAL); + /* Port 1 Rx Buffer Pause Threshold Registers. */ + qla4_83xx_wr_reg_indirect(ha, + QLA83XX_PORT1_RXB_PAUSE_THRS + (i * 0x4), + QLA83XX_SET_PAUSE_VAL); + } + + for (i = 0; i < 4; i++) { + /* Port 0 RxB Traffic Class Max Cell Registers. */ + qla4_83xx_wr_reg_indirect(ha, + QLA83XX_PORT0_RXB_TC_MAX_CELL + (i * 0x4), + QLA83XX_SET_TC_MAX_CELL_VAL); + /* Port 1 RxB Traffic Class Max Cell Registers. */ + qla4_83xx_wr_reg_indirect(ha, + QLA83XX_PORT1_RXB_TC_MAX_CELL + (i * 0x4), + QLA83XX_SET_TC_MAX_CELL_VAL); + } + + qla4_83xx_wr_reg_indirect(ha, QLA83XX_PORT2_IFB_PAUSE_THRS, + QLA83XX_SET_PAUSE_VAL); + qla4_83xx_wr_reg_indirect(ha, QLA83XX_PORT3_IFB_PAUSE_THRS, + QLA83XX_SET_PAUSE_VAL); + + ql4_printk(KERN_INFO, ha, "Disabled pause frames successfully.\n"); +} + +void qla4_83xx_disable_pause(struct scsi_qla_host *ha) +{ + ha->isp_ops->idc_lock(ha); + qla4_83xx_dump_pause_control_regs(ha); + __qla4_83xx_disable_pause(ha); + ha->isp_ops->idc_unlock(ha); +} diff --git a/drivers/scsi/qla4xxx/ql4_83xx.h b/drivers/scsi/qla4xxx/ql4_83xx.h new file mode 100644 index 0000000..6a00f90 --- /dev/null +++ b/drivers/scsi/qla4xxx/ql4_83xx.h @@ -0,0 +1,283 @@ +/* + * QLogic iSCSI HBA Driver + * Copyright (c) 2003-2012 QLogic Corporation + * + * See LICENSE.qla4xxx for copyright and licensing details. + */ + +#ifndef __QL483XX_H +#define __QL483XX_H + +/* Indirectly Mapped Registers */ +#define QLA83XX_FLASH_SPI_STATUS 0x2808E010 +#define QLA83XX_FLASH_SPI_CONTROL 0x2808E014 +#define QLA83XX_FLASH_STATUS 0x42100004 +#define QLA83XX_FLASH_CONTROL 0x42110004 +#define QLA83XX_FLASH_ADDR 0x42110008 +#define QLA83XX_FLASH_WRDATA 0x4211000C +#define QLA83XX_FLASH_RDDATA 0x42110018 +#define QLA83XX_FLASH_DIRECT_WINDOW 0x42110030 +#define QLA83XX_FLASH_DIRECT_DATA(DATA) (0x42150000 | (0x0000FFFF&DATA)) + +/* Directly Mapped Registers in 83xx register table */ + +/* Flash access regs */ +#define QLA83XX_FLASH_LOCK 0x3850 +#define QLA83XX_FLASH_UNLOCK 0x3854 +#define QLA83XX_FLASH_LOCK_ID 0x3500 + +/* Driver Lock regs */ +#define QLA83XX_DRV_LOCK 0x3868 +#define QLA83XX_DRV_UNLOCK 0x386C +#define QLA83XX_DRV_LOCK_ID 0x3504 +#define QLA83XX_DRV_LOCKRECOVERY 0x379C + +/* IDC version */ +#define QLA83XX_IDC_VER_MAJ_VALUE 0x1 +#define QLA83XX_IDC_VER_MIN_VALUE 0x0 + +/* IDC Registers : Driver Coexistence Defines */ +#define QLA83XX_CRB_IDC_VER_MAJOR 0x3780 +#define QLA83XX_CRB_IDC_VER_MINOR 0x3798 +#define QLA83XX_IDC_DRV_CTRL 0x3790 +#define QLA83XX_IDC_DRV_AUDIT 0x3794 +#define QLA83XX_SRE_SHIM_CONTROL 0x0D200284 +#define QLA83XX_PORT0_RXB_PAUSE_THRS 0x0B2003A4 +#define QLA83XX_PORT1_RXB_PAUSE_THRS 0x0B2013A4 +#define QLA83XX_PORT0_RXB_TC_MAX_CELL 0x0B200388 +#define QLA83XX_PORT1_RXB_TC_MAX_CELL 0x0B201388 +#define QLA83XX_PORT0_RXB_TC_STATS 0x0B20039C +#define QLA83XX_PORT1_RXB_TC_STATS 0x0B20139C +#define QLA83XX_PORT2_IFB_PAUSE_THRS 0x0B200704 +#define QLA83XX_PORT3_IFB_PAUSE_THRS 0x0B201704 + +/* set value to pause threshold value */ +#define QLA83XX_SET_PAUSE_VAL 0x0 +#define QLA83XX_SET_TC_MAX_CELL_VAL 0x03FF03FF + +/* qla_83xx_reg_tbl registers */ +#define QLA83XX_PEG_HALT_STATUS1 0x34A8 +#define QLA83XX_PEG_HALT_STATUS2 0x34AC +#define QLA83XX_PEG_ALIVE_COUNTER 0x34B0 /* FW_HEARTBEAT */ +#define QLA83XX_FW_CAPABILITIES 0x3528 +#define QLA83XX_CRB_DRV_ACTIVE 0x3788 /* IDC_DRV_PRESENCE */ +#define QLA83XX_CRB_DEV_STATE 0x3784 /* IDC_DEV_STATE */ +#define QLA83XX_CRB_DRV_STATE 0x378C /* IDC_DRV_ACK */ +#define QLA83XX_CRB_DRV_SCRATCH 0x3548 +#define QLA83XX_CRB_DEV_PART_INFO1 0x37E0 +#define QLA83XX_CRB_DEV_PART_INFO2 0x37E4 + +#define QLA83XX_FW_VER_MAJOR 0x3550 +#define QLA83XX_FW_VER_MINOR 0x3554 +#define QLA83XX_FW_VER_SUB 0x3558 +#define QLA83XX_NPAR_STATE 0x359C +#define QLA83XX_FW_IMAGE_VALID 0x35FC +#define QLA83XX_CMDPEG_STATE 0x3650 +#define QLA83XX_ASIC_TEMP 0x37B4 +#define QLA83XX_FW_API 0x356C +#define QLA83XX_DRV_OP_MODE 0x3570 + +static const uint32_t qla4_83xx_reg_tbl[] = { + QLA83XX_PEG_HALT_STATUS1, + QLA83XX_PEG_HALT_STATUS2, + QLA83XX_PEG_ALIVE_COUNTER, + QLA83XX_CRB_DRV_ACTIVE, + QLA83XX_CRB_DEV_STATE, + QLA83XX_CRB_DRV_STATE, + QLA83XX_CRB_DRV_SCRATCH, + QLA83XX_CRB_DEV_PART_INFO1, + QLA83XX_CRB_IDC_VER_MAJOR, + QLA83XX_FW_VER_MAJOR, + QLA83XX_FW_VER_MINOR, + QLA83XX_FW_VER_SUB, + QLA83XX_CMDPEG_STATE, + QLA83XX_ASIC_TEMP, +}; + +#define QLA83XX_CRB_WIN_BASE 0x3800 +#define QLA83XX_CRB_WIN_FUNC(f) (QLA83XX_CRB_WIN_BASE+((f)*4)) +#define QLA83XX_SEM_LOCK_BASE 0x3840 +#define QLA83XX_SEM_UNLOCK_BASE 0x3844 +#define QLA83XX_SEM_LOCK_FUNC(f) (QLA83XX_SEM_LOCK_BASE+((f)*8)) +#define QLA83XX_SEM_UNLOCK_FUNC(f) (QLA83XX_SEM_UNLOCK_BASE+((f)*8)) +#define QLA83XX_LINK_STATE(f) (0x3698+((f) > 7 ? 4 : 0)) +#define QLA83XX_LINK_SPEED(f) (0x36E0+(((f) >> 2) * 4)) +#define QLA83XX_MAX_LINK_SPEED(f) (0x36F0+(((f) / 4) * 4)) +#define QLA83XX_LINK_SPEED_FACTOR 10 + +/* FLASH API Defines */ +#define QLA83xx_FLASH_MAX_WAIT_USEC 100 +#define QLA83XX_FLASH_LOCK_TIMEOUT 10000 +#define QLA83XX_FLASH_SECTOR_SIZE 65536 +#define QLA83XX_DRV_LOCK_TIMEOUT 2000 +#define QLA83XX_FLASH_SECTOR_ERASE_CMD 0xdeadbeef +#define QLA83XX_FLASH_WRITE_CMD 0xdacdacda +#define QLA83XX_FLASH_BUFFER_WRITE_CMD 0xcadcadca +#define QLA83XX_FLASH_READ_RETRY_COUNT 2000 +#define QLA83XX_FLASH_STATUS_READY 0x6 +#define QLA83XX_FLASH_BUFFER_WRITE_MIN 2 +#define QLA83XX_FLASH_BUFFER_WRITE_MAX 64 +#define QLA83XX_FLASH_STATUS_REG_POLL_DELAY 1 +#define QLA83XX_ERASE_MODE 1 +#define QLA83XX_WRITE_MODE 2 +#define QLA83XX_DWORD_WRITE_MODE 3 + +#define QLA83XX_GLOBAL_RESET 0x38CC +#define QLA83XX_WILDCARD 0x38F0 +#define QLA83XX_INFORMANT 0x38FC +#define QLA83XX_HOST_MBX_CTRL 0x3038 +#define QLA83XX_FW_MBX_CTRL 0x303C +#define QLA83XX_BOOTLOADER_ADDR 0x355C +#define QLA83XX_BOOTLOADER_SIZE 0x3560 +#define QLA83XX_FW_IMAGE_ADDR 0x3564 +#define QLA83XX_MBX_INTR_ENABLE 0x1000 +#define QLA83XX_MBX_INTR_MASK 0x1200 + +/* IDC Control Register bit defines */ +#define DONTRESET_BIT0 0x1 +#define GRACEFUL_RESET_BIT1 0x2 + +#define QLA83XX_HALT_STATUS_INFORMATIONAL (0x1 << 29) +#define QLA83XX_HALT_STATUS_FW_RESET (0x2 << 29) +#define QLA83XX_HALT_STATUS_UNRECOVERABLE (0x4 << 29) + +/* Firmware image definitions */ +#define QLA83XX_BOOTLOADER_FLASH_ADDR 0x10000 +#define QLA83XX_BOOT_FROM_FLASH 0 + +#define QLA83XX_IDC_PARAM_ADDR 0x3e8020 +/* Reset template definitions */ +#define QLA83XX_MAX_RESET_SEQ_ENTRIES 16 +#define QLA83XX_RESTART_TEMPLATE_SIZE 0x2000 +#define QLA83XX_RESET_TEMPLATE_ADDR 0x4F0000 +#define QLA83XX_RESET_SEQ_VERSION 0x0101 + +/* Reset template entry opcodes */ +#define OPCODE_NOP 0x0000 +#define OPCODE_WRITE_LIST 0x0001 +#define OPCODE_READ_WRITE_LIST 0x0002 +#define OPCODE_POLL_LIST 0x0004 +#define OPCODE_POLL_WRITE_LIST 0x0008 +#define OPCODE_READ_MODIFY_WRITE 0x0010 +#define OPCODE_SEQ_PAUSE 0x0020 +#define OPCODE_SEQ_END 0x0040 +#define OPCODE_TMPL_END 0x0080 +#define OPCODE_POLL_READ_LIST 0x0100 + +/* Template Header */ +#define RESET_TMPLT_HDR_SIGNATURE 0xCAFE +struct qla4_83xx_reset_template_hdr { + __le16 version; + __le16 signature; + __le16 size; + __le16 entries; + __le16 hdr_size; + __le16 checksum; + __le16 init_seq_offset; + __le16 start_seq_offset; +} __packed; + +/* Common Entry Header. */ +struct qla4_83xx_reset_entry_hdr { + __le16 cmd; + __le16 size; + __le16 count; + __le16 delay; +} __packed; + +/* Generic poll entry type. */ +struct qla4_83xx_poll { + __le32 test_mask; + __le32 test_value; +} __packed; + +/* Read modify write entry type. */ +struct qla4_83xx_rmw { + __le32 test_mask; + __le32 xor_value; + __le32 or_value; + uint8_t shl; + uint8_t shr; + uint8_t index_a; + uint8_t rsvd; +} __packed; + +/* Generic Entry Item with 2 DWords. */ +struct qla4_83xx_entry { + __le32 arg1; + __le32 arg2; +} __packed; + +/* Generic Entry Item with 4 DWords.*/ +struct qla4_83xx_quad_entry { + __le32 dr_addr; + __le32 dr_value; + __le32 ar_addr; + __le32 ar_value; +} __packed; + +struct qla4_83xx_reset_template { + int seq_index; + int seq_error; + int array_index; + uint32_t array[QLA83XX_MAX_RESET_SEQ_ENTRIES]; + uint8_t *buff; + uint8_t *stop_offset; + uint8_t *start_offset; + uint8_t *init_offset; + struct qla4_83xx_reset_template_hdr *hdr; + uint8_t seq_end; + uint8_t template_end; +}; + +/* POLLRD Entry */ +struct qla83xx_minidump_entry_pollrd { + struct qla8xxx_minidump_entry_hdr h; + uint32_t select_addr; + uint32_t read_addr; + uint32_t select_value; + uint16_t select_value_stride; + uint16_t op_count; + uint32_t poll_wait; + uint32_t poll_mask; + uint32_t data_size; + uint32_t rsvd_1; +}; + +/* RDMUX2 Entry */ +struct qla83xx_minidump_entry_rdmux2 { + struct qla8xxx_minidump_entry_hdr h; + uint32_t select_addr_1; + uint32_t select_addr_2; + uint32_t select_value_1; + uint32_t select_value_2; + uint32_t op_count; + uint32_t select_value_mask; + uint32_t read_addr; + uint8_t select_value_stride; + uint8_t data_size; + uint8_t rsvd[2]; +}; + +/* POLLRDMWR Entry */ +struct qla83xx_minidump_entry_pollrdmwr { + struct qla8xxx_minidump_entry_hdr h; + uint32_t addr_1; + uint32_t addr_2; + uint32_t value_1; + uint32_t value_2; + uint32_t poll_wait; + uint32_t poll_mask; + uint32_t modify_mask; + uint32_t data_size; +}; + +/* IDC additional information */ +struct qla4_83xx_idc_information { + uint32_t request_desc; /* IDC request descriptor */ + uint32_t info1; /* IDC additional info */ + uint32_t info2; /* IDC additional info */ + uint32_t info3; /* IDC additional info */ +}; + +#endif diff --git a/drivers/scsi/qla4xxx/ql4_attr.c b/drivers/scsi/qla4xxx/ql4_attr.c index c681b2a..76819b7 100644 --- a/drivers/scsi/qla4xxx/ql4_attr.c +++ b/drivers/scsi/qla4xxx/ql4_attr.c @@ -17,7 +17,7 @@ qla4_8xxx_sysfs_read_fw_dump(struct file *filep, struct kobject *kobj, struct scsi_qla_host *ha = to_qla_host(dev_to_shost(container_of(kobj, struct device, kobj))); - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) return -EINVAL; if (!test_bit(AF_82XX_DUMP_READING, &ha->flags)) @@ -38,7 +38,7 @@ qla4_8xxx_sysfs_write_fw_dump(struct file *filep, struct kobject *kobj, long reading; int ret = 0; - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) return -EINVAL; if (off != 0) @@ -75,21 +75,21 @@ qla4_8xxx_sysfs_write_fw_dump(struct file *filep, struct kobject *kobj, break; case 2: /* Reset HBA */ - qla4_8xxx_idc_lock(ha); - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); - if (dev_state == QLA82XX_DEV_READY) { + ha->isp_ops->idc_lock(ha); + dev_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE); + if (dev_state == QLA8XXX_DEV_READY) { ql4_printk(KERN_INFO, ha, "%s: Setting Need reset, reset_owner is 0x%x.\n", __func__, ha->func_num); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_NEED_RESET); - set_bit(AF_82XX_RST_OWNER, &ha->flags); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_NEED_RESET); + set_bit(AF_8XXX_RST_OWNER, &ha->flags); } else ql4_printk(KERN_INFO, ha, "%s: Reset not performed as device state is 0x%x\n", __func__, dev_state); - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_unlock(ha); break; default: /* do nothing */ @@ -150,7 +150,7 @@ qla4xxx_fw_version_show(struct device *dev, { struct scsi_qla_host *ha = to_qla_host(class_to_shost(dev)); - if (is_qla8022(ha)) + if (is_qla80XX(ha)) return snprintf(buf, PAGE_SIZE, "%d.%02d.%02d (%x)\n", ha->firmware_version[0], ha->firmware_version[1], @@ -214,7 +214,7 @@ qla4xxx_phy_port_cnt_show(struct device *dev, struct device_attribute *attr, { struct scsi_qla_host *ha = to_qla_host(class_to_shost(dev)); - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) return -ENOSYS; return snprintf(buf, PAGE_SIZE, "0x%04X\n", ha->phy_port_cnt); @@ -226,7 +226,7 @@ qla4xxx_phy_port_num_show(struct device *dev, struct device_attribute *attr, { struct scsi_qla_host *ha = to_qla_host(class_to_shost(dev)); - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) return -ENOSYS; return snprintf(buf, PAGE_SIZE, "0x%04X\n", ha->phy_port_num); @@ -238,7 +238,7 @@ qla4xxx_iscsi_func_cnt_show(struct device *dev, struct device_attribute *attr, { struct scsi_qla_host *ha = to_qla_host(class_to_shost(dev)); - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) return -ENOSYS; return snprintf(buf, PAGE_SIZE, "0x%04X\n", ha->iscsi_pci_func_cnt); diff --git a/drivers/scsi/qla4xxx/ql4_dbg.c b/drivers/scsi/qla4xxx/ql4_dbg.c index 8d58ae2..77b7c59 100644 --- a/drivers/scsi/qla4xxx/ql4_dbg.c +++ b/drivers/scsi/qla4xxx/ql4_dbg.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -37,7 +37,7 @@ void qla4xxx_dump_registers(struct scsi_qla_host *ha) if (is_qla8022(ha)) { for (i = 1; i < MBOX_REG_COUNT; i++) printk(KERN_INFO "mailbox[%d] = 0x%08X\n", - i, readl(&ha->qla4_8xxx_reg->mailbox_in[i])); + i, readl(&ha->qla4_82xx_reg->mailbox_in[i])); return; } @@ -131,3 +131,31 @@ void qla4xxx_dump_registers(struct scsi_qla_host *ha) &ha->reg->ctrl_status); } } + +void qla4_8xxx_dump_peg_reg(struct scsi_qla_host *ha) +{ + uint32_t halt_status1, halt_status2; + + halt_status1 = qla4_8xxx_rd_direct(ha, QLA8XXX_PEG_HALT_STATUS1); + halt_status2 = qla4_8xxx_rd_direct(ha, QLA8XXX_PEG_HALT_STATUS2); + + if (is_qla8022(ha)) { + ql4_printk(KERN_INFO, ha, + "scsi(%ld): %s, ISP8022 Dumping hw/fw registers:\n" + " PEG_HALT_STATUS1: 0x%x, PEG_HALT_STATUS2: 0x%x,\n" + " PEG_NET_0_PC: 0x%x, PEG_NET_1_PC: 0x%x,\n" + " PEG_NET_2_PC: 0x%x, PEG_NET_3_PC: 0x%x,\n" + " PEG_NET_4_PC: 0x%x\n", ha->host_no, + __func__, halt_status1, halt_status2, + qla4_82xx_rd_32(ha, QLA82XX_CRB_PEG_NET_0 + 0x3c), + qla4_82xx_rd_32(ha, QLA82XX_CRB_PEG_NET_1 + 0x3c), + qla4_82xx_rd_32(ha, QLA82XX_CRB_PEG_NET_2 + 0x3c), + qla4_82xx_rd_32(ha, QLA82XX_CRB_PEG_NET_3 + 0x3c), + qla4_82xx_rd_32(ha, QLA82XX_CRB_PEG_NET_4 + 0x3c)); + } else if (is_qla8032(ha)) { + ql4_printk(KERN_INFO, ha, + "scsi(%ld): %s, ISP8324 Dumping hw/fw registers:\n" + " PEG_HALT_STATUS1: 0x%x, PEG_HALT_STATUS2: 0x%x,\n", + ha->host_no, __func__, halt_status1, halt_status2); + } +} diff --git a/drivers/scsi/qla4xxx/ql4_dbg.h b/drivers/scsi/qla4xxx/ql4_dbg.h index abd8360..5b0afc1 100644 --- a/drivers/scsi/qla4xxx/ql4_dbg.h +++ b/drivers/scsi/qla4xxx/ql4_dbg.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla4xxx/ql4_def.h b/drivers/scsi/qla4xxx/ql4_def.h index 7fdba7f..329d553 100644 --- a/drivers/scsi/qla4xxx/ql4_def.h +++ b/drivers/scsi/qla4xxx/ql4_def.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -42,6 +42,7 @@ #include "ql4_nx.h" #include "ql4_fw.h" #include "ql4_nvram.h" +#include "ql4_83xx.h" #ifndef PCI_DEVICE_ID_QLOGIC_ISP4010 #define PCI_DEVICE_ID_QLOGIC_ISP4010 0x4010 @@ -59,6 +60,10 @@ #define PCI_DEVICE_ID_QLOGIC_ISP8022 0x8022 #endif +#ifndef PCI_DEVICE_ID_QLOGIC_ISP8324 +#define PCI_DEVICE_ID_QLOGIC_ISP8324 0x8032 +#endif + #define ISP4XXX_PCI_FN_1 0x1 #define ISP4XXX_PCI_FN_2 0x3 @@ -388,8 +393,10 @@ struct isp_operations { void (*disable_intrs) (struct scsi_qla_host *); void (*enable_intrs) (struct scsi_qla_host *); int (*start_firmware) (struct scsi_qla_host *); + int (*restart_firmware) (struct scsi_qla_host *); irqreturn_t (*intr_handler) (int , void *); void (*interrupt_service_routine) (struct scsi_qla_host *, uint32_t); + int (*need_reset) (struct scsi_qla_host *); int (*reset_chip) (struct scsi_qla_host *); int (*reset_firmware) (struct scsi_qla_host *); void (*queue_iocb) (struct scsi_qla_host *); @@ -397,6 +404,15 @@ struct isp_operations { uint16_t (*rd_shdw_req_q_out) (struct scsi_qla_host *); uint16_t (*rd_shdw_rsp_q_in) (struct scsi_qla_host *); int (*get_sys_info) (struct scsi_qla_host *); + uint32_t (*rd_reg_direct) (struct scsi_qla_host *, ulong); + void (*wr_reg_direct) (struct scsi_qla_host *, ulong, uint32_t); + int (*rd_reg_indirect) (struct scsi_qla_host *, uint32_t, uint32_t *); + int (*wr_reg_indirect) (struct scsi_qla_host *, uint32_t, uint32_t); + int (*idc_lock) (struct scsi_qla_host *); + void (*idc_unlock) (struct scsi_qla_host *); + void (*rom_lock_recovery) (struct scsi_qla_host *); + void (*queue_mailbox_command) (struct scsi_qla_host *, uint32_t *, int); + void (*process_mailbox_interrupt) (struct scsi_qla_host *, int); }; struct ql4_mdump_size_table { @@ -497,8 +513,9 @@ struct scsi_qla_host { #define AF_PCI_CHANNEL_IO_PERM_FAILURE 21 /* 0x00200000 */ #define AF_BUILD_DDB_LIST 22 /* 0x00400000 */ #define AF_82XX_FW_DUMPED 24 /* 0x01000000 */ -#define AF_82XX_RST_OWNER 25 /* 0x02000000 */ +#define AF_8XXX_RST_OWNER 25 /* 0x02000000 */ #define AF_82XX_DUMP_READING 26 /* 0x04000000 */ +#define AF_83XX_NO_FW_DUMP 27 /* 0x08000000 */ unsigned long dpc_flags; @@ -514,7 +531,7 @@ struct scsi_qla_host { #define DPC_RESET_ACTIVE 20 /* 0x00040000 */ #define DPC_HA_UNRECOVERABLE 21 /* 0x00080000 ISP-82xx only*/ #define DPC_HA_NEED_QUIESCENT 22 /* 0x00100000 ISP-82xx only*/ - +#define DPC_POST_IDC_ACK 23 /* 0x00200000 */ struct Scsi_Host *host; /* pointer to host data */ uint32_t tot_ddbs; @@ -647,7 +664,7 @@ struct scsi_qla_host { uint8_t acb_version; /* qla82xx specific fields */ - struct device_reg_82xx __iomem *qla4_8xxx_reg; /* Base I/O address */ + struct device_reg_82xx __iomem *qla4_82xx_reg; /* Base I/O address */ unsigned long nx_pcibase; /* Base I/O address */ uint8_t *nx_db_rd_ptr; /* Doorbell read pointer */ unsigned long nx_db_wr_ptr; /* Door bell write pointer */ @@ -733,6 +750,13 @@ struct scsi_qla_host { #define MAX_MRB 128 struct mrb *active_mrb_array[MAX_MRB]; uint32_t mrb_index; + + uint32_t *reg_tbl; + struct qla4_83xx_reset_template reset_tmplt; + struct device_reg_83xx __iomem *qla4_83xx_reg; /* Base I/O address + for ISP8324 */ + uint32_t pf_bit; + struct qla4_83xx_idc_information idc_info; }; struct ql4_task_data { @@ -752,7 +776,7 @@ struct ql4_task_data { struct qla_endpoint { struct Scsi_Host *host; - struct sockaddr dst_addr; + struct sockaddr_storage dst_addr; }; struct qla_conn { @@ -795,13 +819,20 @@ static inline int is_qla8022(struct scsi_qla_host *ha) return ha->pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8022; } -/* Note: Currently AER/EEH is now supported only for 8022 cards - * This function needs to be updated when AER/EEH is enabled - * for other cards. - */ +static inline int is_qla8032(struct scsi_qla_host *ha) +{ + return ha->pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8324; +} + +static inline int is_qla80XX(struct scsi_qla_host *ha) +{ + return is_qla8022(ha) || is_qla8032(ha); +} + static inline int is_aer_supported(struct scsi_qla_host *ha) { - return ha->pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8022; + return ((ha->pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8022) || + (ha->pdev->device == PCI_DEVICE_ID_QLOGIC_ISP8324)); } static inline int adapter_up(struct scsi_qla_host *ha) @@ -942,6 +973,20 @@ static inline int ql4xxx_reset_active(struct scsi_qla_host *ha) test_bit(DPC_HA_UNRECOVERABLE, &ha->dpc_flags); } + +static inline int qla4_8xxx_rd_direct(struct scsi_qla_host *ha, + const uint32_t crb_reg) +{ + return ha->isp_ops->rd_reg_direct(ha, ha->reg_tbl[crb_reg]); +} + +static inline void qla4_8xxx_wr_direct(struct scsi_qla_host *ha, + const uint32_t crb_reg, + const uint32_t value) +{ + ha->isp_ops->wr_reg_direct(ha, ha->reg_tbl[crb_reg], value); +} + /*---------------------------------------------------------------------------*/ /* Defines for qla4xxx_initialize_adapter() and qla4xxx_recover_adapter() */ diff --git a/drivers/scsi/qla4xxx/ql4_fw.h b/drivers/scsi/qla4xxx/ql4_fw.h index 7240948..1c47950 100644 --- a/drivers/scsi/qla4xxx/ql4_fw.h +++ b/drivers/scsi/qla4xxx/ql4_fw.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -65,6 +65,40 @@ struct device_reg_82xx { #define ISRX_82XX_RISC_INT BIT_0 /* RISC interrupt. */ }; +/* ISP 83xx I/O Register Set structure */ +struct device_reg_83xx { + __le32 mailbox_in[16]; /* 0x0000 */ + __le32 reserve1[496]; /* 0x0040 */ + __le32 mailbox_out[16]; /* 0x0800 */ + __le32 reserve2[496]; + __le32 mbox_int; /* 0x1000 */ + __le32 reserve3[63]; + __le32 req_q_out; /* 0x1100 */ + __le32 reserve4[63]; + + __le32 rsp_q_in; /* 0x1200 */ + __le32 reserve5[1919]; + + __le32 req_q_in; /* 0x3000 */ + __le32 reserve6[3]; + __le32 iocb_int_mask; /* 0x3010 */ + __le32 reserve7[3]; + __le32 rsp_q_out; /* 0x3020 */ + __le32 reserve8[3]; + __le32 anonymousbuff; /* 0x3030 */ + __le32 mb_int_mask; /* 0x3034 */ + + __le32 host_intr; /* 0x3038 - Host Interrupt Register */ + __le32 risc_intr; /* 0x303C - RISC Interrupt Register */ + __le32 reserve9[544]; + __le32 leg_int_ptr; /* 0x38C0 - Legacy Interrupt Pointer Register */ + __le32 leg_int_trig; /* 0x38C4 - Legacy Interrupt Trigger Control */ + __le32 leg_int_mask; /* 0x38C8 - Legacy Interrupt Mask Register */ +}; + +#define INT_ENABLE_FW_MB (1 << 2) +#define INT_MASK_FW_MB (1 << 2) + /* remote register set (access via PCI memory read/write) */ struct isp_reg { #define MBOX_REG_COUNT 8 @@ -356,6 +390,9 @@ struct qla_flt_region { #define LOGOUT_OPTION_CLOSE_SESSION 0x0002 #define LOGOUT_OPTION_RELOGIN 0x0004 #define LOGOUT_OPTION_FREE_DDB 0x0008 +#define MBOX_CMD_SET_PARAM 0x0059 +#define SET_DRVR_VERSION 0x200 +#define MAX_DRVR_VER_LEN 24 #define MBOX_CMD_EXECUTE_IOCB_A64 0x005A #define MBOX_CMD_INITIALIZE_FIRMWARE 0x0060 #define MBOX_CMD_GET_INIT_FW_CTRL_BLOCK 0x0061 @@ -417,6 +454,10 @@ struct qla_flt_region { #define MBOX_CMD_GET_CRASH_RECORD 0x0076 /* 4010 only */ #define MBOX_CMD_GET_CONN_EVENT_LOG 0x0077 +#define MBOX_CMD_IDC_ACK 0x0101 +#define MBOX_CMD_PORT_RESET 0x0120 +#define MBOX_CMD_SET_PORT_CONFIG 0x0122 + /* Mailbox status definitions */ #define MBOX_COMPLETION_STATUS 4 #define MBOX_STS_BUSY 0x0007 @@ -453,6 +494,8 @@ struct qla_flt_region { #define MBOX_ASTS_IPV6_ND_PREFIX_IGNORED 0x802C #define MBOX_ASTS_IPV6_LCL_PREFIX_IGNORED 0x802D #define MBOX_ASTS_ICMPV6_ERROR_MSG_RCVD 0x802E +#define MBOX_ASTS_IDC_COMPLETE 0x8100 +#define MBOX_ASTS_IDC_NOTIFY 0x8101 #define MBOX_ASTS_TXSCVR_INSERTED 0x8130 #define MBOX_ASTS_TXSCVR_REMOVED 0x8131 @@ -1195,9 +1238,12 @@ struct ql_iscsi_stats { uint8_t reserved2[264]; /* 0x0308 - 0x040F */ }; -#define QLA82XX_DBG_STATE_ARRAY_LEN 16 -#define QLA82XX_DBG_CAP_SIZE_ARRAY_LEN 8 -#define QLA82XX_DBG_RSVD_ARRAY_LEN 8 +#define QLA8XXX_DBG_STATE_ARRAY_LEN 16 +#define QLA8XXX_DBG_CAP_SIZE_ARRAY_LEN 8 +#define QLA8XXX_DBG_RSVD_ARRAY_LEN 8 +#define QLA83XX_DBG_OCM_WNDREG_ARRAY_LEN 16 +#define QLA83XX_SS_OCM_WNDREG_INDEX 3 +#define QLA83XX_SS_PCI_INDEX 0 struct qla4_8xxx_minidump_template_hdr { uint32_t entry_type; @@ -1214,8 +1260,9 @@ struct qla4_8xxx_minidump_template_hdr { uint32_t driver_info_word3; uint32_t driver_info_word4; - uint32_t saved_state_array[QLA82XX_DBG_STATE_ARRAY_LEN]; - uint32_t capture_size_array[QLA82XX_DBG_CAP_SIZE_ARRAY_LEN]; + uint32_t saved_state_array[QLA8XXX_DBG_STATE_ARRAY_LEN]; + uint32_t capture_size_array[QLA8XXX_DBG_CAP_SIZE_ARRAY_LEN]; + uint32_t ocm_window_reg[QLA83XX_DBG_OCM_WNDREG_ARRAY_LEN]; }; #endif /* _QLA4X_FW_H */ diff --git a/drivers/scsi/qla4xxx/ql4_glbl.h b/drivers/scsi/qla4xxx/ql4_glbl.h index 5b2525c..57a5a3c 100644 --- a/drivers/scsi/qla4xxx/ql4_glbl.h +++ b/drivers/scsi/qla4xxx/ql4_glbl.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -109,28 +109,28 @@ uint8_t qla4xxx_update_local_ifcb(struct scsi_qla_host *ha, void qla4_8xxx_pci_config(struct scsi_qla_host *); int qla4_8xxx_iospace_config(struct scsi_qla_host *ha); int qla4_8xxx_load_risc(struct scsi_qla_host *); -irqreturn_t qla4_8xxx_intr_handler(int irq, void *dev_id); -void qla4_8xxx_queue_iocb(struct scsi_qla_host *ha); -void qla4_8xxx_complete_iocb(struct scsi_qla_host *ha); +irqreturn_t qla4_82xx_intr_handler(int irq, void *dev_id); +void qla4_82xx_queue_iocb(struct scsi_qla_host *ha); +void qla4_82xx_complete_iocb(struct scsi_qla_host *ha); -int qla4_8xxx_crb_win_lock(struct scsi_qla_host *); -void qla4_8xxx_crb_win_unlock(struct scsi_qla_host *); -int qla4_8xxx_pci_get_crb_addr_2M(struct scsi_qla_host *, ulong *); -void qla4_8xxx_wr_32(struct scsi_qla_host *, ulong, u32); -int qla4_8xxx_rd_32(struct scsi_qla_host *, ulong); -int qla4_8xxx_pci_mem_read_2M(struct scsi_qla_host *, u64, void *, int); -int qla4_8xxx_pci_mem_write_2M(struct scsi_qla_host *ha, u64, void *, int); -int qla4_8xxx_isp_reset(struct scsi_qla_host *ha); -void qla4_8xxx_interrupt_service_routine(struct scsi_qla_host *ha, +int qla4_82xx_crb_win_lock(struct scsi_qla_host *); +void qla4_82xx_crb_win_unlock(struct scsi_qla_host *); +int qla4_82xx_pci_get_crb_addr_2M(struct scsi_qla_host *, ulong *); +void qla4_82xx_wr_32(struct scsi_qla_host *, ulong, u32); +uint32_t qla4_82xx_rd_32(struct scsi_qla_host *, ulong); +int qla4_82xx_pci_mem_read_2M(struct scsi_qla_host *, u64, void *, int); +int qla4_82xx_pci_mem_write_2M(struct scsi_qla_host *ha, u64, void *, int); +int qla4_82xx_isp_reset(struct scsi_qla_host *ha); +void qla4_82xx_interrupt_service_routine(struct scsi_qla_host *ha, uint32_t intr_status); -uint16_t qla4_8xxx_rd_shdw_req_q_out(struct scsi_qla_host *ha); -uint16_t qla4_8xxx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha); +uint16_t qla4_82xx_rd_shdw_req_q_out(struct scsi_qla_host *ha); +uint16_t qla4_82xx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha); int qla4_8xxx_get_sys_info(struct scsi_qla_host *ha); void qla4_8xxx_watchdog(struct scsi_qla_host *ha); int qla4_8xxx_stop_firmware(struct scsi_qla_host *ha); int qla4_8xxx_get_flash_info(struct scsi_qla_host *ha); -void qla4_8xxx_enable_intrs(struct scsi_qla_host *ha); -void qla4_8xxx_disable_intrs(struct scsi_qla_host *ha); +void qla4_82xx_enable_intrs(struct scsi_qla_host *ha); +void qla4_82xx_disable_intrs(struct scsi_qla_host *ha); int qla4_8xxx_enable_msix(struct scsi_qla_host *ha); void qla4_8xxx_disable_msix(struct scsi_qla_host *ha); irqreturn_t qla4_8xxx_msi_handler(int irq, void *dev_id); @@ -138,8 +138,8 @@ irqreturn_t qla4_8xxx_default_intr_handler(int irq, void *dev_id); irqreturn_t qla4_8xxx_msix_rsp_q(int irq, void *dev_id); void qla4xxx_mark_all_devices_missing(struct scsi_qla_host *ha); void qla4xxx_dead_adapter_cleanup(struct scsi_qla_host *ha); -int qla4_8xxx_idc_lock(struct scsi_qla_host *ha); -void qla4_8xxx_idc_unlock(struct scsi_qla_host *ha); +int qla4_82xx_idc_lock(struct scsi_qla_host *ha); +void qla4_82xx_idc_unlock(struct scsi_qla_host *ha); int qla4_8xxx_device_state_handler(struct scsi_qla_host *ha); void qla4_8xxx_need_qsnt_handler(struct scsi_qla_host *ha); void qla4_8xxx_clear_drv_active(struct scsi_qla_host *ha); @@ -203,6 +203,62 @@ int qla4xxx_req_template_size(struct scsi_qla_host *ha); void qla4_8xxx_alloc_sysfs_attr(struct scsi_qla_host *ha); void qla4_8xxx_free_sysfs_attr(struct scsi_qla_host *ha); void qla4xxx_alloc_fw_dump(struct scsi_qla_host *ha); +int qla4_82xx_try_start_fw(struct scsi_qla_host *ha); +int qla4_8xxx_need_reset(struct scsi_qla_host *ha); +int qla4_82xx_md_rd_32(struct scsi_qla_host *ha, uint32_t off, uint32_t *data); +int qla4_82xx_md_wr_32(struct scsi_qla_host *ha, uint32_t off, uint32_t data); +void qla4_82xx_rom_lock_recovery(struct scsi_qla_host *ha); +void qla4_82xx_queue_mbox_cmd(struct scsi_qla_host *ha, uint32_t *mbx_cmd, + int incount); +void qla4_82xx_process_mbox_intr(struct scsi_qla_host *ha, int outcount); +void qla4xxx_queue_mbox_cmd(struct scsi_qla_host *ha, uint32_t *mbx_cmd, + int incount); +void qla4xxx_process_mbox_intr(struct scsi_qla_host *ha, int outcount); +void qla4_8xxx_dump_peg_reg(struct scsi_qla_host *ha); +void qla4_83xx_disable_intrs(struct scsi_qla_host *ha); +void qla4_83xx_enable_intrs(struct scsi_qla_host *ha); +int qla4_83xx_start_firmware(struct scsi_qla_host *ha); +irqreturn_t qla4_83xx_intr_handler(int irq, void *dev_id); +void qla4_83xx_interrupt_service_routine(struct scsi_qla_host *ha, + uint32_t intr_status); +int qla4_83xx_isp_reset(struct scsi_qla_host *ha); +void qla4_83xx_queue_iocb(struct scsi_qla_host *ha); +void qla4_83xx_complete_iocb(struct scsi_qla_host *ha); +uint16_t qla4_83xx_rd_shdw_req_q_out(struct scsi_qla_host *ha); +uint16_t qla4_83xx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha); +uint32_t qla4_83xx_rd_reg(struct scsi_qla_host *ha, ulong addr); +void qla4_83xx_wr_reg(struct scsi_qla_host *ha, ulong addr, uint32_t val); +int qla4_83xx_rd_reg_indirect(struct scsi_qla_host *ha, uint32_t addr, + uint32_t *data); +int qla4_83xx_wr_reg_indirect(struct scsi_qla_host *ha, uint32_t addr, + uint32_t data); +int qla4_83xx_drv_lock(struct scsi_qla_host *ha); +void qla4_83xx_drv_unlock(struct scsi_qla_host *ha); +void qla4_83xx_rom_lock_recovery(struct scsi_qla_host *ha); +void qla4_83xx_queue_mbox_cmd(struct scsi_qla_host *ha, uint32_t *mbx_cmd, + int incount); +void qla4_83xx_process_mbox_intr(struct scsi_qla_host *ha, int outcount); +void qla4_83xx_read_reset_template(struct scsi_qla_host *ha); +void qla4_83xx_set_idc_dontreset(struct scsi_qla_host *ha); +int qla4_83xx_idc_dontreset(struct scsi_qla_host *ha); +int qla4_83xx_lockless_flash_read_u32(struct scsi_qla_host *ha, + uint32_t flash_addr, uint8_t *p_data, + int u32_word_count); +void qla4_83xx_clear_idc_dontreset(struct scsi_qla_host *ha); +void qla4_83xx_need_reset_handler(struct scsi_qla_host *ha); +int qla4_83xx_flash_read_u32(struct scsi_qla_host *ha, uint32_t flash_addr, + uint8_t *p_data, int u32_word_count); +void qla4_83xx_get_idc_param(struct scsi_qla_host *ha); +void qla4_8xxx_set_rst_ready(struct scsi_qla_host *ha); +void qla4_8xxx_clear_rst_ready(struct scsi_qla_host *ha); +int qla4_8xxx_device_bootstrap(struct scsi_qla_host *ha); +void qla4_8xxx_get_minidump(struct scsi_qla_host *ha); +int qla4_8xxx_mbx_intr_disable(struct scsi_qla_host *ha); +int qla4_8xxx_mbx_intr_enable(struct scsi_qla_host *ha); +int qla4_8xxx_set_param(struct scsi_qla_host *ha, int param); +int qla4_8xxx_update_idc_reg(struct scsi_qla_host *ha); +int qla4_83xx_post_idc_ack(struct scsi_qla_host *ha); +void qla4_83xx_disable_pause(struct scsi_qla_host *ha); extern int ql4xextended_error_logging; extern int ql4xdontresethba; diff --git a/drivers/scsi/qla4xxx/ql4_init.c b/drivers/scsi/qla4xxx/ql4_init.c index ddd9472..1aca1b4 100644 --- a/drivers/scsi/qla4xxx/ql4_init.c +++ b/drivers/scsi/qla4xxx/ql4_init.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -102,11 +102,18 @@ int qla4xxx_init_rings(struct scsi_qla_host *ha) if (is_qla8022(ha)) { writel(0, - (unsigned long __iomem *)&ha->qla4_8xxx_reg->req_q_out); + (unsigned long __iomem *)&ha->qla4_82xx_reg->req_q_out); writel(0, - (unsigned long __iomem *)&ha->qla4_8xxx_reg->rsp_q_in); + (unsigned long __iomem *)&ha->qla4_82xx_reg->rsp_q_in); writel(0, - (unsigned long __iomem *)&ha->qla4_8xxx_reg->rsp_q_out); + (unsigned long __iomem *)&ha->qla4_82xx_reg->rsp_q_out); + } else if (is_qla8032(ha)) { + writel(0, + (unsigned long __iomem *)&ha->qla4_83xx_reg->req_q_in); + writel(0, + (unsigned long __iomem *)&ha->qla4_83xx_reg->rsp_q_in); + writel(0, + (unsigned long __iomem *)&ha->qla4_83xx_reg->rsp_q_out); } else { /* * Initialize DMA Shadow registers. The firmware is really @@ -524,7 +531,7 @@ static int qla4xxx_init_firmware(struct scsi_qla_host *ha) /* For 82xx, stop firmware before initializing because if BIOS * has previously initialized firmware, then driver's initialize * firmware will fail. */ - if (is_qla8022(ha)) + if (is_qla80XX(ha)) qla4_8xxx_stop_firmware(ha); ql4_printk(KERN_INFO, ha, "Initializing firmware..\n"); @@ -537,7 +544,7 @@ static int qla4xxx_init_firmware(struct scsi_qla_host *ha) if (!qla4xxx_fw_ready(ha)) return status; - if (is_qla8022(ha) && !test_bit(AF_INIT_DONE, &ha->flags)) + if (is_qla80XX(ha) && !test_bit(AF_INIT_DONE, &ha->flags)) qla4xxx_alloc_fw_dump(ha); return qla4xxx_get_firmware_status(ha); @@ -946,9 +953,9 @@ int qla4xxx_initialize_adapter(struct scsi_qla_host *ha, int is_reset) set_bit(AF_ONLINE, &ha->flags); exit_init_hba: - if (is_qla8022(ha) && (status == QLA_ERROR)) { + if (is_qla80XX(ha) && (status == QLA_ERROR)) { /* Since interrupts are registered in start_firmware for - * 82xx, release them here if initialize_adapter fails */ + * 80XX, release them here if initialize_adapter fails */ qla4xxx_free_irqs(ha); } diff --git a/drivers/scsi/qla4xxx/ql4_inline.h b/drivers/scsi/qla4xxx/ql4_inline.h index 62f90bd..6f4decd 100644 --- a/drivers/scsi/qla4xxx/ql4_inline.h +++ b/drivers/scsi/qla4xxx/ql4_inline.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla4xxx/ql4_iocb.c b/drivers/scsi/qla4xxx/ql4_iocb.c index 2a2022a..f48f37a 100644 --- a/drivers/scsi/qla4xxx/ql4_iocb.c +++ b/drivers/scsi/qla4xxx/ql4_iocb.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -192,35 +192,47 @@ static void qla4xxx_build_scsi_iocbs(struct srb *srb, } } +void qla4_83xx_queue_iocb(struct scsi_qla_host *ha) +{ + writel(ha->request_in, &ha->qla4_83xx_reg->req_q_in); + readl(&ha->qla4_83xx_reg->req_q_in); +} + +void qla4_83xx_complete_iocb(struct scsi_qla_host *ha) +{ + writel(ha->response_out, &ha->qla4_83xx_reg->rsp_q_out); + readl(&ha->qla4_83xx_reg->rsp_q_out); +} + /** - * qla4_8xxx_queue_iocb - Tell ISP it's got new request(s) + * qla4_82xx_queue_iocb - Tell ISP it's got new request(s) * @ha: pointer to host adapter structure. * * This routine notifies the ISP that one or more new request * queue entries have been placed on the request queue. **/ -void qla4_8xxx_queue_iocb(struct scsi_qla_host *ha) +void qla4_82xx_queue_iocb(struct scsi_qla_host *ha) { uint32_t dbval = 0; dbval = 0x14 | (ha->func_num << 5); dbval = dbval | (0 << 8) | (ha->request_in << 16); - qla4_8xxx_wr_32(ha, ha->nx_db_wr_ptr, ha->request_in); + qla4_82xx_wr_32(ha, ha->nx_db_wr_ptr, ha->request_in); } /** - * qla4_8xxx_complete_iocb - Tell ISP we're done with response(s) + * qla4_82xx_complete_iocb - Tell ISP we're done with response(s) * @ha: pointer to host adapter structure. * * This routine notifies the ISP that one or more response/completion * queue entries have been processed by the driver. * This also clears the interrupt. **/ -void qla4_8xxx_complete_iocb(struct scsi_qla_host *ha) +void qla4_82xx_complete_iocb(struct scsi_qla_host *ha) { - writel(ha->response_out, &ha->qla4_8xxx_reg->rsp_q_out); - readl(&ha->qla4_8xxx_reg->rsp_q_out); + writel(ha->response_out, &ha->qla4_82xx_reg->rsp_q_out); + readl(&ha->qla4_82xx_reg->rsp_q_out); } /** diff --git a/drivers/scsi/qla4xxx/ql4_isr.c b/drivers/scsi/qla4xxx/ql4_isr.c index fc542a9..15ea814 100644 --- a/drivers/scsi/qla4xxx/ql4_isr.c +++ b/drivers/scsi/qla4xxx/ql4_isr.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -126,7 +126,7 @@ static void qla4xxx_status_entry(struct scsi_qla_host *ha, ql4_printk(KERN_WARNING, ha, "%s invalid status entry: " "handle=0x%0x, srb=%p\n", __func__, sts_entry->handle, srb); - if (is_qla8022(ha)) + if (is_qla80XX(ha)) set_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags); else set_bit(DPC_RESET_HA, &ha->dpc_flags); @@ -243,56 +243,72 @@ static void qla4xxx_status_entry(struct scsi_qla_host *ha, scsi_set_resid(cmd, residual); - /* - * If there is scsi_status, it takes precedense over - * underflow condition. - */ - if (scsi_status != 0) { - cmd->result = DID_OK << 16 | scsi_status; + if (sts_entry->iscsiFlags & ISCSI_FLAG_RESIDUAL_UNDER) { + + /* Both the firmware and target reported UNDERRUN: + * + * MID-LAYER UNDERFLOW case: + * Some kernels do not properly detect midlayer + * underflow, so we manually check it and return + * ERROR if the minimum required data was not + * received. + * + * ALL OTHER cases: + * Fall thru to check scsi_status + */ + if (!scsi_status && (scsi_bufflen(cmd) - residual) < + cmd->underflow) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "scsi%ld:%d:%d:%d: %s: Mid-layer Data underrun, xferlen = 0x%x,residual = 0x%x\n", + ha->host_no, + cmd->device->channel, + cmd->device->id, + cmd->device->lun, __func__, + scsi_bufflen(cmd), + residual)); - if (scsi_status != SCSI_CHECK_CONDITION) + cmd->result = DID_ERROR << 16; break; + } + + } else if (scsi_status != SAM_STAT_TASK_SET_FULL && + scsi_status != SAM_STAT_BUSY) { - /* Copy Sense Data into sense buffer. */ - qla4xxx_copy_sense(ha, sts_entry, srb); - } else { /* - * If RISC reports underrun and target does not - * report it then we must have a lost frame, so - * tell upper layer to retry it by reporting a - * bus busy. + * The firmware reports UNDERRUN, but the target does + * not report it: + * + * scsi_status | host_byte device_byte + * | (19:16) (7:0) + * ============= | ========= =========== + * TASK_SET_FULL | DID_OK scsi_status + * BUSY | DID_OK scsi_status + * ALL OTHERS | DID_ERROR scsi_status + * + * Note: If scsi_status is task set full or busy, + * then this else if would fall thru to check the + * scsi_status and return DID_OK. */ - if ((sts_entry->iscsiFlags & - ISCSI_FLAG_RESIDUAL_UNDER) == 0) { - cmd->result = DID_BUS_BUSY << 16; - } else if ((scsi_bufflen(cmd) - residual) < - cmd->underflow) { - /* - * Handle mid-layer underflow??? - * - * For kernels less than 2.4, the driver must - * return an error if an underflow is detected. - * For kernels equal-to and above 2.4, the - * mid-layer will appearantly handle the - * underflow by detecting the residual count -- - * unfortunately, we do not see where this is - * actually being done. In the interim, we - * will return DID_ERROR. - */ - DEBUG2(printk("scsi%ld:%d:%d:%d: %s: " - "Mid-layer Data underrun1, " - "xferlen = 0x%x, " - "residual = 0x%x\n", ha->host_no, - cmd->device->channel, - cmd->device->id, - cmd->device->lun, __func__, - scsi_bufflen(cmd), residual)); - cmd->result = DID_ERROR << 16; - } else { - cmd->result = DID_OK << 16; - } + DEBUG2(ql4_printk(KERN_INFO, ha, + "scsi%ld:%d:%d:%d: %s: Dropped frame(s) detected (0x%x of 0x%x bytes).\n", + ha->host_no, + cmd->device->channel, + cmd->device->id, + cmd->device->lun, __func__, + residual, + scsi_bufflen(cmd))); + + cmd->result = DID_ERROR << 16 | scsi_status; + goto check_scsi_status; } + + cmd->result = DID_OK << 16 | scsi_status; + +check_scsi_status: + if (scsi_status == SAM_STAT_CHECK_CONDITION) + qla4xxx_copy_sense(ha, sts_entry, srb); + break; case SCS_DEVICE_LOGGED_OUT: @@ -578,6 +594,14 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, { int i; uint32_t mbox_sts[MBOX_AEN_REG_COUNT]; + __le32 __iomem *mailbox_out; + + if (is_qla8032(ha)) + mailbox_out = &ha->qla4_83xx_reg->mailbox_out[0]; + else if (is_qla8022(ha)) + mailbox_out = &ha->qla4_82xx_reg->mailbox_out[0]; + else + mailbox_out = &ha->reg->mailbox[0]; if ((mbox_status == MBOX_STS_BUSY) || (mbox_status == MBOX_STS_INTERMEDIATE_COMPLETION) || @@ -590,9 +614,7 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, * location and set mailbox command done flag */ for (i = 0; i < ha->mbox_status_count; i++) - ha->mbox_status[i] = is_qla8022(ha) - ? readl(&ha->qla4_8xxx_reg->mailbox_out[i]) - : readl(&ha->reg->mailbox[i]); + ha->mbox_status[i] = readl(&mailbox_out[i]); set_bit(AF_MBOX_COMMAND_DONE, &ha->flags); @@ -601,9 +623,7 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, } } else if (mbox_status >> 12 == MBOX_ASYNC_EVENT_STATUS) { for (i = 0; i < MBOX_AEN_REG_COUNT; i++) - mbox_sts[i] = is_qla8022(ha) - ? readl(&ha->qla4_8xxx_reg->mailbox_out[i]) - : readl(&ha->reg->mailbox[i]); + mbox_sts[i] = readl(&mailbox_out[i]); /* Immediately process the AENs that don't require much work. * Only queue the database_changed AENs */ @@ -619,7 +639,8 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, ql4_printk(KERN_INFO, ha, "%s: System Err\n", __func__); qla4xxx_dump_registers(ha); - if (ql4xdontresethba) { + if ((is_qla8022(ha) && ql4xdontresethba) || + (is_qla8032(ha) && qla4_83xx_idc_dontreset(ha))) { DEBUG2(printk("scsi%ld: %s:Don't Reset HBA\n", ha->host_no, __func__)); } else { @@ -635,7 +656,7 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, case MBOX_ASTS_DHCP_LEASE_EXPIRED: DEBUG2(printk("scsi%ld: AEN %04x, ERROR Status, " "Reset HA\n", ha->host_no, mbox_status)); - if (is_qla8022(ha)) + if (is_qla80XX(ha)) set_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags); else @@ -700,7 +721,7 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, set_bit(DPC_GET_DHCP_IP_ADDR, &ha->dpc_flags); else if ((mbox_sts[3] == ACB_STATE_ACQUIRING) && (mbox_sts[2] == ACB_STATE_VALID)) { - if (is_qla8022(ha)) + if (is_qla80XX(ha)) set_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags); else @@ -785,6 +806,43 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, " removed\n", ha->host_no, mbox_sts[0])); break; + case MBOX_ASTS_IDC_NOTIFY: + { + uint32_t opcode; + if (is_qla8032(ha)) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "scsi%ld: AEN %04x, mbox_sts[1]=%08x, mbox_sts[2]=%08x, mbox_sts[3]=%08x, mbox_sts[4]=%08x\n", + ha->host_no, mbox_sts[0], + mbox_sts[1], mbox_sts[2], + mbox_sts[3], mbox_sts[4])); + opcode = mbox_sts[1] >> 16; + if ((opcode == MBOX_CMD_SET_PORT_CONFIG) || + (opcode == MBOX_CMD_PORT_RESET)) { + set_bit(DPC_POST_IDC_ACK, + &ha->dpc_flags); + ha->idc_info.request_desc = mbox_sts[1]; + ha->idc_info.info1 = mbox_sts[2]; + ha->idc_info.info2 = mbox_sts[3]; + ha->idc_info.info3 = mbox_sts[4]; + qla4xxx_wake_dpc(ha); + } + } + break; + } + + case MBOX_ASTS_IDC_COMPLETE: + if (is_qla8032(ha)) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "scsi%ld: AEN %04x, mbox_sts[1]=%08x, mbox_sts[2]=%08x, mbox_sts[3]=%08x, mbox_sts[4]=%08x\n", + ha->host_no, mbox_sts[0], + mbox_sts[1], mbox_sts[2], + mbox_sts[3], mbox_sts[4])); + DEBUG2(ql4_printk(KERN_INFO, ha, + "scsi:%ld: AEN %04x IDC Complete notification\n", + ha->host_no, mbox_sts[0])); + } + break; + default: DEBUG2(printk(KERN_WARNING "scsi%ld: AEN %04x UNKNOWN\n", @@ -799,14 +857,31 @@ static void qla4xxx_isr_decode_mailbox(struct scsi_qla_host * ha, } } +void qla4_83xx_interrupt_service_routine(struct scsi_qla_host *ha, + uint32_t intr_status) +{ + /* Process mailbox/asynch event interrupt.*/ + if (intr_status) { + qla4xxx_isr_decode_mailbox(ha, + readl(&ha->qla4_83xx_reg->mailbox_out[0])); + /* clear the interrupt */ + writel(0, &ha->qla4_83xx_reg->risc_intr); + } else { + qla4xxx_process_response_queue(ha); + } + + /* clear the interrupt */ + writel(0, &ha->qla4_83xx_reg->mb_int_mask); +} + /** - * qla4_8xxx_interrupt_service_routine - isr + * qla4_82xx_interrupt_service_routine - isr * @ha: pointer to host adapter structure. * * This is the main interrupt service routine. * hardware_lock locked upon entry. runs in interrupt context. **/ -void qla4_8xxx_interrupt_service_routine(struct scsi_qla_host *ha, +void qla4_82xx_interrupt_service_routine(struct scsi_qla_host *ha, uint32_t intr_status) { /* Process response queue interrupt. */ @@ -816,11 +891,11 @@ void qla4_8xxx_interrupt_service_routine(struct scsi_qla_host *ha, /* Process mailbox/asynch event interrupt.*/ if (intr_status & HSRX_RISC_MB_INT) qla4xxx_isr_decode_mailbox(ha, - readl(&ha->qla4_8xxx_reg->mailbox_out[0])); + readl(&ha->qla4_82xx_reg->mailbox_out[0])); /* clear the interrupt */ - writel(0, &ha->qla4_8xxx_reg->host_int); - readl(&ha->qla4_8xxx_reg->host_int); + writel(0, &ha->qla4_82xx_reg->host_int); + readl(&ha->qla4_82xx_reg->host_int); } /** @@ -850,12 +925,12 @@ void qla4xxx_interrupt_service_routine(struct scsi_qla_host * ha, } /** - * qla4_8xxx_spurious_interrupt - processes spurious interrupt + * qla4_82xx_spurious_interrupt - processes spurious interrupt * @ha: pointer to host adapter structure. * @reqs_count: . * **/ -static void qla4_8xxx_spurious_interrupt(struct scsi_qla_host *ha, +static void qla4_82xx_spurious_interrupt(struct scsi_qla_host *ha, uint8_t reqs_count) { if (reqs_count) @@ -863,9 +938,9 @@ static void qla4_8xxx_spurious_interrupt(struct scsi_qla_host *ha, DEBUG2(ql4_printk(KERN_INFO, ha, "Spurious Interrupt\n")); if (is_qla8022(ha)) { - writel(0, &ha->qla4_8xxx_reg->host_int); + writel(0, &ha->qla4_82xx_reg->host_int); if (test_bit(AF_INTx_ENABLED, &ha->flags)) - qla4_8xxx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0xfbff); } ha->spurious_int_count++; @@ -968,11 +1043,11 @@ irqreturn_t qla4xxx_intr_handler(int irq, void *dev_id) } /** - * qla4_8xxx_intr_handler - hardware interrupt handler. + * qla4_82xx_intr_handler - hardware interrupt handler. * @irq: Unused * @dev_id: Pointer to host adapter structure **/ -irqreturn_t qla4_8xxx_intr_handler(int irq, void *dev_id) +irqreturn_t qla4_82xx_intr_handler(int irq, void *dev_id) { struct scsi_qla_host *ha = dev_id; uint32_t intr_status; @@ -984,11 +1059,11 @@ irqreturn_t qla4_8xxx_intr_handler(int irq, void *dev_id) return IRQ_HANDLED; ha->isr_count++; - status = qla4_8xxx_rd_32(ha, ISR_INT_VECTOR); + status = qla4_82xx_rd_32(ha, ISR_INT_VECTOR); if (!(status & ha->nx_legacy_intr.int_vec_bit)) return IRQ_NONE; - status = qla4_8xxx_rd_32(ha, ISR_INT_STATE_REG); + status = qla4_82xx_rd_32(ha, ISR_INT_STATE_REG); if (!ISR_IS_LEGACY_INTR_TRIGGERED(status)) { DEBUG2(ql4_printk(KERN_INFO, ha, "%s legacy Int not triggered\n", __func__)); @@ -996,30 +1071,30 @@ irqreturn_t qla4_8xxx_intr_handler(int irq, void *dev_id) } /* clear the interrupt */ - qla4_8xxx_wr_32(ha, ha->nx_legacy_intr.tgt_status_reg, 0xffffffff); + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_status_reg, 0xffffffff); /* read twice to ensure write is flushed */ - qla4_8xxx_rd_32(ha, ISR_INT_VECTOR); - qla4_8xxx_rd_32(ha, ISR_INT_VECTOR); + qla4_82xx_rd_32(ha, ISR_INT_VECTOR); + qla4_82xx_rd_32(ha, ISR_INT_VECTOR); spin_lock_irqsave(&ha->hardware_lock, flags); while (1) { - if (!(readl(&ha->qla4_8xxx_reg->host_int) & + if (!(readl(&ha->qla4_82xx_reg->host_int) & ISRX_82XX_RISC_INT)) { - qla4_8xxx_spurious_interrupt(ha, reqs_count); + qla4_82xx_spurious_interrupt(ha, reqs_count); break; } - intr_status = readl(&ha->qla4_8xxx_reg->host_status); + intr_status = readl(&ha->qla4_82xx_reg->host_status); if ((intr_status & (HSRX_RISC_MB_INT | HSRX_RISC_IOCB_INT)) == 0) { - qla4_8xxx_spurious_interrupt(ha, reqs_count); + qla4_82xx_spurious_interrupt(ha, reqs_count); break; } ha->isp_ops->interrupt_service_routine(ha, intr_status); /* Enable Interrupt */ - qla4_8xxx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0xfbff); + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0xfbff); if (++reqs_count == MAX_REQS_SERVICED_PER_INTR) break; @@ -1029,6 +1104,59 @@ irqreturn_t qla4_8xxx_intr_handler(int irq, void *dev_id) return IRQ_HANDLED; } +#define LEG_INT_PTR_B31 (1 << 31) +#define LEG_INT_PTR_B30 (1 << 30) +#define PF_BITS_MASK (0xF << 16) + +/** + * qla4_83xx_intr_handler - hardware interrupt handler. + * @irq: Unused + * @dev_id: Pointer to host adapter structure + **/ +irqreturn_t qla4_83xx_intr_handler(int irq, void *dev_id) +{ + struct scsi_qla_host *ha = dev_id; + uint32_t leg_int_ptr = 0; + unsigned long flags = 0; + + ha->isr_count++; + leg_int_ptr = readl(&ha->qla4_83xx_reg->leg_int_ptr); + + /* Legacy interrupt is valid if bit31 of leg_int_ptr is set */ + if (!(leg_int_ptr & LEG_INT_PTR_B31)) { + ql4_printk(KERN_ERR, ha, + "%s: Legacy Interrupt Bit 31 not set, spurious interrupt!\n", + __func__); + return IRQ_NONE; + } + + /* Validate the PCIE function ID set in leg_int_ptr bits [19..16] */ + if ((leg_int_ptr & PF_BITS_MASK) != ha->pf_bit) { + ql4_printk(KERN_ERR, ha, + "%s: Incorrect function ID 0x%x in legacy interrupt register, ha->pf_bit = 0x%x\n", + __func__, (leg_int_ptr & PF_BITS_MASK), ha->pf_bit); + return IRQ_NONE; + } + + /* To de-assert legacy interrupt, write 0 to Legacy Interrupt Trigger + * Control register and poll till Legacy Interrupt Pointer register + * bit30 is 0. + */ + writel(0, &ha->qla4_83xx_reg->leg_int_trig); + do { + leg_int_ptr = readl(&ha->qla4_83xx_reg->leg_int_ptr); + if ((leg_int_ptr & PF_BITS_MASK) != ha->pf_bit) + break; + } while (leg_int_ptr & LEG_INT_PTR_B30); + + spin_lock_irqsave(&ha->hardware_lock, flags); + leg_int_ptr = readl(&ha->qla4_83xx_reg->risc_intr); + ha->isp_ops->interrupt_service_routine(ha, leg_int_ptr); + spin_unlock_irqrestore(&ha->hardware_lock, flags); + + return IRQ_HANDLED; +} + irqreturn_t qla4_8xxx_msi_handler(int irq, void *dev_id) { @@ -1043,15 +1171,46 @@ qla4_8xxx_msi_handler(int irq, void *dev_id) ha->isr_count++; /* clear the interrupt */ - qla4_8xxx_wr_32(ha, ha->nx_legacy_intr.tgt_status_reg, 0xffffffff); + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_status_reg, 0xffffffff); /* read twice to ensure write is flushed */ - qla4_8xxx_rd_32(ha, ISR_INT_VECTOR); - qla4_8xxx_rd_32(ha, ISR_INT_VECTOR); + qla4_82xx_rd_32(ha, ISR_INT_VECTOR); + qla4_82xx_rd_32(ha, ISR_INT_VECTOR); return qla4_8xxx_default_intr_handler(irq, dev_id); } +static irqreturn_t qla4_83xx_mailbox_intr_handler(int irq, void *dev_id) +{ + struct scsi_qla_host *ha = dev_id; + unsigned long flags; + uint32_t ival = 0; + + spin_lock_irqsave(&ha->hardware_lock, flags); + + ival = readl(&ha->qla4_83xx_reg->risc_intr); + if (ival == 0) { + ql4_printk(KERN_INFO, ha, + "%s: It is a spurious mailbox interrupt!\n", + __func__); + ival = readl(&ha->qla4_83xx_reg->mb_int_mask); + ival &= ~INT_MASK_FW_MB; + writel(ival, &ha->qla4_83xx_reg->mb_int_mask); + goto exit; + } + + qla4xxx_isr_decode_mailbox(ha, + readl(&ha->qla4_83xx_reg->mailbox_out[0])); + writel(0, &ha->qla4_83xx_reg->risc_intr); + ival = readl(&ha->qla4_83xx_reg->mb_int_mask); + ival &= ~INT_MASK_FW_MB; + writel(ival, &ha->qla4_83xx_reg->mb_int_mask); + ha->isr_count++; +exit: + spin_unlock_irqrestore(&ha->hardware_lock, flags); + return IRQ_HANDLED; +} + /** * qla4_8xxx_default_intr_handler - hardware interrupt handler. * @irq: Unused @@ -1068,29 +1227,32 @@ qla4_8xxx_default_intr_handler(int irq, void *dev_id) uint32_t intr_status; uint8_t reqs_count = 0; - spin_lock_irqsave(&ha->hardware_lock, flags); - while (1) { - if (!(readl(&ha->qla4_8xxx_reg->host_int) & - ISRX_82XX_RISC_INT)) { - qla4_8xxx_spurious_interrupt(ha, reqs_count); - break; - } + if (is_qla8032(ha)) { + qla4_83xx_mailbox_intr_handler(irq, dev_id); + } else { + spin_lock_irqsave(&ha->hardware_lock, flags); + while (1) { + if (!(readl(&ha->qla4_82xx_reg->host_int) & + ISRX_82XX_RISC_INT)) { + qla4_82xx_spurious_interrupt(ha, reqs_count); + break; + } - intr_status = readl(&ha->qla4_8xxx_reg->host_status); - if ((intr_status & - (HSRX_RISC_MB_INT | HSRX_RISC_IOCB_INT)) == 0) { - qla4_8xxx_spurious_interrupt(ha, reqs_count); - break; - } + intr_status = readl(&ha->qla4_82xx_reg->host_status); + if ((intr_status & + (HSRX_RISC_MB_INT | HSRX_RISC_IOCB_INT)) == 0) { + qla4_82xx_spurious_interrupt(ha, reqs_count); + break; + } - ha->isp_ops->interrupt_service_routine(ha, intr_status); + ha->isp_ops->interrupt_service_routine(ha, intr_status); - if (++reqs_count == MAX_REQS_SERVICED_PER_INTR) - break; + if (++reqs_count == MAX_REQS_SERVICED_PER_INTR) + break; + } + ha->isr_count++; + spin_unlock_irqrestore(&ha->hardware_lock, flags); } - - ha->isr_count++; - spin_unlock_irqrestore(&ha->hardware_lock, flags); return IRQ_HANDLED; } @@ -1099,13 +1261,25 @@ qla4_8xxx_msix_rsp_q(int irq, void *dev_id) { struct scsi_qla_host *ha = dev_id; unsigned long flags; + uint32_t ival = 0; spin_lock_irqsave(&ha->hardware_lock, flags); - qla4xxx_process_response_queue(ha); - writel(0, &ha->qla4_8xxx_reg->host_int); - spin_unlock_irqrestore(&ha->hardware_lock, flags); - + if (is_qla8032(ha)) { + ival = readl(&ha->qla4_83xx_reg->iocb_int_mask); + if (ival == 0) { + ql4_printk(KERN_INFO, ha, "%s: It is a spurious iocb interrupt!\n", + __func__); + goto exit_msix_rsp_q; + } + qla4xxx_process_response_queue(ha); + writel(0, &ha->qla4_83xx_reg->iocb_int_mask); + } else { + qla4xxx_process_response_queue(ha); + writel(0, &ha->qla4_82xx_reg->host_int); + } ha->isr_count++; +exit_msix_rsp_q: + spin_unlock_irqrestore(&ha->hardware_lock, flags); return IRQ_HANDLED; } @@ -1177,11 +1351,18 @@ int qla4xxx_request_irqs(struct scsi_qla_host *ha) { int ret; - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) goto try_intx; - if (ql4xenablemsix == 2) + if (ql4xenablemsix == 2) { + /* Note: MSI Interrupts not supported for ISP8324 */ + if (is_qla8032(ha)) { + ql4_printk(KERN_INFO, ha, "%s: MSI Interrupts not supported for ISP8324, Falling back-to INTx mode\n", + __func__); + goto try_intx; + } goto try_msi; + } if (ql4xenablemsix == 0 || ql4xenablemsix != 1) goto try_intx; @@ -1192,6 +1373,12 @@ int qla4xxx_request_irqs(struct scsi_qla_host *ha) DEBUG2(ql4_printk(KERN_INFO, ha, "MSI-X: Enabled (0x%X).\n", ha->revision_id)); goto irq_attached; + } else { + if (is_qla8032(ha)) { + ql4_printk(KERN_INFO, ha, "%s: ISP8324: MSI-X: Falling back-to INTx mode. ret = %d\n", + __func__, ret); + goto try_intx; + } } ql4_printk(KERN_WARNING, ha, @@ -1214,9 +1401,15 @@ try_msi: pci_disable_msi(ha->pdev); } } - ql4_printk(KERN_WARNING, ha, - "MSI: Falling back-to INTx mode -- %d.\n", ret); + /* + * Prevent interrupts from falling back to INTx mode in cases where + * interrupts cannot get acquired through MSI-X or MSI mode. + */ + if (is_qla8022(ha)) { + ql4_printk(KERN_WARNING, ha, "IRQ not attached -- %d.\n", ret); + goto irq_not_attached; + } try_intx: /* Trying INTx */ ret = request_irq(ha->pdev->irq, ha->isp_ops->intr_handler, @@ -1230,7 +1423,7 @@ try_intx: ql4_printk(KERN_WARNING, ha, "INTx: Failed to reserve interrupt %d already in" " use.\n", ha->pdev->irq); - return ret; + goto irq_not_attached; } irq_attached: @@ -1238,6 +1431,7 @@ irq_attached: ha->host->irq = ha->pdev->irq; ql4_printk(KERN_INFO, ha, "%s: irq %d attached\n", __func__, ha->pdev->irq); +irq_not_attached: return ret; } diff --git a/drivers/scsi/qla4xxx/ql4_mbx.c b/drivers/scsi/qla4xxx/ql4_mbx.c index cab8f66..3d41034 100644 --- a/drivers/scsi/qla4xxx/ql4_mbx.c +++ b/drivers/scsi/qla4xxx/ql4_mbx.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -9,7 +9,39 @@ #include "ql4_glbl.h" #include "ql4_dbg.h" #include "ql4_inline.h" +#include "ql4_version.h" +void qla4xxx_queue_mbox_cmd(struct scsi_qla_host *ha, uint32_t *mbx_cmd, + int in_count) +{ + int i; + + /* Load all mailbox registers, except mailbox 0. */ + for (i = 1; i < in_count; i++) + writel(mbx_cmd[i], &ha->reg->mailbox[i]); + + /* Wakeup firmware */ + writel(mbx_cmd[0], &ha->reg->mailbox[0]); + readl(&ha->reg->mailbox[0]); + writel(set_rmask(CSR_INTR_RISC), &ha->reg->ctrl_status); + readl(&ha->reg->ctrl_status); +} + +void qla4xxx_process_mbox_intr(struct scsi_qla_host *ha, int out_count) +{ + int intr_status; + + intr_status = readl(&ha->reg->ctrl_status); + if (intr_status & INTR_PENDING) { + /* + * Service the interrupt. + * The ISR will save the mailbox status registers + * to a temporary storage location in the adapter structure. + */ + ha->mbox_status_count = out_count; + ha->isp_ops->interrupt_service_routine(ha, intr_status); + } +} /** * qla4xxx_mailbox_command - issues mailbox commands @@ -30,7 +62,6 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, int status = QLA_ERROR; uint8_t i; u_long wait_count; - uint32_t intr_status; unsigned long flags = 0; uint32_t dev_state; @@ -77,7 +108,7 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, msleep(10); } - if (is_qla8022(ha)) { + if (is_qla80XX(ha)) { if (test_bit(AF_FW_RECOVERY, &ha->flags)) { DEBUG2(ql4_printk(KERN_WARNING, ha, "scsi%ld: %s: prematurely completing mbx cmd as firmware recovery detected\n", @@ -85,10 +116,10 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, goto mbox_exit; } /* Do not send any mbx cmd if h/w is in failed state*/ - qla4_8xxx_idc_lock(ha); - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); - qla4_8xxx_idc_unlock(ha); - if (dev_state == QLA82XX_DEV_FAILED) { + ha->isp_ops->idc_lock(ha); + dev_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE); + ha->isp_ops->idc_unlock(ha); + if (dev_state == QLA8XXX_DEV_FAILED) { ql4_printk(KERN_WARNING, ha, "scsi%ld: %s: H/W is in failed state, do not send any mailbox commands\n", ha->host_no, __func__); @@ -102,30 +133,8 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, for (i = 0; i < outCount; i++) ha->mbox_status[i] = 0; - if (is_qla8022(ha)) { - /* Load all mailbox registers, except mailbox 0. */ - DEBUG5( - printk("scsi%ld: %s: Cmd ", ha->host_no, __func__); - for (i = 0; i < inCount; i++) - printk("mb%d=%04x ", i, mbx_cmd[i]); - printk("\n")); - - for (i = 1; i < inCount; i++) - writel(mbx_cmd[i], &ha->qla4_8xxx_reg->mailbox_in[i]); - writel(mbx_cmd[0], &ha->qla4_8xxx_reg->mailbox_in[0]); - readl(&ha->qla4_8xxx_reg->mailbox_in[0]); - writel(HINT_MBX_INT_PENDING, &ha->qla4_8xxx_reg->hint); - } else { - /* Load all mailbox registers, except mailbox 0. */ - for (i = 1; i < inCount; i++) - writel(mbx_cmd[i], &ha->reg->mailbox[i]); - - /* Wakeup firmware */ - writel(mbx_cmd[0], &ha->reg->mailbox[0]); - readl(&ha->reg->mailbox[0]); - writel(set_rmask(CSR_INTR_RISC), &ha->reg->ctrl_status); - readl(&ha->reg->ctrl_status); - } + /* Queue the mailbox command to the firmware */ + ha->isp_ops->queue_mailbox_command(ha, mbx_cmd, inCount); spin_unlock_irqrestore(&ha->hardware_lock, flags); @@ -167,37 +176,7 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, */ spin_lock_irqsave(&ha->hardware_lock, flags); - if (is_qla8022(ha)) { - intr_status = - readl(&ha->qla4_8xxx_reg->host_int); - if (intr_status & ISRX_82XX_RISC_INT) { - ha->mbox_status_count = outCount; - intr_status = - readl(&ha->qla4_8xxx_reg->host_status); - ha->isp_ops->interrupt_service_routine( - ha, intr_status); - if (test_bit(AF_INTERRUPTS_ON, - &ha->flags) && - test_bit(AF_INTx_ENABLED, - &ha->flags)) - qla4_8xxx_wr_32(ha, - ha->nx_legacy_intr.tgt_mask_reg, - 0xfbff); - } - } else { - intr_status = readl(&ha->reg->ctrl_status); - if (intr_status & INTR_PENDING) { - /* - * Service the interrupt. - * The ISR will save the mailbox status - * registers to a temporary storage - * location in the adapter structure. - */ - ha->mbox_status_count = outCount; - ha->isp_ops->interrupt_service_routine( - ha, intr_status); - } - } + ha->isp_ops->process_mailbox_interrupt(ha, outCount); spin_unlock_irqrestore(&ha->hardware_lock, flags); msleep(10); } @@ -205,7 +184,7 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, /* Check for mailbox timeout. */ if (!test_bit(AF_MBOX_COMMAND_DONE, &ha->flags)) { - if (is_qla8022(ha) && + if (is_qla80XX(ha) && test_bit(AF_FW_RECOVERY, &ha->flags)) { DEBUG2(ql4_printk(KERN_INFO, ha, "scsi%ld: %s: prematurely completing mbx cmd as " @@ -222,9 +201,13 @@ int qla4xxx_mailbox_command(struct scsi_qla_host *ha, uint8_t inCount, if (is_qla8022(ha)) { ql4_printk(KERN_INFO, ha, "disabling pause transmit on port 0 & 1.\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x98, + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x98, CRB_NIU_XG_PAUSE_CTL_P0 | CRB_NIU_XG_PAUSE_CTL_P1); + } else if (is_qla8032(ha)) { + ql4_printk(KERN_INFO, ha, " %s: disabling pause transmit on port 0 & 1.\n", + __func__); + qla4_83xx_disable_pause(ha); } goto mbox_exit; } @@ -373,7 +356,7 @@ qla4xxx_set_ifcb(struct scsi_qla_host *ha, uint32_t *mbox_cmd, memset(mbox_sts, 0, sizeof(mbox_sts[0]) * MBOX_REG_COUNT); if (is_qla8022(ha)) - qla4_8xxx_wr_32(ha, ha->nx_db_wr_ptr, 0); + qla4_82xx_wr_32(ha, ha->nx_db_wr_ptr, 0); mbox_cmd[0] = MBOX_CMD_INITIALIZE_FIRMWARE; mbox_cmd[1] = 0; @@ -566,7 +549,7 @@ int qla4xxx_initialize_fw_cb(struct scsi_qla_host * ha) __constant_cpu_to_le16(FWOPT_SESSION_MODE | FWOPT_INITIATOR_MODE); - if (is_qla8022(ha)) + if (is_qla80XX(ha)) init_fw_cb->fw_options |= __constant_cpu_to_le16(FWOPT_ENABLE_CRBDB); @@ -1695,7 +1678,7 @@ int qla4xxx_set_param_ddbentry(struct scsi_qla_host *ha, conn = cls_conn->dd_data; qla_conn = conn->dd_data; sess = conn->session; - dst_addr = &qla_conn->qla_ep->dst_addr; + dst_addr = (struct sockaddr *)&qla_conn->qla_ep->dst_addr; if (dst_addr->sa_family == AF_INET6) options |= IPV6_DEFAULT_DDB_ENTRY; @@ -1953,3 +1936,72 @@ int qla4xxx_restore_factory_defaults(struct scsi_qla_host *ha, } return status; } + +/** + * qla4_8xxx_set_param - set driver version in firmware. + * @ha: Pointer to host adapter structure. + * @param: Parameter to set i.e driver version + **/ +int qla4_8xxx_set_param(struct scsi_qla_host *ha, int param) +{ + uint32_t mbox_cmd[MBOX_REG_COUNT]; + uint32_t mbox_sts[MBOX_REG_COUNT]; + uint32_t status; + + memset(&mbox_cmd, 0, sizeof(mbox_cmd)); + memset(&mbox_sts, 0, sizeof(mbox_sts)); + + mbox_cmd[0] = MBOX_CMD_SET_PARAM; + if (param == SET_DRVR_VERSION) { + mbox_cmd[1] = SET_DRVR_VERSION; + strncpy((char *)&mbox_cmd[2], QLA4XXX_DRIVER_VERSION, + MAX_DRVR_VER_LEN); + } else { + ql4_printk(KERN_ERR, ha, "%s: invalid parameter 0x%x\n", + __func__, param); + status = QLA_ERROR; + goto exit_set_param; + } + + status = qla4xxx_mailbox_command(ha, MBOX_REG_COUNT, 2, mbox_cmd, + mbox_sts); + if (status == QLA_ERROR) + ql4_printk(KERN_ERR, ha, "%s: failed status %04X\n", + __func__, mbox_sts[0]); + +exit_set_param: + return status; +} + +/** + * qla4_83xx_post_idc_ack - post IDC ACK + * @ha: Pointer to host adapter structure. + * + * Posts IDC ACK for IDC Request Notification AEN. + **/ +int qla4_83xx_post_idc_ack(struct scsi_qla_host *ha) +{ + uint32_t mbox_cmd[MBOX_REG_COUNT]; + uint32_t mbox_sts[MBOX_REG_COUNT]; + int status; + + memset(&mbox_cmd, 0, sizeof(mbox_cmd)); + memset(&mbox_sts, 0, sizeof(mbox_sts)); + + mbox_cmd[0] = MBOX_CMD_IDC_ACK; + mbox_cmd[1] = ha->idc_info.request_desc; + mbox_cmd[2] = ha->idc_info.info1; + mbox_cmd[3] = ha->idc_info.info2; + mbox_cmd[4] = ha->idc_info.info3; + + status = qla4xxx_mailbox_command(ha, MBOX_REG_COUNT, MBOX_REG_COUNT, + mbox_cmd, mbox_sts); + if (status == QLA_ERROR) + ql4_printk(KERN_ERR, ha, "%s: failed status %04X\n", __func__, + mbox_sts[0]); + else + DEBUG2(ql4_printk(KERN_INFO, ha, "%s: IDC ACK posted\n", + __func__)); + + return status; +} diff --git a/drivers/scsi/qla4xxx/ql4_nvram.c b/drivers/scsi/qla4xxx/ql4_nvram.c index 7851f31..325db1f 100644 --- a/drivers/scsi/qla4xxx/ql4_nvram.c +++ b/drivers/scsi/qla4xxx/ql4_nvram.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla4xxx/ql4_nvram.h b/drivers/scsi/qla4xxx/ql4_nvram.h index 945cc32..dba0514 100644 --- a/drivers/scsi/qla4xxx/ql4_nvram.h +++ b/drivers/scsi/qla4xxx/ql4_nvram.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ diff --git a/drivers/scsi/qla4xxx/ql4_nx.c b/drivers/scsi/qla4xxx/ql4_nx.c index 807bf76..499a92d 100644 --- a/drivers/scsi/qla4xxx/ql4_nx.c +++ b/drivers/scsi/qla4xxx/ql4_nx.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -10,6 +10,7 @@ #include <linux/ratelimit.h> #include "ql4_def.h" #include "ql4_glbl.h" +#include "ql4_inline.h" #include <asm-generic/io-64-nonatomic-lo-hi.h> @@ -27,7 +28,7 @@ #define CRB_BLK(off) ((off >> 20) & 0x3f) #define CRB_SUBBLK(off) ((off >> 16) & 0xf) #define CRB_WINDOW_2M (0x130060) -#define CRB_HI(off) ((qla4_8xxx_crb_hub_agt[CRB_BLK(off)] << 20) | \ +#define CRB_HI(off) ((qla4_82xx_crb_hub_agt[CRB_BLK(off)] << 20) | \ ((off) & 0xf0000)) #define QLA82XX_PCI_CAMQM_2M_END (0x04800800UL) #define QLA82XX_PCI_CAMQM_2M_BASE (0x000ff800UL) @@ -51,7 +52,7 @@ static int qla4_8xxx_crb_table_initialized; (crb_addr_xform[QLA82XX_HW_PX_MAP_CRB_##name] = \ QLA82XX_HW_CRB_HUB_AGT_ADR_##name << 20) static void -qla4_8xxx_crb_addr_transform_setup(void) +qla4_82xx_crb_addr_transform_setup(void) { qla4_8xxx_crb_addr_transform(XDMA); qla4_8xxx_crb_addr_transform(TIMR); @@ -268,7 +269,7 @@ static struct crb_128M_2M_block_map crb_128M_2M_map[64] = { /* * top 12 bits of crb internal address (hub, agent) */ -static unsigned qla4_8xxx_crb_hub_agt[64] = { +static unsigned qla4_82xx_crb_hub_agt[64] = { 0, QLA82XX_HW_CRB_HUB_AGT_ADR_PS, QLA82XX_HW_CRB_HUB_AGT_ADR_MN, @@ -353,7 +354,7 @@ static char *qdev_state[] = { * side effect: lock crb window */ static void -qla4_8xxx_pci_set_crbwindow_2M(struct scsi_qla_host *ha, ulong *off) +qla4_82xx_pci_set_crbwindow_2M(struct scsi_qla_host *ha, ulong *off) { u32 win_read; @@ -373,96 +374,115 @@ qla4_8xxx_pci_set_crbwindow_2M(struct scsi_qla_host *ha, ulong *off) } void -qla4_8xxx_wr_32(struct scsi_qla_host *ha, ulong off, u32 data) +qla4_82xx_wr_32(struct scsi_qla_host *ha, ulong off, u32 data) { unsigned long flags = 0; int rv; - rv = qla4_8xxx_pci_get_crb_addr_2M(ha, &off); + rv = qla4_82xx_pci_get_crb_addr_2M(ha, &off); BUG_ON(rv == -1); if (rv == 1) { write_lock_irqsave(&ha->hw_lock, flags); - qla4_8xxx_crb_win_lock(ha); - qla4_8xxx_pci_set_crbwindow_2M(ha, &off); + qla4_82xx_crb_win_lock(ha); + qla4_82xx_pci_set_crbwindow_2M(ha, &off); } writel(data, (void __iomem *)off); if (rv == 1) { - qla4_8xxx_crb_win_unlock(ha); + qla4_82xx_crb_win_unlock(ha); write_unlock_irqrestore(&ha->hw_lock, flags); } } -int -qla4_8xxx_rd_32(struct scsi_qla_host *ha, ulong off) +uint32_t qla4_82xx_rd_32(struct scsi_qla_host *ha, ulong off) { unsigned long flags = 0; int rv; u32 data; - rv = qla4_8xxx_pci_get_crb_addr_2M(ha, &off); + rv = qla4_82xx_pci_get_crb_addr_2M(ha, &off); BUG_ON(rv == -1); if (rv == 1) { write_lock_irqsave(&ha->hw_lock, flags); - qla4_8xxx_crb_win_lock(ha); - qla4_8xxx_pci_set_crbwindow_2M(ha, &off); + qla4_82xx_crb_win_lock(ha); + qla4_82xx_pci_set_crbwindow_2M(ha, &off); } data = readl((void __iomem *)off); if (rv == 1) { - qla4_8xxx_crb_win_unlock(ha); + qla4_82xx_crb_win_unlock(ha); write_unlock_irqrestore(&ha->hw_lock, flags); } return data; } /* Minidump related functions */ -static int qla4_8xxx_md_rw_32(struct scsi_qla_host *ha, uint32_t off, - u32 data, uint8_t flag) +int qla4_82xx_md_rd_32(struct scsi_qla_host *ha, uint32_t off, uint32_t *data) { - uint32_t win_read, off_value, rval = QLA_SUCCESS; + uint32_t win_read, off_value; + int rval = QLA_SUCCESS; off_value = off & 0xFFFF0000; writel(off_value, (void __iomem *)(CRB_WINDOW_2M + ha->nx_pcibase)); - /* Read back value to make sure write has gone through before trying + /* + * Read back value to make sure write has gone through before trying * to use it. */ win_read = readl((void __iomem *)(CRB_WINDOW_2M + ha->nx_pcibase)); if (win_read != off_value) { DEBUG2(ql4_printk(KERN_INFO, ha, "%s: Written (0x%x) != Read (0x%x), off=0x%x\n", - __func__, off_value, win_read, off)); - return QLA_ERROR; + __func__, off_value, win_read, off)); + rval = QLA_ERROR; + } else { + off_value = off & 0x0000FFFF; + *data = readl((void __iomem *)(off_value + CRB_INDIRECT_2M + + ha->nx_pcibase)); } + return rval; +} - off_value = off & 0x0000FFFF; +int qla4_82xx_md_wr_32(struct scsi_qla_host *ha, uint32_t off, uint32_t data) +{ + uint32_t win_read, off_value; + int rval = QLA_SUCCESS; + + off_value = off & 0xFFFF0000; + writel(off_value, (void __iomem *)(CRB_WINDOW_2M + ha->nx_pcibase)); - if (flag) + /* Read back value to make sure write has gone through before trying + * to use it. + */ + win_read = readl((void __iomem *)(CRB_WINDOW_2M + ha->nx_pcibase)); + if (win_read != off_value) { + DEBUG2(ql4_printk(KERN_INFO, ha, + "%s: Written (0x%x) != Read (0x%x), off=0x%x\n", + __func__, off_value, win_read, off)); + rval = QLA_ERROR; + } else { + off_value = off & 0x0000FFFF; writel(data, (void __iomem *)(off_value + CRB_INDIRECT_2M + ha->nx_pcibase)); - else - rval = readl((void __iomem *)(off_value + CRB_INDIRECT_2M + - ha->nx_pcibase)); - + } return rval; } #define CRB_WIN_LOCK_TIMEOUT 100000000 -int qla4_8xxx_crb_win_lock(struct scsi_qla_host *ha) +int qla4_82xx_crb_win_lock(struct scsi_qla_host *ha) { int i; int done = 0, timeout = 0; while (!done) { /* acquire semaphore3 from PCI HW block */ - done = qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM7_LOCK)); + done = qla4_82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM7_LOCK)); if (done == 1) break; if (timeout >= CRB_WIN_LOCK_TIMEOUT) @@ -478,32 +498,32 @@ int qla4_8xxx_crb_win_lock(struct scsi_qla_host *ha) cpu_relax(); /*This a nop instr on i386*/ } } - qla4_8xxx_wr_32(ha, QLA82XX_CRB_WIN_LOCK_ID, ha->func_num); + qla4_82xx_wr_32(ha, QLA82XX_CRB_WIN_LOCK_ID, ha->func_num); return 0; } -void qla4_8xxx_crb_win_unlock(struct scsi_qla_host *ha) +void qla4_82xx_crb_win_unlock(struct scsi_qla_host *ha) { - qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM7_UNLOCK)); + qla4_82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM7_UNLOCK)); } #define IDC_LOCK_TIMEOUT 100000000 /** - * qla4_8xxx_idc_lock - hw_lock + * qla4_82xx_idc_lock - hw_lock * @ha: pointer to adapter structure * * General purpose lock used to synchronize access to * CRB_DEV_STATE, CRB_DEV_REF_COUNT, etc. **/ -int qla4_8xxx_idc_lock(struct scsi_qla_host *ha) +int qla4_82xx_idc_lock(struct scsi_qla_host *ha) { int i; int done = 0, timeout = 0; while (!done) { /* acquire semaphore5 from PCI HW block */ - done = qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM5_LOCK)); + done = qla4_82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM5_LOCK)); if (done == 1) break; if (timeout >= IDC_LOCK_TIMEOUT) @@ -522,13 +542,13 @@ int qla4_8xxx_idc_lock(struct scsi_qla_host *ha) return 0; } -void qla4_8xxx_idc_unlock(struct scsi_qla_host *ha) +void qla4_82xx_idc_unlock(struct scsi_qla_host *ha) { - qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM5_UNLOCK)); + qla4_82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM5_UNLOCK)); } int -qla4_8xxx_pci_get_crb_addr_2M(struct scsi_qla_host *ha, ulong *off) +qla4_82xx_pci_get_crb_addr_2M(struct scsi_qla_host *ha, ulong *off) { struct crb_128M_2M_sub_block_map *m; @@ -562,44 +582,40 @@ qla4_8xxx_pci_get_crb_addr_2M(struct scsi_qla_host *ha, ulong *off) return 1; } -/* PCI Windowing for DDR regions. */ -#define QLA82XX_ADDR_IN_RANGE(addr, low, high) \ - (((addr) <= (high)) && ((addr) >= (low))) - /* * check memory access boundary. * used by test agent. support ddr access only for now */ static unsigned long -qla4_8xxx_pci_mem_bound_check(struct scsi_qla_host *ha, +qla4_82xx_pci_mem_bound_check(struct scsi_qla_host *ha, unsigned long long addr, int size) { - if (!QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_DDR_NET, - QLA82XX_ADDR_DDR_NET_MAX) || - !QLA82XX_ADDR_IN_RANGE(addr + size - 1, - QLA82XX_ADDR_DDR_NET, QLA82XX_ADDR_DDR_NET_MAX) || + if (!QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_DDR_NET, + QLA8XXX_ADDR_DDR_NET_MAX) || + !QLA8XXX_ADDR_IN_RANGE(addr + size - 1, + QLA8XXX_ADDR_DDR_NET, QLA8XXX_ADDR_DDR_NET_MAX) || ((size != 1) && (size != 2) && (size != 4) && (size != 8))) { return 0; } return 1; } -static int qla4_8xxx_pci_set_window_warning_count; +static int qla4_82xx_pci_set_window_warning_count; static unsigned long -qla4_8xxx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) +qla4_82xx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) { int window; u32 win_read; - if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_DDR_NET, - QLA82XX_ADDR_DDR_NET_MAX)) { + if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_DDR_NET, + QLA8XXX_ADDR_DDR_NET_MAX)) { /* DDR network side */ window = MN_WIN(addr); ha->ddr_mn_window = window; - qla4_8xxx_wr_32(ha, ha->mn_win_crb | + qla4_82xx_wr_32(ha, ha->mn_win_crb | QLA82XX_PCI_CRBSPACE, window); - win_read = qla4_8xxx_rd_32(ha, ha->mn_win_crb | + win_read = qla4_82xx_rd_32(ha, ha->mn_win_crb | QLA82XX_PCI_CRBSPACE); if ((win_read << 17) != window) { ql4_printk(KERN_WARNING, ha, @@ -607,8 +623,8 @@ qla4_8xxx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) __func__, window, win_read); } addr = GET_MEM_OFFS_2M(addr) + QLA82XX_PCI_DDR_NET; - } else if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_OCM0, - QLA82XX_ADDR_OCM0_MAX)) { + } else if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_OCM0, + QLA8XXX_ADDR_OCM0_MAX)) { unsigned int temp1; /* if bits 19:18&17:11 are on */ if ((addr & 0x00ff800) == 0xff800) { @@ -618,9 +634,9 @@ qla4_8xxx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) window = OCM_WIN(addr); ha->ddr_mn_window = window; - qla4_8xxx_wr_32(ha, ha->mn_win_crb | + qla4_82xx_wr_32(ha, ha->mn_win_crb | QLA82XX_PCI_CRBSPACE, window); - win_read = qla4_8xxx_rd_32(ha, ha->mn_win_crb | + win_read = qla4_82xx_rd_32(ha, ha->mn_win_crb | QLA82XX_PCI_CRBSPACE); temp1 = ((window & 0x1FF) << 7) | ((window & 0x0FFFE0000) >> 17); @@ -630,14 +646,14 @@ qla4_8xxx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) } addr = GET_MEM_OFFS_2M(addr) + QLA82XX_PCI_OCM0_2M; - } else if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_QDR_NET, + } else if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_QDR_NET, QLA82XX_P3_ADDR_QDR_NET_MAX)) { /* QDR network side */ window = MS_WIN(addr); ha->qdr_sn_window = window; - qla4_8xxx_wr_32(ha, ha->ms_win_crb | + qla4_82xx_wr_32(ha, ha->ms_win_crb | QLA82XX_PCI_CRBSPACE, window); - win_read = qla4_8xxx_rd_32(ha, + win_read = qla4_82xx_rd_32(ha, ha->ms_win_crb | QLA82XX_PCI_CRBSPACE); if (win_read != window) { printk("%s: Written MSwin (0x%x) != Read " @@ -650,8 +666,8 @@ qla4_8xxx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) * peg gdb frequently accesses memory that doesn't exist, * this limits the chit chat so debugging isn't slowed down. */ - if ((qla4_8xxx_pci_set_window_warning_count++ < 8) || - (qla4_8xxx_pci_set_window_warning_count%64 == 0)) { + if ((qla4_82xx_pci_set_window_warning_count++ < 8) || + (qla4_82xx_pci_set_window_warning_count%64 == 0)) { printk("%s: Warning:%s Unknown address range!\n", __func__, DRIVER_NAME); } @@ -661,7 +677,7 @@ qla4_8xxx_pci_set_window(struct scsi_qla_host *ha, unsigned long long addr) } /* check if address is in the same windows as the previous access */ -static int qla4_8xxx_pci_is_same_window(struct scsi_qla_host *ha, +static int qla4_82xx_pci_is_same_window(struct scsi_qla_host *ha, unsigned long long addr) { int window; @@ -669,20 +685,20 @@ static int qla4_8xxx_pci_is_same_window(struct scsi_qla_host *ha, qdr_max = QLA82XX_P3_ADDR_QDR_NET_MAX; - if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_DDR_NET, - QLA82XX_ADDR_DDR_NET_MAX)) { + if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_DDR_NET, + QLA8XXX_ADDR_DDR_NET_MAX)) { /* DDR network side */ BUG(); /* MN access can not come here */ - } else if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_OCM0, - QLA82XX_ADDR_OCM0_MAX)) { + } else if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_OCM0, + QLA8XXX_ADDR_OCM0_MAX)) { return 1; - } else if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_OCM1, - QLA82XX_ADDR_OCM1_MAX)) { + } else if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_OCM1, + QLA8XXX_ADDR_OCM1_MAX)) { return 1; - } else if (QLA82XX_ADDR_IN_RANGE(addr, QLA82XX_ADDR_QDR_NET, + } else if (QLA8XXX_ADDR_IN_RANGE(addr, QLA8XXX_ADDR_QDR_NET, qdr_max)) { /* QDR network side */ - window = ((addr - QLA82XX_ADDR_QDR_NET) >> 22) & 0x3f; + window = ((addr - QLA8XXX_ADDR_QDR_NET) >> 22) & 0x3f; if (ha->qdr_sn_window == window) return 1; } @@ -690,7 +706,7 @@ static int qla4_8xxx_pci_is_same_window(struct scsi_qla_host *ha, return 0; } -static int qla4_8xxx_pci_mem_read_direct(struct scsi_qla_host *ha, +static int qla4_82xx_pci_mem_read_direct(struct scsi_qla_host *ha, u64 off, void *data, int size) { unsigned long flags; @@ -707,9 +723,9 @@ static int qla4_8xxx_pci_mem_read_direct(struct scsi_qla_host *ha, * If attempting to access unknown address or straddle hw windows, * do not access. */ - start = qla4_8xxx_pci_set_window(ha, off); + start = qla4_82xx_pci_set_window(ha, off); if ((start == -1UL) || - (qla4_8xxx_pci_is_same_window(ha, off + size - 1) == 0)) { + (qla4_82xx_pci_is_same_window(ha, off + size - 1) == 0)) { write_unlock_irqrestore(&ha->hw_lock, flags); printk(KERN_ERR"%s out of bound pci memory access. " "offset is 0x%llx\n", DRIVER_NAME, off); @@ -763,7 +779,7 @@ static int qla4_8xxx_pci_mem_read_direct(struct scsi_qla_host *ha, } static int -qla4_8xxx_pci_mem_write_direct(struct scsi_qla_host *ha, u64 off, +qla4_82xx_pci_mem_write_direct(struct scsi_qla_host *ha, u64 off, void *data, int size) { unsigned long flags; @@ -780,9 +796,9 @@ qla4_8xxx_pci_mem_write_direct(struct scsi_qla_host *ha, u64 off, * If attempting to access unknown address or straddle hw windows, * do not access. */ - start = qla4_8xxx_pci_set_window(ha, off); + start = qla4_82xx_pci_set_window(ha, off); if ((start == -1UL) || - (qla4_8xxx_pci_is_same_window(ha, off + size - 1) == 0)) { + (qla4_82xx_pci_is_same_window(ha, off + size - 1) == 0)) { write_unlock_irqrestore(&ha->hw_lock, flags); printk(KERN_ERR"%s out of bound pci memory access. " "offset is 0x%llx\n", DRIVER_NAME, off); @@ -835,13 +851,13 @@ qla4_8xxx_pci_mem_write_direct(struct scsi_qla_host *ha, u64 off, #define MTU_FUDGE_FACTOR 100 static unsigned long -qla4_8xxx_decode_crb_addr(unsigned long addr) +qla4_82xx_decode_crb_addr(unsigned long addr) { int i; unsigned long base_addr, offset, pci_base; if (!qla4_8xxx_crb_table_initialized) - qla4_8xxx_crb_addr_transform_setup(); + qla4_82xx_crb_addr_transform_setup(); pci_base = ADDR_ERROR; base_addr = addr & 0xfff00000; @@ -860,10 +876,10 @@ qla4_8xxx_decode_crb_addr(unsigned long addr) } static long rom_max_timeout = 100; -static long qla4_8xxx_rom_lock_timeout = 100; +static long qla4_82xx_rom_lock_timeout = 100; static int -qla4_8xxx_rom_lock(struct scsi_qla_host *ha) +qla4_82xx_rom_lock(struct scsi_qla_host *ha) { int i; int done = 0, timeout = 0; @@ -871,10 +887,10 @@ qla4_8xxx_rom_lock(struct scsi_qla_host *ha) while (!done) { /* acquire semaphore2 from PCI HW block */ - done = qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM2_LOCK)); + done = qla4_82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM2_LOCK)); if (done == 1) break; - if (timeout >= qla4_8xxx_rom_lock_timeout) + if (timeout >= qla4_82xx_rom_lock_timeout) return -1; timeout++; @@ -887,24 +903,24 @@ qla4_8xxx_rom_lock(struct scsi_qla_host *ha) cpu_relax(); /*This a nop instr on i386*/ } } - qla4_8xxx_wr_32(ha, QLA82XX_ROM_LOCK_ID, ROM_LOCK_DRIVER); + qla4_82xx_wr_32(ha, QLA82XX_ROM_LOCK_ID, ROM_LOCK_DRIVER); return 0; } static void -qla4_8xxx_rom_unlock(struct scsi_qla_host *ha) +qla4_82xx_rom_unlock(struct scsi_qla_host *ha) { - qla4_8xxx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM2_UNLOCK)); + qla4_82xx_rd_32(ha, QLA82XX_PCIE_REG(PCIE_SEM2_UNLOCK)); } static int -qla4_8xxx_wait_rom_done(struct scsi_qla_host *ha) +qla4_82xx_wait_rom_done(struct scsi_qla_host *ha) { long timeout = 0; long done = 0 ; while (done == 0) { - done = qla4_8xxx_rd_32(ha, QLA82XX_ROMUSB_GLB_STATUS); + done = qla4_82xx_rd_32(ha, QLA82XX_ROMUSB_GLB_STATUS); done &= 2; timeout++; if (timeout >= rom_max_timeout) { @@ -917,40 +933,41 @@ qla4_8xxx_wait_rom_done(struct scsi_qla_host *ha) } static int -qla4_8xxx_do_rom_fast_read(struct scsi_qla_host *ha, int addr, int *valp) +qla4_82xx_do_rom_fast_read(struct scsi_qla_host *ha, int addr, int *valp) { - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_ROM_ADDRESS, addr); - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_ROM_DUMMY_BYTE_CNT, 0); - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_ROM_ABYTE_CNT, 3); - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_ROM_INSTR_OPCODE, 0xb); - if (qla4_8xxx_wait_rom_done(ha)) { + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_ROM_ADDRESS, addr); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_ROM_DUMMY_BYTE_CNT, 0); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_ROM_ABYTE_CNT, 3); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_ROM_INSTR_OPCODE, 0xb); + if (qla4_82xx_wait_rom_done(ha)) { printk("%s: Error waiting for rom done\n", DRIVER_NAME); return -1; } /* reset abyte_cnt and dummy_byte_cnt */ - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_ROM_DUMMY_BYTE_CNT, 0); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_ROM_DUMMY_BYTE_CNT, 0); udelay(10); - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_ROM_ABYTE_CNT, 0); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_ROM_ABYTE_CNT, 0); - *valp = qla4_8xxx_rd_32(ha, QLA82XX_ROMUSB_ROM_RDATA); + *valp = qla4_82xx_rd_32(ha, QLA82XX_ROMUSB_ROM_RDATA); return 0; } static int -qla4_8xxx_rom_fast_read(struct scsi_qla_host *ha, int addr, int *valp) +qla4_82xx_rom_fast_read(struct scsi_qla_host *ha, int addr, int *valp) { int ret, loops = 0; - while ((qla4_8xxx_rom_lock(ha) != 0) && (loops < 50000)) { + while ((qla4_82xx_rom_lock(ha) != 0) && (loops < 50000)) { udelay(100); loops++; } if (loops >= 50000) { - printk("%s: qla4_8xxx_rom_lock failed\n", DRIVER_NAME); + ql4_printk(KERN_WARNING, ha, "%s: qla4_82xx_rom_lock failed\n", + DRIVER_NAME); return -1; } - ret = qla4_8xxx_do_rom_fast_read(ha, addr, valp); - qla4_8xxx_rom_unlock(ha); + ret = qla4_82xx_do_rom_fast_read(ha, addr, valp); + qla4_82xx_rom_unlock(ha); return ret; } @@ -959,7 +976,7 @@ qla4_8xxx_rom_fast_read(struct scsi_qla_host *ha, int addr, int *valp) * to put the ISP into operational state **/ static int -qla4_8xxx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) +qla4_82xx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) { int addr, val; int i ; @@ -973,68 +990,68 @@ qla4_8xxx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) }; /* Halt all the indiviual PEGs and other blocks of the ISP */ - qla4_8xxx_rom_lock(ha); + qla4_82xx_rom_lock(ha); /* disable all I2Q */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_I2Q + 0x10, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_I2Q + 0x14, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_I2Q + 0x18, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_I2Q + 0x1c, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_I2Q + 0x20, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_I2Q + 0x24, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_I2Q + 0x10, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_I2Q + 0x14, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_I2Q + 0x18, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_I2Q + 0x1c, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_I2Q + 0x20, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_I2Q + 0x24, 0x0); /* disable all niu interrupts */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x40, 0xff); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x40, 0xff); /* disable xge rx/tx */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x70000, 0x00); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x70000, 0x00); /* disable xg1 rx/tx */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x80000, 0x00); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x80000, 0x00); /* disable sideband mac */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x90000, 0x00); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x90000, 0x00); /* disable ap0 mac */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0xa0000, 0x00); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0xa0000, 0x00); /* disable ap1 mac */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0xb0000, 0x00); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0xb0000, 0x00); /* halt sre */ - val = qla4_8xxx_rd_32(ha, QLA82XX_CRB_SRE + 0x1000); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_SRE + 0x1000, val & (~(0x1))); + val = qla4_82xx_rd_32(ha, QLA82XX_CRB_SRE + 0x1000); + qla4_82xx_wr_32(ha, QLA82XX_CRB_SRE + 0x1000, val & (~(0x1))); /* halt epg */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_EPG + 0x1300, 0x1); + qla4_82xx_wr_32(ha, QLA82XX_CRB_EPG + 0x1300, 0x1); /* halt timers */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_TIMER + 0x0, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_TIMER + 0x8, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_TIMER + 0x10, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_TIMER + 0x18, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_TIMER + 0x100, 0x0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_TIMER + 0x200, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_TIMER + 0x0, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_TIMER + 0x8, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_TIMER + 0x10, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_TIMER + 0x18, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_TIMER + 0x100, 0x0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_TIMER + 0x200, 0x0); /* halt pegs */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_0 + 0x3c, 1); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_1 + 0x3c, 1); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_2 + 0x3c, 1); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_3 + 0x3c, 1); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_4 + 0x3c, 1); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_0 + 0x3c, 1); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_1 + 0x3c, 1); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_2 + 0x3c, 1); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_3 + 0x3c, 1); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_4 + 0x3c, 1); msleep(5); /* big hammer */ if (test_bit(DPC_RESET_HA, &ha->dpc_flags)) /* don't reset CAM block on reset */ - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, 0xfeffffff); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, 0xfeffffff); else - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, 0xffffffff); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, 0xffffffff); - qla4_8xxx_rom_unlock(ha); + qla4_82xx_rom_unlock(ha); /* Read the signature value from the flash. * Offset 0: Contain signature (0xcafecafe) * Offset 4: Offset and number of addr/value pairs * that present in CRB initialize sequence */ - if (qla4_8xxx_rom_fast_read(ha, 0, &n) != 0 || n != 0xcafecafeUL || - qla4_8xxx_rom_fast_read(ha, 4, &n) != 0) { + if (qla4_82xx_rom_fast_read(ha, 0, &n) != 0 || n != 0xcafecafeUL || + qla4_82xx_rom_fast_read(ha, 4, &n) != 0) { ql4_printk(KERN_WARNING, ha, "[ERROR] Reading crb_init area: n: %08x\n", n); return -1; @@ -1065,8 +1082,8 @@ qla4_8xxx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) } for (i = 0; i < n; i++) { - if (qla4_8xxx_rom_fast_read(ha, 8*i + 4*offset, &val) != 0 || - qla4_8xxx_rom_fast_read(ha, 8*i + 4*offset + 4, &addr) != + if (qla4_82xx_rom_fast_read(ha, 8*i + 4*offset, &val) != 0 || + qla4_82xx_rom_fast_read(ha, 8*i + 4*offset + 4, &addr) != 0) { kfree(buf); return -1; @@ -1080,7 +1097,7 @@ qla4_8xxx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) /* Translate internal CRB initialization * address to PCI bus address */ - off = qla4_8xxx_decode_crb_addr((unsigned long)buf[i].addr) + + off = qla4_82xx_decode_crb_addr((unsigned long)buf[i].addr) + QLA82XX_PCI_CRBSPACE; /* Not all CRB addr/value pair to be written, * some of them are skipped @@ -1125,7 +1142,7 @@ qla4_8xxx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) continue; } - qla4_8xxx_wr_32(ha, off, buf[i].data); + qla4_82xx_wr_32(ha, off, buf[i].data); /* ISP requires much bigger delay to settle down, * else crb_window returns 0xffffffff @@ -1142,25 +1159,25 @@ qla4_8xxx_pinit_from_rom(struct scsi_qla_host *ha, int verbose) kfree(buf); /* Resetting the data and instruction cache */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_D+0xec, 0x1e); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_D+0x4c, 8); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_I+0x4c, 8); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_D+0xec, 0x1e); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_D+0x4c, 8); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_I+0x4c, 8); /* Clear all protocol processing engines */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_0+0x8, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_0+0xc, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_1+0x8, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_1+0xc, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_2+0x8, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_2+0xc, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_3+0x8, 0); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_3+0xc, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_0+0x8, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_0+0xc, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_1+0x8, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_1+0xc, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_2+0x8, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_2+0xc, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_3+0x8, 0); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_3+0xc, 0); return 0; } static int -qla4_8xxx_load_from_flash(struct scsi_qla_host *ha, uint32_t image_start) +qla4_82xx_load_from_flash(struct scsi_qla_host *ha, uint32_t image_start) { int i, rval = 0; long size = 0; @@ -1175,14 +1192,14 @@ qla4_8xxx_load_from_flash(struct scsi_qla_host *ha, uint32_t image_start) ha->host_no, __func__, flashaddr, image_start)); for (i = 0; i < size; i++) { - if ((qla4_8xxx_rom_fast_read(ha, flashaddr, (int *)&low)) || - (qla4_8xxx_rom_fast_read(ha, flashaddr + 4, + if ((qla4_82xx_rom_fast_read(ha, flashaddr, (int *)&low)) || + (qla4_82xx_rom_fast_read(ha, flashaddr + 4, (int *)&high))) { rval = -1; goto exit_load_from_flash; } data = ((u64)high << 32) | low ; - rval = qla4_8xxx_pci_mem_write_2M(ha, memaddr, &data, 8); + rval = qla4_82xx_pci_mem_write_2M(ha, memaddr, &data, 8); if (rval) goto exit_load_from_flash; @@ -1197,20 +1214,20 @@ qla4_8xxx_load_from_flash(struct scsi_qla_host *ha, uint32_t image_start) udelay(100); read_lock(&ha->hw_lock); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_PEG_NET_0 + 0x18, 0x1020); - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, 0x80001e); + qla4_82xx_wr_32(ha, QLA82XX_CRB_PEG_NET_0 + 0x18, 0x1020); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, 0x80001e); read_unlock(&ha->hw_lock); exit_load_from_flash: return rval; } -static int qla4_8xxx_load_fw(struct scsi_qla_host *ha, uint32_t image_start) +static int qla4_82xx_load_fw(struct scsi_qla_host *ha, uint32_t image_start) { u32 rst; - qla4_8xxx_wr_32(ha, CRB_CMDPEG_STATE, 0); - if (qla4_8xxx_pinit_from_rom(ha, 0) != QLA_SUCCESS) { + qla4_82xx_wr_32(ha, CRB_CMDPEG_STATE, 0); + if (qla4_82xx_pinit_from_rom(ha, 0) != QLA_SUCCESS) { printk(KERN_WARNING "%s: Error during CRB Initialization\n", __func__); return QLA_ERROR; @@ -1223,12 +1240,12 @@ static int qla4_8xxx_load_fw(struct scsi_qla_host *ha, uint32_t image_start) * To get around this, QM is brought out of reset. */ - rst = qla4_8xxx_rd_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET); + rst = qla4_82xx_rd_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET); /* unreset qm */ rst &= ~(1 << 28); - qla4_8xxx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, rst); + qla4_82xx_wr_32(ha, QLA82XX_ROMUSB_GLB_SW_RESET, rst); - if (qla4_8xxx_load_from_flash(ha, image_start)) { + if (qla4_82xx_load_from_flash(ha, image_start)) { printk("%s: Error trying to load fw from flash!\n", __func__); return QLA_ERROR; } @@ -1237,7 +1254,7 @@ static int qla4_8xxx_load_fw(struct scsi_qla_host *ha, uint32_t image_start) } int -qla4_8xxx_pci_mem_read_2M(struct scsi_qla_host *ha, +qla4_82xx_pci_mem_read_2M(struct scsi_qla_host *ha, u64 off, void *data, int size) { int i, j = 0, k, start, end, loop, sz[2], off0[2]; @@ -1249,12 +1266,12 @@ qla4_8xxx_pci_mem_read_2M(struct scsi_qla_host *ha, * If not MN, go check for MS or invalid. */ - if (off >= QLA82XX_ADDR_QDR_NET && off <= QLA82XX_P3_ADDR_QDR_NET_MAX) + if (off >= QLA8XXX_ADDR_QDR_NET && off <= QLA82XX_P3_ADDR_QDR_NET_MAX) mem_crb = QLA82XX_CRB_QDR_NET; else { mem_crb = QLA82XX_CRB_DDR_NET; - if (qla4_8xxx_pci_mem_bound_check(ha, off, size) == 0) - return qla4_8xxx_pci_mem_read_direct(ha, + if (qla4_82xx_pci_mem_bound_check(ha, off, size) == 0) + return qla4_82xx_pci_mem_read_direct(ha, off, data, size); } @@ -1270,16 +1287,16 @@ qla4_8xxx_pci_mem_read_2M(struct scsi_qla_host *ha, for (i = 0; i < loop; i++) { temp = off8 + (i << shift_amount); - qla4_8xxx_wr_32(ha, mem_crb + MIU_TEST_AGT_ADDR_LO, temp); + qla4_82xx_wr_32(ha, mem_crb + MIU_TEST_AGT_ADDR_LO, temp); temp = 0; - qla4_8xxx_wr_32(ha, mem_crb + MIU_TEST_AGT_ADDR_HI, temp); + qla4_82xx_wr_32(ha, mem_crb + MIU_TEST_AGT_ADDR_HI, temp); temp = MIU_TA_CTL_ENABLE; - qla4_8xxx_wr_32(ha, mem_crb + MIU_TEST_AGT_CTRL, temp); - temp = MIU_TA_CTL_START | MIU_TA_CTL_ENABLE; - qla4_8xxx_wr_32(ha, mem_crb + MIU_TEST_AGT_CTRL, temp); + qla4_82xx_wr_32(ha, mem_crb + MIU_TEST_AGT_CTRL, temp); + temp = MIU_TA_CTL_START_ENABLE; + qla4_82xx_wr_32(ha, mem_crb + MIU_TEST_AGT_CTRL, temp); for (j = 0; j < MAX_CTL_CHECK; j++) { - temp = qla4_8xxx_rd_32(ha, mem_crb + MIU_TEST_AGT_CTRL); + temp = qla4_82xx_rd_32(ha, mem_crb + MIU_TEST_AGT_CTRL); if ((temp & MIU_TA_CTL_BUSY) == 0) break; } @@ -1294,7 +1311,7 @@ qla4_8xxx_pci_mem_read_2M(struct scsi_qla_host *ha, start = off0[i] >> 2; end = (off0[i] + sz[i] - 1) >> 2; for (k = start; k <= end; k++) { - temp = qla4_8xxx_rd_32(ha, + temp = qla4_82xx_rd_32(ha, mem_crb + MIU_TEST_AGT_RDDATA(k)); word[i] |= ((uint64_t)temp << (32 * (k & 1))); } @@ -1328,7 +1345,7 @@ qla4_8xxx_pci_mem_read_2M(struct scsi_qla_host *ha, } int -qla4_8xxx_pci_mem_write_2M(struct scsi_qla_host *ha, +qla4_82xx_pci_mem_write_2M(struct scsi_qla_host *ha, u64 off, void *data, int size) { int i, j, ret = 0, loop, sz[2], off0; @@ -1339,12 +1356,12 @@ qla4_8xxx_pci_mem_write_2M(struct scsi_qla_host *ha, /* * If not MN, go check for MS or invalid. */ - if (off >= QLA82XX_ADDR_QDR_NET && off <= QLA82XX_P3_ADDR_QDR_NET_MAX) + if (off >= QLA8XXX_ADDR_QDR_NET && off <= QLA82XX_P3_ADDR_QDR_NET_MAX) mem_crb = QLA82XX_CRB_QDR_NET; else { mem_crb = QLA82XX_CRB_DDR_NET; - if (qla4_8xxx_pci_mem_bound_check(ha, off, size) == 0) - return qla4_8xxx_pci_mem_write_direct(ha, + if (qla4_82xx_pci_mem_bound_check(ha, off, size) == 0) + return qla4_82xx_pci_mem_write_direct(ha, off, data, size); } @@ -1359,7 +1376,7 @@ qla4_8xxx_pci_mem_write_2M(struct scsi_qla_host *ha, startword = (off & 0xf)/8; for (i = 0; i < loop; i++) { - if (qla4_8xxx_pci_mem_read_2M(ha, off8 + + if (qla4_82xx_pci_mem_read_2M(ha, off8 + (i << shift_amount), &word[i * scale], 8)) return -1; } @@ -1395,27 +1412,27 @@ qla4_8xxx_pci_mem_write_2M(struct scsi_qla_host *ha, for (i = 0; i < loop; i++) { temp = off8 + (i << shift_amount); - qla4_8xxx_wr_32(ha, mem_crb+MIU_TEST_AGT_ADDR_LO, temp); + qla4_82xx_wr_32(ha, mem_crb+MIU_TEST_AGT_ADDR_LO, temp); temp = 0; - qla4_8xxx_wr_32(ha, mem_crb+MIU_TEST_AGT_ADDR_HI, temp); + qla4_82xx_wr_32(ha, mem_crb+MIU_TEST_AGT_ADDR_HI, temp); temp = word[i * scale] & 0xffffffff; - qla4_8xxx_wr_32(ha, mem_crb+MIU_TEST_AGT_WRDATA_LO, temp); + qla4_82xx_wr_32(ha, mem_crb+MIU_TEST_AGT_WRDATA_LO, temp); temp = (word[i * scale] >> 32) & 0xffffffff; - qla4_8xxx_wr_32(ha, mem_crb+MIU_TEST_AGT_WRDATA_HI, temp); + qla4_82xx_wr_32(ha, mem_crb+MIU_TEST_AGT_WRDATA_HI, temp); temp = word[i*scale + 1] & 0xffffffff; - qla4_8xxx_wr_32(ha, mem_crb + MIU_TEST_AGT_WRDATA_UPPER_LO, + qla4_82xx_wr_32(ha, mem_crb + MIU_TEST_AGT_WRDATA_UPPER_LO, temp); temp = (word[i*scale + 1] >> 32) & 0xffffffff; - qla4_8xxx_wr_32(ha, mem_crb + MIU_TEST_AGT_WRDATA_UPPER_HI, + qla4_82xx_wr_32(ha, mem_crb + MIU_TEST_AGT_WRDATA_UPPER_HI, temp); - temp = MIU_TA_CTL_ENABLE | MIU_TA_CTL_WRITE; - qla4_8xxx_wr_32(ha, mem_crb+MIU_TEST_AGT_CTRL, temp); - temp = MIU_TA_CTL_START | MIU_TA_CTL_ENABLE | MIU_TA_CTL_WRITE; - qla4_8xxx_wr_32(ha, mem_crb+MIU_TEST_AGT_CTRL, temp); + temp = MIU_TA_CTL_WRITE_ENABLE; + qla4_82xx_wr_32(ha, mem_crb+MIU_TEST_AGT_CTRL, temp); + temp = MIU_TA_CTL_WRITE_START; + qla4_82xx_wr_32(ha, mem_crb+MIU_TEST_AGT_CTRL, temp); for (j = 0; j < MAX_CTL_CHECK; j++) { - temp = qla4_8xxx_rd_32(ha, mem_crb + MIU_TEST_AGT_CTRL); + temp = qla4_82xx_rd_32(ha, mem_crb + MIU_TEST_AGT_CTRL); if ((temp & MIU_TA_CTL_BUSY) == 0) break; } @@ -1433,14 +1450,14 @@ qla4_8xxx_pci_mem_write_2M(struct scsi_qla_host *ha, return ret; } -static int qla4_8xxx_cmdpeg_ready(struct scsi_qla_host *ha, int pegtune_val) +static int qla4_82xx_cmdpeg_ready(struct scsi_qla_host *ha, int pegtune_val) { u32 val = 0; int retries = 60; if (!pegtune_val) { do { - val = qla4_8xxx_rd_32(ha, CRB_CMDPEG_STATE); + val = qla4_82xx_rd_32(ha, CRB_CMDPEG_STATE); if ((val == PHAN_INITIALIZE_COMPLETE) || (val == PHAN_INITIALIZE_ACK)) return 0; @@ -1450,7 +1467,7 @@ static int qla4_8xxx_cmdpeg_ready(struct scsi_qla_host *ha, int pegtune_val) } while (--retries); if (!retries) { - pegtune_val = qla4_8xxx_rd_32(ha, + pegtune_val = qla4_82xx_rd_32(ha, QLA82XX_ROMUSB_GLB_PEGTUNE_DONE); printk(KERN_WARNING "%s: init failed, " "pegtune_val = %x\n", __func__, pegtune_val); @@ -1460,21 +1477,21 @@ static int qla4_8xxx_cmdpeg_ready(struct scsi_qla_host *ha, int pegtune_val) return 0; } -static int qla4_8xxx_rcvpeg_ready(struct scsi_qla_host *ha) +static int qla4_82xx_rcvpeg_ready(struct scsi_qla_host *ha) { uint32_t state = 0; int loops = 0; /* Window 1 call */ read_lock(&ha->hw_lock); - state = qla4_8xxx_rd_32(ha, CRB_RCVPEG_STATE); + state = qla4_82xx_rd_32(ha, CRB_RCVPEG_STATE); read_unlock(&ha->hw_lock); while ((state != PHAN_PEG_RCV_INITIALIZED) && (loops < 30000)) { udelay(100); /* Window 1 call */ read_lock(&ha->hw_lock); - state = qla4_8xxx_rd_32(ha, CRB_RCVPEG_STATE); + state = qla4_82xx_rd_32(ha, CRB_RCVPEG_STATE); read_unlock(&ha->hw_lock); loops++; @@ -1494,11 +1511,21 @@ qla4_8xxx_set_drv_active(struct scsi_qla_host *ha) { uint32_t drv_active; - drv_active = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); - drv_active |= (1 << (ha->func_num * 4)); + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + + /* + * For ISP8324, drv_active register has 1 bit per function, + * shift 1 by func_num to set a bit for the function. + * For ISP8022, drv_active has 4 bits per function + */ + if (is_qla8032(ha)) + drv_active |= (1 << ha->func_num); + else + drv_active |= (1 << (ha->func_num * 4)); + ql4_printk(KERN_INFO, ha, "%s(%ld): drv_active: 0x%08x\n", __func__, ha->host_no, drv_active); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_ACTIVE, drv_active); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_ACTIVE, drv_active); } void @@ -1506,50 +1533,87 @@ qla4_8xxx_clear_drv_active(struct scsi_qla_host *ha) { uint32_t drv_active; - drv_active = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); - drv_active &= ~(1 << (ha->func_num * 4)); + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + + /* + * For ISP8324, drv_active register has 1 bit per function, + * shift 1 by func_num to set a bit for the function. + * For ISP8022, drv_active has 4 bits per function + */ + if (is_qla8032(ha)) + drv_active &= ~(1 << (ha->func_num)); + else + drv_active &= ~(1 << (ha->func_num * 4)); + ql4_printk(KERN_INFO, ha, "%s(%ld): drv_active: 0x%08x\n", __func__, ha->host_no, drv_active); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_ACTIVE, drv_active); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_ACTIVE, drv_active); } -static inline int -qla4_8xxx_need_reset(struct scsi_qla_host *ha) +inline int qla4_8xxx_need_reset(struct scsi_qla_host *ha) { uint32_t drv_state, drv_active; int rval; - drv_active = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); - drv_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_STATE); - rval = drv_state & (1 << (ha->func_num * 4)); + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + drv_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_STATE); + + /* + * For ISP8324, drv_active register has 1 bit per function, + * shift 1 by func_num to set a bit for the function. + * For ISP8022, drv_active has 4 bits per function + */ + if (is_qla8032(ha)) + rval = drv_state & (1 << ha->func_num); + else + rval = drv_state & (1 << (ha->func_num * 4)); + if ((test_bit(AF_EEH_BUSY, &ha->flags)) && drv_active) rval = 1; return rval; } -static inline void -qla4_8xxx_set_rst_ready(struct scsi_qla_host *ha) +void qla4_8xxx_set_rst_ready(struct scsi_qla_host *ha) { uint32_t drv_state; - drv_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_STATE); - drv_state |= (1 << (ha->func_num * 4)); + drv_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_STATE); + + /* + * For ISP8324, drv_active register has 1 bit per function, + * shift 1 by func_num to set a bit for the function. + * For ISP8022, drv_active has 4 bits per function + */ + if (is_qla8032(ha)) + drv_state |= (1 << ha->func_num); + else + drv_state |= (1 << (ha->func_num * 4)); + ql4_printk(KERN_INFO, ha, "%s(%ld): drv_state: 0x%08x\n", __func__, ha->host_no, drv_state); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_STATE, drv_state); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_STATE, drv_state); } -static inline void -qla4_8xxx_clear_rst_ready(struct scsi_qla_host *ha) +void qla4_8xxx_clear_rst_ready(struct scsi_qla_host *ha) { uint32_t drv_state; - drv_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_STATE); - drv_state &= ~(1 << (ha->func_num * 4)); + drv_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_STATE); + + /* + * For ISP8324, drv_active register has 1 bit per function, + * shift 1 by func_num to set a bit for the function. + * For ISP8022, drv_active has 4 bits per function + */ + if (is_qla8032(ha)) + drv_state &= ~(1 << ha->func_num); + else + drv_state &= ~(1 << (ha->func_num * 4)); + ql4_printk(KERN_INFO, ha, "%s(%ld): drv_state: 0x%08x\n", __func__, ha->host_no, drv_state); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_STATE, drv_state); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_STATE, drv_state); } static inline void @@ -1557,33 +1621,43 @@ qla4_8xxx_set_qsnt_ready(struct scsi_qla_host *ha) { uint32_t qsnt_state; - qsnt_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_STATE); - qsnt_state |= (2 << (ha->func_num * 4)); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_STATE, qsnt_state); + qsnt_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_STATE); + + /* + * For ISP8324, drv_active register has 1 bit per function, + * shift 1 by func_num to set a bit for the function. + * For ISP8022, drv_active has 4 bits per function. + */ + if (is_qla8032(ha)) + qsnt_state |= (1 << ha->func_num); + else + qsnt_state |= (2 << (ha->func_num * 4)); + + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_STATE, qsnt_state); } static int -qla4_8xxx_start_firmware(struct scsi_qla_host *ha, uint32_t image_start) +qla4_82xx_start_firmware(struct scsi_qla_host *ha, uint32_t image_start) { uint16_t lnk; /* scrub dma mask expansion register */ - qla4_8xxx_wr_32(ha, CRB_DMA_SHIFT, 0x55555555); + qla4_82xx_wr_32(ha, CRB_DMA_SHIFT, 0x55555555); /* Overwrite stale initialization register values */ - qla4_8xxx_wr_32(ha, CRB_CMDPEG_STATE, 0); - qla4_8xxx_wr_32(ha, CRB_RCVPEG_STATE, 0); - qla4_8xxx_wr_32(ha, QLA82XX_PEG_HALT_STATUS1, 0); - qla4_8xxx_wr_32(ha, QLA82XX_PEG_HALT_STATUS2, 0); + qla4_82xx_wr_32(ha, CRB_CMDPEG_STATE, 0); + qla4_82xx_wr_32(ha, CRB_RCVPEG_STATE, 0); + qla4_82xx_wr_32(ha, QLA82XX_PEG_HALT_STATUS1, 0); + qla4_82xx_wr_32(ha, QLA82XX_PEG_HALT_STATUS2, 0); - if (qla4_8xxx_load_fw(ha, image_start) != QLA_SUCCESS) { + if (qla4_82xx_load_fw(ha, image_start) != QLA_SUCCESS) { printk("%s: Error trying to start fw!\n", __func__); return QLA_ERROR; } /* Handshake with the card before we register the devices. */ - if (qla4_8xxx_cmdpeg_ready(ha, 0) != QLA_SUCCESS) { + if (qla4_82xx_cmdpeg_ready(ha, 0) != QLA_SUCCESS) { printk("%s: Error during card handshake!\n", __func__); return QLA_ERROR; } @@ -1593,11 +1667,10 @@ qla4_8xxx_start_firmware(struct scsi_qla_host *ha, uint32_t image_start) ha->link_width = (lnk >> 4) & 0x3f; /* Synchronize with Receive peg */ - return qla4_8xxx_rcvpeg_ready(ha); + return qla4_82xx_rcvpeg_ready(ha); } -static int -qla4_8xxx_try_start_fw(struct scsi_qla_host *ha) +int qla4_82xx_try_start_fw(struct scsi_qla_host *ha) { int rval = QLA_ERROR; @@ -1615,7 +1688,7 @@ qla4_8xxx_try_start_fw(struct scsi_qla_host *ha) ql4_printk(KERN_INFO, ha, "FW: Attempting to load firmware from flash...\n"); - rval = qla4_8xxx_start_firmware(ha, ha->hw.flt_region_fw); + rval = qla4_82xx_start_firmware(ha, ha->hw.flt_region_fw); if (rval != QLA_SUCCESS) { ql4_printk(KERN_ERR, ha, "FW: Load firmware from flash" @@ -1626,9 +1699,9 @@ qla4_8xxx_try_start_fw(struct scsi_qla_host *ha) return rval; } -static void qla4_8xxx_rom_lock_recovery(struct scsi_qla_host *ha) +void qla4_82xx_rom_lock_recovery(struct scsi_qla_host *ha) { - if (qla4_8xxx_rom_lock(ha)) { + if (qla4_82xx_rom_lock(ha)) { /* Someone else is holding the lock. */ dev_info(&ha->pdev->dev, "Resetting rom_lock\n"); } @@ -1638,25 +1711,25 @@ static void qla4_8xxx_rom_lock_recovery(struct scsi_qla_host *ha) * else died while holding it. * In either case, unlock. */ - qla4_8xxx_rom_unlock(ha); + qla4_82xx_rom_unlock(ha); } static void qla4_8xxx_minidump_process_rdcrb(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t r_addr, r_stride, loop_cnt, i, r_value; - struct qla82xx_minidump_entry_crb *crb_hdr; + struct qla8xxx_minidump_entry_crb *crb_hdr; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - crb_hdr = (struct qla82xx_minidump_entry_crb *)entry_hdr; + crb_hdr = (struct qla8xxx_minidump_entry_crb *)entry_hdr; r_addr = crb_hdr->addr; r_stride = crb_hdr->crb_strd.addr_stride; loop_cnt = crb_hdr->op_count; for (i = 0; i < loop_cnt; i++) { - r_value = qla4_8xxx_md_rw_32(ha, r_addr, 0, 0); + ha->isp_ops->rd_reg_indirect(ha, r_addr, &r_value); *data_ptr++ = cpu_to_le32(r_addr); *data_ptr++ = cpu_to_le32(r_value); r_addr += r_stride; @@ -1665,19 +1738,19 @@ static void qla4_8xxx_minidump_process_rdcrb(struct scsi_qla_host *ha, } static int qla4_8xxx_minidump_process_l2tag(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t addr, r_addr, c_addr, t_r_addr; uint32_t i, k, loop_count, t_value, r_cnt, r_value; unsigned long p_wait, w_time, p_mask; uint32_t c_value_w, c_value_r; - struct qla82xx_minidump_entry_cache *cache_hdr; + struct qla8xxx_minidump_entry_cache *cache_hdr; int rval = QLA_ERROR; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - cache_hdr = (struct qla82xx_minidump_entry_cache *)entry_hdr; + cache_hdr = (struct qla8xxx_minidump_entry_cache *)entry_hdr; loop_count = cache_hdr->op_count; r_addr = cache_hdr->read_addr; @@ -1691,16 +1764,16 @@ static int qla4_8xxx_minidump_process_l2tag(struct scsi_qla_host *ha, p_mask = cache_hdr->cache_ctrl.poll_mask; for (i = 0; i < loop_count; i++) { - qla4_8xxx_md_rw_32(ha, t_r_addr, t_value, 1); + ha->isp_ops->wr_reg_indirect(ha, t_r_addr, t_value); if (c_value_w) - qla4_8xxx_md_rw_32(ha, c_addr, c_value_w, 1); + ha->isp_ops->wr_reg_indirect(ha, c_addr, c_value_w); if (p_mask) { w_time = jiffies + p_wait; do { - c_value_r = qla4_8xxx_md_rw_32(ha, c_addr, - 0, 0); + ha->isp_ops->rd_reg_indirect(ha, c_addr, + &c_value_r); if ((c_value_r & p_mask) == 0) { break; } else if (time_after_eq(jiffies, w_time)) { @@ -1712,7 +1785,7 @@ static int qla4_8xxx_minidump_process_l2tag(struct scsi_qla_host *ha, addr = r_addr; for (k = 0; k < r_cnt; k++) { - r_value = qla4_8xxx_md_rw_32(ha, addr, 0, 0); + ha->isp_ops->rd_reg_indirect(ha, addr, &r_value); *data_ptr++ = cpu_to_le32(r_value); addr += cache_hdr->read_ctrl.read_addr_stride; } @@ -1724,9 +1797,9 @@ static int qla4_8xxx_minidump_process_l2tag(struct scsi_qla_host *ha, } static int qla4_8xxx_minidump_process_control(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr) + struct qla8xxx_minidump_entry_hdr *entry_hdr) { - struct qla82xx_minidump_entry_crb *crb_entry; + struct qla8xxx_minidump_entry_crb *crb_entry; uint32_t read_value, opcode, poll_time, addr, index, rval = QLA_SUCCESS; uint32_t crb_addr; unsigned long wtime; @@ -1736,58 +1809,59 @@ static int qla4_8xxx_minidump_process_control(struct scsi_qla_host *ha, DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); tmplt_hdr = (struct qla4_8xxx_minidump_template_hdr *) ha->fw_dump_tmplt_hdr; - crb_entry = (struct qla82xx_minidump_entry_crb *)entry_hdr; + crb_entry = (struct qla8xxx_minidump_entry_crb *)entry_hdr; crb_addr = crb_entry->addr; for (i = 0; i < crb_entry->op_count; i++) { opcode = crb_entry->crb_ctrl.opcode; - if (opcode & QLA82XX_DBG_OPCODE_WR) { - qla4_8xxx_md_rw_32(ha, crb_addr, - crb_entry->value_1, 1); - opcode &= ~QLA82XX_DBG_OPCODE_WR; + if (opcode & QLA8XXX_DBG_OPCODE_WR) { + ha->isp_ops->wr_reg_indirect(ha, crb_addr, + crb_entry->value_1); + opcode &= ~QLA8XXX_DBG_OPCODE_WR; } - if (opcode & QLA82XX_DBG_OPCODE_RW) { - read_value = qla4_8xxx_md_rw_32(ha, crb_addr, 0, 0); - qla4_8xxx_md_rw_32(ha, crb_addr, read_value, 1); - opcode &= ~QLA82XX_DBG_OPCODE_RW; + if (opcode & QLA8XXX_DBG_OPCODE_RW) { + ha->isp_ops->rd_reg_indirect(ha, crb_addr, &read_value); + ha->isp_ops->wr_reg_indirect(ha, crb_addr, read_value); + opcode &= ~QLA8XXX_DBG_OPCODE_RW; } - if (opcode & QLA82XX_DBG_OPCODE_AND) { - read_value = qla4_8xxx_md_rw_32(ha, crb_addr, 0, 0); + if (opcode & QLA8XXX_DBG_OPCODE_AND) { + ha->isp_ops->rd_reg_indirect(ha, crb_addr, &read_value); read_value &= crb_entry->value_2; - opcode &= ~QLA82XX_DBG_OPCODE_AND; - if (opcode & QLA82XX_DBG_OPCODE_OR) { + opcode &= ~QLA8XXX_DBG_OPCODE_AND; + if (opcode & QLA8XXX_DBG_OPCODE_OR) { read_value |= crb_entry->value_3; - opcode &= ~QLA82XX_DBG_OPCODE_OR; + opcode &= ~QLA8XXX_DBG_OPCODE_OR; } - qla4_8xxx_md_rw_32(ha, crb_addr, read_value, 1); + ha->isp_ops->wr_reg_indirect(ha, crb_addr, read_value); } - if (opcode & QLA82XX_DBG_OPCODE_OR) { - read_value = qla4_8xxx_md_rw_32(ha, crb_addr, 0, 0); + if (opcode & QLA8XXX_DBG_OPCODE_OR) { + ha->isp_ops->rd_reg_indirect(ha, crb_addr, &read_value); read_value |= crb_entry->value_3; - qla4_8xxx_md_rw_32(ha, crb_addr, read_value, 1); - opcode &= ~QLA82XX_DBG_OPCODE_OR; + ha->isp_ops->wr_reg_indirect(ha, crb_addr, read_value); + opcode &= ~QLA8XXX_DBG_OPCODE_OR; } - if (opcode & QLA82XX_DBG_OPCODE_POLL) { + if (opcode & QLA8XXX_DBG_OPCODE_POLL) { poll_time = crb_entry->crb_strd.poll_timeout; wtime = jiffies + poll_time; - read_value = qla4_8xxx_md_rw_32(ha, crb_addr, 0, 0); + ha->isp_ops->rd_reg_indirect(ha, crb_addr, &read_value); do { if ((read_value & crb_entry->value_2) == - crb_entry->value_1) + crb_entry->value_1) { break; - else if (time_after_eq(jiffies, wtime)) { + } else if (time_after_eq(jiffies, wtime)) { /* capturing dump failed */ rval = QLA_ERROR; break; - } else - read_value = qla4_8xxx_md_rw_32(ha, - crb_addr, 0, 0); + } else { + ha->isp_ops->rd_reg_indirect(ha, + crb_addr, &read_value); + } } while (1); - opcode &= ~QLA82XX_DBG_OPCODE_POLL; + opcode &= ~QLA8XXX_DBG_OPCODE_POLL; } - if (opcode & QLA82XX_DBG_OPCODE_RDSTATE) { + if (opcode & QLA8XXX_DBG_OPCODE_RDSTATE) { if (crb_entry->crb_strd.state_index_a) { index = crb_entry->crb_strd.state_index_a; addr = tmplt_hdr->saved_state_array[index]; @@ -1795,13 +1869,13 @@ static int qla4_8xxx_minidump_process_control(struct scsi_qla_host *ha, addr = crb_addr; } - read_value = qla4_8xxx_md_rw_32(ha, addr, 0, 0); + ha->isp_ops->rd_reg_indirect(ha, addr, &read_value); index = crb_entry->crb_ctrl.state_index_v; tmplt_hdr->saved_state_array[index] = read_value; - opcode &= ~QLA82XX_DBG_OPCODE_RDSTATE; + opcode &= ~QLA8XXX_DBG_OPCODE_RDSTATE; } - if (opcode & QLA82XX_DBG_OPCODE_WRSTATE) { + if (opcode & QLA8XXX_DBG_OPCODE_WRSTATE) { if (crb_entry->crb_strd.state_index_a) { index = crb_entry->crb_strd.state_index_a; addr = tmplt_hdr->saved_state_array[index]; @@ -1817,11 +1891,11 @@ static int qla4_8xxx_minidump_process_control(struct scsi_qla_host *ha, read_value = crb_entry->value_1; } - qla4_8xxx_md_rw_32(ha, addr, read_value, 1); - opcode &= ~QLA82XX_DBG_OPCODE_WRSTATE; + ha->isp_ops->wr_reg_indirect(ha, addr, read_value); + opcode &= ~QLA8XXX_DBG_OPCODE_WRSTATE; } - if (opcode & QLA82XX_DBG_OPCODE_MDSTATE) { + if (opcode & QLA8XXX_DBG_OPCODE_MDSTATE) { index = crb_entry->crb_ctrl.state_index_v; read_value = tmplt_hdr->saved_state_array[index]; read_value <<= crb_entry->crb_ctrl.shl; @@ -1831,7 +1905,7 @@ static int qla4_8xxx_minidump_process_control(struct scsi_qla_host *ha, read_value |= crb_entry->value_3; read_value += crb_entry->value_1; tmplt_hdr->saved_state_array[index] = read_value; - opcode &= ~QLA82XX_DBG_OPCODE_MDSTATE; + opcode &= ~QLA8XXX_DBG_OPCODE_MDSTATE; } crb_addr += crb_entry->crb_strd.addr_stride; } @@ -1840,15 +1914,15 @@ static int qla4_8xxx_minidump_process_control(struct scsi_qla_host *ha, } static void qla4_8xxx_minidump_process_rdocm(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t r_addr, r_stride, loop_cnt, i, r_value; - struct qla82xx_minidump_entry_rdocm *ocm_hdr; + struct qla8xxx_minidump_entry_rdocm *ocm_hdr; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - ocm_hdr = (struct qla82xx_minidump_entry_rdocm *)entry_hdr; + ocm_hdr = (struct qla8xxx_minidump_entry_rdocm *)entry_hdr; r_addr = ocm_hdr->read_addr; r_stride = ocm_hdr->read_addr_stride; loop_cnt = ocm_hdr->op_count; @@ -1863,20 +1937,20 @@ static void qla4_8xxx_minidump_process_rdocm(struct scsi_qla_host *ha, r_addr += r_stride; } DEBUG2(ql4_printk(KERN_INFO, ha, "Leaving fn: %s datacount: 0x%lx\n", - __func__, (loop_cnt * sizeof(uint32_t)))); + __func__, (long unsigned int) (loop_cnt * sizeof(uint32_t)))); *d_ptr = data_ptr; } static void qla4_8xxx_minidump_process_rdmux(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t r_addr, s_stride, s_addr, s_value, loop_cnt, i, r_value; - struct qla82xx_minidump_entry_mux *mux_hdr; + struct qla8xxx_minidump_entry_mux *mux_hdr; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - mux_hdr = (struct qla82xx_minidump_entry_mux *)entry_hdr; + mux_hdr = (struct qla8xxx_minidump_entry_mux *)entry_hdr; r_addr = mux_hdr->read_addr; s_addr = mux_hdr->select_addr; s_stride = mux_hdr->select_value_stride; @@ -1884,8 +1958,8 @@ static void qla4_8xxx_minidump_process_rdmux(struct scsi_qla_host *ha, loop_cnt = mux_hdr->op_count; for (i = 0; i < loop_cnt; i++) { - qla4_8xxx_md_rw_32(ha, s_addr, s_value, 1); - r_value = qla4_8xxx_md_rw_32(ha, r_addr, 0, 0); + ha->isp_ops->wr_reg_indirect(ha, s_addr, s_value); + ha->isp_ops->rd_reg_indirect(ha, r_addr, &r_value); *data_ptr++ = cpu_to_le32(s_value); *data_ptr++ = cpu_to_le32(r_value); s_value += s_stride; @@ -1894,16 +1968,16 @@ static void qla4_8xxx_minidump_process_rdmux(struct scsi_qla_host *ha, } static void qla4_8xxx_minidump_process_l1cache(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t addr, r_addr, c_addr, t_r_addr; uint32_t i, k, loop_count, t_value, r_cnt, r_value; uint32_t c_value_w; - struct qla82xx_minidump_entry_cache *cache_hdr; + struct qla8xxx_minidump_entry_cache *cache_hdr; uint32_t *data_ptr = *d_ptr; - cache_hdr = (struct qla82xx_minidump_entry_cache *)entry_hdr; + cache_hdr = (struct qla8xxx_minidump_entry_cache *)entry_hdr; loop_count = cache_hdr->op_count; r_addr = cache_hdr->read_addr; c_addr = cache_hdr->control_addr; @@ -1914,11 +1988,11 @@ static void qla4_8xxx_minidump_process_l1cache(struct scsi_qla_host *ha, r_cnt = cache_hdr->read_ctrl.read_addr_cnt; for (i = 0; i < loop_count; i++) { - qla4_8xxx_md_rw_32(ha, t_r_addr, t_value, 1); - qla4_8xxx_md_rw_32(ha, c_addr, c_value_w, 1); + ha->isp_ops->wr_reg_indirect(ha, t_r_addr, t_value); + ha->isp_ops->wr_reg_indirect(ha, c_addr, c_value_w); addr = r_addr; for (k = 0; k < r_cnt; k++) { - r_value = qla4_8xxx_md_rw_32(ha, addr, 0, 0); + ha->isp_ops->rd_reg_indirect(ha, addr, &r_value); *data_ptr++ = cpu_to_le32(r_value); addr += cache_hdr->read_ctrl.read_addr_stride; } @@ -1928,27 +2002,27 @@ static void qla4_8xxx_minidump_process_l1cache(struct scsi_qla_host *ha, } static void qla4_8xxx_minidump_process_queue(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t s_addr, r_addr; uint32_t r_stride, r_value, r_cnt, qid = 0; uint32_t i, k, loop_cnt; - struct qla82xx_minidump_entry_queue *q_hdr; + struct qla8xxx_minidump_entry_queue *q_hdr; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - q_hdr = (struct qla82xx_minidump_entry_queue *)entry_hdr; + q_hdr = (struct qla8xxx_minidump_entry_queue *)entry_hdr; s_addr = q_hdr->select_addr; r_cnt = q_hdr->rd_strd.read_addr_cnt; r_stride = q_hdr->rd_strd.read_addr_stride; loop_cnt = q_hdr->op_count; for (i = 0; i < loop_cnt; i++) { - qla4_8xxx_md_rw_32(ha, s_addr, qid, 1); + ha->isp_ops->wr_reg_indirect(ha, s_addr, qid); r_addr = q_hdr->read_addr; for (k = 0; k < r_cnt; k++) { - r_value = qla4_8xxx_md_rw_32(ha, r_addr, 0, 0); + ha->isp_ops->rd_reg_indirect(ha, r_addr, &r_value); *data_ptr++ = cpu_to_le32(r_value); r_addr += r_stride; } @@ -1960,17 +2034,17 @@ static void qla4_8xxx_minidump_process_queue(struct scsi_qla_host *ha, #define MD_DIRECT_ROM_WINDOW 0x42110030 #define MD_DIRECT_ROM_READ_BASE 0x42150000 -static void qla4_8xxx_minidump_process_rdrom(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, +static void qla4_82xx_minidump_process_rdrom(struct scsi_qla_host *ha, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t r_addr, r_value; uint32_t i, loop_cnt; - struct qla82xx_minidump_entry_rdrom *rom_hdr; + struct qla8xxx_minidump_entry_rdrom *rom_hdr; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - rom_hdr = (struct qla82xx_minidump_entry_rdrom *)entry_hdr; + rom_hdr = (struct qla8xxx_minidump_entry_rdrom *)entry_hdr; r_addr = rom_hdr->read_addr; loop_cnt = rom_hdr->read_data_size/sizeof(uint32_t); @@ -1979,11 +2053,11 @@ static void qla4_8xxx_minidump_process_rdrom(struct scsi_qla_host *ha, __func__, r_addr, loop_cnt)); for (i = 0; i < loop_cnt; i++) { - qla4_8xxx_md_rw_32(ha, MD_DIRECT_ROM_WINDOW, - (r_addr & 0xFFFF0000), 1); - r_value = qla4_8xxx_md_rw_32(ha, - MD_DIRECT_ROM_READ_BASE + - (r_addr & 0x0000FFFF), 0, 0); + ha->isp_ops->wr_reg_indirect(ha, MD_DIRECT_ROM_WINDOW, + (r_addr & 0xFFFF0000)); + ha->isp_ops->rd_reg_indirect(ha, + MD_DIRECT_ROM_READ_BASE + (r_addr & 0x0000FFFF), + &r_value); *data_ptr++ = cpu_to_le32(r_value); r_addr += sizeof(uint32_t); } @@ -1995,17 +2069,17 @@ static void qla4_8xxx_minidump_process_rdrom(struct scsi_qla_host *ha, #define MD_MIU_TEST_AGT_ADDR_HI 0x41000098 static int qla4_8xxx_minidump_process_rdmem(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, + struct qla8xxx_minidump_entry_hdr *entry_hdr, uint32_t **d_ptr) { uint32_t r_addr, r_value, r_data; uint32_t i, j, loop_cnt; - struct qla82xx_minidump_entry_rdmem *m_hdr; + struct qla8xxx_minidump_entry_rdmem *m_hdr; unsigned long flags; uint32_t *data_ptr = *d_ptr; DEBUG2(ql4_printk(KERN_INFO, ha, "Entering fn: %s\n", __func__)); - m_hdr = (struct qla82xx_minidump_entry_rdmem *)entry_hdr; + m_hdr = (struct qla8xxx_minidump_entry_rdmem *)entry_hdr; r_addr = m_hdr->read_addr; loop_cnt = m_hdr->read_data_size/16; @@ -2033,17 +2107,19 @@ static int qla4_8xxx_minidump_process_rdmem(struct scsi_qla_host *ha, write_lock_irqsave(&ha->hw_lock, flags); for (i = 0; i < loop_cnt; i++) { - qla4_8xxx_md_rw_32(ha, MD_MIU_TEST_AGT_ADDR_LO, r_addr, 1); + ha->isp_ops->wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_LO, + r_addr); r_value = 0; - qla4_8xxx_md_rw_32(ha, MD_MIU_TEST_AGT_ADDR_HI, r_value, 1); + ha->isp_ops->wr_reg_indirect(ha, MD_MIU_TEST_AGT_ADDR_HI, + r_value); r_value = MIU_TA_CTL_ENABLE; - qla4_8xxx_md_rw_32(ha, MD_MIU_TEST_AGT_CTRL, r_value, 1); - r_value = MIU_TA_CTL_START | MIU_TA_CTL_ENABLE; - qla4_8xxx_md_rw_32(ha, MD_MIU_TEST_AGT_CTRL, r_value, 1); + ha->isp_ops->wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL, r_value); + r_value = MIU_TA_CTL_START_ENABLE; + ha->isp_ops->wr_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL, r_value); for (j = 0; j < MAX_CTL_CHECK; j++) { - r_value = qla4_8xxx_md_rw_32(ha, MD_MIU_TEST_AGT_CTRL, - 0, 0); + ha->isp_ops->rd_reg_indirect(ha, MD_MIU_TEST_AGT_CTRL, + &r_value); if ((r_value & MIU_TA_CTL_BUSY) == 0) break; } @@ -2057,9 +2133,9 @@ static int qla4_8xxx_minidump_process_rdmem(struct scsi_qla_host *ha, } for (j = 0; j < 4; j++) { - r_data = qla4_8xxx_md_rw_32(ha, - MD_MIU_TEST_AGT_RDDATA[j], - 0, 0); + ha->isp_ops->rd_reg_indirect(ha, + MD_MIU_TEST_AGT_RDDATA[j], + &r_data); *data_ptr++ = cpu_to_le32(r_data); } @@ -2074,25 +2150,215 @@ static int qla4_8xxx_minidump_process_rdmem(struct scsi_qla_host *ha, return QLA_SUCCESS; } -static void ql4_8xxx_mark_entry_skipped(struct scsi_qla_host *ha, - struct qla82xx_minidump_entry_hdr *entry_hdr, +static void qla4_8xxx_mark_entry_skipped(struct scsi_qla_host *ha, + struct qla8xxx_minidump_entry_hdr *entry_hdr, int index) { - entry_hdr->d_ctrl.driver_flags |= QLA82XX_DBG_SKIPPED_FLAG; + entry_hdr->d_ctrl.driver_flags |= QLA8XXX_DBG_SKIPPED_FLAG; DEBUG2(ql4_printk(KERN_INFO, ha, "scsi(%ld): Skipping entry[%d]: ETYPE[0x%x]-ELEVEL[0x%x]\n", ha->host_no, index, entry_hdr->entry_type, entry_hdr->d_ctrl.entry_capture_mask)); } +/* ISP83xx functions to process new minidump entries... */ +static uint32_t qla83xx_minidump_process_pollrd(struct scsi_qla_host *ha, + struct qla8xxx_minidump_entry_hdr *entry_hdr, + uint32_t **d_ptr) +{ + uint32_t r_addr, s_addr, s_value, r_value, poll_wait, poll_mask; + uint16_t s_stride, i; + uint32_t *data_ptr = *d_ptr; + uint32_t rval = QLA_SUCCESS; + struct qla83xx_minidump_entry_pollrd *pollrd_hdr; + + pollrd_hdr = (struct qla83xx_minidump_entry_pollrd *)entry_hdr; + s_addr = le32_to_cpu(pollrd_hdr->select_addr); + r_addr = le32_to_cpu(pollrd_hdr->read_addr); + s_value = le32_to_cpu(pollrd_hdr->select_value); + s_stride = le32_to_cpu(pollrd_hdr->select_value_stride); + + poll_wait = le32_to_cpu(pollrd_hdr->poll_wait); + poll_mask = le32_to_cpu(pollrd_hdr->poll_mask); + + for (i = 0; i < le32_to_cpu(pollrd_hdr->op_count); i++) { + ha->isp_ops->wr_reg_indirect(ha, s_addr, s_value); + poll_wait = le32_to_cpu(pollrd_hdr->poll_wait); + while (1) { + ha->isp_ops->rd_reg_indirect(ha, s_addr, &r_value); + + if ((r_value & poll_mask) != 0) { + break; + } else { + msleep(1); + if (--poll_wait == 0) { + ql4_printk(KERN_ERR, ha, "%s: TIMEOUT\n", + __func__); + rval = QLA_ERROR; + goto exit_process_pollrd; + } + } + } + ha->isp_ops->rd_reg_indirect(ha, r_addr, &r_value); + *data_ptr++ = cpu_to_le32(s_value); + *data_ptr++ = cpu_to_le32(r_value); + s_value += s_stride; + } + + *d_ptr = data_ptr; + +exit_process_pollrd: + return rval; +} + +static void qla83xx_minidump_process_rdmux2(struct scsi_qla_host *ha, + struct qla8xxx_minidump_entry_hdr *entry_hdr, + uint32_t **d_ptr) +{ + uint32_t sel_val1, sel_val2, t_sel_val, data, i; + uint32_t sel_addr1, sel_addr2, sel_val_mask, read_addr; + struct qla83xx_minidump_entry_rdmux2 *rdmux2_hdr; + uint32_t *data_ptr = *d_ptr; + + rdmux2_hdr = (struct qla83xx_minidump_entry_rdmux2 *)entry_hdr; + sel_val1 = le32_to_cpu(rdmux2_hdr->select_value_1); + sel_val2 = le32_to_cpu(rdmux2_hdr->select_value_2); + sel_addr1 = le32_to_cpu(rdmux2_hdr->select_addr_1); + sel_addr2 = le32_to_cpu(rdmux2_hdr->select_addr_2); + sel_val_mask = le32_to_cpu(rdmux2_hdr->select_value_mask); + read_addr = le32_to_cpu(rdmux2_hdr->read_addr); + + for (i = 0; i < rdmux2_hdr->op_count; i++) { + ha->isp_ops->wr_reg_indirect(ha, sel_addr1, sel_val1); + t_sel_val = sel_val1 & sel_val_mask; + *data_ptr++ = cpu_to_le32(t_sel_val); + + ha->isp_ops->wr_reg_indirect(ha, sel_addr2, t_sel_val); + ha->isp_ops->rd_reg_indirect(ha, read_addr, &data); + + *data_ptr++ = cpu_to_le32(data); + + ha->isp_ops->wr_reg_indirect(ha, sel_addr1, sel_val2); + t_sel_val = sel_val2 & sel_val_mask; + *data_ptr++ = cpu_to_le32(t_sel_val); + + ha->isp_ops->wr_reg_indirect(ha, sel_addr2, t_sel_val); + ha->isp_ops->rd_reg_indirect(ha, read_addr, &data); + + *data_ptr++ = cpu_to_le32(data); + + sel_val1 += rdmux2_hdr->select_value_stride; + sel_val2 += rdmux2_hdr->select_value_stride; + } + + *d_ptr = data_ptr; +} + +static uint32_t qla83xx_minidump_process_pollrdmwr(struct scsi_qla_host *ha, + struct qla8xxx_minidump_entry_hdr *entry_hdr, + uint32_t **d_ptr) +{ + uint32_t poll_wait, poll_mask, r_value, data; + uint32_t addr_1, addr_2, value_1, value_2; + uint32_t *data_ptr = *d_ptr; + uint32_t rval = QLA_SUCCESS; + struct qla83xx_minidump_entry_pollrdmwr *poll_hdr; + + poll_hdr = (struct qla83xx_minidump_entry_pollrdmwr *)entry_hdr; + addr_1 = le32_to_cpu(poll_hdr->addr_1); + addr_2 = le32_to_cpu(poll_hdr->addr_2); + value_1 = le32_to_cpu(poll_hdr->value_1); + value_2 = le32_to_cpu(poll_hdr->value_2); + poll_mask = le32_to_cpu(poll_hdr->poll_mask); + + ha->isp_ops->wr_reg_indirect(ha, addr_1, value_1); + + poll_wait = le32_to_cpu(poll_hdr->poll_wait); + while (1) { + ha->isp_ops->rd_reg_indirect(ha, addr_1, &r_value); + + if ((r_value & poll_mask) != 0) { + break; + } else { + msleep(1); + if (--poll_wait == 0) { + ql4_printk(KERN_ERR, ha, "%s: TIMEOUT_1\n", + __func__); + rval = QLA_ERROR; + goto exit_process_pollrdmwr; + } + } + } + + ha->isp_ops->rd_reg_indirect(ha, addr_2, &data); + data &= le32_to_cpu(poll_hdr->modify_mask); + ha->isp_ops->wr_reg_indirect(ha, addr_2, data); + ha->isp_ops->wr_reg_indirect(ha, addr_1, value_2); + + poll_wait = le32_to_cpu(poll_hdr->poll_wait); + while (1) { + ha->isp_ops->rd_reg_indirect(ha, addr_1, &r_value); + + if ((r_value & poll_mask) != 0) { + break; + } else { + msleep(1); + if (--poll_wait == 0) { + ql4_printk(KERN_ERR, ha, "%s: TIMEOUT_2\n", + __func__); + rval = QLA_ERROR; + goto exit_process_pollrdmwr; + } + } + } + + *data_ptr++ = cpu_to_le32(addr_2); + *data_ptr++ = cpu_to_le32(data); + *d_ptr = data_ptr; + +exit_process_pollrdmwr: + return rval; +} + +static uint32_t qla4_83xx_minidump_process_rdrom(struct scsi_qla_host *ha, + struct qla8xxx_minidump_entry_hdr *entry_hdr, + uint32_t **d_ptr) +{ + uint32_t fl_addr, u32_count, rval; + struct qla8xxx_minidump_entry_rdrom *rom_hdr; + uint32_t *data_ptr = *d_ptr; + + rom_hdr = (struct qla8xxx_minidump_entry_rdrom *)entry_hdr; + fl_addr = le32_to_cpu(rom_hdr->read_addr); + u32_count = le32_to_cpu(rom_hdr->read_data_size)/sizeof(uint32_t); + + DEBUG2(ql4_printk(KERN_INFO, ha, "[%s]: fl_addr: 0x%x, count: 0x%x\n", + __func__, fl_addr, u32_count)); + + rval = qla4_83xx_lockless_flash_read_u32(ha, fl_addr, + (u8 *)(data_ptr), u32_count); + + if (rval == QLA_ERROR) { + ql4_printk(KERN_ERR, ha, "%s: Flash Read Error,Count=%d\n", + __func__, u32_count); + goto exit_process_rdrom; + } + + data_ptr += u32_count; + *d_ptr = data_ptr; + +exit_process_rdrom: + return rval; +} + /** - * qla82xx_collect_md_data - Retrieve firmware minidump data. + * qla4_8xxx_collect_md_data - Retrieve firmware minidump data. * @ha: pointer to adapter structure **/ static int qla4_8xxx_collect_md_data(struct scsi_qla_host *ha) { int num_entry_hdr = 0; - struct qla82xx_minidump_entry_hdr *entry_hdr; + struct qla8xxx_minidump_entry_hdr *entry_hdr; struct qla4_8xxx_minidump_template_hdr *tmplt_hdr; uint32_t *data_ptr; uint32_t data_collected = 0; @@ -2128,10 +2394,14 @@ static int qla4_8xxx_collect_md_data(struct scsi_qla_host *ha) timestamp = (u32)(jiffies_to_msecs(now) / 1000); tmplt_hdr->driver_timestamp = timestamp; - entry_hdr = (struct qla82xx_minidump_entry_hdr *) + entry_hdr = (struct qla8xxx_minidump_entry_hdr *) (((uint8_t *)ha->fw_dump_tmplt_hdr) + tmplt_hdr->first_entry_offset); + if (is_qla8032(ha)) + tmplt_hdr->saved_state_array[QLA83XX_SS_OCM_WNDREG_INDEX] = + tmplt_hdr->ocm_window_reg[ha->func_num]; + /* Walk through the entry headers - validate/perform required action */ for (i = 0; i < num_entry_hdr; i++) { if (data_collected >= ha->fw_dump_size) { @@ -2144,7 +2414,7 @@ static int qla4_8xxx_collect_md_data(struct scsi_qla_host *ha) if (!(entry_hdr->d_ctrl.entry_capture_mask & ha->fw_dump_capture_mask)) { entry_hdr->d_ctrl.driver_flags |= - QLA82XX_DBG_SKIPPED_FLAG; + QLA8XXX_DBG_SKIPPED_FLAG; goto skip_nxt_entry; } @@ -2157,65 +2427,105 @@ static int qla4_8xxx_collect_md_data(struct scsi_qla_host *ha) * debug data */ switch (entry_hdr->entry_type) { - case QLA82XX_RDEND: - ql4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + case QLA8XXX_RDEND: + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); break; - case QLA82XX_CNTRL: + case QLA8XXX_CNTRL: rval = qla4_8xxx_minidump_process_control(ha, entry_hdr); if (rval != QLA_SUCCESS) { - ql4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); goto md_failed; } break; - case QLA82XX_RDCRB: + case QLA8XXX_RDCRB: qla4_8xxx_minidump_process_rdcrb(ha, entry_hdr, &data_ptr); break; - case QLA82XX_RDMEM: + case QLA8XXX_RDMEM: rval = qla4_8xxx_minidump_process_rdmem(ha, entry_hdr, &data_ptr); if (rval != QLA_SUCCESS) { - ql4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); goto md_failed; } break; - case QLA82XX_BOARD: - case QLA82XX_RDROM: - qla4_8xxx_minidump_process_rdrom(ha, entry_hdr, - &data_ptr); + case QLA8XXX_BOARD: + case QLA8XXX_RDROM: + if (is_qla8022(ha)) { + qla4_82xx_minidump_process_rdrom(ha, entry_hdr, + &data_ptr); + } else if (is_qla8032(ha)) { + rval = qla4_83xx_minidump_process_rdrom(ha, + entry_hdr, + &data_ptr); + if (rval != QLA_SUCCESS) + qla4_8xxx_mark_entry_skipped(ha, + entry_hdr, + i); + } break; - case QLA82XX_L2DTG: - case QLA82XX_L2ITG: - case QLA82XX_L2DAT: - case QLA82XX_L2INS: + case QLA8XXX_L2DTG: + case QLA8XXX_L2ITG: + case QLA8XXX_L2DAT: + case QLA8XXX_L2INS: rval = qla4_8xxx_minidump_process_l2tag(ha, entry_hdr, &data_ptr); if (rval != QLA_SUCCESS) { - ql4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); goto md_failed; } break; - case QLA82XX_L1DAT: - case QLA82XX_L1INS: + case QLA8XXX_L1DTG: + case QLA8XXX_L1ITG: + case QLA8XXX_L1DAT: + case QLA8XXX_L1INS: qla4_8xxx_minidump_process_l1cache(ha, entry_hdr, &data_ptr); break; - case QLA82XX_RDOCM: + case QLA8XXX_RDOCM: qla4_8xxx_minidump_process_rdocm(ha, entry_hdr, &data_ptr); break; - case QLA82XX_RDMUX: + case QLA8XXX_RDMUX: qla4_8xxx_minidump_process_rdmux(ha, entry_hdr, &data_ptr); break; - case QLA82XX_QUEUE: + case QLA8XXX_QUEUE: qla4_8xxx_minidump_process_queue(ha, entry_hdr, &data_ptr); break; - case QLA82XX_RDNOP: + case QLA83XX_POLLRD: + if (!is_qla8032(ha)) { + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + break; + } + rval = qla83xx_minidump_process_pollrd(ha, entry_hdr, + &data_ptr); + if (rval != QLA_SUCCESS) + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + break; + case QLA83XX_RDMUX2: + if (!is_qla8032(ha)) { + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + break; + } + qla83xx_minidump_process_rdmux2(ha, entry_hdr, + &data_ptr); + break; + case QLA83XX_POLLRDMWR: + if (!is_qla8032(ha)) { + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + break; + } + rval = qla83xx_minidump_process_pollrdmwr(ha, entry_hdr, + &data_ptr); + if (rval != QLA_SUCCESS) + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + break; + case QLA8XXX_RDNOP: default: - ql4_8xxx_mark_entry_skipped(ha, entry_hdr, i); + qla4_8xxx_mark_entry_skipped(ha, entry_hdr, i); break; } @@ -2224,7 +2534,7 @@ static int qla4_8xxx_collect_md_data(struct scsi_qla_host *ha) ha->fw_dump_tmplt_size)); skip_nxt_entry: /* next entry in the template */ - entry_hdr = (struct qla82xx_minidump_entry_hdr *) + entry_hdr = (struct qla8xxx_minidump_entry_hdr *) (((uint8_t *)entry_hdr) + entry_hdr->entry_size); } @@ -2264,33 +2574,45 @@ static void qla4_8xxx_uevent_emit(struct scsi_qla_host *ha, u32 code) kobject_uevent_env(&(&ha->pdev->dev)->kobj, KOBJ_CHANGE, envp); } +void qla4_8xxx_get_minidump(struct scsi_qla_host *ha) +{ + if (ql4xenablemd && test_bit(AF_FW_RECOVERY, &ha->flags) && + !test_bit(AF_82XX_FW_DUMPED, &ha->flags)) { + if (!qla4_8xxx_collect_md_data(ha)) { + qla4_8xxx_uevent_emit(ha, QL4_UEVENT_CODE_FW_DUMP); + set_bit(AF_82XX_FW_DUMPED, &ha->flags); + } else { + ql4_printk(KERN_INFO, ha, "%s: Unable to collect minidump\n", + __func__); + } + } +} + /** * qla4_8xxx_device_bootstrap - Initialize device, set DEV_READY, start fw * @ha: pointer to adapter structure * * Note: IDC lock must be held upon entry **/ -static int -qla4_8xxx_device_bootstrap(struct scsi_qla_host *ha) +int qla4_8xxx_device_bootstrap(struct scsi_qla_host *ha) { int rval = QLA_ERROR; int i, timeout; - uint32_t old_count, count; + uint32_t old_count, count, idc_ctrl; int need_reset = 0, peg_stuck = 1; - need_reset = qla4_8xxx_need_reset(ha); - - old_count = qla4_8xxx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER); + need_reset = ha->isp_ops->need_reset(ha); + old_count = qla4_8xxx_rd_direct(ha, QLA8XXX_PEG_ALIVE_COUNTER); for (i = 0; i < 10; i++) { timeout = msleep_interruptible(200); if (timeout) { - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); return rval; } - count = qla4_8xxx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER); + count = qla4_8xxx_rd_direct(ha, QLA8XXX_PEG_ALIVE_COUNTER); if (count != old_count) peg_stuck = 0; } @@ -2298,13 +2620,13 @@ qla4_8xxx_device_bootstrap(struct scsi_qla_host *ha) if (need_reset) { /* We are trying to perform a recovery here. */ if (peg_stuck) - qla4_8xxx_rom_lock_recovery(ha); + ha->isp_ops->rom_lock_recovery(ha); goto dev_initialize; } else { /* Start of day for this ha context. */ if (peg_stuck) { /* Either we are the first or recovery in progress. */ - qla4_8xxx_rom_lock_recovery(ha); + ha->isp_ops->rom_lock_recovery(ha); goto dev_initialize; } else { /* Firmware already running. */ @@ -2316,46 +2638,53 @@ qla4_8xxx_device_bootstrap(struct scsi_qla_host *ha) dev_initialize: /* set to DEV_INITIALIZING */ ql4_printk(KERN_INFO, ha, "HW State: INITIALIZING\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_INITIALIZING); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_INITIALIZING); - /* Driver that sets device state to initializating sets IDC version */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_IDC_VERSION, QLA82XX_IDC_VERSION); - - qla4_8xxx_idc_unlock(ha); - if (ql4xenablemd && test_bit(AF_FW_RECOVERY, &ha->flags) && - !test_and_set_bit(AF_82XX_FW_DUMPED, &ha->flags)) { - if (!qla4_8xxx_collect_md_data(ha)) { - qla4_8xxx_uevent_emit(ha, QL4_UEVENT_CODE_FW_DUMP); - } else { - ql4_printk(KERN_INFO, ha, "Unable to collect minidump\n"); - clear_bit(AF_82XX_FW_DUMPED, &ha->flags); + /* + * For ISP8324, if IDC_CTRL GRACEFUL_RESET_BIT1 is set, reset it after + * device goes to INIT state. + */ + if (is_qla8032(ha)) { + idc_ctrl = qla4_83xx_rd_reg(ha, QLA83XX_IDC_DRV_CTRL); + if (idc_ctrl & GRACEFUL_RESET_BIT1) { + qla4_83xx_wr_reg(ha, QLA83XX_IDC_DRV_CTRL, + (idc_ctrl & ~GRACEFUL_RESET_BIT1)); + set_bit(AF_83XX_NO_FW_DUMP, &ha->flags); } } - rval = qla4_8xxx_try_start_fw(ha); - qla4_8xxx_idc_lock(ha); + + ha->isp_ops->idc_unlock(ha); + + if (is_qla8022(ha)) + qla4_8xxx_get_minidump(ha); + + rval = ha->isp_ops->restart_firmware(ha); + ha->isp_ops->idc_lock(ha); if (rval != QLA_SUCCESS) { ql4_printk(KERN_INFO, ha, "HW State: FAILED\n"); qla4_8xxx_clear_drv_active(ha); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_FAILED); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); return rval; } dev_ready: ql4_printk(KERN_INFO, ha, "HW State: READY\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_READY); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, QLA8XXX_DEV_READY); return rval; } /** - * qla4_8xxx_need_reset_handler - Code to start reset sequence + * qla4_82xx_need_reset_handler - Code to start reset sequence * @ha: pointer to adapter structure * * Note: IDC lock must be held upon entry **/ static void -qla4_8xxx_need_reset_handler(struct scsi_qla_host *ha) +qla4_82xx_need_reset_handler(struct scsi_qla_host *ha) { uint32_t dev_state, drv_state, drv_active; uint32_t active_mask = 0xFFFFFFFF; @@ -2365,12 +2694,12 @@ qla4_8xxx_need_reset_handler(struct scsi_qla_host *ha) "Performing ISP error recovery\n"); if (test_and_clear_bit(AF_ONLINE, &ha->flags)) { - qla4_8xxx_idc_unlock(ha); + qla4_82xx_idc_unlock(ha); ha->isp_ops->disable_intrs(ha); - qla4_8xxx_idc_lock(ha); + qla4_82xx_idc_lock(ha); } - if (!test_bit(AF_82XX_RST_OWNER, &ha->flags)) { + if (!test_bit(AF_8XXX_RST_OWNER, &ha->flags)) { DEBUG2(ql4_printk(KERN_INFO, ha, "%s(%ld): reset acknowledged\n", __func__, ha->host_no)); @@ -2382,8 +2711,8 @@ qla4_8xxx_need_reset_handler(struct scsi_qla_host *ha) /* wait for 10 seconds for reset ack from all functions */ reset_timeout = jiffies + (ha->nx_reset_timeout * HZ); - drv_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_STATE); - drv_active = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); + drv_state = qla4_82xx_rd_32(ha, QLA82XX_CRB_DRV_STATE); + drv_active = qla4_82xx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); ql4_printk(KERN_INFO, ha, "%s(%ld): drv_state = 0x%x, drv_active = 0x%x\n", @@ -2401,31 +2730,31 @@ qla4_8xxx_need_reset_handler(struct scsi_qla_host *ha) * When reset_owner times out, check which functions * acked/did not ack */ - if (test_bit(AF_82XX_RST_OWNER, &ha->flags)) { + if (test_bit(AF_8XXX_RST_OWNER, &ha->flags)) { ql4_printk(KERN_INFO, ha, "%s(%ld): drv_state = 0x%x, drv_active = 0x%x\n", __func__, ha->host_no, drv_state, drv_active); } - qla4_8xxx_idc_unlock(ha); + qla4_82xx_idc_unlock(ha); msleep(1000); - qla4_8xxx_idc_lock(ha); + qla4_82xx_idc_lock(ha); - drv_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_STATE); - drv_active = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); + drv_state = qla4_82xx_rd_32(ha, QLA82XX_CRB_DRV_STATE); + drv_active = qla4_82xx_rd_32(ha, QLA82XX_CRB_DRV_ACTIVE); } /* Clear RESET OWNER as we are not going to use it any further */ - clear_bit(AF_82XX_RST_OWNER, &ha->flags); + clear_bit(AF_8XXX_RST_OWNER, &ha->flags); - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); + dev_state = qla4_82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); ql4_printk(KERN_INFO, ha, "Device state is 0x%x = %s\n", dev_state, dev_state < MAX_STATES ? qdev_state[dev_state] : "Unknown"); /* Force to DEV_COLD unless someone else is starting a reset */ - if (dev_state != QLA82XX_DEV_INITIALIZING) { + if (dev_state != QLA8XXX_DEV_INITIALIZING) { ql4_printk(KERN_INFO, ha, "HW State: COLD/RE-INIT\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA82XX_DEV_COLD); + qla4_82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, QLA8XXX_DEV_COLD); qla4_8xxx_set_rst_ready(ha); } } @@ -2437,9 +2766,104 @@ qla4_8xxx_need_reset_handler(struct scsi_qla_host *ha) void qla4_8xxx_need_qsnt_handler(struct scsi_qla_host *ha) { - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); qla4_8xxx_set_qsnt_ready(ha); - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_unlock(ha); +} + +static void qla4_82xx_set_idc_ver(struct scsi_qla_host *ha) +{ + int idc_ver; + uint32_t drv_active; + + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + if (drv_active == (1 << (ha->func_num * 4))) { + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_IDC_VERSION, + QLA82XX_IDC_VERSION); + ql4_printk(KERN_INFO, ha, + "%s: IDC version updated to %d\n", __func__, + QLA82XX_IDC_VERSION); + } else { + idc_ver = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_IDC_VERSION); + if (QLA82XX_IDC_VERSION != idc_ver) { + ql4_printk(KERN_INFO, ha, + "%s: qla4xxx driver IDC version %d is not compatible with IDC version %d of other drivers!\n", + __func__, QLA82XX_IDC_VERSION, idc_ver); + } + } +} + +static int qla4_83xx_set_idc_ver(struct scsi_qla_host *ha) +{ + int idc_ver; + uint32_t drv_active; + int rval = QLA_SUCCESS; + + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + if (drv_active == (1 << ha->func_num)) { + idc_ver = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_IDC_VERSION); + idc_ver &= (~0xFF); + idc_ver |= QLA83XX_IDC_VER_MAJ_VALUE; + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_IDC_VERSION, idc_ver); + ql4_printk(KERN_INFO, ha, + "%s: IDC version updated to %d\n", __func__, + idc_ver); + } else { + idc_ver = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_IDC_VERSION); + idc_ver &= 0xFF; + if (QLA83XX_IDC_VER_MAJ_VALUE != idc_ver) { + ql4_printk(KERN_INFO, ha, + "%s: qla4xxx driver IDC version %d is not compatible with IDC version %d of other drivers!\n", + __func__, QLA83XX_IDC_VER_MAJ_VALUE, + idc_ver); + rval = QLA_ERROR; + goto exit_set_idc_ver; + } + } + + /* Update IDC_MINOR_VERSION */ + idc_ver = qla4_83xx_rd_reg(ha, QLA83XX_CRB_IDC_VER_MINOR); + idc_ver &= ~(0x03 << (ha->func_num * 2)); + idc_ver |= (QLA83XX_IDC_VER_MIN_VALUE << (ha->func_num * 2)); + qla4_83xx_wr_reg(ha, QLA83XX_CRB_IDC_VER_MINOR, idc_ver); + +exit_set_idc_ver: + return rval; +} + +int qla4_8xxx_update_idc_reg(struct scsi_qla_host *ha) +{ + uint32_t drv_active; + int rval = QLA_SUCCESS; + + if (test_bit(AF_INIT_DONE, &ha->flags)) + goto exit_update_idc_reg; + + ha->isp_ops->idc_lock(ha); + qla4_8xxx_set_drv_active(ha); + + /* + * If we are the first driver to load and + * ql4xdontresethba is not set, clear IDC_CTRL BIT0. + */ + if (is_qla8032(ha)) { + drv_active = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DRV_ACTIVE); + if ((drv_active == (1 << ha->func_num)) && !ql4xdontresethba) + qla4_83xx_clear_idc_dontreset(ha); + } + + if (is_qla8022(ha)) { + qla4_82xx_set_idc_ver(ha); + } else if (is_qla8032(ha)) { + rval = qla4_83xx_set_idc_ver(ha); + if (rval == QLA_ERROR) + qla4_8xxx_clear_drv_active(ha); + } + + ha->isp_ops->idc_unlock(ha); + +exit_update_idc_reg: + return rval; } /** @@ -2454,13 +2878,11 @@ int qla4_8xxx_device_state_handler(struct scsi_qla_host *ha) int rval = QLA_SUCCESS; unsigned long dev_init_timeout; - if (!test_bit(AF_INIT_DONE, &ha->flags)) { - qla4_8xxx_idc_lock(ha); - qla4_8xxx_set_drv_active(ha); - qla4_8xxx_idc_unlock(ha); - } + rval = qla4_8xxx_update_idc_reg(ha); + if (rval == QLA_ERROR) + goto exit_state_handler; - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); + dev_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE); DEBUG2(ql4_printk(KERN_INFO, ha, "Device state is 0x%x = %s\n", dev_state, dev_state < MAX_STATES ? qdev_state[dev_state] : "Unknown")); @@ -2468,7 +2890,7 @@ int qla4_8xxx_device_state_handler(struct scsi_qla_host *ha) /* wait for 30 seconds for device to go ready */ dev_init_timeout = jiffies + (ha->nx_dev_init_timeout * HZ); - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); while (1) { if (time_after_eq(jiffies, dev_init_timeout)) { @@ -2477,65 +2899,75 @@ int qla4_8xxx_device_state_handler(struct scsi_qla_host *ha) DRIVER_NAME, dev_state, dev_state < MAX_STATES ? qdev_state[dev_state] : "Unknown"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); } - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); + dev_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE); ql4_printk(KERN_INFO, ha, "Device state is 0x%x = %s\n", dev_state, dev_state < MAX_STATES ? qdev_state[dev_state] : "Unknown"); /* NOTE: Make sure idc unlocked upon exit of switch statement */ switch (dev_state) { - case QLA82XX_DEV_READY: + case QLA8XXX_DEV_READY: goto exit; - case QLA82XX_DEV_COLD: + case QLA8XXX_DEV_COLD: rval = qla4_8xxx_device_bootstrap(ha); goto exit; - case QLA82XX_DEV_INITIALIZING: - qla4_8xxx_idc_unlock(ha); + case QLA8XXX_DEV_INITIALIZING: + ha->isp_ops->idc_unlock(ha); msleep(1000); - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); break; - case QLA82XX_DEV_NEED_RESET: - if (!ql4xdontresethba) { - qla4_8xxx_need_reset_handler(ha); - /* Update timeout value after need - * reset handler */ - dev_init_timeout = jiffies + - (ha->nx_dev_init_timeout * HZ); - } else { - qla4_8xxx_idc_unlock(ha); - msleep(1000); - qla4_8xxx_idc_lock(ha); + case QLA8XXX_DEV_NEED_RESET: + /* + * For ISP8324, if NEED_RESET is set by any driver, + * it should be honored, irrespective of IDC_CTRL + * DONTRESET_BIT0 + */ + if (is_qla8032(ha)) { + qla4_83xx_need_reset_handler(ha); + } else if (is_qla8022(ha)) { + if (!ql4xdontresethba) { + qla4_82xx_need_reset_handler(ha); + /* Update timeout value after need + * reset handler */ + dev_init_timeout = jiffies + + (ha->nx_dev_init_timeout * HZ); + } else { + ha->isp_ops->idc_unlock(ha); + msleep(1000); + ha->isp_ops->idc_lock(ha); + } } break; - case QLA82XX_DEV_NEED_QUIESCENT: + case QLA8XXX_DEV_NEED_QUIESCENT: /* idc locked/unlocked in handler */ qla4_8xxx_need_qsnt_handler(ha); break; - case QLA82XX_DEV_QUIESCENT: - qla4_8xxx_idc_unlock(ha); + case QLA8XXX_DEV_QUIESCENT: + ha->isp_ops->idc_unlock(ha); msleep(1000); - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); break; - case QLA82XX_DEV_FAILED: - qla4_8xxx_idc_unlock(ha); + case QLA8XXX_DEV_FAILED: + ha->isp_ops->idc_unlock(ha); qla4xxx_dead_adapter_cleanup(ha); rval = QLA_ERROR; - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); goto exit; default: - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_unlock(ha); qla4xxx_dead_adapter_cleanup(ha); rval = QLA_ERROR; - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); goto exit; } } exit: - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_unlock(ha); +exit_state_handler: return rval; } @@ -2544,8 +2976,13 @@ int qla4_8xxx_load_risc(struct scsi_qla_host *ha) int retval; /* clear the interrupt */ - writel(0, &ha->qla4_8xxx_reg->host_int); - readl(&ha->qla4_8xxx_reg->host_int); + if (is_qla8032(ha)) { + writel(0, &ha->qla4_83xx_reg->risc_intr); + readl(&ha->qla4_83xx_reg->risc_intr); + } else if (is_qla8022(ha)) { + writel(0, &ha->qla4_82xx_reg->host_int); + readl(&ha->qla4_82xx_reg->host_int); + } retval = qla4_8xxx_device_state_handler(ha); @@ -2579,13 +3016,13 @@ flash_data_addr(struct ql82xx_hw_data *hw, uint32_t faddr) } static uint32_t * -qla4_8xxx_read_flash_data(struct scsi_qla_host *ha, uint32_t *dwptr, +qla4_82xx_read_flash_data(struct scsi_qla_host *ha, uint32_t *dwptr, uint32_t faddr, uint32_t length) { uint32_t i; uint32_t val; int loops = 0; - while ((qla4_8xxx_rom_lock(ha) != 0) && (loops < 50000)) { + while ((qla4_82xx_rom_lock(ha) != 0) && (loops < 50000)) { udelay(100); cond_resched(); loops++; @@ -2597,7 +3034,7 @@ qla4_8xxx_read_flash_data(struct scsi_qla_host *ha, uint32_t *dwptr, /* Dword reads to flash. */ for (i = 0; i < length/4; i++, faddr += 4) { - if (qla4_8xxx_do_rom_fast_read(ha, faddr, &val)) { + if (qla4_82xx_do_rom_fast_read(ha, faddr, &val)) { ql4_printk(KERN_WARNING, ha, "Do ROM fast read failed\n"); goto done_read; @@ -2606,7 +3043,7 @@ qla4_8xxx_read_flash_data(struct scsi_qla_host *ha, uint32_t *dwptr, } done_read: - qla4_8xxx_rom_unlock(ha); + qla4_82xx_rom_unlock(ha); return dwptr; } @@ -2614,10 +3051,10 @@ done_read: * Address and length are byte address **/ static uint8_t * -qla4_8xxx_read_optrom_data(struct scsi_qla_host *ha, uint8_t *buf, +qla4_82xx_read_optrom_data(struct scsi_qla_host *ha, uint8_t *buf, uint32_t offset, uint32_t length) { - qla4_8xxx_read_flash_data(ha, (uint32_t *)buf, offset, length); + qla4_82xx_read_flash_data(ha, (uint32_t *)buf, offset, length); return buf; } @@ -2644,7 +3081,7 @@ qla4_8xxx_get_flt_info(struct scsi_qla_host *ha, uint32_t flt_addr) const char *loc, *locations[] = { "DEF", "FLT" }; uint16_t *wptr; uint16_t cnt, chksum; - uint32_t start; + uint32_t start, status; struct qla_flt_header *flt; struct qla_flt_region *region; struct ql82xx_hw_data *hw = &ha->hw; @@ -2653,8 +3090,18 @@ qla4_8xxx_get_flt_info(struct scsi_qla_host *ha, uint32_t flt_addr) wptr = (uint16_t *)ha->request_ring; flt = (struct qla_flt_header *)ha->request_ring; region = (struct qla_flt_region *)&flt[1]; - qla4_8xxx_read_optrom_data(ha, (uint8_t *)ha->request_ring, - flt_addr << 2, OPTROM_BURST_SIZE); + + if (is_qla8022(ha)) { + qla4_82xx_read_optrom_data(ha, (uint8_t *)ha->request_ring, + flt_addr << 2, OPTROM_BURST_SIZE); + } else if (is_qla8032(ha)) { + status = qla4_83xx_flash_read_u32(ha, flt_addr << 2, + (uint8_t *)ha->request_ring, + 0x400); + if (status != QLA_SUCCESS) + goto no_flash_data; + } + if (*wptr == __constant_cpu_to_le16(0xffff)) goto no_flash_data; if (flt->version != __constant_cpu_to_le16(1)) { @@ -2730,7 +3177,7 @@ done: } static void -qla4_8xxx_get_fdt_info(struct scsi_qla_host *ha) +qla4_82xx_get_fdt_info(struct scsi_qla_host *ha) { #define FLASH_BLK_SIZE_4K 0x1000 #define FLASH_BLK_SIZE_32K 0x8000 @@ -2748,7 +3195,7 @@ qla4_8xxx_get_fdt_info(struct scsi_qla_host *ha) wptr = (uint16_t *)ha->request_ring; fdt = (struct qla_fdt_layout *)ha->request_ring; - qla4_8xxx_read_optrom_data(ha, (uint8_t *)ha->request_ring, + qla4_82xx_read_optrom_data(ha, (uint8_t *)ha->request_ring, hw->flt_region_fdt << 2, OPTROM_BURST_SIZE); if (*wptr == __constant_cpu_to_le16(0xffff)) @@ -2797,7 +3244,7 @@ done: } static void -qla4_8xxx_get_idc_param(struct scsi_qla_host *ha) +qla4_82xx_get_idc_param(struct scsi_qla_host *ha) { #define QLA82XX_IDC_PARAM_ADDR 0x003e885c uint32_t *wptr; @@ -2805,7 +3252,7 @@ qla4_8xxx_get_idc_param(struct scsi_qla_host *ha) if (!is_qla8022(ha)) return; wptr = (uint32_t *)ha->request_ring; - qla4_8xxx_read_optrom_data(ha, (uint8_t *)ha->request_ring, + qla4_82xx_read_optrom_data(ha, (uint8_t *)ha->request_ring, QLA82XX_IDC_PARAM_ADDR , 8); if (*wptr == __constant_cpu_to_le32(0xffffffff)) { @@ -2823,6 +3270,39 @@ qla4_8xxx_get_idc_param(struct scsi_qla_host *ha) return; } +void qla4_82xx_queue_mbox_cmd(struct scsi_qla_host *ha, uint32_t *mbx_cmd, + int in_count) +{ + int i; + + /* Load all mailbox registers, except mailbox 0. */ + for (i = 1; i < in_count; i++) + writel(mbx_cmd[i], &ha->qla4_82xx_reg->mailbox_in[i]); + + /* Wakeup firmware */ + writel(mbx_cmd[0], &ha->qla4_82xx_reg->mailbox_in[0]); + readl(&ha->qla4_82xx_reg->mailbox_in[0]); + writel(HINT_MBX_INT_PENDING, &ha->qla4_82xx_reg->hint); + readl(&ha->qla4_82xx_reg->hint); +} + +void qla4_82xx_process_mbox_intr(struct scsi_qla_host *ha, int out_count) +{ + int intr_status; + + intr_status = readl(&ha->qla4_82xx_reg->host_int); + if (intr_status & ISRX_82XX_RISC_INT) { + ha->mbox_status_count = out_count; + intr_status = readl(&ha->qla4_82xx_reg->host_status); + ha->isp_ops->interrupt_service_routine(ha, intr_status); + + if (test_bit(AF_INTERRUPTS_ON, &ha->flags) && + test_bit(AF_INTx_ENABLED, &ha->flags)) + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, + 0xfbff); + } +} + int qla4_8xxx_get_flash_info(struct scsi_qla_host *ha) { @@ -2834,8 +3314,12 @@ qla4_8xxx_get_flash_info(struct scsi_qla_host *ha) return ret; qla4_8xxx_get_flt_info(ha, flt_addr); - qla4_8xxx_get_fdt_info(ha); - qla4_8xxx_get_idc_param(ha); + if (is_qla8022(ha)) { + qla4_82xx_get_fdt_info(ha); + qla4_82xx_get_idc_param(ha); + } else if (is_qla8032(ha)) { + qla4_83xx_get_idc_param(ha); + } return QLA_SUCCESS; } @@ -2869,36 +3353,36 @@ qla4_8xxx_stop_firmware(struct scsi_qla_host *ha) } /** - * qla4_8xxx_isp_reset - Resets ISP and aborts all outstanding commands. + * qla4_82xx_isp_reset - Resets ISP and aborts all outstanding commands. * @ha: pointer to host adapter structure. **/ int -qla4_8xxx_isp_reset(struct scsi_qla_host *ha) +qla4_82xx_isp_reset(struct scsi_qla_host *ha) { int rval; uint32_t dev_state; - qla4_8xxx_idc_lock(ha); - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); + qla4_82xx_idc_lock(ha); + dev_state = qla4_82xx_rd_32(ha, QLA82XX_CRB_DEV_STATE); - if (dev_state == QLA82XX_DEV_READY) { + if (dev_state == QLA8XXX_DEV_READY) { ql4_printk(KERN_INFO, ha, "HW State: NEED RESET\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_NEED_RESET); - set_bit(AF_82XX_RST_OWNER, &ha->flags); + qla4_82xx_wr_32(ha, QLA82XX_CRB_DEV_STATE, + QLA8XXX_DEV_NEED_RESET); + set_bit(AF_8XXX_RST_OWNER, &ha->flags); } else ql4_printk(KERN_INFO, ha, "HW State: DEVICE INITIALIZING\n"); - qla4_8xxx_idc_unlock(ha); + qla4_82xx_idc_unlock(ha); rval = qla4_8xxx_device_state_handler(ha); - qla4_8xxx_idc_lock(ha); + qla4_82xx_idc_lock(ha); qla4_8xxx_clear_rst_ready(ha); - qla4_8xxx_idc_unlock(ha); + qla4_82xx_idc_unlock(ha); if (rval == QLA_SUCCESS) { - ql4_printk(KERN_INFO, ha, "Clearing AF_RECOVERY in qla4_8xxx_isp_reset\n"); + ql4_printk(KERN_INFO, ha, "Clearing AF_RECOVERY in qla4_82xx_isp_reset\n"); clear_bit(AF_FW_RECOVERY, &ha->flags); } @@ -2979,8 +3463,7 @@ exit_validate_mac82: /* Interrupt handling helpers. */ -static int -qla4_8xxx_mbx_intr_enable(struct scsi_qla_host *ha) +int qla4_8xxx_mbx_intr_enable(struct scsi_qla_host *ha) { uint32_t mbox_cmd[MBOX_REG_COUNT]; uint32_t mbox_sts[MBOX_REG_COUNT]; @@ -3001,8 +3484,7 @@ qla4_8xxx_mbx_intr_enable(struct scsi_qla_host *ha) return QLA_SUCCESS; } -static int -qla4_8xxx_mbx_intr_disable(struct scsi_qla_host *ha) +int qla4_8xxx_mbx_intr_disable(struct scsi_qla_host *ha) { uint32_t mbox_cmd[MBOX_REG_COUNT]; uint32_t mbox_sts[MBOX_REG_COUNT]; @@ -3025,26 +3507,26 @@ qla4_8xxx_mbx_intr_disable(struct scsi_qla_host *ha) } void -qla4_8xxx_enable_intrs(struct scsi_qla_host *ha) +qla4_82xx_enable_intrs(struct scsi_qla_host *ha) { qla4_8xxx_mbx_intr_enable(ha); spin_lock_irq(&ha->hardware_lock); /* BIT 10 - reset */ - qla4_8xxx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0xfbff); + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0xfbff); spin_unlock_irq(&ha->hardware_lock); set_bit(AF_INTERRUPTS_ON, &ha->flags); } void -qla4_8xxx_disable_intrs(struct scsi_qla_host *ha) +qla4_82xx_disable_intrs(struct scsi_qla_host *ha) { if (test_and_clear_bit(AF_INTERRUPTS_ON, &ha->flags)) qla4_8xxx_mbx_intr_disable(ha); spin_lock_irq(&ha->hardware_lock); /* BIT 10 - set */ - qla4_8xxx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0x0400); + qla4_82xx_wr_32(ha, ha->nx_legacy_intr.tgt_mask_reg, 0x0400); spin_unlock_irq(&ha->hardware_lock); } diff --git a/drivers/scsi/qla4xxx/ql4_nx.h b/drivers/scsi/qla4xxx/ql4_nx.h index 3025847..9dc0bbf 100644 --- a/drivers/scsi/qla4xxx/ql4_nx.h +++ b/drivers/scsi/qla4xxx/ql4_nx.h @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -25,6 +25,8 @@ #define CRB_RCVPEG_STATE QLA82XX_REG(0x13c) #define CRB_DMA_SHIFT QLA82XX_REG(0xcc) #define CRB_TEMP_STATE QLA82XX_REG(0x1b4) +#define CRB_CMDPEG_CHECK_RETRY_COUNT 60 +#define CRB_CMDPEG_CHECK_DELAY 500 #define qla82xx_get_temp_val(x) ((x) >> 16) #define qla82xx_get_temp_state(x) ((x) & 0xffff) @@ -490,8 +492,8 @@ enum { * Base addresses of major components on-chip. * ====================== BASE ADDRESSES ON-CHIP ====================== */ -#define QLA82XX_ADDR_DDR_NET (0x0000000000000000ULL) -#define QLA82XX_ADDR_DDR_NET_MAX (0x000000000fffffffULL) +#define QLA8XXX_ADDR_DDR_NET (0x0000000000000000ULL) +#define QLA8XXX_ADDR_DDR_NET_MAX (0x000000000fffffffULL) /* Imbus address bit used to indicate a host address. This bit is * eliminated by the pcie bar and bar select before presentation @@ -500,14 +502,15 @@ enum { #define QLA82XX_P2_ADDR_PCIE (0x0000000800000000ULL) #define QLA82XX_P3_ADDR_PCIE (0x0000008000000000ULL) #define QLA82XX_ADDR_PCIE_MAX (0x0000000FFFFFFFFFULL) -#define QLA82XX_ADDR_OCM0 (0x0000000200000000ULL) -#define QLA82XX_ADDR_OCM0_MAX (0x00000002000fffffULL) -#define QLA82XX_ADDR_OCM1 (0x0000000200400000ULL) -#define QLA82XX_ADDR_OCM1_MAX (0x00000002004fffffULL) -#define QLA82XX_ADDR_QDR_NET (0x0000000300000000ULL) +#define QLA8XXX_ADDR_OCM0 (0x0000000200000000ULL) +#define QLA8XXX_ADDR_OCM0_MAX (0x00000002000fffffULL) +#define QLA8XXX_ADDR_OCM1 (0x0000000200400000ULL) +#define QLA8XXX_ADDR_OCM1_MAX (0x00000002004fffffULL) +#define QLA8XXX_ADDR_QDR_NET (0x0000000300000000ULL) #define QLA82XX_P2_ADDR_QDR_NET_MAX (0x00000003001fffffULL) #define QLA82XX_P3_ADDR_QDR_NET_MAX (0x0000000303ffffffULL) +#define QLA8XXX_ADDR_QDR_NET_MAX (0x0000000307ffffffULL) #define QLA82XX_PCI_CRBSPACE (unsigned long)0x06000000 #define QLA82XX_PCI_DIRECT_CRB (unsigned long)0x04400000 @@ -517,6 +520,10 @@ enum { #define QLA82XX_PCI_QDR_NET (unsigned long)0x04000000 #define QLA82XX_PCI_QDR_NET_MAX (unsigned long)0x043fffff +/* PCI Windowing for DDR regions. */ +#define QLA8XXX_ADDR_IN_RANGE(addr, low, high) \ + (((addr) <= (high)) && ((addr) >= (low))) + /* * Register offsets for MN */ @@ -540,6 +547,11 @@ enum { #define MIU_TA_CTL_WRITE 4 #define MIU_TA_CTL_BUSY 8 +#define MIU_TA_CTL_WRITE_ENABLE (MIU_TA_CTL_WRITE | MIU_TA_CTL_ENABLE) +#define MIU_TA_CTL_WRITE_START (MIU_TA_CTL_WRITE | MIU_TA_CTL_ENABLE |\ + MIU_TA_CTL_START) +#define MIU_TA_CTL_START_ENABLE (MIU_TA_CTL_START | MIU_TA_CTL_ENABLE) + /*CAM RAM */ # define QLA82XX_CAM_RAM_BASE (QLA82XX_CRB_CAM + 0x02000) # define QLA82XX_CAM_RAM(reg) (QLA82XX_CAM_RAM_BASE + (reg)) @@ -565,20 +577,53 @@ enum { /* Driver Coexistence Defines */ #define QLA82XX_CRB_DRV_ACTIVE (QLA82XX_CAM_RAM(0x138)) #define QLA82XX_CRB_DEV_STATE (QLA82XX_CAM_RAM(0x140)) -#define QLA82XX_CRB_DEV_PART_INFO (QLA82XX_CAM_RAM(0x14c)) -#define QLA82XX_CRB_DRV_IDC_VERSION (QLA82XX_CAM_RAM(0x174)) #define QLA82XX_CRB_DRV_STATE (QLA82XX_CAM_RAM(0x144)) #define QLA82XX_CRB_DRV_SCRATCH (QLA82XX_CAM_RAM(0x148)) #define QLA82XX_CRB_DEV_PART_INFO (QLA82XX_CAM_RAM(0x14c)) +#define QLA82XX_CRB_DRV_IDC_VERSION (QLA82XX_CAM_RAM(0x174)) + +enum qla_regs { + QLA8XXX_PEG_HALT_STATUS1 = 0, + QLA8XXX_PEG_HALT_STATUS2, + QLA8XXX_PEG_ALIVE_COUNTER, + QLA8XXX_CRB_DRV_ACTIVE, + QLA8XXX_CRB_DEV_STATE, + QLA8XXX_CRB_DRV_STATE, + QLA8XXX_CRB_DRV_SCRATCH, + QLA8XXX_CRB_DEV_PART_INFO, + QLA8XXX_CRB_DRV_IDC_VERSION, + QLA8XXX_FW_VERSION_MAJOR, + QLA8XXX_FW_VERSION_MINOR, + QLA8XXX_FW_VERSION_SUB, + QLA8XXX_CRB_CMDPEG_STATE, + QLA8XXX_CRB_TEMP_STATE, +}; + +static const uint32_t qla4_82xx_reg_tbl[] = { + QLA82XX_PEG_HALT_STATUS1, + QLA82XX_PEG_HALT_STATUS2, + QLA82XX_PEG_ALIVE_COUNTER, + QLA82XX_CRB_DRV_ACTIVE, + QLA82XX_CRB_DEV_STATE, + QLA82XX_CRB_DRV_STATE, + QLA82XX_CRB_DRV_SCRATCH, + QLA82XX_CRB_DEV_PART_INFO, + QLA82XX_CRB_DRV_IDC_VERSION, + QLA82XX_FW_VERSION_MAJOR, + QLA82XX_FW_VERSION_MINOR, + QLA82XX_FW_VERSION_SUB, + CRB_CMDPEG_STATE, + CRB_TEMP_STATE, +}; /* Every driver should use these Device State */ -#define QLA82XX_DEV_COLD 1 -#define QLA82XX_DEV_INITIALIZING 2 -#define QLA82XX_DEV_READY 3 -#define QLA82XX_DEV_NEED_RESET 4 -#define QLA82XX_DEV_NEED_QUIESCENT 5 -#define QLA82XX_DEV_FAILED 6 -#define QLA82XX_DEV_QUIESCENT 7 +#define QLA8XXX_DEV_COLD 1 +#define QLA8XXX_DEV_INITIALIZING 2 +#define QLA8XXX_DEV_READY 3 +#define QLA8XXX_DEV_NEED_RESET 4 +#define QLA8XXX_DEV_NEED_QUIESCENT 5 +#define QLA8XXX_DEV_FAILED 6 +#define QLA8XXX_DEV_QUIESCENT 7 #define MAX_STATES 8 /* Increment if new state added */ #define QLA82XX_IDC_VERSION 0x1 @@ -795,47 +840,51 @@ struct crb_addr_pair { /* Minidump related */ /* Entry Type Defines */ -#define QLA82XX_RDNOP 0 -#define QLA82XX_RDCRB 1 -#define QLA82XX_RDMUX 2 -#define QLA82XX_QUEUE 3 -#define QLA82XX_BOARD 4 -#define QLA82XX_RDOCM 6 -#define QLA82XX_PREGS 7 -#define QLA82XX_L1DTG 8 -#define QLA82XX_L1ITG 9 -#define QLA82XX_L1DAT 11 -#define QLA82XX_L1INS 12 -#define QLA82XX_L2DTG 21 -#define QLA82XX_L2ITG 22 -#define QLA82XX_L2DAT 23 -#define QLA82XX_L2INS 24 -#define QLA82XX_RDROM 71 -#define QLA82XX_RDMEM 72 -#define QLA82XX_CNTRL 98 -#define QLA82XX_RDEND 255 +#define QLA8XXX_RDNOP 0 +#define QLA8XXX_RDCRB 1 +#define QLA8XXX_RDMUX 2 +#define QLA8XXX_QUEUE 3 +#define QLA8XXX_BOARD 4 +#define QLA8XXX_RDOCM 6 +#define QLA8XXX_PREGS 7 +#define QLA8XXX_L1DTG 8 +#define QLA8XXX_L1ITG 9 +#define QLA8XXX_L1DAT 11 +#define QLA8XXX_L1INS 12 +#define QLA8XXX_L2DTG 21 +#define QLA8XXX_L2ITG 22 +#define QLA8XXX_L2DAT 23 +#define QLA8XXX_L2INS 24 +#define QLA83XX_POLLRD 35 +#define QLA83XX_RDMUX2 36 +#define QLA83XX_POLLRDMWR 37 +#define QLA8XXX_RDROM 71 +#define QLA8XXX_RDMEM 72 +#define QLA8XXX_CNTRL 98 +#define QLA83XX_TLHDR 99 +#define QLA8XXX_RDEND 255 /* Opcodes for Control Entries. * These Flags are bit fields. */ -#define QLA82XX_DBG_OPCODE_WR 0x01 -#define QLA82XX_DBG_OPCODE_RW 0x02 -#define QLA82XX_DBG_OPCODE_AND 0x04 -#define QLA82XX_DBG_OPCODE_OR 0x08 -#define QLA82XX_DBG_OPCODE_POLL 0x10 -#define QLA82XX_DBG_OPCODE_RDSTATE 0x20 -#define QLA82XX_DBG_OPCODE_WRSTATE 0x40 -#define QLA82XX_DBG_OPCODE_MDSTATE 0x80 +#define QLA8XXX_DBG_OPCODE_WR 0x01 +#define QLA8XXX_DBG_OPCODE_RW 0x02 +#define QLA8XXX_DBG_OPCODE_AND 0x04 +#define QLA8XXX_DBG_OPCODE_OR 0x08 +#define QLA8XXX_DBG_OPCODE_POLL 0x10 +#define QLA8XXX_DBG_OPCODE_RDSTATE 0x20 +#define QLA8XXX_DBG_OPCODE_WRSTATE 0x40 +#define QLA8XXX_DBG_OPCODE_MDSTATE 0x80 /* Driver Flags */ -#define QLA82XX_DBG_SKIPPED_FLAG 0x80 /* driver skipped this entry */ -#define QLA82XX_DBG_SIZE_ERR_FLAG 0x40 /* Entry vs Capture size +#define QLA8XXX_DBG_SKIPPED_FLAG 0x80 /* driver skipped this entry */ +#define QLA8XXX_DBG_SIZE_ERR_FLAG 0x40 /* Entry vs Capture size * mismatch */ /* Driver_code is for driver to write some info about the entry * currently not used. */ -struct qla82xx_minidump_entry_hdr { +struct qla8xxx_minidump_entry_hdr { uint32_t entry_type; uint32_t entry_size; uint32_t entry_capture_size; @@ -848,8 +897,8 @@ struct qla82xx_minidump_entry_hdr { }; /* Read CRB entry header */ -struct qla82xx_minidump_entry_crb { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_crb { + struct qla8xxx_minidump_entry_hdr h; uint32_t addr; struct { uint8_t addr_stride; @@ -871,8 +920,8 @@ struct qla82xx_minidump_entry_crb { uint32_t value_3; }; -struct qla82xx_minidump_entry_cache { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_cache { + struct qla8xxx_minidump_entry_hdr h; uint32_t tag_reg_addr; struct { uint16_t tag_value_stride; @@ -895,8 +944,8 @@ struct qla82xx_minidump_entry_cache { }; /* Read OCM */ -struct qla82xx_minidump_entry_rdocm { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_rdocm { + struct qla8xxx_minidump_entry_hdr h; uint32_t rsvd_0; uint32_t rsvd_1; uint32_t data_size; @@ -908,24 +957,24 @@ struct qla82xx_minidump_entry_rdocm { }; /* Read Memory */ -struct qla82xx_minidump_entry_rdmem { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_rdmem { + struct qla8xxx_minidump_entry_hdr h; uint32_t rsvd[6]; uint32_t read_addr; uint32_t read_data_size; }; /* Read ROM */ -struct qla82xx_minidump_entry_rdrom { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_rdrom { + struct qla8xxx_minidump_entry_hdr h; uint32_t rsvd[6]; uint32_t read_addr; uint32_t read_data_size; }; /* Mux entry */ -struct qla82xx_minidump_entry_mux { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_mux { + struct qla8xxx_minidump_entry_hdr h; uint32_t select_addr; uint32_t rsvd_0; uint32_t data_size; @@ -937,8 +986,8 @@ struct qla82xx_minidump_entry_mux { }; /* Queue entry */ -struct qla82xx_minidump_entry_queue { - struct qla82xx_minidump_entry_hdr h; +struct qla8xxx_minidump_entry_queue { + struct qla8xxx_minidump_entry_hdr h; uint32_t select_addr; struct { uint16_t queue_id_stride; @@ -956,23 +1005,6 @@ struct qla82xx_minidump_entry_queue { } rd_strd; }; -#define QLA82XX_MINIDUMP_OCM0_SIZE (256 * 1024) -#define QLA82XX_MINIDUMP_L1C_SIZE (256 * 1024) -#define QLA82XX_MINIDUMP_L2C_SIZE 1572864 -#define QLA82XX_MINIDUMP_COMMON_STR_SIZE 0 -#define QLA82XX_MINIDUMP_FCOE_STR_SIZE 0 -#define QLA82XX_MINIDUMP_MEM_SIZE 0 -#define QLA82XX_MAX_ENTRY_HDR 4 - -struct qla82xx_minidump { - uint32_t md_ocm0_data[QLA82XX_MINIDUMP_OCM0_SIZE]; - uint32_t md_l1c_data[QLA82XX_MINIDUMP_L1C_SIZE]; - uint32_t md_l2c_data[QLA82XX_MINIDUMP_L2C_SIZE]; - uint32_t md_cs_data[QLA82XX_MINIDUMP_COMMON_STR_SIZE]; - uint32_t md_fcoes_data[QLA82XX_MINIDUMP_FCOE_STR_SIZE]; - uint32_t md_mem_data[QLA82XX_MINIDUMP_MEM_SIZE]; -}; - #define MBC_DIAGNOSTIC_MINIDUMP_TEMPLATE 0x129 #define RQST_TMPLT_SIZE 0x0 #define RQST_TMPLT 0x1 @@ -982,6 +1014,16 @@ struct qla82xx_minidump { #define MD_MIU_TEST_AGT_ADDR_LO 0x41000094 #define MD_MIU_TEST_AGT_ADDR_HI 0x41000098 +#define MD_MIU_TEST_AGT_WRDATA_LO 0x410000A0 +#define MD_MIU_TEST_AGT_WRDATA_HI 0x410000A4 +#define MD_MIU_TEST_AGT_WRDATA_ULO 0x410000B0 +#define MD_MIU_TEST_AGT_WRDATA_UHI 0x410000B4 + +#define MD_MIU_TEST_AGT_RDDATA_LO 0x410000A8 +#define MD_MIU_TEST_AGT_RDDATA_HI 0x410000AC +#define MD_MIU_TEST_AGT_RDDATA_ULO 0x410000B8 +#define MD_MIU_TEST_AGT_RDDATA_UHI 0x410000BC + static const int MD_MIU_TEST_AGT_RDDATA[] = { 0x410000A8, 0x410000AC, 0x410000B8, 0x410000BC }; #endif diff --git a/drivers/scsi/qla4xxx/ql4_os.c b/drivers/scsi/qla4xxx/ql4_os.c index 79243b7..fbc546e 100644 --- a/drivers/scsi/qla4xxx/ql4_os.c +++ b/drivers/scsi/qla4xxx/ql4_os.c @@ -1,6 +1,6 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ @@ -18,6 +18,7 @@ #include "ql4_glbl.h" #include "ql4_dbg.h" #include "ql4_inline.h" +#include "ql4_83xx.h" /* * Driver version @@ -160,7 +161,7 @@ static int qla4xxx_eh_host_reset(struct scsi_cmnd *cmd); static int qla4xxx_slave_alloc(struct scsi_device *device); static int qla4xxx_slave_configure(struct scsi_device *device); static void qla4xxx_slave_destroy(struct scsi_device *sdev); -static umode_t ql4_attr_is_visible(int param_type, int param); +static umode_t qla4_attr_is_visible(int param_type, int param); static int qla4xxx_host_reset(struct Scsi_Host *shost, int reset_type); static int qla4xxx_change_queue_depth(struct scsi_device *sdev, int qdepth, int reason); @@ -203,7 +204,7 @@ static struct iscsi_transport qla4xxx_iscsi_transport = { CAP_DATA_PATH_OFFLOAD | CAP_HDRDGST | CAP_DATADGST | CAP_LOGIN_OFFLOAD | CAP_MULTI_R2T, - .attr_is_visible = ql4_attr_is_visible, + .attr_is_visible = qla4_attr_is_visible, .create_session = qla4xxx_session_create, .destroy_session = qla4xxx_session_destroy, .start_conn = qla4xxx_conn_start, @@ -315,7 +316,7 @@ exit_send_ping: return rval; } -static umode_t ql4_attr_is_visible(int param_type, int param) +static umode_t qla4_attr_is_visible(int param_type, int param) { switch (param_type) { case ISCSI_HOST_PARAM: @@ -1366,7 +1367,7 @@ static int qla4xxx_conn_get_param(struct iscsi_cls_conn *cls_conn, conn = cls_conn->dd_data; qla_conn = conn->dd_data; - dst_addr = &qla_conn->qla_ep->dst_addr; + dst_addr = (struct sockaddr *)&qla_conn->qla_ep->dst_addr; switch (param) { case ISCSI_PARAM_CONN_PORT: @@ -2315,8 +2316,17 @@ static void qla4xxx_mem_free(struct scsi_qla_host *ha) if (ha->nx_pcibase) iounmap( (struct device_reg_82xx __iomem *)ha->nx_pcibase); - } else if (ha->reg) + } else if (is_qla8032(ha)) { + if (ha->nx_pcibase) + iounmap( + (struct device_reg_83xx __iomem *)ha->nx_pcibase); + } else if (ha->reg) { iounmap(ha->reg); + } + + if (ha->reset_tmplt.buff) + vfree(ha->reset_tmplt.buff); + pci_release_regions(ha->pdev); } @@ -2420,7 +2430,7 @@ static int qla4_8xxx_check_temp(struct scsi_qla_host *ha) uint32_t temp, temp_state, temp_val; int status = QLA_SUCCESS; - temp = qla4_8xxx_rd_32(ha, CRB_TEMP_STATE); + temp = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_TEMP_STATE); temp_state = qla82xx_get_temp_state(temp); temp_val = qla82xx_get_temp_val(temp); @@ -2456,7 +2466,8 @@ static int qla4_8xxx_check_fw_alive(struct scsi_qla_host *ha) uint32_t fw_heartbeat_counter; int status = QLA_SUCCESS; - fw_heartbeat_counter = qla4_8xxx_rd_32(ha, QLA82XX_PEG_ALIVE_COUNTER); + fw_heartbeat_counter = qla4_8xxx_rd_direct(ha, + QLA8XXX_PEG_ALIVE_COUNTER); /* If PEG_ALIVE_COUNTER is 0xffffffff, AER/EEH is in progress, ignore */ if (fw_heartbeat_counter == 0xffffffff) { DEBUG2(printk(KERN_WARNING "scsi%ld: %s: Device in frozen " @@ -2470,28 +2481,7 @@ static int qla4_8xxx_check_fw_alive(struct scsi_qla_host *ha) /* FW not alive after 2 seconds */ if (ha->seconds_since_last_heartbeat == 2) { ha->seconds_since_last_heartbeat = 0; - - ql4_printk(KERN_INFO, ha, - "scsi(%ld): %s, Dumping hw/fw registers:\n " - " PEG_HALT_STATUS1: 0x%x, PEG_HALT_STATUS2:" - " 0x%x,\n PEG_NET_0_PC: 0x%x, PEG_NET_1_PC:" - " 0x%x,\n PEG_NET_2_PC: 0x%x, PEG_NET_3_PC:" - " 0x%x,\n PEG_NET_4_PC: 0x%x\n", - ha->host_no, __func__, - qla4_8xxx_rd_32(ha, - QLA82XX_PEG_HALT_STATUS1), - qla4_8xxx_rd_32(ha, - QLA82XX_PEG_HALT_STATUS2), - qla4_8xxx_rd_32(ha, QLA82XX_CRB_PEG_NET_0 + - 0x3c), - qla4_8xxx_rd_32(ha, QLA82XX_CRB_PEG_NET_1 + - 0x3c), - qla4_8xxx_rd_32(ha, QLA82XX_CRB_PEG_NET_2 + - 0x3c), - qla4_8xxx_rd_32(ha, QLA82XX_CRB_PEG_NET_3 + - 0x3c), - qla4_8xxx_rd_32(ha, QLA82XX_CRB_PEG_NET_4 + - 0x3c)); + qla4_8xxx_dump_peg_reg(ha); status = QLA_ERROR; } } else @@ -2501,6 +2491,48 @@ static int qla4_8xxx_check_fw_alive(struct scsi_qla_host *ha) return status; } +static void qla4_8xxx_process_fw_error(struct scsi_qla_host *ha) +{ + uint32_t halt_status; + int halt_status_unrecoverable = 0; + + halt_status = qla4_8xxx_rd_direct(ha, QLA8XXX_PEG_HALT_STATUS1); + + if (is_qla8022(ha)) { + ql4_printk(KERN_INFO, ha, "%s: disabling pause transmit on port 0 & 1.\n", + __func__); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x98, + CRB_NIU_XG_PAUSE_CTL_P0 | + CRB_NIU_XG_PAUSE_CTL_P1); + + if (QLA82XX_FWERROR_CODE(halt_status) == 0x67) + ql4_printk(KERN_ERR, ha, "%s: Firmware aborted with error code 0x00006700. Device is being reset\n", + __func__); + if (halt_status & HALT_STATUS_UNRECOVERABLE) + halt_status_unrecoverable = 1; + } else if (is_qla8032(ha)) { + if (halt_status & QLA83XX_HALT_STATUS_FW_RESET) + ql4_printk(KERN_ERR, ha, "%s: Firmware error detected device is being reset\n", + __func__); + else if (halt_status & QLA83XX_HALT_STATUS_UNRECOVERABLE) + halt_status_unrecoverable = 1; + } + + /* + * Since we cannot change dev_state in interrupt context, + * set appropriate DPC flag then wakeup DPC + */ + if (halt_status_unrecoverable) { + set_bit(DPC_HA_UNRECOVERABLE, &ha->dpc_flags); + } else { + ql4_printk(KERN_INFO, ha, "%s: detect abort needed!\n", + __func__); + set_bit(DPC_RESET_HA, &ha->dpc_flags); + } + qla4xxx_mailbox_premature_completion(ha); + qla4xxx_wake_dpc(ha); +} + /** * qla4_8xxx_watchdog - Poll dev state * @ha: Pointer to host adapter structure. @@ -2509,31 +2541,33 @@ static int qla4_8xxx_check_fw_alive(struct scsi_qla_host *ha) **/ void qla4_8xxx_watchdog(struct scsi_qla_host *ha) { - uint32_t dev_state, halt_status; + uint32_t dev_state; /* don't poll if reset is going on */ if (!(test_bit(DPC_RESET_ACTIVE, &ha->dpc_flags) || test_bit(DPC_RESET_HA, &ha->dpc_flags) || test_bit(DPC_RETRY_RESET_HA, &ha->dpc_flags))) { - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); + dev_state = qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE); if (qla4_8xxx_check_temp(ha)) { - ql4_printk(KERN_INFO, ha, "disabling pause" - " transmit on port 0 & 1.\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x98, - CRB_NIU_XG_PAUSE_CTL_P0 | - CRB_NIU_XG_PAUSE_CTL_P1); + if (is_qla8022(ha)) { + ql4_printk(KERN_INFO, ha, "disabling pause transmit on port 0 & 1.\n"); + qla4_82xx_wr_32(ha, QLA82XX_CRB_NIU + 0x98, + CRB_NIU_XG_PAUSE_CTL_P0 | + CRB_NIU_XG_PAUSE_CTL_P1); + } set_bit(DPC_HA_UNRECOVERABLE, &ha->dpc_flags); qla4xxx_wake_dpc(ha); - } else if (dev_state == QLA82XX_DEV_NEED_RESET && - !test_bit(DPC_RESET_HA, &ha->dpc_flags)) { - if (!ql4xdontresethba) { + } else if (dev_state == QLA8XXX_DEV_NEED_RESET && + !test_bit(DPC_RESET_HA, &ha->dpc_flags)) { + if (is_qla8032(ha) || + (is_qla8022(ha) && !ql4xdontresethba)) { ql4_printk(KERN_INFO, ha, "%s: HW State: " "NEED RESET!\n", __func__); set_bit(DPC_RESET_HA, &ha->dpc_flags); qla4xxx_wake_dpc(ha); } - } else if (dev_state == QLA82XX_DEV_NEED_QUIESCENT && + } else if (dev_state == QLA8XXX_DEV_NEED_QUIESCENT && !test_bit(DPC_HA_NEED_QUIESCENT, &ha->dpc_flags)) { ql4_printk(KERN_INFO, ha, "%s: HW State: NEED QUIES!\n", __func__); @@ -2541,36 +2575,8 @@ void qla4_8xxx_watchdog(struct scsi_qla_host *ha) qla4xxx_wake_dpc(ha); } else { /* Check firmware health */ - if (qla4_8xxx_check_fw_alive(ha)) { - ql4_printk(KERN_INFO, ha, "disabling pause" - " transmit on port 0 & 1.\n"); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_NIU + 0x98, - CRB_NIU_XG_PAUSE_CTL_P0 | - CRB_NIU_XG_PAUSE_CTL_P1); - halt_status = qla4_8xxx_rd_32(ha, - QLA82XX_PEG_HALT_STATUS1); - - if (QLA82XX_FWERROR_CODE(halt_status) == 0x67) - ql4_printk(KERN_ERR, ha, "%s:" - " Firmware aborted with" - " error code 0x00006700." - " Device is being reset\n", - __func__); - - /* Since we cannot change dev_state in interrupt - * context, set appropriate DPC flag then wakeup - * DPC */ - if (halt_status & HALT_STATUS_UNRECOVERABLE) - set_bit(DPC_HA_UNRECOVERABLE, - &ha->dpc_flags); - else { - ql4_printk(KERN_INFO, ha, "%s: detect " - "abort needed!\n", __func__); - set_bit(DPC_RESET_HA, &ha->dpc_flags); - } - qla4xxx_mailbox_premature_completion(ha); - qla4xxx_wake_dpc(ha); - } + if (qla4_8xxx_check_fw_alive(ha)) + qla4_8xxx_process_fw_error(ha); } } } @@ -2652,11 +2658,10 @@ static void qla4xxx_timer(struct scsi_qla_host *ha) if (!pci_channel_offline(ha->pdev)) pci_read_config_word(ha->pdev, PCI_VENDOR_ID, &w); - if (is_qla8022(ha)) { + if (is_qla80XX(ha)) qla4_8xxx_watchdog(ha); - } - if (!is_qla8022(ha)) { + if (is_qla40XX(ha)) { /* Check for heartbeat interval. */ if (ha->firmware_options & FWOPT_HEARTBEAT_ENABLE && ha->heartbeat_interval != 0) { @@ -2941,6 +2946,14 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha) set_bit(DPC_RESET_ACTIVE, &ha->dpc_flags); + if (is_qla8032(ha) && + !test_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags)) { + ql4_printk(KERN_INFO, ha, "%s: disabling pause transmit on port 0 & 1.\n", + __func__); + /* disable pause frame for ISP83xx */ + qla4_83xx_disable_pause(ha); + } + iscsi_host_for_each_session(ha->host, qla4xxx_fail_session); if (test_bit(DPC_RESET_HA, &ha->dpc_flags)) @@ -2953,9 +2966,9 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha) goto recover_ha_init_adapter; } - /* For the ISP-82xx adapter, issue a stop_firmware if invoked + /* For the ISP-8xxx adapter, issue a stop_firmware if invoked * from eh_host_reset or ioctl module */ - if (is_qla8022(ha) && !reset_chip && + if (is_qla80XX(ha) && !reset_chip && test_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags)) { DEBUG2(ql4_printk(KERN_INFO, ha, @@ -2978,13 +2991,13 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha) } /* Issue full chip reset if recovering from a catastrophic error, - * or if stop_firmware fails for ISP-82xx. + * or if stop_firmware fails for ISP-8xxx. * This is the default case for ISP-4xxx */ - if (!is_qla8022(ha) || reset_chip) { - if (!is_qla8022(ha)) + if (is_qla40XX(ha) || reset_chip) { + if (is_qla40XX(ha)) goto chip_reset; - /* Check if 82XX firmware is alive or not + /* Check if 8XXX firmware is alive or not * We may have arrived here from NEED_RESET * detection only */ if (test_bit(AF_FW_RECOVERY, &ha->flags)) @@ -3000,10 +3013,10 @@ static int qla4xxx_recover_adapter(struct scsi_qla_host *ha) set_current_state(TASK_UNINTERRUPTIBLE); schedule_timeout(HZ); } - +chip_reset: if (!test_bit(AF_FW_RECOVERY, &ha->flags)) qla4xxx_cmd_wait(ha); -chip_reset: + qla4xxx_process_aen(ha, FLUSH_DDB_CHANGED_AENS); qla4xxx_abort_active_cmds(ha, DID_RESET << 16); DEBUG2(ql4_printk(KERN_INFO, ha, @@ -3021,7 +3034,7 @@ recover_ha_init_adapter: /* For ISP-4xxx, force function 1 to always initialize * before function 3 to prevent both funcions from * stepping on top of the other */ - if (!is_qla8022(ha) && (ha->mac_index == 3)) + if (is_qla40XX(ha) && (ha->mac_index == 3)) ssleep(6); /* NOTE: AF_ONLINE flag set upon successful completion of @@ -3039,11 +3052,12 @@ recover_ha_init_adapter: * Since we don't want to block the DPC for too long * with multiple resets in the same thread, * utilize DPC to retry */ - if (is_qla8022(ha)) { - qla4_8xxx_idc_lock(ha); - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); - qla4_8xxx_idc_unlock(ha); - if (dev_state == QLA82XX_DEV_FAILED) { + if (is_qla80XX(ha)) { + ha->isp_ops->idc_lock(ha); + dev_state = qla4_8xxx_rd_direct(ha, + QLA8XXX_CRB_DEV_STATE); + ha->isp_ops->idc_unlock(ha); + if (dev_state == QLA8XXX_DEV_FAILED) { ql4_printk(KERN_INFO, ha, "%s: don't retry " "recover adapter. H/W is in Failed " "state\n", __func__); @@ -3168,6 +3182,7 @@ int qla4xxx_unblock_ddb(struct iscsi_cls_session *cls_session) struct iscsi_session *sess; struct ddb_entry *ddb_entry; struct scsi_qla_host *ha; + int status = QLA_SUCCESS; sess = cls_session->dd_data; ddb_entry = sess->dd_data; @@ -3175,11 +3190,20 @@ int qla4xxx_unblock_ddb(struct iscsi_cls_session *cls_session) ql4_printk(KERN_INFO, ha, "scsi%ld: %s: ddb[%d]" " unblock user space session\n", ha->host_no, __func__, ddb_entry->fw_ddb_index); - iscsi_conn_start(ddb_entry->conn); - iscsi_conn_login_event(ddb_entry->conn, - ISCSI_CONN_STATE_LOGGED_IN); - return QLA_SUCCESS; + if (!iscsi_is_session_online(cls_session)) { + iscsi_conn_start(ddb_entry->conn); + iscsi_conn_login_event(ddb_entry->conn, + ISCSI_CONN_STATE_LOGGED_IN); + } else { + ql4_printk(KERN_INFO, ha, + "scsi%ld: %s: ddb[%d] session [%d] already logged in\n", + ha->host_no, __func__, ddb_entry->fw_ddb_index, + cls_session->sid); + status = QLA_ERROR; + } + + return status; } static void qla4xxx_relogin_all_devices(struct scsi_qla_host *ha) @@ -3373,15 +3397,26 @@ static void qla4xxx_do_dpc(struct work_struct *work) /* post events to application */ qla4xxx_do_work(ha); - if (is_qla8022(ha)) { + if (is_qla80XX(ha)) { if (test_bit(DPC_HA_UNRECOVERABLE, &ha->dpc_flags)) { - qla4_8xxx_idc_lock(ha); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); - qla4_8xxx_idc_unlock(ha); + if (is_qla8032(ha)) { + ql4_printk(KERN_INFO, ha, "%s: disabling pause transmit on port 0 & 1.\n", + __func__); + /* disable pause frame for ISP83xx */ + qla4_83xx_disable_pause(ha); + } + + ha->isp_ops->idc_lock(ha); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); + ha->isp_ops->idc_unlock(ha); ql4_printk(KERN_INFO, ha, "HW State: FAILED\n"); qla4_8xxx_device_state_handler(ha); } + + if (test_and_clear_bit(DPC_POST_IDC_ACK, &ha->dpc_flags)) + qla4_83xx_post_idc_ack(ha); + if (test_and_clear_bit(DPC_HA_NEED_QUIESCENT, &ha->dpc_flags)) { qla4_8xxx_need_qsnt_handler(ha); } @@ -3391,7 +3426,8 @@ static void qla4xxx_do_dpc(struct work_struct *work) (test_bit(DPC_RESET_HA, &ha->dpc_flags) || test_bit(DPC_RESET_HA_INTR, &ha->dpc_flags) || test_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags))) { - if (ql4xdontresethba) { + if ((is_qla8022(ha) && ql4xdontresethba) || + (is_qla8032(ha) && qla4_83xx_idc_dontreset(ha))) { DEBUG2(printk("scsi%ld: %s: Don't Reset HBA\n", ha->host_no, __func__)); clear_bit(DPC_RESET_HA, &ha->dpc_flags); @@ -3477,6 +3513,18 @@ static void qla4xxx_free_adapter(struct scsi_qla_host *ha) ha->isp_ops->disable_intrs(ha); } + if (is_qla40XX(ha)) { + writel(set_rmask(CSR_SCSI_PROCESSOR_INTR), + &ha->reg->ctrl_status); + readl(&ha->reg->ctrl_status); + } else if (is_qla8022(ha)) { + writel(0, &ha->qla4_82xx_reg->host_int); + readl(&ha->qla4_82xx_reg->host_int); + } else if (is_qla8032(ha)) { + writel(0, &ha->qla4_83xx_reg->risc_intr); + readl(&ha->qla4_83xx_reg->risc_intr); + } + /* Remove timer thread, if present */ if (ha->timer_active) qla4xxx_stop_timer(ha); @@ -3492,10 +3540,10 @@ static void qla4xxx_free_adapter(struct scsi_qla_host *ha) /* Put firmware in known state */ ha->isp_ops->reset_firmware(ha); - if (is_qla8022(ha)) { - qla4_8xxx_idc_lock(ha); + if (is_qla80XX(ha)) { + ha->isp_ops->idc_lock(ha); qla4_8xxx_clear_drv_active(ha); - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_unlock(ha); } /* Detach interrupts */ @@ -3542,16 +3590,20 @@ int qla4_8xxx_iospace_config(struct scsi_qla_host *ha) /* Mapping of IO base pointer, door bell read and write pointer */ /* mapping of IO base pointer */ - ha->qla4_8xxx_reg = - (struct device_reg_82xx __iomem *)((uint8_t *)ha->nx_pcibase + - 0xbc000 + (ha->pdev->devfn << 11)); + if (is_qla8022(ha)) { + ha->qla4_82xx_reg = (struct device_reg_82xx __iomem *) + ((uint8_t *)ha->nx_pcibase + 0xbc000 + + (ha->pdev->devfn << 11)); + ha->nx_db_wr_ptr = (ha->pdev->devfn == 4 ? QLA82XX_CAM_RAM_DB1 : + QLA82XX_CAM_RAM_DB2); + } else if (is_qla8032(ha)) { + ha->qla4_83xx_reg = (struct device_reg_83xx __iomem *) + ((uint8_t *)ha->nx_pcibase); + } db_base = pci_resource_start(pdev, 4); /* doorbell is on bar 4 */ db_len = pci_resource_len(pdev, 4); - ha->nx_db_wr_ptr = (ha->pdev->devfn == 4 ? QLA82XX_CAM_RAM_DB1 : - QLA82XX_CAM_RAM_DB2); - return 0; iospace_error_exit: return -ENOMEM; @@ -3639,23 +3691,64 @@ static struct isp_operations qla4xxx_isp_ops = { .rd_shdw_req_q_out = qla4xxx_rd_shdw_req_q_out, .rd_shdw_rsp_q_in = qla4xxx_rd_shdw_rsp_q_in, .get_sys_info = qla4xxx_get_sys_info, + .queue_mailbox_command = qla4xxx_queue_mbox_cmd, + .process_mailbox_interrupt = qla4xxx_process_mbox_intr, }; -static struct isp_operations qla4_8xxx_isp_ops = { +static struct isp_operations qla4_82xx_isp_ops = { .iospace_config = qla4_8xxx_iospace_config, .pci_config = qla4_8xxx_pci_config, - .disable_intrs = qla4_8xxx_disable_intrs, - .enable_intrs = qla4_8xxx_enable_intrs, + .disable_intrs = qla4_82xx_disable_intrs, + .enable_intrs = qla4_82xx_enable_intrs, .start_firmware = qla4_8xxx_load_risc, - .intr_handler = qla4_8xxx_intr_handler, - .interrupt_service_routine = qla4_8xxx_interrupt_service_routine, - .reset_chip = qla4_8xxx_isp_reset, + .restart_firmware = qla4_82xx_try_start_fw, + .intr_handler = qla4_82xx_intr_handler, + .interrupt_service_routine = qla4_82xx_interrupt_service_routine, + .need_reset = qla4_8xxx_need_reset, + .reset_chip = qla4_82xx_isp_reset, .reset_firmware = qla4_8xxx_stop_firmware, - .queue_iocb = qla4_8xxx_queue_iocb, - .complete_iocb = qla4_8xxx_complete_iocb, - .rd_shdw_req_q_out = qla4_8xxx_rd_shdw_req_q_out, - .rd_shdw_rsp_q_in = qla4_8xxx_rd_shdw_rsp_q_in, + .queue_iocb = qla4_82xx_queue_iocb, + .complete_iocb = qla4_82xx_complete_iocb, + .rd_shdw_req_q_out = qla4_82xx_rd_shdw_req_q_out, + .rd_shdw_rsp_q_in = qla4_82xx_rd_shdw_rsp_q_in, .get_sys_info = qla4_8xxx_get_sys_info, + .rd_reg_direct = qla4_82xx_rd_32, + .wr_reg_direct = qla4_82xx_wr_32, + .rd_reg_indirect = qla4_82xx_md_rd_32, + .wr_reg_indirect = qla4_82xx_md_wr_32, + .idc_lock = qla4_82xx_idc_lock, + .idc_unlock = qla4_82xx_idc_unlock, + .rom_lock_recovery = qla4_82xx_rom_lock_recovery, + .queue_mailbox_command = qla4_82xx_queue_mbox_cmd, + .process_mailbox_interrupt = qla4_82xx_process_mbox_intr, +}; + +static struct isp_operations qla4_83xx_isp_ops = { + .iospace_config = qla4_8xxx_iospace_config, + .pci_config = qla4_8xxx_pci_config, + .disable_intrs = qla4_83xx_disable_intrs, + .enable_intrs = qla4_83xx_enable_intrs, + .start_firmware = qla4_8xxx_load_risc, + .restart_firmware = qla4_83xx_start_firmware, + .intr_handler = qla4_83xx_intr_handler, + .interrupt_service_routine = qla4_83xx_interrupt_service_routine, + .need_reset = qla4_8xxx_need_reset, + .reset_chip = qla4_83xx_isp_reset, + .reset_firmware = qla4_8xxx_stop_firmware, + .queue_iocb = qla4_83xx_queue_iocb, + .complete_iocb = qla4_83xx_complete_iocb, + .rd_shdw_req_q_out = qla4_83xx_rd_shdw_req_q_out, + .rd_shdw_rsp_q_in = qla4_83xx_rd_shdw_rsp_q_in, + .get_sys_info = qla4_8xxx_get_sys_info, + .rd_reg_direct = qla4_83xx_rd_reg, + .wr_reg_direct = qla4_83xx_wr_reg, + .rd_reg_indirect = qla4_83xx_rd_reg_indirect, + .wr_reg_indirect = qla4_83xx_wr_reg_indirect, + .idc_lock = qla4_83xx_drv_lock, + .idc_unlock = qla4_83xx_drv_unlock, + .rom_lock_recovery = qla4_83xx_rom_lock_recovery, + .queue_mailbox_command = qla4_83xx_queue_mbox_cmd, + .process_mailbox_interrupt = qla4_83xx_process_mbox_intr, }; uint16_t qla4xxx_rd_shdw_req_q_out(struct scsi_qla_host *ha) @@ -3663,9 +3756,14 @@ uint16_t qla4xxx_rd_shdw_req_q_out(struct scsi_qla_host *ha) return (uint16_t)le32_to_cpu(ha->shadow_regs->req_q_out); } -uint16_t qla4_8xxx_rd_shdw_req_q_out(struct scsi_qla_host *ha) +uint16_t qla4_82xx_rd_shdw_req_q_out(struct scsi_qla_host *ha) { - return (uint16_t)le32_to_cpu(readl(&ha->qla4_8xxx_reg->req_q_out)); + return (uint16_t)le32_to_cpu(readl(&ha->qla4_82xx_reg->req_q_out)); +} + +uint16_t qla4_83xx_rd_shdw_req_q_out(struct scsi_qla_host *ha) +{ + return (uint16_t)le32_to_cpu(readl(&ha->qla4_83xx_reg->req_q_out)); } uint16_t qla4xxx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha) @@ -3673,9 +3771,14 @@ uint16_t qla4xxx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha) return (uint16_t)le32_to_cpu(ha->shadow_regs->rsp_q_in); } -uint16_t qla4_8xxx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha) +uint16_t qla4_82xx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha) { - return (uint16_t)le32_to_cpu(readl(&ha->qla4_8xxx_reg->rsp_q_in)); + return (uint16_t)le32_to_cpu(readl(&ha->qla4_82xx_reg->rsp_q_in)); +} + +uint16_t qla4_83xx_rd_shdw_rsp_q_in(struct scsi_qla_host *ha) +{ + return (uint16_t)le32_to_cpu(readl(&ha->qla4_83xx_reg->rsp_q_in)); } static ssize_t qla4xxx_show_boot_eth_info(void *data, int type, char *buf) @@ -5050,30 +5153,36 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, ha->pdev = pdev; ha->host = host; ha->host_no = host->host_no; + ha->func_num = PCI_FUNC(ha->pdev->devfn); pci_enable_pcie_error_reporting(pdev); /* Setup Runtime configurable options */ if (is_qla8022(ha)) { - ha->isp_ops = &qla4_8xxx_isp_ops; - rwlock_init(&ha->hw_lock); + ha->isp_ops = &qla4_82xx_isp_ops; + ha->reg_tbl = (uint32_t *) qla4_82xx_reg_tbl; ha->qdr_sn_window = -1; ha->ddr_mn_window = -1; ha->curr_window = 255; - ha->func_num = PCI_FUNC(ha->pdev->devfn); nx_legacy_intr = &legacy_intr[ha->func_num]; ha->nx_legacy_intr.int_vec_bit = nx_legacy_intr->int_vec_bit; ha->nx_legacy_intr.tgt_status_reg = nx_legacy_intr->tgt_status_reg; ha->nx_legacy_intr.tgt_mask_reg = nx_legacy_intr->tgt_mask_reg; ha->nx_legacy_intr.pci_int_reg = nx_legacy_intr->pci_int_reg; + } else if (is_qla8032(ha)) { + ha->isp_ops = &qla4_83xx_isp_ops; + ha->reg_tbl = (uint32_t *)qla4_83xx_reg_tbl; } else { ha->isp_ops = &qla4xxx_isp_ops; } - /* Set EEH reset type to fundamental if required by hba */ - if (is_qla8022(ha)) + if (is_qla80XX(ha)) { + rwlock_init(&ha->hw_lock); + ha->pf_bit = ha->func_num << 16; + /* Set EEH reset type to fundamental if required by hba */ pdev->needs_freset = 1; + } /* Configure PCI I/O space. */ ret = ha->isp_ops->iospace_config(ha); @@ -5094,6 +5203,7 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, init_completion(&ha->disable_acb_comp); spin_lock_init(&ha->hardware_lock); + spin_lock_init(&ha->work_lock); /* Initialize work list */ INIT_LIST_HEAD(&ha->work_list); @@ -5128,8 +5238,20 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, if (ret) goto probe_failed; - if (is_qla8022(ha)) - (void) qla4_8xxx_get_flash_info(ha); + if (is_qla80XX(ha)) + qla4_8xxx_get_flash_info(ha); + + if (is_qla8032(ha)) { + qla4_83xx_read_reset_template(ha); + /* + * NOTE: If ql4dontresethba==1, set IDC_CTRL DONTRESET_BIT0. + * If DONRESET_BIT0 is set, drivers should not set dev_state + * to NEED_RESET. But if NEED_RESET is set, drivers should + * should honor the reset. + */ + if (ql4xdontresethba == 1) + qla4_83xx_set_idc_dontreset(ha); + } /* * Initialize the Host adapter request/response queues and @@ -5137,14 +5259,20 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, * NOTE: interrupts enabled upon successful completion */ status = qla4xxx_initialize_adapter(ha, INIT_ADAPTER); + + /* Dont retry adapter initialization if IRQ allocation failed */ + if (!test_bit(AF_IRQ_ATTACHED, &ha->flags)) + goto skip_retry_init; + while ((!test_bit(AF_ONLINE, &ha->flags)) && init_retry_count++ < MAX_INIT_RETRIES) { - if (is_qla8022(ha)) { - qla4_8xxx_idc_lock(ha); - dev_state = qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE); - qla4_8xxx_idc_unlock(ha); - if (dev_state == QLA82XX_DEV_FAILED) { + if (is_qla80XX(ha)) { + ha->isp_ops->idc_lock(ha); + dev_state = qla4_8xxx_rd_direct(ha, + QLA82XX_CRB_DEV_STATE); + ha->isp_ops->idc_unlock(ha); + if (dev_state == QLA8XXX_DEV_FAILED) { ql4_printk(KERN_WARNING, ha, "%s: don't retry " "initialize adapter. H/W is in failed state\n", __func__); @@ -5160,16 +5288,18 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, status = qla4xxx_initialize_adapter(ha, INIT_ADAPTER); } +skip_retry_init: if (!test_bit(AF_ONLINE, &ha->flags)) { ql4_printk(KERN_WARNING, ha, "Failed to initialize adapter\n"); - if (is_qla8022(ha) && ql4xdontresethba) { + if ((is_qla8022(ha) && ql4xdontresethba) || + (is_qla8032(ha) && qla4_83xx_idc_dontreset(ha))) { /* Put the device in failed state. */ DEBUG2(printk(KERN_ERR "HW STATE: FAILED\n")); - qla4_8xxx_idc_lock(ha); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_lock(ha); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); + ha->isp_ops->idc_unlock(ha); } ret = -ENODEV; goto remove_host; @@ -5195,12 +5325,13 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, goto remove_host; } - /* For ISP-82XX, request_irqs is called in qla4_8xxx_load_risc + /* + * For ISP-8XXX, request_irqs is called in qla4_8xxx_load_risc * (which is called indirectly by qla4xxx_initialize_adapter), * so that irqs will be registered after crbinit but before * mbx_intr_enable. */ - if (!is_qla8022(ha)) { + if (is_qla40XX(ha)) { ret = qla4xxx_request_irqs(ha); if (ret) { ql4_printk(KERN_WARNING, ha, "Failed to reserve " @@ -5226,6 +5357,10 @@ static int __devinit qla4xxx_probe_adapter(struct pci_dev *pdev, ha->host_no, ha->firmware_version[0], ha->firmware_version[1], ha->patch_number, ha->build_number); + /* Set the driver version */ + if (is_qla80XX(ha)) + qla4_8xxx_set_param(ha, SET_DRVR_VERSION); + if (qla4xxx_setup_boot_info(ha)) ql4_printk(KERN_ERR, ha, "%s: No iSCSI boot target configured\n", __func__); @@ -5333,9 +5468,16 @@ static void __devexit qla4xxx_remove_adapter(struct pci_dev *pdev) { struct scsi_qla_host *ha; + /* + * If the PCI device is disabled then it means probe_adapter had + * failed and resources already cleaned up on probe_adapter exit. + */ + if (!pci_is_enabled(pdev)) + return; + ha = pci_get_drvdata(pdev); - if (!is_qla8022(ha)) + if (is_qla40XX(ha)) qla4xxx_prevent_other_port_reinit(ha); /* destroy iface from sysfs */ @@ -5755,7 +5897,16 @@ static int qla4xxx_eh_host_reset(struct scsi_cmnd *cmd) ha = to_qla_host(cmd->device->host); - if (ql4xdontresethba) { + if (is_qla8032(ha) && ql4xdontresethba) + qla4_83xx_set_idc_dontreset(ha); + + /* + * For ISP8324, if IDC_CTRL DONTRESET_BIT0 is set by other + * protocol drivers, we should not set device_state to + * NEED_RESET + */ + if (ql4xdontresethba || + (is_qla8032(ha) && qla4_83xx_idc_dontreset(ha))) { DEBUG2(printk("scsi%ld: %s: Don't Reset HBA\n", ha->host_no, __func__)); @@ -5779,7 +5930,7 @@ static int qla4xxx_eh_host_reset(struct scsi_cmnd *cmd) } if (!test_bit(DPC_RESET_HA, &ha->dpc_flags)) { - if (is_qla8022(ha)) + if (is_qla80XX(ha)) set_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags); else set_bit(DPC_RESET_HA, &ha->dpc_flags); @@ -5874,7 +6025,7 @@ static int qla4xxx_host_reset(struct Scsi_Host *shost, int reset_type) break; case SCSI_FIRMWARE_RESET: if (!test_bit(DPC_RESET_HA, &ha->dpc_flags)) { - if (is_qla8022(ha)) + if (is_qla80XX(ha)) /* set firmware context reset */ set_bit(DPC_RESET_HA_FW_CONTEXT, &ha->dpc_flags); @@ -6013,32 +6164,43 @@ static uint32_t qla4_8xxx_error_recovery(struct scsi_qla_host *ha) "0x%x is the owner\n", ha->host_no, __func__, ha->pdev->devfn); - qla4_8xxx_idc_lock(ha); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_COLD); - - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_IDC_VERSION, - QLA82XX_IDC_VERSION); + ha->isp_ops->idc_lock(ha); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_COLD); + ha->isp_ops->idc_unlock(ha); + + rval = qla4_8xxx_update_idc_reg(ha); + if (rval == QLA_ERROR) { + ql4_printk(KERN_INFO, ha, "scsi%ld: %s: HW State: FAILED\n", + ha->host_no, __func__); + ha->isp_ops->idc_lock(ha); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); + ha->isp_ops->idc_unlock(ha); + goto exit_error_recovery; + } - qla4_8xxx_idc_unlock(ha); clear_bit(AF_FW_RECOVERY, &ha->flags); rval = qla4xxx_initialize_adapter(ha, RESET_ADAPTER); - qla4_8xxx_idc_lock(ha); if (rval != QLA_SUCCESS) { ql4_printk(KERN_INFO, ha, "scsi%ld: %s: HW State: " "FAILED\n", ha->host_no, __func__); + ha->isp_ops->idc_lock(ha); qla4_8xxx_clear_drv_active(ha); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_FAILED); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_FAILED); + ha->isp_ops->idc_unlock(ha); } else { ql4_printk(KERN_INFO, ha, "scsi%ld: %s: HW State: " "READY\n", ha->host_no, __func__); - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DEV_STATE, - QLA82XX_DEV_READY); + ha->isp_ops->idc_lock(ha); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DEV_STATE, + QLA8XXX_DEV_READY); /* Clear driver state register */ - qla4_8xxx_wr_32(ha, QLA82XX_CRB_DRV_STATE, 0); + qla4_8xxx_wr_direct(ha, QLA8XXX_CRB_DRV_STATE, 0); qla4_8xxx_set_drv_active(ha); + ha->isp_ops->idc_unlock(ha); ret = qla4xxx_request_irqs(ha); if (ret) { ql4_printk(KERN_WARNING, ha, "Failed to " @@ -6050,13 +6212,12 @@ static uint32_t qla4_8xxx_error_recovery(struct scsi_qla_host *ha) rval = QLA_SUCCESS; } } - qla4_8xxx_idc_unlock(ha); } else { ql4_printk(KERN_INFO, ha, "scsi%ld: %s: devfn 0x%x is not " "the reset owner\n", ha->host_no, __func__, ha->pdev->devfn); - if ((qla4_8xxx_rd_32(ha, QLA82XX_CRB_DEV_STATE) == - QLA82XX_DEV_READY)) { + if ((qla4_8xxx_rd_direct(ha, QLA8XXX_CRB_DEV_STATE) == + QLA8XXX_DEV_READY)) { clear_bit(AF_FW_RECOVERY, &ha->flags); rval = qla4xxx_initialize_adapter(ha, RESET_ADAPTER); if (rval == QLA_SUCCESS) { @@ -6071,11 +6232,12 @@ static uint32_t qla4_8xxx_error_recovery(struct scsi_qla_host *ha) rval = QLA_SUCCESS; } } - qla4_8xxx_idc_lock(ha); + ha->isp_ops->idc_lock(ha); qla4_8xxx_set_drv_active(ha); - qla4_8xxx_idc_unlock(ha); + ha->isp_ops->idc_unlock(ha); } } +exit_error_recovery: clear_bit(DPC_RESET_ACTIVE, &ha->dpc_flags); return rval; } @@ -6114,7 +6276,7 @@ qla4xxx_pci_slot_reset(struct pci_dev *pdev) ha->isp_ops->disable_intrs(ha); - if (is_qla8022(ha)) { + if (is_qla80XX(ha)) { if (qla4_8xxx_error_recovery(ha) == QLA_SUCCESS) { ret = PCI_ERS_RESULT_RECOVERED; goto exit_slot_reset; @@ -6180,6 +6342,12 @@ static struct pci_device_id qla4xxx_pci_tbl[] = { .subvendor = PCI_ANY_ID, .subdevice = PCI_ANY_ID, }, + { + .vendor = PCI_VENDOR_ID_QLOGIC, + .device = PCI_DEVICE_ID_QLOGIC_ISP8324, + .subvendor = PCI_ANY_ID, + .subdevice = PCI_ANY_ID, + }, {0, 0}, }; MODULE_DEVICE_TABLE(pci, qla4xxx_pci_tbl); diff --git a/drivers/scsi/qla4xxx/ql4_version.h b/drivers/scsi/qla4xxx/ql4_version.h index 725034f..f6df2ea 100644 --- a/drivers/scsi/qla4xxx/ql4_version.h +++ b/drivers/scsi/qla4xxx/ql4_version.h @@ -1,8 +1,8 @@ /* * QLogic iSCSI HBA Driver - * Copyright (c) 2003-2010 QLogic Corporation + * Copyright (c) 2003-2012 QLogic Corporation * * See LICENSE.qla4xxx for copyright and licensing details. */ -#define QLA4XXX_DRIVER_VERSION "5.02.00-k18" +#define QLA4XXX_DRIVER_VERSION "5.03.00-k1" diff --git a/drivers/scsi/scsi_debug.c b/drivers/scsi/scsi_debug.c index 182d5a5..57fbd5a 100644 --- a/drivers/scsi/scsi_debug.c +++ b/drivers/scsi/scsi_debug.c @@ -109,6 +109,7 @@ static const char * scsi_debug_version_date = "20100324"; #define DEF_OPT_BLKS 64 #define DEF_PHYSBLK_EXP 0 #define DEF_PTYPE 0 +#define DEF_REMOVABLE false #define DEF_SCSI_LEVEL 5 /* INQUIRY, byte2 [5->SPC-3] */ #define DEF_SECTOR_SIZE 512 #define DEF_UNMAP_ALIGNMENT 0 @@ -193,11 +194,11 @@ static unsigned int scsi_debug_unmap_granularity = DEF_UNMAP_GRANULARITY; static unsigned int scsi_debug_unmap_max_blocks = DEF_UNMAP_MAX_BLOCKS; static unsigned int scsi_debug_unmap_max_desc = DEF_UNMAP_MAX_DESC; static unsigned int scsi_debug_write_same_length = DEF_WRITESAME_LENGTH; +static bool scsi_debug_removable = DEF_REMOVABLE; static int scsi_debug_cmnd_count = 0; #define DEV_READONLY(TGT) (0) -#define DEV_REMOVEABLE(TGT) (0) static unsigned int sdebug_store_sectors; static sector_t sdebug_capacity; /* in sectors */ @@ -919,7 +920,7 @@ static int resp_inquiry(struct scsi_cmnd * scp, int target, return ret; } /* drops through here for a standard inquiry */ - arr[1] = DEV_REMOVEABLE(target) ? 0x80 : 0; /* Removable disk */ + arr[1] = scsi_debug_removable ? 0x80 : 0; /* Removable disk */ arr[2] = scsi_debug_scsi_level; arr[3] = 2; /* response_data_format==2 */ arr[4] = SDEBUG_LONG_INQ_SZ - 5; @@ -1211,7 +1212,7 @@ static int resp_format_pg(unsigned char * p, int pcontrol, int target) p[11] = sdebug_sectors_per & 0xff; p[12] = (scsi_debug_sector_size >> 8) & 0xff; p[13] = scsi_debug_sector_size & 0xff; - if (DEV_REMOVEABLE(target)) + if (scsi_debug_removable) p[20] |= 0x20; /* should agree with INQUIRY */ if (1 == pcontrol) memset(p + 2, 0, sizeof(format_pg) - 2); @@ -2754,6 +2755,7 @@ module_param_named(opt_blks, scsi_debug_opt_blks, int, S_IRUGO); module_param_named(opts, scsi_debug_opts, int, S_IRUGO | S_IWUSR); module_param_named(physblk_exp, scsi_debug_physblk_exp, int, S_IRUGO); module_param_named(ptype, scsi_debug_ptype, int, S_IRUGO | S_IWUSR); +module_param_named(removable, scsi_debug_removable, bool, S_IRUGO | S_IWUSR); module_param_named(scsi_level, scsi_debug_scsi_level, int, S_IRUGO); module_param_named(sector_size, scsi_debug_sector_size, int, S_IRUGO); module_param_named(unmap_alignment, scsi_debug_unmap_alignment, int, S_IRUGO); @@ -2796,6 +2798,7 @@ MODULE_PARM_DESC(opt_blks, "optimal transfer length in block (def=64)"); MODULE_PARM_DESC(opts, "1->noise, 2->medium_err, 4->timeout, 8->recovered_err... (def=0)"); MODULE_PARM_DESC(physblk_exp, "physical block exponent (def=0)"); MODULE_PARM_DESC(ptype, "SCSI peripheral type(def=0[disk])"); +MODULE_PARM_DESC(removable, "claim to have removable media (def=0)"); MODULE_PARM_DESC(scsi_level, "SCSI level to simulate(def=5[SPC-3])"); MODULE_PARM_DESC(sector_size, "logical block size in bytes (def=512)"); MODULE_PARM_DESC(unmap_alignment, "lowest aligned thin provisioning lba (def=0)"); @@ -3205,6 +3208,25 @@ static ssize_t sdebug_map_show(struct device_driver *ddp, char *buf) } DRIVER_ATTR(map, S_IRUGO, sdebug_map_show, NULL); +static ssize_t sdebug_removable_show(struct device_driver *ddp, + char *buf) +{ + return scnprintf(buf, PAGE_SIZE, "%d\n", scsi_debug_removable ? 1 : 0); +} +static ssize_t sdebug_removable_store(struct device_driver *ddp, + const char *buf, size_t count) +{ + int n; + + if ((count > 0) && (1 == sscanf(buf, "%d", &n)) && (n >= 0)) { + scsi_debug_removable = (n > 0); + return count; + } + return -EINVAL; +} +DRIVER_ATTR(removable, S_IRUGO | S_IWUSR, sdebug_removable_show, + sdebug_removable_store); + /* Note: The following function creates attribute files in the /sys/bus/pseudo/drivers/scsi_debug directory. The advantage of these @@ -3230,6 +3252,7 @@ static int do_create_driverfs_files(void) ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_num_tgts); ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_ptype); ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_opts); + ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_removable); ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_scsi_level); ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_virtual_gb); ret |= driver_create_file(&sdebug_driverfs_driver, &driver_attr_vpd_use_hostno); @@ -3255,6 +3278,7 @@ static void do_remove_driverfs_files(void) driver_remove_file(&sdebug_driverfs_driver, &driver_attr_scsi_level); driver_remove_file(&sdebug_driverfs_driver, &driver_attr_opts); driver_remove_file(&sdebug_driverfs_driver, &driver_attr_ptype); + driver_remove_file(&sdebug_driverfs_driver, &driver_attr_removable); driver_remove_file(&sdebug_driverfs_driver, &driver_attr_num_tgts); driver_remove_file(&sdebug_driverfs_driver, &driver_attr_num_parts); driver_remove_file(&sdebug_driverfs_driver, &driver_attr_no_uld); diff --git a/drivers/scsi/scsi_devinfo.c b/drivers/scsi/scsi_devinfo.c index cf8dfab..43fca91 100644 --- a/drivers/scsi/scsi_devinfo.c +++ b/drivers/scsi/scsi_devinfo.c @@ -172,6 +172,7 @@ static struct { {"HITACHI", "DF400", "*", BLIST_REPORTLUN2}, {"HITACHI", "DF500", "*", BLIST_REPORTLUN2}, {"HITACHI", "DISK-SUBSYSTEM", "*", BLIST_REPORTLUN2}, + {"HITACHI", "HUS1530", "*", BLIST_NO_DIF}, {"HITACHI", "OPEN-", "*", BLIST_REPORTLUN2}, {"HITACHI", "OP-C-", "*", BLIST_SPARSELUN | BLIST_LARGELUN}, {"HITACHI", "3380-", "*", BLIST_SPARSELUN | BLIST_LARGELUN}, diff --git a/drivers/scsi/scsi_lib.c b/drivers/scsi/scsi_lib.c index faa790f..da36a3a 100644 --- a/drivers/scsi/scsi_lib.c +++ b/drivers/scsi/scsi_lib.c @@ -2473,7 +2473,8 @@ scsi_internal_device_unblock(struct scsi_device *sdev, * Try to transition the scsi device to SDEV_RUNNING or one of the * offlined states and goose the device queue if successful. */ - if (sdev->sdev_state == SDEV_BLOCK) + if ((sdev->sdev_state == SDEV_BLOCK) || + (sdev->sdev_state == SDEV_TRANSPORT_OFFLINE)) sdev->sdev_state = new_state; else if (sdev->sdev_state == SDEV_CREATED_BLOCK) { if (new_state == SDEV_TRANSPORT_OFFLINE || diff --git a/drivers/scsi/scsi_scan.c b/drivers/scsi/scsi_scan.c index d947ffc..3e58b22 100644 --- a/drivers/scsi/scsi_scan.c +++ b/drivers/scsi/scsi_scan.c @@ -921,6 +921,9 @@ static int scsi_add_lun(struct scsi_device *sdev, unsigned char *inq_result, if (*bflags & BLIST_RETRY_HWERROR) sdev->retry_hwerror = 1; + if (*bflags & BLIST_NO_DIF) + sdev->no_dif = 1; + transport_configure_device(&sdev->sdev_gendev); if (sdev->host->hostt->slave_configure) { diff --git a/drivers/scsi/scsi_sysfs.c b/drivers/scsi/scsi_sysfs.c index 093d4f6..ce5224c 100644 --- a/drivers/scsi/scsi_sysfs.c +++ b/drivers/scsi/scsi_sysfs.c @@ -1031,33 +1031,31 @@ static void __scsi_remove_target(struct scsi_target *starget) void scsi_remove_target(struct device *dev) { struct Scsi_Host *shost = dev_to_shost(dev->parent); - struct scsi_target *starget, *found; + struct scsi_target *starget, *last = NULL; unsigned long flags; - restart: - found = NULL; + /* remove targets being careful to lookup next entry before + * deleting the last + */ spin_lock_irqsave(shost->host_lock, flags); list_for_each_entry(starget, &shost->__targets, siblings) { if (starget->state == STARGET_DEL) continue; if (starget->dev.parent == dev || &starget->dev == dev) { - found = starget; - found->reap_ref++; - break; + /* assuming new targets arrive at the end */ + starget->reap_ref++; + spin_unlock_irqrestore(shost->host_lock, flags); + if (last) + scsi_target_reap(last); + last = starget; + __scsi_remove_target(starget); + spin_lock_irqsave(shost->host_lock, flags); } } spin_unlock_irqrestore(shost->host_lock, flags); - if (found) { - __scsi_remove_target(found); - scsi_target_reap(found); - /* in the case where @dev has multiple starget children, - * continue removing. - * - * FIXME: does such a case exist? - */ - goto restart; - } + if (last) + scsi_target_reap(last); } EXPORT_SYMBOL(scsi_remove_target); diff --git a/drivers/scsi/sd.c b/drivers/scsi/sd.c index 4df73e5..12f6fdf 100644 --- a/drivers/scsi/sd.c +++ b/drivers/scsi/sd.c @@ -262,6 +262,28 @@ sd_show_protection_type(struct device *dev, struct device_attribute *attr, } static ssize_t +sd_store_protection_type(struct device *dev, struct device_attribute *attr, + const char *buf, size_t count) +{ + struct scsi_disk *sdkp = to_scsi_disk(dev); + unsigned int val; + int err; + + if (!capable(CAP_SYS_ADMIN)) + return -EACCES; + + err = kstrtouint(buf, 10, &val); + + if (err) + return err; + + if (val >= 0 && val <= SD_DIF_TYPE3_PROTECTION) + sdkp->protection_type = val; + + return count; +} + +static ssize_t sd_show_protection_mode(struct device *dev, struct device_attribute *attr, char *buf) { @@ -381,7 +403,8 @@ static struct device_attribute sd_disk_attrs[] = { sd_store_allow_restart), __ATTR(manage_start_stop, S_IRUGO|S_IWUSR, sd_show_manage_start_stop, sd_store_manage_start_stop), - __ATTR(protection_type, S_IRUGO, sd_show_protection_type, NULL), + __ATTR(protection_type, S_IRUGO|S_IWUSR, sd_show_protection_type, + sd_store_protection_type), __ATTR(protection_mode, S_IRUGO, sd_show_protection_mode, NULL), __ATTR(app_tag_own, S_IRUGO, sd_show_app_tag_own, NULL), __ATTR(thin_provisioning, S_IRUGO, sd_show_thin_provisioning, NULL), @@ -804,9 +827,8 @@ static int sd_prep_fn(struct request_queue *q, struct request *rq) SCpnt->cmnd[0] = WRITE_6; SCpnt->sc_data_direction = DMA_TO_DEVICE; - if (blk_integrity_rq(rq) && - sd_dif_prepare(rq, block, sdp->sector_size) == -EIO) - goto out; + if (blk_integrity_rq(rq)) + sd_dif_prepare(rq, block, sdp->sector_size); } else if (rq_data_dir(rq) == READ) { SCpnt->cmnd[0] = READ_6; @@ -1671,34 +1693,42 @@ sd_spinup_disk(struct scsi_disk *sdkp) /* * Determine whether disk supports Data Integrity Field. */ -static void sd_read_protection_type(struct scsi_disk *sdkp, unsigned char *buffer) +static int sd_read_protection_type(struct scsi_disk *sdkp, unsigned char *buffer) { struct scsi_device *sdp = sdkp->device; u8 type; + int ret = 0; if (scsi_device_protection(sdp) == 0 || (buffer[12] & 1) == 0) - return; + return ret; type = ((buffer[12] >> 1) & 7) + 1; /* P_TYPE 0 = Type 1 */ - if (type == sdkp->protection_type || !sdkp->first_scan) - return; + if (type > SD_DIF_TYPE3_PROTECTION) + ret = -ENODEV; + else if (scsi_host_dif_capable(sdp->host, type)) + ret = 1; + + if (sdkp->first_scan || type != sdkp->protection_type) + switch (ret) { + case -ENODEV: + sd_printk(KERN_ERR, sdkp, "formatted with unsupported" \ + " protection type %u. Disabling disk!\n", + type); + break; + case 1: + sd_printk(KERN_NOTICE, sdkp, + "Enabling DIF Type %u protection\n", type); + break; + case 0: + sd_printk(KERN_NOTICE, sdkp, + "Disabling DIF Type %u protection\n", type); + break; + } sdkp->protection_type = type; - if (type > SD_DIF_TYPE3_PROTECTION) { - sd_printk(KERN_ERR, sdkp, "formatted with unsupported " \ - "protection type %u. Disabling disk!\n", type); - sdkp->capacity = 0; - return; - } - - if (scsi_host_dif_capable(sdp->host, type)) - sd_printk(KERN_NOTICE, sdkp, - "Enabling DIF Type %u protection\n", type); - else - sd_printk(KERN_NOTICE, sdkp, - "Disabling DIF Type %u protection\n", type); + return ret; } static void read_capacity_error(struct scsi_disk *sdkp, struct scsi_device *sdp, @@ -1794,7 +1824,10 @@ static int read_capacity_16(struct scsi_disk *sdkp, struct scsi_device *sdp, sector_size = get_unaligned_be32(&buffer[8]); lba = get_unaligned_be64(&buffer[0]); - sd_read_protection_type(sdkp, buffer); + if (sd_read_protection_type(sdkp, buffer) < 0) { + sdkp->capacity = 0; + return -ENODEV; + } if ((sizeof(sdkp->capacity) == 4) && (lba >= 0xffffffffULL)) { sd_printk(KERN_ERR, sdkp, "Too big for this kernel. Use a " @@ -2632,7 +2665,8 @@ static void sd_probe_async(void *data, async_cookie_t cookie) } add_disk(gd); - sd_dif_config_host(sdkp); + if (sdkp->capacity) + sd_dif_config_host(sdkp); sd_revalidate_disk(gd); diff --git a/drivers/scsi/sd.h b/drivers/scsi/sd.h index f703f48..47c52a6 100644 --- a/drivers/scsi/sd.h +++ b/drivers/scsi/sd.h @@ -156,7 +156,7 @@ struct sd_dif_tuple { #ifdef CONFIG_BLK_DEV_INTEGRITY extern void sd_dif_config_host(struct scsi_disk *); -extern int sd_dif_prepare(struct request *rq, sector_t, unsigned int); +extern void sd_dif_prepare(struct request *rq, sector_t, unsigned int); extern void sd_dif_complete(struct scsi_cmnd *, unsigned int); #else /* CONFIG_BLK_DEV_INTEGRITY */ diff --git a/drivers/scsi/sd_dif.c b/drivers/scsi/sd_dif.c index e52d5bc..04998f3 100644 --- a/drivers/scsi/sd_dif.c +++ b/drivers/scsi/sd_dif.c @@ -366,7 +366,8 @@ void sd_dif_config_host(struct scsi_disk *sdkp) * * Type 3 does not have a reference tag so no remapping is required. */ -int sd_dif_prepare(struct request *rq, sector_t hw_sector, unsigned int sector_sz) +void sd_dif_prepare(struct request *rq, sector_t hw_sector, + unsigned int sector_sz) { const int tuple_sz = sizeof(struct sd_dif_tuple); struct bio *bio; @@ -378,7 +379,7 @@ int sd_dif_prepare(struct request *rq, sector_t hw_sector, unsigned int sector_s sdkp = rq->bio->bi_bdev->bd_disk->private_data; if (sdkp->protection_type == SD_DIF_TYPE3_PROTECTION) - return 0; + return; phys = hw_sector & 0xffffffff; @@ -397,10 +398,9 @@ int sd_dif_prepare(struct request *rq, sector_t hw_sector, unsigned int sector_s for (j = 0 ; j < iv->bv_len ; j += tuple_sz, sdt++) { - if (be32_to_cpu(sdt->ref_tag) != virt) - goto error; + if (be32_to_cpu(sdt->ref_tag) == virt) + sdt->ref_tag = cpu_to_be32(phys); - sdt->ref_tag = cpu_to_be32(phys); virt++; phys++; } @@ -410,16 +410,6 @@ int sd_dif_prepare(struct request *rq, sector_t hw_sector, unsigned int sector_s bio->bi_flags |= (1 << BIO_MAPPED_INTEGRITY); } - - return 0; - -error: - kunmap_atomic(sdt); - sd_printk(KERN_ERR, sdkp, "%s: virt %u, phys %u, ref %u, app %4x\n", - __func__, virt, phys, be32_to_cpu(sdt->ref_tag), - be16_to_cpu(sdt->app_tag)); - - return -EILSEQ; } /* @@ -463,10 +453,7 @@ void sd_dif_complete(struct scsi_cmnd *scmd, unsigned int good_bytes) return; } - if (be32_to_cpu(sdt->ref_tag) != phys && - sdt->app_tag != 0xffff) - sdt->ref_tag = 0xffffffff; /* Bad ref */ - else + if (be32_to_cpu(sdt->ref_tag) == phys) sdt->ref_tag = cpu_to_be32(virt); virt++; diff --git a/drivers/scsi/st.c b/drivers/scsi/st.c index e41998c..98156a9 100644 --- a/drivers/scsi/st.c +++ b/drivers/scsi/st.c @@ -37,6 +37,7 @@ static const char *verstr = "20101219"; #include <linux/blkdev.h> #include <linux/moduleparam.h> #include <linux/cdev.h> +#include <linux/idr.h> #include <linux/delay.h> #include <linux/mutex.h> @@ -74,17 +75,14 @@ static const char *verstr = "20101219"; #include "st_options.h" #include "st.h" -static DEFINE_MUTEX(st_mutex); static int buffer_kbs; static int max_sg_segs; static int try_direct_io = TRY_DIRECT_IO; static int try_rdio = 1; static int try_wdio = 1; -static int st_dev_max; -static int st_nr_dev; - -static struct class *st_sysfs_class; +static struct class st_sysfs_class; +static struct device_attribute st_dev_attrs[]; MODULE_AUTHOR("Kai Makisara"); MODULE_DESCRIPTION("SCSI tape (st) driver"); @@ -173,13 +171,9 @@ static int debugging = DEBUG; 24 bits) */ #define SET_DENS_AND_BLK 0x10001 -static DEFINE_RWLOCK(st_dev_arr_lock); - static int st_fixed_buffer_size = ST_FIXED_BUFFER_SIZE; static int st_max_sg_segs = ST_MAX_SG; -static struct scsi_tape **scsi_tapes = NULL; - static int modes_defined; static int enlarge_buffer(struct st_buffer *, int, int); @@ -198,7 +192,6 @@ static int st_remove(struct device *); static int do_create_sysfs_files(void); static void do_remove_sysfs_files(void); -static int do_create_class_files(struct scsi_tape *, int, int); static struct scsi_driver st_template = { .owner = THIS_MODULE, @@ -221,6 +214,10 @@ static void scsi_tape_release(struct kref *); #define to_scsi_tape(obj) container_of(obj, struct scsi_tape, kref) static DEFINE_MUTEX(st_ref_mutex); +static DEFINE_SPINLOCK(st_index_lock); +static DEFINE_SPINLOCK(st_use_lock); +static DEFINE_IDR(st_index_idr); + #include "osst_detect.h" @@ -238,10 +235,9 @@ static struct scsi_tape *scsi_tape_get(int dev) struct scsi_tape *STp = NULL; mutex_lock(&st_ref_mutex); - write_lock(&st_dev_arr_lock); + spin_lock(&st_index_lock); - if (dev < st_dev_max && scsi_tapes != NULL) - STp = scsi_tapes[dev]; + STp = idr_find(&st_index_idr, dev); if (!STp) goto out; kref_get(&STp->kref); @@ -258,7 +254,7 @@ out_put: kref_put(&STp->kref, scsi_tape_release); STp = NULL; out: - write_unlock(&st_dev_arr_lock); + spin_unlock(&st_index_lock); mutex_unlock(&st_ref_mutex); return STp; } @@ -1188,7 +1184,6 @@ static int st_open(struct inode *inode, struct file *filp) int dev = TAPE_NR(inode); char *name; - mutex_lock(&st_mutex); /* * We really want to do nonseekable_open(inode, filp); here, but some * versions of tar incorrectly call lseek on tapes and bail out if that @@ -1197,24 +1192,22 @@ static int st_open(struct inode *inode, struct file *filp) filp->f_mode &= ~(FMODE_PREAD | FMODE_PWRITE); if (!(STp = scsi_tape_get(dev))) { - mutex_unlock(&st_mutex); return -ENXIO; } - write_lock(&st_dev_arr_lock); filp->private_data = STp; name = tape_name(STp); + spin_lock(&st_use_lock); if (STp->in_use) { - write_unlock(&st_dev_arr_lock); + spin_unlock(&st_use_lock); scsi_tape_put(STp); - mutex_unlock(&st_mutex); DEB( printk(ST_DEB_MSG "%s: Device already in use.\n", name); ) return (-EBUSY); } STp->in_use = 1; - write_unlock(&st_dev_arr_lock); + spin_unlock(&st_use_lock); STp->rew_at_close = STp->autorew_dev = (iminor(inode) & 0x80) == 0; if (scsi_autopm_get_device(STp->device) < 0) { @@ -1262,16 +1255,16 @@ static int st_open(struct inode *inode, struct file *filp) retval = (-EIO); goto err_out; } - mutex_unlock(&st_mutex); return 0; err_out: normalize_buffer(STp->buffer); + spin_lock(&st_use_lock); STp->in_use = 0; + spin_unlock(&st_use_lock); scsi_tape_put(STp); if (resumed) scsi_autopm_put_device(STp->device); - mutex_unlock(&st_mutex); return retval; } @@ -1403,9 +1396,9 @@ static int st_release(struct inode *inode, struct file *filp) do_door_lock(STp, 0); normalize_buffer(STp->buffer); - write_lock(&st_dev_arr_lock); + spin_lock(&st_use_lock); STp->in_use = 0; - write_unlock(&st_dev_arr_lock); + spin_unlock(&st_use_lock); scsi_autopm_put_device(STp->device); scsi_tape_put(STp); @@ -3992,16 +3985,98 @@ static const struct file_operations st_fops = .llseek = noop_llseek, }; +static int create_one_cdev(struct scsi_tape *tape, int mode, int rew) +{ + int i, error; + dev_t cdev_devno; + struct cdev *cdev; + struct device *dev; + struct st_modedef *STm = &(tape->modes[mode]); + char name[10]; + int dev_num = tape->index; + + cdev_devno = MKDEV(SCSI_TAPE_MAJOR, TAPE_MINOR(dev_num, mode, rew)); + + cdev = cdev_alloc(); + if (!cdev) { + pr_err("st%d: out of memory. Device not attached.\n", dev_num); + error = -ENOMEM; + goto out; + } + cdev->owner = THIS_MODULE; + cdev->ops = &st_fops; + + error = cdev_add(cdev, cdev_devno, 1); + if (error) { + pr_err("st%d: Can't add %s-rewind mode %d\n", dev_num, + rew ? "non" : "auto", mode); + pr_err("st%d: Device not attached.\n", dev_num); + goto out_free; + } + STm->cdevs[rew] = cdev; + + i = mode << (4 - ST_NBR_MODE_BITS); + snprintf(name, 10, "%s%s%s", rew ? "n" : "", + tape->disk->disk_name, st_formats[i]); + + dev = device_create(&st_sysfs_class, &tape->device->sdev_gendev, + cdev_devno, &tape->modes[mode], "%s", name); + if (IS_ERR(dev)) { + pr_err("st%d: device_create failed\n", dev_num); + error = PTR_ERR(dev); + goto out_free; + } + + STm->devs[rew] = dev; + + return 0; +out_free: + cdev_del(STm->cdevs[rew]); + STm->cdevs[rew] = NULL; +out: + return error; +} + +static int create_cdevs(struct scsi_tape *tape) +{ + int mode, error; + for (mode = 0; mode < ST_NBR_MODES; ++mode) { + error = create_one_cdev(tape, mode, 0); + if (error) + return error; + error = create_one_cdev(tape, mode, 1); + if (error) + return error; + } + + return sysfs_create_link(&tape->device->sdev_gendev.kobj, + &tape->modes[0].devs[0]->kobj, "tape"); +} + +static void remove_cdevs(struct scsi_tape *tape) +{ + int mode, rew; + sysfs_remove_link(&tape->device->sdev_gendev.kobj, "tape"); + for (mode = 0; mode < ST_NBR_MODES; mode++) { + struct st_modedef *STm = &(tape->modes[mode]); + for (rew = 0; rew < 2; rew++) { + if (STm->cdevs[rew]) + cdev_del(STm->cdevs[rew]); + if (STm->devs[rew]) + device_unregister(STm->devs[rew]); + } + } +} + static int st_probe(struct device *dev) { struct scsi_device *SDp = to_scsi_device(dev); struct gendisk *disk = NULL; - struct cdev *cdev = NULL; struct scsi_tape *tpnt = NULL; struct st_modedef *STm; struct st_partstat *STps; struct st_buffer *buffer; - int i, j, mode, dev_num, error; + int i, dev_num, error; char *stp; if (SDp->type != TYPE_TAPE) @@ -4028,58 +4103,16 @@ static int st_probe(struct device *dev) goto out_buffer_free; } - write_lock(&st_dev_arr_lock); - if (st_nr_dev >= st_dev_max) { - struct scsi_tape **tmp_da; - int tmp_dev_max; - - tmp_dev_max = max(st_nr_dev * 2, 8); - if (tmp_dev_max > ST_MAX_TAPES) - tmp_dev_max = ST_MAX_TAPES; - if (tmp_dev_max <= st_nr_dev) { - write_unlock(&st_dev_arr_lock); - printk(KERN_ERR "st: Too many tape devices (max. %d).\n", - ST_MAX_TAPES); - goto out_put_disk; - } - - tmp_da = kzalloc(tmp_dev_max * sizeof(struct scsi_tape *), GFP_ATOMIC); - if (tmp_da == NULL) { - write_unlock(&st_dev_arr_lock); - printk(KERN_ERR "st: Can't extend device array.\n"); - goto out_put_disk; - } - - if (scsi_tapes != NULL) { - memcpy(tmp_da, scsi_tapes, - st_dev_max * sizeof(struct scsi_tape *)); - kfree(scsi_tapes); - } - scsi_tapes = tmp_da; - - st_dev_max = tmp_dev_max; - } - - for (i = 0; i < st_dev_max; i++) - if (scsi_tapes[i] == NULL) - break; - if (i >= st_dev_max) - panic("scsi_devices corrupt (st)"); - tpnt = kzalloc(sizeof(struct scsi_tape), GFP_ATOMIC); if (tpnt == NULL) { - write_unlock(&st_dev_arr_lock); printk(KERN_ERR "st: Can't allocate device descriptor.\n"); goto out_put_disk; } kref_init(&tpnt->kref); tpnt->disk = disk; - sprintf(disk->disk_name, "st%d", i); disk->private_data = &tpnt->driver; disk->queue = SDp->request_queue; tpnt->driver = &st_template; - scsi_tapes[i] = tpnt; - dev_num = i; tpnt->device = SDp; if (SDp->scsi_level <= 2) @@ -4125,6 +4158,7 @@ static int st_probe(struct device *dev) STm->default_compression = ST_DONT_TOUCH; STm->default_blksize = (-1); /* No forced size */ STm->default_density = (-1); /* No forced density */ + STm->tape = tpnt; } for (i = 0; i < ST_NBR_PARTITIONS; i++) { @@ -4144,38 +4178,34 @@ static int st_probe(struct device *dev) tpnt->blksize_changed = 0; mutex_init(&tpnt->lock); - st_nr_dev++; - write_unlock(&st_dev_arr_lock); + if (!idr_pre_get(&st_index_idr, GFP_KERNEL)) { + pr_warn("st: idr expansion failed\n"); + error = -ENOMEM; + goto out_put_disk; + } - for (mode = 0; mode < ST_NBR_MODES; ++mode) { - STm = &(tpnt->modes[mode]); - for (j=0; j < 2; j++) { - cdev = cdev_alloc(); - if (!cdev) { - printk(KERN_ERR - "st%d: out of memory. Device not attached.\n", - dev_num); - goto out_free_tape; - } - cdev->owner = THIS_MODULE; - cdev->ops = &st_fops; - - error = cdev_add(cdev, - MKDEV(SCSI_TAPE_MAJOR, TAPE_MINOR(dev_num, mode, j)), - 1); - if (error) { - printk(KERN_ERR "st%d: Can't add %s-rewind mode %d\n", - dev_num, j ? "non" : "auto", mode); - printk(KERN_ERR "st%d: Device not attached.\n", dev_num); - goto out_free_tape; - } - STm->cdevs[j] = cdev; + spin_lock(&st_index_lock); + error = idr_get_new(&st_index_idr, tpnt, &dev_num); + spin_unlock(&st_index_lock); + if (error) { + pr_warn("st: idr allocation failed: %d\n", error); + goto out_put_disk; + } - } - error = do_create_class_files(tpnt, dev_num, mode); - if (error) - goto out_free_tape; + if (dev_num > ST_MAX_TAPES) { + pr_err("st: Too many tape devices (max. %d).\n", ST_MAX_TAPES); + goto out_put_index; } + + tpnt->index = dev_num; + sprintf(disk->disk_name, "st%d", dev_num); + + dev_set_drvdata(dev, tpnt); + + + error = create_cdevs(tpnt); + if (error) + goto out_remove_devs; scsi_autopm_put_device(SDp); sdev_printk(KERN_NOTICE, SDp, @@ -4186,28 +4216,12 @@ static int st_probe(struct device *dev) return 0; -out_free_tape: - for (mode=0; mode < ST_NBR_MODES; mode++) { - STm = &(tpnt->modes[mode]); - sysfs_remove_link(&tpnt->device->sdev_gendev.kobj, - "tape"); - for (j=0; j < 2; j++) { - if (STm->cdevs[j]) { - if (cdev == STm->cdevs[j]) - cdev = NULL; - device_destroy(st_sysfs_class, - MKDEV(SCSI_TAPE_MAJOR, - TAPE_MINOR(i, mode, j))); - cdev_del(STm->cdevs[j]); - } - } - } - if (cdev) - cdev_del(cdev); - write_lock(&st_dev_arr_lock); - scsi_tapes[dev_num] = NULL; - st_nr_dev--; - write_unlock(&st_dev_arr_lock); +out_remove_devs: + remove_cdevs(tpnt); +out_put_index: + spin_lock(&st_index_lock); + idr_remove(&st_index_idr, dev_num); + spin_unlock(&st_index_lock); out_put_disk: put_disk(disk); kfree(tpnt); @@ -4220,38 +4234,18 @@ out: static int st_remove(struct device *dev) { - struct scsi_device *SDp = to_scsi_device(dev); - struct scsi_tape *tpnt; - int i, j, mode; - - scsi_autopm_get_device(SDp); - write_lock(&st_dev_arr_lock); - for (i = 0; i < st_dev_max; i++) { - tpnt = scsi_tapes[i]; - if (tpnt != NULL && tpnt->device == SDp) { - scsi_tapes[i] = NULL; - st_nr_dev--; - write_unlock(&st_dev_arr_lock); - sysfs_remove_link(&tpnt->device->sdev_gendev.kobj, - "tape"); - for (mode = 0; mode < ST_NBR_MODES; ++mode) { - for (j=0; j < 2; j++) { - device_destroy(st_sysfs_class, - MKDEV(SCSI_TAPE_MAJOR, - TAPE_MINOR(i, mode, j))); - cdev_del(tpnt->modes[mode].cdevs[j]); - tpnt->modes[mode].cdevs[j] = NULL; - } - } + struct scsi_tape *tpnt = dev_get_drvdata(dev); + int index = tpnt->index; - mutex_lock(&st_ref_mutex); - kref_put(&tpnt->kref, scsi_tape_release); - mutex_unlock(&st_ref_mutex); - return 0; - } - } + scsi_autopm_get_device(to_scsi_device(dev)); + remove_cdevs(tpnt); - write_unlock(&st_dev_arr_lock); + mutex_lock(&st_ref_mutex); + kref_put(&tpnt->kref, scsi_tape_release); + mutex_unlock(&st_ref_mutex); + spin_lock(&st_index_lock); + idr_remove(&st_index_idr, index); + spin_unlock(&st_index_lock); return 0; } @@ -4283,6 +4277,11 @@ static void scsi_tape_release(struct kref *kref) return; } +static struct class st_sysfs_class = { + .name = "scsi_tape", + .dev_attrs = st_dev_attrs, +}; + static int __init init_st(void) { int err; @@ -4292,10 +4291,10 @@ static int __init init_st(void) printk(KERN_INFO "st: Version %s, fixed bufsize %d, s/g segs %d\n", verstr, st_fixed_buffer_size, st_max_sg_segs); - st_sysfs_class = class_create(THIS_MODULE, "scsi_tape"); - if (IS_ERR(st_sysfs_class)) { - printk(KERN_ERR "Unable create sysfs class for SCSI tapes\n"); - return PTR_ERR(st_sysfs_class); + err = class_register(&st_sysfs_class); + if (err) { + pr_err("Unable register sysfs class for SCSI tapes\n"); + return err; } err = register_chrdev_region(MKDEV(SCSI_TAPE_MAJOR, 0), @@ -4322,7 +4321,7 @@ err_chrdev: unregister_chrdev_region(MKDEV(SCSI_TAPE_MAJOR, 0), ST_MAX_TAPE_ENTRIES); err_class: - class_destroy(st_sysfs_class); + class_unregister(&st_sysfs_class); return err; } @@ -4332,8 +4331,7 @@ static void __exit exit_st(void) scsi_unregister_driver(&st_template.gendrv); unregister_chrdev_region(MKDEV(SCSI_TAPE_MAJOR, 0), ST_MAX_TAPE_ENTRIES); - class_destroy(st_sysfs_class); - kfree(scsi_tapes); + class_unregister(&st_sysfs_class); printk(KERN_INFO "st: Unloaded.\n"); } @@ -4405,10 +4403,9 @@ static void do_remove_sysfs_files(void) driver_remove_file(sysfs, &driver_attr_try_direct_io); } - /* The sysfs simple class interface */ static ssize_t -st_defined_show(struct device *dev, struct device_attribute *attr, char *buf) +defined_show(struct device *dev, struct device_attribute *attr, char *buf) { struct st_modedef *STm = dev_get_drvdata(dev); ssize_t l = 0; @@ -4417,10 +4414,9 @@ st_defined_show(struct device *dev, struct device_attribute *attr, char *buf) return l; } -DEVICE_ATTR(defined, S_IRUGO, st_defined_show, NULL); - static ssize_t -st_defblk_show(struct device *dev, struct device_attribute *attr, char *buf) +default_blksize_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct st_modedef *STm = dev_get_drvdata(dev); ssize_t l = 0; @@ -4429,10 +4425,10 @@ st_defblk_show(struct device *dev, struct device_attribute *attr, char *buf) return l; } -DEVICE_ATTR(default_blksize, S_IRUGO, st_defblk_show, NULL); static ssize_t -st_defdensity_show(struct device *dev, struct device_attribute *attr, char *buf) +default_density_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct st_modedef *STm = dev_get_drvdata(dev); ssize_t l = 0; @@ -4443,11 +4439,9 @@ st_defdensity_show(struct device *dev, struct device_attribute *attr, char *buf) return l; } -DEVICE_ATTR(default_density, S_IRUGO, st_defdensity_show, NULL); - static ssize_t -st_defcompression_show(struct device *dev, struct device_attribute *attr, - char *buf) +default_compression_show(struct device *dev, struct device_attribute *attr, + char *buf) { struct st_modedef *STm = dev_get_drvdata(dev); ssize_t l = 0; @@ -4456,28 +4450,14 @@ st_defcompression_show(struct device *dev, struct device_attribute *attr, return l; } -DEVICE_ATTR(default_compression, S_IRUGO, st_defcompression_show, NULL); - static ssize_t -st_options_show(struct device *dev, struct device_attribute *attr, char *buf) +options_show(struct device *dev, struct device_attribute *attr, char *buf) { struct st_modedef *STm = dev_get_drvdata(dev); - struct scsi_tape *STp; - int i, j, options; + struct scsi_tape *STp = STm->tape; + int options; ssize_t l = 0; - for (i=0; i < st_dev_max; i++) { - for (j=0; j < ST_NBR_MODES; j++) - if (&scsi_tapes[i]->modes[j] == STm) - break; - if (j < ST_NBR_MODES) - break; - } - if (i == st_dev_max) - return 0; /* should never happen */ - - STp = scsi_tapes[i]; - options = STm->do_buffer_writes ? MT_ST_BUFFER_WRITES : 0; options |= STm->do_async_writes ? MT_ST_ASYNC_WRITES : 0; options |= STm->do_read_ahead ? MT_ST_READ_AHEAD : 0; @@ -4498,66 +4478,14 @@ st_options_show(struct device *dev, struct device_attribute *attr, char *buf) return l; } -DEVICE_ATTR(options, S_IRUGO, st_options_show, NULL); - -static int do_create_class_files(struct scsi_tape *STp, int dev_num, int mode) -{ - int i, rew, error; - char name[10]; - struct device *st_class_member; - - for (rew=0; rew < 2; rew++) { - /* Make sure that the minor numbers corresponding to the four - first modes always get the same names */ - i = mode << (4 - ST_NBR_MODE_BITS); - snprintf(name, 10, "%s%s%s", rew ? "n" : "", - STp->disk->disk_name, st_formats[i]); - st_class_member = - device_create(st_sysfs_class, &STp->device->sdev_gendev, - MKDEV(SCSI_TAPE_MAJOR, - TAPE_MINOR(dev_num, mode, rew)), - &STp->modes[mode], "%s", name); - if (IS_ERR(st_class_member)) { - printk(KERN_WARNING "st%d: device_create failed\n", - dev_num); - error = PTR_ERR(st_class_member); - goto out; - } - - error = device_create_file(st_class_member, - &dev_attr_defined); - if (error) goto out; - error = device_create_file(st_class_member, - &dev_attr_default_blksize); - if (error) goto out; - error = device_create_file(st_class_member, - &dev_attr_default_density); - if (error) goto out; - error = device_create_file(st_class_member, - &dev_attr_default_compression); - if (error) goto out; - error = device_create_file(st_class_member, - &dev_attr_options); - if (error) goto out; - - if (mode == 0 && rew == 0) { - error = sysfs_create_link(&STp->device->sdev_gendev.kobj, - &st_class_member->kobj, - "tape"); - if (error) { - printk(KERN_ERR - "st%d: Can't create sysfs link from SCSI device.\n", - dev_num); - goto out; - } - } - } - - return 0; - -out: - return error; -} +static struct device_attribute st_dev_attrs[] = { + __ATTR_RO(defined), + __ATTR_RO(default_blksize), + __ATTR_RO(default_density), + __ATTR_RO(default_compression), + __ATTR_RO(options), + __ATTR_NULL, +}; /* The following functions may be useful for a larger audience. */ static int sgl_map_user_pages(struct st_buffer *STbp, diff --git a/drivers/scsi/st.h b/drivers/scsi/st.h index b548923..f3eee0f 100644 --- a/drivers/scsi/st.h +++ b/drivers/scsi/st.h @@ -66,6 +66,8 @@ struct st_modedef { unsigned char default_compression; /* 0 = don't touch, etc */ short default_density; /* Forced density, -1 = no value */ int default_blksize; /* Forced blocksize, -1 = no value */ + struct scsi_tape *tape; + struct device *devs[2]; /* Auto-rewind and non-rewind devices */ struct cdev *cdevs[2]; /* Auto-rewind and non-rewind devices */ }; @@ -76,7 +78,7 @@ struct st_modedef { #define ST_MODE_SHIFT (7 - ST_NBR_MODE_BITS) #define ST_MODE_MASK ((ST_NBR_MODES - 1) << ST_MODE_SHIFT) -#define ST_MAX_TAPES 128 +#define ST_MAX_TAPES (1 << (20 - (ST_NBR_MODE_BITS + 1))) #define ST_MAX_TAPE_ENTRIES (ST_MAX_TAPES << (ST_NBR_MODE_BITS + 1)) /* The status related to each partition */ @@ -99,6 +101,7 @@ struct scsi_tape { struct mutex lock; /* For serialization */ struct completion wait; /* For SCSI commands */ struct st_buffer *buffer; + int index; /* Drive characteristics */ unsigned char omit_blklims; diff --git a/drivers/staging/android/binder.c b/drivers/staging/android/binder.c index b1937ca..7b0ba92 100644 --- a/drivers/staging/android/binder.c +++ b/drivers/staging/android/binder.c @@ -362,71 +362,22 @@ struct binder_transaction { static void binder_defer_work(struct binder_proc *proc, enum binder_deferred_state defer); -/* - * copied from get_unused_fd_flags - */ static int task_get_unused_fd_flags(struct binder_proc *proc, int flags) { struct files_struct *files = proc->files; - int fd, error; - struct fdtable *fdt; unsigned long rlim_cur; unsigned long irqs; if (files == NULL) return -ESRCH; - error = -EMFILE; - spin_lock(&files->file_lock); - -repeat: - fdt = files_fdtable(files); - fd = find_next_zero_bit(fdt->open_fds, fdt->max_fds, files->next_fd); - - /* - * N.B. For clone tasks sharing a files structure, this test - * will limit the total number of files that can be opened. - */ - rlim_cur = 0; - if (lock_task_sighand(proc->tsk, &irqs)) { - rlim_cur = proc->tsk->signal->rlim[RLIMIT_NOFILE].rlim_cur; - unlock_task_sighand(proc->tsk, &irqs); - } - if (fd >= rlim_cur) - goto out; - - /* Do we need to expand the fd array or fd set? */ - error = expand_files(files, fd); - if (error < 0) - goto out; - - if (error) { - /* - * If we needed to expand the fs array we - * might have blocked - try again. - */ - error = -EMFILE; - goto repeat; - } - - __set_open_fd(fd, fdt); - if (flags & O_CLOEXEC) - __set_close_on_exec(fd, fdt); - else - __clear_close_on_exec(fd, fdt); - files->next_fd = fd + 1; - - /* Sanity check */ - if (fdt->fd[fd] != NULL) { - pr_warn("get_unused_fd: slot %d not NULL!\n", fd); - fdt->fd[fd] = NULL; - } + if (!lock_task_sighand(proc->tsk, &irqs)) + return -EMFILE; - error = fd; + rlim_cur = task_rlimit(proc->tsk, RLIMIT_NOFILE); + unlock_task_sighand(proc->tsk, &irqs); -out: - spin_unlock(&files->file_lock); - return error; + return __alloc_fd(files, 0, rlim_cur, flags); } /* @@ -435,28 +386,8 @@ out: static void task_fd_install( struct binder_proc *proc, unsigned int fd, struct file *file) { - struct files_struct *files = proc->files; - struct fdtable *fdt; - - if (files == NULL) - return; - - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - BUG_ON(fdt->fd[fd] != NULL); - rcu_assign_pointer(fdt->fd[fd], file); - spin_unlock(&files->file_lock); -} - -/* - * copied from __put_unused_fd in open.c - */ -static void __put_unused_fd(struct files_struct *files, unsigned int fd) -{ - struct fdtable *fdt = files_fdtable(files); - __clear_open_fd(fd, fdt); - if (fd < files->next_fd) - files->next_fd = fd; + if (proc->files) + __fd_install(proc->files, fd, file); } /* @@ -464,27 +395,12 @@ static void __put_unused_fd(struct files_struct *files, unsigned int fd) */ static long task_close_fd(struct binder_proc *proc, unsigned int fd) { - struct file *filp; - struct files_struct *files = proc->files; - struct fdtable *fdt; int retval; - if (files == NULL) + if (proc->files == NULL) return -ESRCH; - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - if (fd >= fdt->max_fds) - goto out_unlock; - filp = fdt->fd[fd]; - if (!filp) - goto out_unlock; - rcu_assign_pointer(fdt->fd[fd], NULL); - __clear_close_on_exec(fd, fdt); - __put_unused_fd(files, fd); - spin_unlock(&files->file_lock); - retval = filp_close(filp, files); - + retval = __close_fd(proc->files, fd); /* can't restart close syscall because file table entry was cleared */ if (unlikely(retval == -ERESTARTSYS || retval == -ERESTARTNOINTR || @@ -493,10 +409,6 @@ static long task_close_fd(struct binder_proc *proc, unsigned int fd) retval = -EINTR; return retval; - -out_unlock: - spin_unlock(&files->file_lock); - return -EBADF; } static void binder_set_nice(long nice) @@ -2793,6 +2705,9 @@ static int binder_mmap(struct file *filp, struct vm_area_struct *vma) const char *failure_string; struct binder_buffer *buffer; + if (proc->tsk != current) + return -EINVAL; + if ((vma->vm_end - vma->vm_start) > SZ_4M) vma->vm_end = vma->vm_start + SZ_4M; @@ -2857,7 +2772,7 @@ static int binder_mmap(struct file *filp, struct vm_area_struct *vma) binder_insert_free_buffer(proc, buffer); proc->free_async_space = proc->buffer_size / 2; barrier(); - proc->files = get_files_struct(proc->tsk); + proc->files = get_files_struct(current); proc->vma = vma; proc->vma_vm_mm = vma->vm_mm; diff --git a/drivers/staging/omapdrm/omap_gem.c b/drivers/staging/omapdrm/omap_gem.c index c828743..3434e6e 100644 --- a/drivers/staging/omapdrm/omap_gem.c +++ b/drivers/staging/omapdrm/omap_gem.c @@ -592,9 +592,8 @@ int omap_gem_mmap_obj(struct drm_gem_object *obj, * in particular in the case of mmap'd dmabufs) */ fput(vma->vm_file); - get_file(obj->filp); vma->vm_pgoff = 0; - vma->vm_file = obj->filp; + vma->vm_file = get_file(obj->filp); vma->vm_page_prot = vm_get_page_prot(vma->vm_flags); } diff --git a/drivers/tty/hvc/hvc_xen.c b/drivers/tty/hvc/hvc_xen.c index 1e456dc..2944ff8 100644 --- a/drivers/tty/hvc/hvc_xen.c +++ b/drivers/tty/hvc/hvc_xen.c @@ -21,6 +21,7 @@ #include <linux/console.h> #include <linux/delay.h> #include <linux/err.h> +#include <linux/irq.h> #include <linux/init.h> #include <linux/types.h> #include <linux/list.h> @@ -35,6 +36,7 @@ #include <xen/page.h> #include <xen/events.h> #include <xen/interface/io/console.h> +#include <xen/interface/sched.h> #include <xen/hvc-console.h> #include <xen/xenbus.h> diff --git a/drivers/tty/tty_io.c b/drivers/tty/tty_io.c index 8a5a8b0..2ea176b 100644 --- a/drivers/tty/tty_io.c +++ b/drivers/tty/tty_io.c @@ -1166,10 +1166,8 @@ ssize_t redirected_tty_write(struct file *file, const char __user *buf, struct file *p = NULL; spin_lock(&redirect_lock); - if (redirect) { - get_file(redirect); - p = redirect; - } + if (redirect) + p = get_file(redirect); spin_unlock(&redirect_lock); if (p) { @@ -2264,8 +2262,7 @@ static int tioccons(struct file *file) spin_unlock(&redirect_lock); return -EBUSY; } - get_file(file); - redirect = file; + redirect = get_file(file); spin_unlock(&redirect_lock); return 0; } @@ -2809,6 +2806,13 @@ static long tty_compat_ioctl(struct file *file, unsigned int cmd, } #endif +static int this_tty(const void *t, struct file *file, unsigned fd) +{ + if (likely(file->f_op->read != tty_read)) + return 0; + return file_tty(file) != t ? 0 : fd + 1; +} + /* * This implements the "Secure Attention Key" --- the idea is to * prevent trojan horses by killing all processes associated with this @@ -2836,8 +2840,6 @@ void __do_SAK(struct tty_struct *tty) struct task_struct *g, *p; struct pid *session; int i; - struct file *filp; - struct fdtable *fdt; if (!tty) return; @@ -2867,27 +2869,12 @@ void __do_SAK(struct tty_struct *tty) continue; } task_lock(p); - if (p->files) { - /* - * We don't take a ref to the file, so we must - * hold ->file_lock instead. - */ - spin_lock(&p->files->file_lock); - fdt = files_fdtable(p->files); - for (i = 0; i < fdt->max_fds; i++) { - filp = fcheck_files(p->files, i); - if (!filp) - continue; - if (filp->f_op->read == tty_read && - file_tty(filp) == tty) { - printk(KERN_NOTICE "SAK: killed process %d" - " (%s): fd#%d opened to the tty\n", - task_pid_nr(p), p->comm, i); - force_sig(SIGKILL, p); - break; - } - } - spin_unlock(&p->files->file_lock); + i = iterate_fd(p->files, 0, this_tty, tty); + if (i != 0) { + printk(KERN_NOTICE "SAK: killed process %d" + " (%s): fd#%d opened to the tty\n", + task_pid_nr(p), p->comm, i - 1); + force_sig(SIGKILL, p); } task_unlock(p); } while_each_thread(g, p); diff --git a/drivers/usb/gadget/f_fs.c b/drivers/usb/gadget/f_fs.c index a26c43a1..64c4ec1 100644 --- a/drivers/usb/gadget/f_fs.c +++ b/drivers/usb/gadget/f_fs.c @@ -340,7 +340,7 @@ ffs_sb_create_file(struct super_block *sb, const char *name, void *data, static int ffs_mutex_lock(struct mutex *mutex, unsigned nonblock) __attribute__((warn_unused_result, nonnull)); -static char *ffs_prepare_buffer(const char * __user buf, size_t len) +static char *ffs_prepare_buffer(const char __user *buf, size_t len) __attribute__((warn_unused_result, nonnull)); @@ -2445,7 +2445,7 @@ static int ffs_mutex_lock(struct mutex *mutex, unsigned nonblock) : mutex_lock_interruptible(mutex); } -static char *ffs_prepare_buffer(const char * __user buf, size_t len) +static char *ffs_prepare_buffer(const char __user *buf, size_t len) { char *data; diff --git a/drivers/vfio/vfio.c b/drivers/vfio/vfio.c index 17830c9..56097c6 100644 --- a/drivers/vfio/vfio.c +++ b/drivers/vfio/vfio.c @@ -1014,7 +1014,7 @@ static void vfio_group_try_dissolve_container(struct vfio_group *group) static int vfio_group_set_container(struct vfio_group *group, int container_fd) { - struct file *filep; + struct fd f; struct vfio_container *container; struct vfio_iommu_driver *driver; int ret = 0; @@ -1022,17 +1022,17 @@ static int vfio_group_set_container(struct vfio_group *group, int container_fd) if (atomic_read(&group->container_users)) return -EINVAL; - filep = fget(container_fd); - if (!filep) + f = fdget(container_fd); + if (!f.file) return -EBADF; /* Sanity check, is this really our fd? */ - if (filep->f_op != &vfio_fops) { - fput(filep); + if (f.file->f_op != &vfio_fops) { + fdput(f); return -EINVAL; } - container = filep->private_data; + container = f.file->private_data; WARN_ON(!container); /* fget ensures we don't race vfio_release */ mutex_lock(&container->group_lock); @@ -1054,8 +1054,7 @@ static int vfio_group_set_container(struct vfio_group *group, int container_fd) unlock_out: mutex_unlock(&container->group_lock); - fput(filep); - + fdput(f); return ret; } diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c index ef82a0d..99ac2cb 100644 --- a/drivers/vhost/vhost.c +++ b/drivers/vhost/vhost.c @@ -636,8 +636,8 @@ static long vhost_set_memory(struct vhost_dev *d, struct vhost_memory __user *m) static long vhost_set_vring(struct vhost_dev *d, int ioctl, void __user *argp) { - struct file *eventfp, *filep = NULL, - *pollstart = NULL, *pollstop = NULL; + struct file *eventfp, *filep = NULL; + bool pollstart = false, pollstop = false; struct eventfd_ctx *ctx = NULL; u32 __user *idxp = argp; struct vhost_virtqueue *vq; @@ -763,8 +763,8 @@ static long vhost_set_vring(struct vhost_dev *d, int ioctl, void __user *argp) break; } if (eventfp != vq->kick) { - pollstop = filep = vq->kick; - pollstart = vq->kick = eventfp; + pollstop = (filep = vq->kick) != NULL; + pollstart = (vq->kick = eventfp) != NULL; } else filep = eventfp; break; diff --git a/drivers/video/msm/mdp.c b/drivers/video/msm/mdp.c index d1f881e..2e0f3ba 100644 --- a/drivers/video/msm/mdp.c +++ b/drivers/video/msm/mdp.c @@ -257,19 +257,17 @@ int get_img(struct mdp_img *img, struct fb_info *info, unsigned long *start, unsigned long *len, struct file **filep) { - int put_needed, ret = 0; - struct file *file; - - file = fget_light(img->memory_id, &put_needed); - if (file == NULL) + int ret = 0; + struct fd f = fdget(img->memory_id); + if (f.file == NULL) return -1; - if (MAJOR(file->f_dentry->d_inode->i_rdev) == FB_MAJOR) { + if (MAJOR(f.file->f_dentry->d_inode->i_rdev) == FB_MAJOR) { *start = info->fix.smem_start; *len = info->fix.smem_len; } else ret = -1; - fput_light(file, put_needed); + fdput(f); return ret; } diff --git a/drivers/xen/events.c b/drivers/xen/events.c index 7595581..c60d162 100644 --- a/drivers/xen/events.c +++ b/drivers/xen/events.c @@ -373,11 +373,22 @@ static void unmask_evtchn(int port) { struct shared_info *s = HYPERVISOR_shared_info; unsigned int cpu = get_cpu(); + int do_hypercall = 0, evtchn_pending = 0; BUG_ON(!irqs_disabled()); - /* Slow path (hypercall) if this is a non-local port. */ - if (unlikely(cpu != cpu_from_evtchn(port))) { + if (unlikely((cpu != cpu_from_evtchn(port)))) + do_hypercall = 1; + else + evtchn_pending = sync_test_bit(port, &s->evtchn_pending[0]); + + if (unlikely(evtchn_pending && xen_hvm_domain())) + do_hypercall = 1; + + /* Slow path (hypercall) if this is a non-local port or if this is + * an hvm domain and an event is pending (hvm domains don't have + * their own implementation of irq_enable). */ + if (do_hypercall) { struct evtchn_unmask unmask = { .port = port }; (void)HYPERVISOR_event_channel_op(EVTCHNOP_unmask, &unmask); } else { @@ -390,7 +401,7 @@ static void unmask_evtchn(int port) * 'hw_resend_irq'. Just like a real IO-APIC we 'lose * the interrupt edge' if the channel is masked. */ - if (sync_test_bit(port, &s->evtchn_pending[0]) && + if (evtchn_pending && !sync_test_and_set_bit(port / BITS_PER_LONG, &vcpu_info->evtchn_pending_sel)) vcpu_info->evtchn_upcall_pending = 1; @@ -831,6 +842,7 @@ int bind_evtchn_to_irq(unsigned int evtchn) struct irq_info *info = info_for_irq(irq); WARN_ON(info == NULL || info->type != IRQT_EVTCHN); } + irq_clear_status_flags(irq, IRQ_NOREQUEST|IRQ_NOAUTOEN); out: mutex_unlock(&irq_mapping_update_lock); diff --git a/drivers/xen/gntdev.c b/drivers/xen/gntdev.c index 7f12416..5df9fd8 100644 --- a/drivers/xen/gntdev.c +++ b/drivers/xen/gntdev.c @@ -446,7 +446,7 @@ static void mn_release(struct mmu_notifier *mn, spin_unlock(&priv->lock); } -struct mmu_notifier_ops gntdev_mmu_ops = { +static struct mmu_notifier_ops gntdev_mmu_ops = { .release = mn_release, .invalidate_page = mn_invl_page, .invalidate_range_start = mn_invl_range_start, diff --git a/drivers/xen/grant-table.c b/drivers/xen/grant-table.c index 0067266..b2b0a37 100644 --- a/drivers/xen/grant-table.c +++ b/drivers/xen/grant-table.c @@ -38,6 +38,7 @@ #include <linux/vmalloc.h> #include <linux/uaccess.h> #include <linux/io.h> +#include <linux/delay.h> #include <linux/hardirq.h> #include <xen/xen.h> @@ -47,6 +48,7 @@ #include <xen/interface/memory.h> #include <xen/hvc-console.h> #include <asm/xen/hypercall.h> +#include <asm/xen/interface.h> #include <asm/pgtable.h> #include <asm/sync_bitops.h> @@ -285,10 +287,9 @@ int gnttab_grant_foreign_access(domid_t domid, unsigned long frame, } EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access); -void gnttab_update_subpage_entry_v2(grant_ref_t ref, domid_t domid, - unsigned long frame, int flags, - unsigned page_off, - unsigned length) +static void gnttab_update_subpage_entry_v2(grant_ref_t ref, domid_t domid, + unsigned long frame, int flags, + unsigned page_off, unsigned length) { gnttab_shared.v2[ref].sub_page.frame = frame; gnttab_shared.v2[ref].sub_page.page_off = page_off; @@ -345,9 +346,9 @@ bool gnttab_subpage_grants_available(void) } EXPORT_SYMBOL_GPL(gnttab_subpage_grants_available); -void gnttab_update_trans_entry_v2(grant_ref_t ref, domid_t domid, - int flags, domid_t trans_domid, - grant_ref_t trans_gref) +static void gnttab_update_trans_entry_v2(grant_ref_t ref, domid_t domid, + int flags, domid_t trans_domid, + grant_ref_t trans_gref) { gnttab_shared.v2[ref].transitive.trans_domid = trans_domid; gnttab_shared.v2[ref].transitive.gref = trans_gref; @@ -823,6 +824,52 @@ unsigned int gnttab_max_grant_frames(void) } EXPORT_SYMBOL_GPL(gnttab_max_grant_frames); +/* Handling of paged out grant targets (GNTST_eagain) */ +#define MAX_DELAY 256 +static inline void +gnttab_retry_eagain_gop(unsigned int cmd, void *gop, int16_t *status, + const char *func) +{ + unsigned delay = 1; + + do { + BUG_ON(HYPERVISOR_grant_table_op(cmd, gop, 1)); + if (*status == GNTST_eagain) + msleep(delay++); + } while ((*status == GNTST_eagain) && (delay < MAX_DELAY)); + + if (delay >= MAX_DELAY) { + printk(KERN_ERR "%s: %s eagain grant\n", func, current->comm); + *status = GNTST_bad_page; + } +} + +void gnttab_batch_map(struct gnttab_map_grant_ref *batch, unsigned count) +{ + struct gnttab_map_grant_ref *op; + + if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, batch, count)) + BUG(); + for (op = batch; op < batch + count; op++) + if (op->status == GNTST_eagain) + gnttab_retry_eagain_gop(GNTTABOP_map_grant_ref, op, + &op->status, __func__); +} +EXPORT_SYMBOL_GPL(gnttab_batch_map); + +void gnttab_batch_copy(struct gnttab_copy *batch, unsigned count) +{ + struct gnttab_copy *op; + + if (HYPERVISOR_grant_table_op(GNTTABOP_copy, batch, count)) + BUG(); + for (op = batch; op < batch + count; op++) + if (op->status == GNTST_eagain) + gnttab_retry_eagain_gop(GNTTABOP_copy, op, + &op->status, __func__); +} +EXPORT_SYMBOL_GPL(gnttab_batch_copy); + int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, struct gnttab_map_grant_ref *kmap_ops, struct page **pages, unsigned int count) @@ -836,6 +883,12 @@ int gnttab_map_refs(struct gnttab_map_grant_ref *map_ops, if (ret) return ret; + /* Retry eagain maps */ + for (i = 0; i < count; i++) + if (map_ops[i].status == GNTST_eagain) + gnttab_retry_eagain_gop(GNTTABOP_map_grant_ref, map_ops + i, + &map_ops[i].status, __func__); + if (xen_feature(XENFEAT_auto_translated_physmap)) return ret; diff --git a/drivers/xen/privcmd.c b/drivers/xen/privcmd.c index ccee0f1..ef63895 100644 --- a/drivers/xen/privcmd.c +++ b/drivers/xen/privcmd.c @@ -76,7 +76,7 @@ static void free_page_list(struct list_head *pages) */ static int gather_array(struct list_head *pagelist, unsigned nelem, size_t size, - void __user *data) + const void __user *data) { unsigned pageidx; void *pagedata; @@ -246,61 +246,117 @@ struct mmap_batch_state { domid_t domain; unsigned long va; struct vm_area_struct *vma; - int err; - - xen_pfn_t __user *user; + /* A tristate: + * 0 for no errors + * 1 if at least one error has happened (and no + * -ENOENT errors have happened) + * -ENOENT if at least 1 -ENOENT has happened. + */ + int global_error; + /* An array for individual errors */ + int *err; + + /* User-space mfn array to store errors in the second pass for V1. */ + xen_pfn_t __user *user_mfn; }; static int mmap_batch_fn(void *data, void *state) { xen_pfn_t *mfnp = data; struct mmap_batch_state *st = state; + int ret; + + ret = xen_remap_domain_mfn_range(st->vma, st->va & PAGE_MASK, *mfnp, 1, + st->vma->vm_page_prot, st->domain); - if (xen_remap_domain_mfn_range(st->vma, st->va & PAGE_MASK, *mfnp, 1, - st->vma->vm_page_prot, st->domain) < 0) { - *mfnp |= 0xf0000000U; - st->err++; + /* Store error code for second pass. */ + *(st->err++) = ret; + + /* And see if it affects the global_error. */ + if (ret < 0) { + if (ret == -ENOENT) + st->global_error = -ENOENT; + else { + /* Record that at least one error has happened. */ + if (st->global_error == 0) + st->global_error = 1; + } } st->va += PAGE_SIZE; return 0; } -static int mmap_return_errors(void *data, void *state) +static int mmap_return_errors_v1(void *data, void *state) { xen_pfn_t *mfnp = data; struct mmap_batch_state *st = state; - - return put_user(*mfnp, st->user++); + int err = *(st->err++); + + /* + * V1 encodes the error codes in the 32bit top nibble of the + * mfn (with its known limitations vis-a-vis 64 bit callers). + */ + *mfnp |= (err == -ENOENT) ? + PRIVCMD_MMAPBATCH_PAGED_ERROR : + PRIVCMD_MMAPBATCH_MFN_ERROR; + return __put_user(*mfnp, st->user_mfn++); } static struct vm_operations_struct privcmd_vm_ops; -static long privcmd_ioctl_mmap_batch(void __user *udata) +static long privcmd_ioctl_mmap_batch(void __user *udata, int version) { int ret; - struct privcmd_mmapbatch m; + struct privcmd_mmapbatch_v2 m; struct mm_struct *mm = current->mm; struct vm_area_struct *vma; unsigned long nr_pages; LIST_HEAD(pagelist); + int *err_array = NULL; struct mmap_batch_state state; if (!xen_initial_domain()) return -EPERM; - if (copy_from_user(&m, udata, sizeof(m))) - return -EFAULT; + switch (version) { + case 1: + if (copy_from_user(&m, udata, sizeof(struct privcmd_mmapbatch))) + return -EFAULT; + /* Returns per-frame error in m.arr. */ + m.err = NULL; + if (!access_ok(VERIFY_WRITE, m.arr, m.num * sizeof(*m.arr))) + return -EFAULT; + break; + case 2: + if (copy_from_user(&m, udata, sizeof(struct privcmd_mmapbatch_v2))) + return -EFAULT; + /* Returns per-frame error code in m.err. */ + if (!access_ok(VERIFY_WRITE, m.err, m.num * (sizeof(*m.err)))) + return -EFAULT; + break; + default: + return -EINVAL; + } nr_pages = m.num; if ((m.num <= 0) || (nr_pages > (LONG_MAX >> PAGE_SHIFT))) return -EINVAL; - ret = gather_array(&pagelist, m.num, sizeof(xen_pfn_t), - m.arr); + ret = gather_array(&pagelist, m.num, sizeof(xen_pfn_t), m.arr); - if (ret || list_empty(&pagelist)) + if (ret) goto out; + if (list_empty(&pagelist)) { + ret = -EINVAL; + goto out; + } + + err_array = kcalloc(m.num, sizeof(int), GFP_KERNEL); + if (err_array == NULL) { + ret = -ENOMEM; + goto out; + } down_write(&mm->mmap_sem); @@ -315,24 +371,37 @@ static long privcmd_ioctl_mmap_batch(void __user *udata) goto out; } - state.domain = m.dom; - state.vma = vma; - state.va = m.addr; - state.err = 0; + state.domain = m.dom; + state.vma = vma; + state.va = m.addr; + state.global_error = 0; + state.err = err_array; - ret = traverse_pages(m.num, sizeof(xen_pfn_t), - &pagelist, mmap_batch_fn, &state); + /* mmap_batch_fn guarantees ret == 0 */ + BUG_ON(traverse_pages(m.num, sizeof(xen_pfn_t), + &pagelist, mmap_batch_fn, &state)); up_write(&mm->mmap_sem); - if (state.err > 0) { - state.user = m.arr; + if (state.global_error && (version == 1)) { + /* Write back errors in second pass. */ + state.user_mfn = (xen_pfn_t *)m.arr; + state.err = err_array; ret = traverse_pages(m.num, sizeof(xen_pfn_t), - &pagelist, - mmap_return_errors, &state); + &pagelist, mmap_return_errors_v1, &state); + } else if (version == 2) { + ret = __copy_to_user(m.err, err_array, m.num * sizeof(int)); + if (ret) + ret = -EFAULT; } + /* If we have not had any EFAULT-like global errors then set the global + * error to -ENOENT if necessary. */ + if ((ret == 0) && (state.global_error == -ENOENT)) + ret = -ENOENT; + out: + kfree(err_array); free_page_list(&pagelist); return ret; @@ -354,7 +423,11 @@ static long privcmd_ioctl(struct file *file, break; case IOCTL_PRIVCMD_MMAPBATCH: - ret = privcmd_ioctl_mmap_batch(udata); + ret = privcmd_ioctl_mmap_batch(udata, 1); + break; + + case IOCTL_PRIVCMD_MMAPBATCH_V2: + ret = privcmd_ioctl_mmap_batch(udata, 2); break; default: @@ -380,10 +453,6 @@ static struct vm_operations_struct privcmd_vm_ops = { static int privcmd_mmap(struct file *file, struct vm_area_struct *vma) { - /* Unsupported for auto-translate guests. */ - if (xen_feature(XENFEAT_auto_translated_physmap)) - return -ENOSYS; - /* DONTCOPY is essential for Xen because copy_page_range doesn't know * how to recreate these mappings */ vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY | VM_PFNMAP; diff --git a/drivers/xen/swiotlb-xen.c b/drivers/xen/swiotlb-xen.c index 4d51948..58db6df 100644 --- a/drivers/xen/swiotlb-xen.c +++ b/drivers/xen/swiotlb-xen.c @@ -52,7 +52,7 @@ static unsigned long xen_io_tlb_nslabs; * Quick lookup value of the bus address of the IOTLB. */ -u64 start_dma_addr; +static u64 start_dma_addr; static dma_addr_t xen_phys_to_bus(phys_addr_t paddr) { @@ -144,31 +144,72 @@ xen_swiotlb_fixup(void *buf, size_t size, unsigned long nslabs) } while (i < nslabs); return 0; } +static unsigned long xen_set_nslabs(unsigned long nr_tbl) +{ + if (!nr_tbl) { + xen_io_tlb_nslabs = (64 * 1024 * 1024 >> IO_TLB_SHIFT); + xen_io_tlb_nslabs = ALIGN(xen_io_tlb_nslabs, IO_TLB_SEGSIZE); + } else + xen_io_tlb_nslabs = nr_tbl; -void __init xen_swiotlb_init(int verbose) + return xen_io_tlb_nslabs << IO_TLB_SHIFT; +} + +enum xen_swiotlb_err { + XEN_SWIOTLB_UNKNOWN = 0, + XEN_SWIOTLB_ENOMEM, + XEN_SWIOTLB_EFIXUP +}; + +static const char *xen_swiotlb_error(enum xen_swiotlb_err err) +{ + switch (err) { + case XEN_SWIOTLB_ENOMEM: + return "Cannot allocate Xen-SWIOTLB buffer\n"; + case XEN_SWIOTLB_EFIXUP: + return "Failed to get contiguous memory for DMA from Xen!\n"\ + "You either: don't have the permissions, do not have"\ + " enough free memory under 4GB, or the hypervisor memory"\ + " is too fragmented!"; + default: + break; + } + return ""; +} +int __ref xen_swiotlb_init(int verbose, bool early) { - unsigned long bytes; + unsigned long bytes, order; int rc = -ENOMEM; - unsigned long nr_tbl; - char *m = NULL; + enum xen_swiotlb_err m_ret = XEN_SWIOTLB_UNKNOWN; unsigned int repeat = 3; - nr_tbl = swiotlb_nr_tbl(); - if (nr_tbl) - xen_io_tlb_nslabs = nr_tbl; - else { - xen_io_tlb_nslabs = (64 * 1024 * 1024 >> IO_TLB_SHIFT); - xen_io_tlb_nslabs = ALIGN(xen_io_tlb_nslabs, IO_TLB_SEGSIZE); - } + xen_io_tlb_nslabs = swiotlb_nr_tbl(); retry: - bytes = xen_io_tlb_nslabs << IO_TLB_SHIFT; - + bytes = xen_set_nslabs(xen_io_tlb_nslabs); + order = get_order(xen_io_tlb_nslabs << IO_TLB_SHIFT); /* * Get IO TLB memory from any location. */ - xen_io_tlb_start = alloc_bootmem_pages(PAGE_ALIGN(bytes)); + if (early) + xen_io_tlb_start = alloc_bootmem_pages(PAGE_ALIGN(bytes)); + else { +#define SLABS_PER_PAGE (1 << (PAGE_SHIFT - IO_TLB_SHIFT)) +#define IO_TLB_MIN_SLABS ((1<<20) >> IO_TLB_SHIFT) + while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) { + xen_io_tlb_start = (void *)__get_free_pages(__GFP_NOWARN, order); + if (xen_io_tlb_start) + break; + order--; + } + if (order != get_order(bytes)) { + pr_warn("Warning: only able to allocate %ld MB " + "for software IO TLB\n", (PAGE_SIZE << order) >> 20); + xen_io_tlb_nslabs = SLABS_PER_PAGE << order; + bytes = xen_io_tlb_nslabs << IO_TLB_SHIFT; + } + } if (!xen_io_tlb_start) { - m = "Cannot allocate Xen-SWIOTLB buffer!\n"; + m_ret = XEN_SWIOTLB_ENOMEM; goto error; } xen_io_tlb_end = xen_io_tlb_start + bytes; @@ -179,17 +220,22 @@ retry: bytes, xen_io_tlb_nslabs); if (rc) { - free_bootmem(__pa(xen_io_tlb_start), PAGE_ALIGN(bytes)); - m = "Failed to get contiguous memory for DMA from Xen!\n"\ - "You either: don't have the permissions, do not have"\ - " enough free memory under 4GB, or the hypervisor memory"\ - "is too fragmented!"; + if (early) + free_bootmem(__pa(xen_io_tlb_start), PAGE_ALIGN(bytes)); + else { + free_pages((unsigned long)xen_io_tlb_start, order); + xen_io_tlb_start = NULL; + } + m_ret = XEN_SWIOTLB_EFIXUP; goto error; } start_dma_addr = xen_virt_to_bus(xen_io_tlb_start); - swiotlb_init_with_tbl(xen_io_tlb_start, xen_io_tlb_nslabs, verbose); - - return; + if (early) { + swiotlb_init_with_tbl(xen_io_tlb_start, xen_io_tlb_nslabs, verbose); + rc = 0; + } else + rc = swiotlb_late_init_with_tbl(xen_io_tlb_start, xen_io_tlb_nslabs); + return rc; error: if (repeat--) { xen_io_tlb_nslabs = max(1024UL, /* Min is 2MB */ @@ -198,10 +244,13 @@ error: (xen_io_tlb_nslabs << IO_TLB_SHIFT) >> 20); goto retry; } - xen_raw_printk("%s (rc:%d)", m, rc); - panic("%s (rc:%d)", m, rc); + pr_err("%s (rc:%d)", xen_swiotlb_error(m_ret), rc); + if (early) + panic("%s (rc:%d)", xen_swiotlb_error(m_ret), rc); + else + free_pages((unsigned long)xen_io_tlb_start, order); + return rc; } - void * xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size, dma_addr_t *dma_handle, gfp_t flags, @@ -466,14 +515,6 @@ xen_swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, } EXPORT_SYMBOL_GPL(xen_swiotlb_map_sg_attrs); -int -xen_swiotlb_map_sg(struct device *hwdev, struct scatterlist *sgl, int nelems, - enum dma_data_direction dir) -{ - return xen_swiotlb_map_sg_attrs(hwdev, sgl, nelems, dir, NULL); -} -EXPORT_SYMBOL_GPL(xen_swiotlb_map_sg); - /* * Unmap a set of streaming mode DMA translations. Again, cpu read rules * concerning calls here are the same as for swiotlb_unmap_page() above. @@ -494,14 +535,6 @@ xen_swiotlb_unmap_sg_attrs(struct device *hwdev, struct scatterlist *sgl, } EXPORT_SYMBOL_GPL(xen_swiotlb_unmap_sg_attrs); -void -xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sgl, int nelems, - enum dma_data_direction dir) -{ - return xen_swiotlb_unmap_sg_attrs(hwdev, sgl, nelems, dir, NULL); -} -EXPORT_SYMBOL_GPL(xen_swiotlb_unmap_sg); - /* * Make physical memory consistent for a set of streaming mode DMA translations * after a transfer. diff --git a/drivers/xen/sys-hypervisor.c b/drivers/xen/sys-hypervisor.c index fdb6d22..5e5ad7e2 100644 --- a/drivers/xen/sys-hypervisor.c +++ b/drivers/xen/sys-hypervisor.c @@ -114,7 +114,7 @@ static void xen_sysfs_version_destroy(void) /* UUID */ -static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer) +static ssize_t uuid_show_fallback(struct hyp_sysfs_attr *attr, char *buffer) { char *vm, *val; int ret; @@ -135,6 +135,17 @@ static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer) return ret; } +static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer) +{ + xen_domain_handle_t uuid; + int ret; + ret = HYPERVISOR_xen_version(XENVER_guest_handle, uuid); + if (ret) + return uuid_show_fallback(attr, buffer); + ret = sprintf(buffer, "%pU\n", uuid); + return ret; +} + HYPERVISOR_ATTR_RO(uuid); static int __init xen_sysfs_uuid_init(void) diff --git a/drivers/xen/tmem.c b/drivers/xen/tmem.c index 89f264c..144564e 100644 --- a/drivers/xen/tmem.c +++ b/drivers/xen/tmem.c @@ -21,6 +21,7 @@ #include <asm/xen/hypercall.h> #include <asm/xen/page.h> #include <asm/xen/hypervisor.h> +#include <xen/tmem.h> #define TMEM_CONTROL 0 #define TMEM_NEW_POOL 1 diff --git a/drivers/xen/xen-acpi-processor.c b/drivers/xen/xen-acpi-processor.c index b590ee0..316df65 100644 --- a/drivers/xen/xen-acpi-processor.c +++ b/drivers/xen/xen-acpi-processor.c @@ -98,7 +98,6 @@ static int push_cxx_to_hypervisor(struct acpi_processor *_pr) dst_cx->type = cx->type; dst_cx->latency = cx->latency; - dst_cx->power = cx->power; dst_cx->dpcnt = 0; set_xen_guest_handle(dst_cx->dp, NULL); diff --git a/drivers/xen/xen-pciback/pci_stub.c b/drivers/xen/xen-pciback/pci_stub.c index 92ff01d..961d664 100644 --- a/drivers/xen/xen-pciback/pci_stub.c +++ b/drivers/xen/xen-pciback/pci_stub.c @@ -362,6 +362,7 @@ static int __devinit pcistub_init_device(struct pci_dev *dev) else { dev_dbg(&dev->dev, "reseting (FLR, D3, etc) the device\n"); __pci_reset_function_locked(dev); + pci_restore_state(dev); } /* Now disable the device (this also ensures some private device * data is setup before we export) @@ -681,14 +682,14 @@ static pci_ers_result_t xen_pcibk_slot_reset(struct pci_dev *dev) dev_err(&dev->dev, DRV_NAME " device is not connected or owned" " by HVM, kill it\n"); kill_domain_by_device(psdev); - goto release; + goto end; } if (!test_bit(_XEN_PCIB_AERHANDLER, (unsigned long *)&psdev->pdev->sh_info->flags)) { dev_err(&dev->dev, "guest with no AER driver should have been killed\n"); - goto release; + goto end; } result = common_process(psdev, 1, XEN_PCI_OP_aer_slotreset, result); @@ -698,9 +699,9 @@ static pci_ers_result_t xen_pcibk_slot_reset(struct pci_dev *dev) "No AER slot_reset service or disconnected!\n"); kill_domain_by_device(psdev); } -release: - pcistub_device_put(psdev); end: + if (psdev) + pcistub_device_put(psdev); up_write(&pcistub_sem); return result; @@ -739,14 +740,14 @@ static pci_ers_result_t xen_pcibk_mmio_enabled(struct pci_dev *dev) dev_err(&dev->dev, DRV_NAME " device is not connected or owned" " by HVM, kill it\n"); kill_domain_by_device(psdev); - goto release; + goto end; } if (!test_bit(_XEN_PCIB_AERHANDLER, (unsigned long *)&psdev->pdev->sh_info->flags)) { dev_err(&dev->dev, "guest with no AER driver should have been killed\n"); - goto release; + goto end; } result = common_process(psdev, 1, XEN_PCI_OP_aer_mmio, result); @@ -756,9 +757,9 @@ static pci_ers_result_t xen_pcibk_mmio_enabled(struct pci_dev *dev) "No AER mmio_enabled service or disconnected!\n"); kill_domain_by_device(psdev); } -release: - pcistub_device_put(psdev); end: + if (psdev) + pcistub_device_put(psdev); up_write(&pcistub_sem); return result; } @@ -797,7 +798,7 @@ static pci_ers_result_t xen_pcibk_error_detected(struct pci_dev *dev, dev_err(&dev->dev, DRV_NAME " device is not connected or owned" " by HVM, kill it\n"); kill_domain_by_device(psdev); - goto release; + goto end; } /*Guest owns the device yet no aer handler regiested, kill guest*/ @@ -805,7 +806,7 @@ static pci_ers_result_t xen_pcibk_error_detected(struct pci_dev *dev, (unsigned long *)&psdev->pdev->sh_info->flags)) { dev_dbg(&dev->dev, "guest may have no aer driver, kill it\n"); kill_domain_by_device(psdev); - goto release; + goto end; } result = common_process(psdev, error, XEN_PCI_OP_aer_detected, result); @@ -815,9 +816,9 @@ static pci_ers_result_t xen_pcibk_error_detected(struct pci_dev *dev, "No AER error_detected service or disconnected!\n"); kill_domain_by_device(psdev); } -release: - pcistub_device_put(psdev); end: + if (psdev) + pcistub_device_put(psdev); up_write(&pcistub_sem); return result; } @@ -851,7 +852,7 @@ static void xen_pcibk_error_resume(struct pci_dev *dev) dev_err(&dev->dev, DRV_NAME " device is not connected or owned" " by HVM, kill it\n"); kill_domain_by_device(psdev); - goto release; + goto end; } if (!test_bit(_XEN_PCIB_AERHANDLER, @@ -859,13 +860,13 @@ static void xen_pcibk_error_resume(struct pci_dev *dev) dev_err(&dev->dev, "guest with no AER driver should have been killed\n"); kill_domain_by_device(psdev); - goto release; + goto end; } common_process(psdev, 1, XEN_PCI_OP_aer_resume, PCI_ERS_RESULT_RECOVERED); -release: - pcistub_device_put(psdev); end: + if (psdev) + pcistub_device_put(psdev); up_write(&pcistub_sem); return; } @@ -897,17 +898,41 @@ static inline int str_to_slot(const char *buf, int *domain, int *bus, int *slot, int *func) { int err; + char wc = '*'; err = sscanf(buf, " %x:%x:%x.%x", domain, bus, slot, func); - if (err == 4) + switch (err) { + case 3: + *func = -1; + err = sscanf(buf, " %x:%x:%x.%c", domain, bus, slot, &wc); + break; + case 2: + *slot = *func = -1; + err = sscanf(buf, " %x:%x:*.%c", domain, bus, &wc); + if (err >= 2) + ++err; + break; + } + if (err == 4 && wc == '*') return 0; else if (err < 0) return -EINVAL; /* try again without domain */ *domain = 0; + wc = '*'; err = sscanf(buf, " %x:%x.%x", bus, slot, func); - if (err == 3) + switch (err) { + case 2: + *func = -1; + err = sscanf(buf, " %x:%x.%c", bus, slot, &wc); + break; + case 1: + *slot = *func = -1; + err = sscanf(buf, " %x:*.%c", bus, &wc) + 1; + break; + } + if (err == 3 && wc == '*') return 0; return -EINVAL; @@ -930,6 +955,19 @@ static int pcistub_device_id_add(int domain, int bus, int slot, int func) { struct pcistub_device_id *pci_dev_id; unsigned long flags; + int rc = 0; + + if (slot < 0) { + for (slot = 0; !rc && slot < 32; ++slot) + rc = pcistub_device_id_add(domain, bus, slot, func); + return rc; + } + + if (func < 0) { + for (func = 0; !rc && func < 8; ++func) + rc = pcistub_device_id_add(domain, bus, slot, func); + return rc; + } pci_dev_id = kmalloc(sizeof(*pci_dev_id), GFP_KERNEL); if (!pci_dev_id) @@ -952,15 +990,15 @@ static int pcistub_device_id_add(int domain, int bus, int slot, int func) static int pcistub_device_id_remove(int domain, int bus, int slot, int func) { struct pcistub_device_id *pci_dev_id, *t; - int devfn = PCI_DEVFN(slot, func); int err = -ENOENT; unsigned long flags; spin_lock_irqsave(&device_ids_lock, flags); list_for_each_entry_safe(pci_dev_id, t, &pcistub_device_ids, slot_list) { - if (pci_dev_id->domain == domain - && pci_dev_id->bus == bus && pci_dev_id->devfn == devfn) { + if (pci_dev_id->domain == domain && pci_dev_id->bus == bus + && (slot < 0 || PCI_SLOT(pci_dev_id->devfn) == slot) + && (func < 0 || PCI_FUNC(pci_dev_id->devfn) == func)) { /* Don't break; here because it's possible the same * slot could be in the list more than once */ @@ -987,7 +1025,7 @@ static int pcistub_reg_add(int domain, int bus, int slot, int func, int reg, struct config_field *field; psdev = pcistub_device_find(domain, bus, slot, func); - if (!psdev || !psdev->dev) { + if (!psdev) { err = -ENODEV; goto out; } @@ -1011,6 +1049,8 @@ static int pcistub_reg_add(int domain, int bus, int slot, int func, int reg, if (err) kfree(field); out: + if (psdev) + pcistub_device_put(psdev); return err; } @@ -1115,10 +1155,9 @@ static ssize_t pcistub_irq_handler_switch(struct device_driver *drv, err = str_to_slot(buf, &domain, &bus, &slot, &func); if (err) - goto out; + return err; psdev = pcistub_device_find(domain, bus, slot, func); - if (!psdev) goto out; @@ -1134,6 +1173,8 @@ static ssize_t pcistub_irq_handler_switch(struct device_driver *drv, if (dev_data->isr_on) dev_data->ack_intr = 1; out: + if (psdev) + pcistub_device_put(psdev); if (!err) err = count; return err; @@ -1216,15 +1257,16 @@ static ssize_t permissive_add(struct device_driver *drv, const char *buf, err = str_to_slot(buf, &domain, &bus, &slot, &func); if (err) goto out; + if (slot < 0 || func < 0) { + err = -EINVAL; + goto out; + } psdev = pcistub_device_find(domain, bus, slot, func); if (!psdev) { err = -ENODEV; goto out; } - if (!psdev->dev) { - err = -ENODEV; - goto release; - } + dev_data = pci_get_drvdata(psdev->dev); /* the driver data for a device should never be null at this point */ if (!dev_data) { @@ -1297,17 +1339,51 @@ static int __init pcistub_init(void) if (pci_devs_to_hide && *pci_devs_to_hide) { do { + char wc = '*'; + parsed = 0; err = sscanf(pci_devs_to_hide + pos, " (%x:%x:%x.%x) %n", &domain, &bus, &slot, &func, &parsed); - if (err != 4) { + switch (err) { + case 3: + func = -1; + err = sscanf(pci_devs_to_hide + pos, + " (%x:%x:%x.%c) %n", + &domain, &bus, &slot, &wc, + &parsed); + break; + case 2: + slot = func = -1; + err = sscanf(pci_devs_to_hide + pos, + " (%x:%x:*.%c) %n", + &domain, &bus, &wc, &parsed) + 1; + break; + } + + if (err != 4 || wc != '*') { domain = 0; + wc = '*'; err = sscanf(pci_devs_to_hide + pos, " (%x:%x.%x) %n", &bus, &slot, &func, &parsed); - if (err != 3) + switch (err) { + case 2: + func = -1; + err = sscanf(pci_devs_to_hide + pos, + " (%x:%x.%c) %n", + &bus, &slot, &wc, + &parsed); + break; + case 1: + slot = func = -1; + err = sscanf(pci_devs_to_hide + pos, + " (%x:*.%c) %n", + &bus, &wc, &parsed) + 1; + break; + } + if (err != 3 || wc != '*') goto parse_error; } diff --git a/drivers/xen/xenbus/xenbus_client.c b/drivers/xen/xenbus/xenbus_client.c index b3e146e..bcf3ba4 100644 --- a/drivers/xen/xenbus/xenbus_client.c +++ b/drivers/xen/xenbus/xenbus_client.c @@ -490,8 +490,7 @@ static int xenbus_map_ring_valloc_pv(struct xenbus_device *dev, op.host_addr = arbitrary_virt_to_machine(pte).maddr; - if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1)) - BUG(); + gnttab_batch_map(&op, 1); if (op.status != GNTST_okay) { free_vm_area(area); @@ -572,8 +571,7 @@ int xenbus_map_ring(struct xenbus_device *dev, int gnt_ref, gnttab_set_map_op(&op, (unsigned long)vaddr, GNTMAP_host_map, gnt_ref, dev->otherend_id); - if (HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1)) - BUG(); + gnttab_batch_map(&op, 1); if (op.status != GNTST_okay) { xenbus_dev_fatal(dev, op.status, diff --git a/drivers/xen/xenbus/xenbus_comms.c b/drivers/xen/xenbus/xenbus_comms.c index 52fe7ad..c5aa55c 100644 --- a/drivers/xen/xenbus/xenbus_comms.c +++ b/drivers/xen/xenbus/xenbus_comms.c @@ -224,7 +224,7 @@ int xb_init_comms(void) int err; err = bind_evtchn_to_irqhandler(xen_store_evtchn, wake_waiting, 0, "xenbus", &xb_waitq); - if (err <= 0) { + if (err < 0) { printk(KERN_ERR "XENBUS request irq failed %i\n", err); return err; } diff --git a/drivers/xen/xenbus/xenbus_dev_backend.c b/drivers/xen/xenbus/xenbus_dev_backend.c index be738c4..d730008 100644 --- a/drivers/xen/xenbus/xenbus_dev_backend.c +++ b/drivers/xen/xenbus/xenbus_dev_backend.c @@ -107,7 +107,7 @@ static int xenbus_backend_mmap(struct file *file, struct vm_area_struct *vma) return 0; } -const struct file_operations xenbus_backend_fops = { +static const struct file_operations xenbus_backend_fops = { .open = xenbus_backend_open, .mmap = xenbus_backend_mmap, .unlocked_ioctl = xenbus_backend_ioctl, diff --git a/drivers/xen/xenbus/xenbus_probe.c b/drivers/xen/xenbus/xenbus_probe.c index b793723..038b71d 100644 --- a/drivers/xen/xenbus/xenbus_probe.c +++ b/drivers/xen/xenbus/xenbus_probe.c @@ -324,8 +324,8 @@ static int cmp_dev(struct device *dev, void *data) return 0; } -struct xenbus_device *xenbus_device_find(const char *nodename, - struct bus_type *bus) +static struct xenbus_device *xenbus_device_find(const char *nodename, + struct bus_type *bus) { struct xb_find_info info = { .dev = NULL, .nodename = nodename }; @@ -719,17 +719,47 @@ static int __init xenstored_local_init(void) return err; } +enum xenstore_init { + UNKNOWN, + PV, + HVM, + LOCAL, +}; static int __init xenbus_init(void) { int err = 0; + enum xenstore_init usage = UNKNOWN; + uint64_t v = 0; if (!xen_domain()) return -ENODEV; xenbus_ring_ops_init(); - if (xen_hvm_domain()) { - uint64_t v = 0; + if (xen_pv_domain()) + usage = PV; + if (xen_hvm_domain()) + usage = HVM; + if (xen_hvm_domain() && xen_initial_domain()) + usage = LOCAL; + if (xen_pv_domain() && !xen_start_info->store_evtchn) + usage = LOCAL; + if (xen_pv_domain() && xen_start_info->store_evtchn) + xenstored_ready = 1; + + switch (usage) { + case LOCAL: + err = xenstored_local_init(); + if (err) + goto out_error; + xen_store_interface = mfn_to_virt(xen_store_mfn); + break; + case PV: + xen_store_evtchn = xen_start_info->store_evtchn; + xen_store_mfn = xen_start_info->store_mfn; + xen_store_interface = mfn_to_virt(xen_store_mfn); + break; + case HVM: err = hvm_get_parameter(HVM_PARAM_STORE_EVTCHN, &v); if (err) goto out_error; @@ -738,18 +768,12 @@ static int __init xenbus_init(void) if (err) goto out_error; xen_store_mfn = (unsigned long)v; - xen_store_interface = ioremap(xen_store_mfn << PAGE_SHIFT, PAGE_SIZE); - } else { - xen_store_evtchn = xen_start_info->store_evtchn; - xen_store_mfn = xen_start_info->store_mfn; - if (xen_store_evtchn) - xenstored_ready = 1; - else { - err = xenstored_local_init(); - if (err) - goto out_error; - } - xen_store_interface = mfn_to_virt(xen_store_mfn); + xen_store_interface = + ioremap(xen_store_mfn << PAGE_SHIFT, PAGE_SIZE); + break; + default: + pr_warn("Xenstore state unknown\n"); + break; } /* Initialize the interface to xenstore. */ diff --git a/drivers/xen/xenbus/xenbus_probe_frontend.c b/drivers/xen/xenbus/xenbus_probe_frontend.c index a31b54d..3159a37 100644 --- a/drivers/xen/xenbus/xenbus_probe_frontend.c +++ b/drivers/xen/xenbus/xenbus_probe_frontend.c @@ -21,6 +21,7 @@ #include <xen/xenbus.h> #include <xen/events.h> #include <xen/page.h> +#include <xen/xen.h> #include <xen/platform_pci.h> diff --git a/drivers/xen/xenbus/xenbus_xs.c b/drivers/xen/xenbus/xenbus_xs.c index bce15cf..131dec0 100644 --- a/drivers/xen/xenbus/xenbus_xs.c +++ b/drivers/xen/xenbus/xenbus_xs.c @@ -44,6 +44,7 @@ #include <linux/rwsem.h> #include <linux/module.h> #include <linux/mutex.h> +#include <asm/xen/hypervisor.h> #include <xen/xenbus.h> #include <xen/xen.h> #include "xenbus_comms.h" @@ -622,7 +623,7 @@ static void xs_reset_watches(void) { int err, supported = 0; - if (!xen_hvm_domain()) + if (!xen_hvm_domain() || xen_initial_domain()) return; err = xenbus_scanf(XBT_NIL, "control", diff --git a/fs/9p/v9fs.c b/fs/9p/v9fs.c index b85efa7..392c5da 100644 --- a/fs/9p/v9fs.c +++ b/fs/9p/v9fs.c @@ -560,6 +560,11 @@ static int v9fs_init_inode_cache(void) */ static void v9fs_destroy_inode_cache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(v9fs_inode_cache); } diff --git a/fs/Makefile b/fs/Makefile index 2fb9779..8938f82 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -11,7 +11,7 @@ obj-y := open.o read_write.o file_table.o super.o \ attr.o bad_inode.o file.o filesystems.o namespace.o \ seq_file.o xattr.o libfs.o fs-writeback.o \ pnode.o drop_caches.o splice.o sync.o utimes.o \ - stack.o fs_struct.o statfs.o + stack.o fs_struct.o statfs.o coredump.o ifeq ($(CONFIG_BLOCK),y) obj-y += buffer.o bio.o block_dev.o direct-io.o mpage.o ioprio.o diff --git a/fs/adfs/super.c b/fs/adfs/super.c index 22a0d7e..d571229 100644 --- a/fs/adfs/super.c +++ b/fs/adfs/super.c @@ -280,6 +280,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(adfs_inode_cachep); } diff --git a/fs/affs/super.c b/fs/affs/super.c index 1f03082..b84dc73 100644 --- a/fs/affs/super.c +++ b/fs/affs/super.c @@ -147,6 +147,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(affs_inode_cachep); } diff --git a/fs/afs/super.c b/fs/afs/super.c index df8c604..4316500 100644 --- a/fs/afs/super.c +++ b/fs/afs/super.c @@ -123,6 +123,11 @@ void __exit afs_fs_exit(void) BUG(); } + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(afs_inode_cachep); _leave(""); } @@ -14,6 +14,7 @@ #include <linux/fcntl.h> #include <linux/security.h> #include <linux/evm.h> +#include <linux/ima.h> /** * inode_change_ok - check if attribute changes to an inode are allowed @@ -247,6 +248,7 @@ int notify_change(struct dentry * dentry, struct iattr * attr) if (!error) { fsnotify_change(dentry, ia_valid); + ima_inode_post_setattr(dentry); evm_inode_post_setattr(dentry, ia_valid); } diff --git a/fs/autofs4/dev-ioctl.c b/fs/autofs4/dev-ioctl.c index abf645c..a162141 100644 --- a/fs/autofs4/dev-ioctl.c +++ b/fs/autofs4/dev-ioctl.c @@ -221,20 +221,6 @@ static int test_by_type(struct path *path, void *p) return ino && ino->sbi->type & *(unsigned *)p; } -static void autofs_dev_ioctl_fd_install(unsigned int fd, struct file *file) -{ - struct files_struct *files = current->files; - struct fdtable *fdt; - - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - BUG_ON(fdt->fd[fd] != NULL); - rcu_assign_pointer(fdt->fd[fd], file); - __set_close_on_exec(fd, fdt); - spin_unlock(&files->file_lock); -} - - /* * Open a file descriptor on the autofs mount point corresponding * to the given path and device number (aka. new_encode_dev(sb->s_dev)). @@ -243,7 +229,7 @@ static int autofs_dev_ioctl_open_mountpoint(const char *name, dev_t devid) { int err, fd; - fd = get_unused_fd(); + fd = get_unused_fd_flags(O_CLOEXEC); if (likely(fd >= 0)) { struct file *filp; struct path path; @@ -264,7 +250,7 @@ static int autofs_dev_ioctl_open_mountpoint(const char *name, dev_t devid) goto out; } - autofs_dev_ioctl_fd_install(fd, filp); + fd_install(fd, filp); } return fd; diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c index da8876d..dce436e 100644 --- a/fs/autofs4/waitq.c +++ b/fs/autofs4/waitq.c @@ -175,8 +175,7 @@ static void autofs4_notify_daemon(struct autofs_sb_info *sbi, return; } - pipe = sbi->pipe; - get_file(pipe); + pipe = get_file(sbi->pipe); mutex_unlock(&sbi->wq_mutex); diff --git a/fs/befs/linuxvfs.c b/fs/befs/linuxvfs.c index 7f73a69..2b3bda8 100644 --- a/fs/befs/linuxvfs.c +++ b/fs/befs/linuxvfs.c @@ -457,6 +457,11 @@ befs_init_inodecache(void) static void befs_destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(befs_inode_cachep); } diff --git a/fs/bfs/inode.c b/fs/bfs/inode.c index b242beb..737aaa3 100644 --- a/fs/bfs/inode.c +++ b/fs/bfs/inode.c @@ -280,6 +280,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(bfs_inode_cachep); } diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index 1b52956..0225fdd 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1696,30 +1696,19 @@ static int elf_note_info_init(struct elf_note_info *info) return 0; info->psinfo = kmalloc(sizeof(*info->psinfo), GFP_KERNEL); if (!info->psinfo) - goto notes_free; + return 0; info->prstatus = kmalloc(sizeof(*info->prstatus), GFP_KERNEL); if (!info->prstatus) - goto psinfo_free; + return 0; info->fpu = kmalloc(sizeof(*info->fpu), GFP_KERNEL); if (!info->fpu) - goto prstatus_free; + return 0; #ifdef ELF_CORE_COPY_XFPREGS info->xfpu = kmalloc(sizeof(*info->xfpu), GFP_KERNEL); if (!info->xfpu) - goto fpu_free; + return 0; #endif return 1; -#ifdef ELF_CORE_COPY_XFPREGS - fpu_free: - kfree(info->fpu); -#endif - prstatus_free: - kfree(info->prstatus); - psinfo_free: - kfree(info->psinfo); - notes_free: - kfree(info->notes); - return 0; } static int fill_note_info(struct elfhdr *elf, int phdrs, diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 4c87847..b08ea47 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -107,6 +107,12 @@ void extent_io_exit(void) list_del(&eb->leak_list); kmem_cache_free(extent_buffer_cache, eb); } + + /* + * Make sure all delayed rcu free are flushed before we + * destroy caches. + */ + rcu_barrier(); if (extent_state_cache) kmem_cache_destroy(extent_state_cache); if (extent_buffer_cache) diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 2a028a5..a6ed694 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -7076,6 +7076,11 @@ static void init_once(void *foo) void btrfs_destroy_cachep(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); if (btrfs_inode_cachep) kmem_cache_destroy(btrfs_inode_cachep); if (btrfs_trans_handle_cachep) diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 27bfce5..47127c1 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1397,7 +1397,6 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file, u64 *transid, bool readonly, struct btrfs_qgroup_inherit **inherit) { - struct file *src_file; int namelen; int ret = 0; @@ -1421,25 +1420,24 @@ static noinline int btrfs_ioctl_snap_create_transid(struct file *file, ret = btrfs_mksubvol(&file->f_path, name, namelen, NULL, transid, readonly, inherit); } else { + struct fd src = fdget(fd); struct inode *src_inode; - src_file = fget(fd); - if (!src_file) { + if (!src.file) { ret = -EINVAL; goto out_drop_write; } - src_inode = src_file->f_path.dentry->d_inode; + src_inode = src.file->f_path.dentry->d_inode; if (src_inode->i_sb != file->f_path.dentry->d_inode->i_sb) { printk(KERN_INFO "btrfs: Snapshot src from " "another FS\n"); ret = -EINVAL; - fput(src_file); - goto out_drop_write; + } else { + ret = btrfs_mksubvol(&file->f_path, name, namelen, + BTRFS_I(src_inode)->root, + transid, readonly, inherit); } - ret = btrfs_mksubvol(&file->f_path, name, namelen, - BTRFS_I(src_inode)->root, - transid, readonly, inherit); - fput(src_file); + fdput(src); } out_drop_write: mnt_drop_write_file(file); @@ -2341,7 +2339,7 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, { struct inode *inode = fdentry(file)->d_inode; struct btrfs_root *root = BTRFS_I(inode)->root; - struct file *src_file; + struct fd src_file; struct inode *src; struct btrfs_trans_handle *trans; struct btrfs_path *path; @@ -2376,24 +2374,24 @@ static noinline long btrfs_ioctl_clone(struct file *file, unsigned long srcfd, if (ret) return ret; - src_file = fget(srcfd); - if (!src_file) { + src_file = fdget(srcfd); + if (!src_file.file) { ret = -EBADF; goto out_drop_write; } ret = -EXDEV; - if (src_file->f_path.mnt != file->f_path.mnt) + if (src_file.file->f_path.mnt != file->f_path.mnt) goto out_fput; - src = src_file->f_dentry->d_inode; + src = src_file.file->f_dentry->d_inode; ret = -EINVAL; if (src == inode) goto out_fput; /* the src must be open for reading */ - if (!(src_file->f_mode & FMODE_READ)) + if (!(src_file.file->f_mode & FMODE_READ)) goto out_fput; /* don't make the dst file partly checksummed */ @@ -2724,7 +2722,7 @@ out_unlock: vfree(buf); btrfs_free_path(path); out_fput: - fput(src_file); + fdput(src_file); out_drop_write: mnt_drop_write_file(file); return ret; diff --git a/fs/btrfs/reada.c b/fs/btrfs/reada.c index 48a4882..a955669 100644 --- a/fs/btrfs/reada.c +++ b/fs/btrfs/reada.c @@ -68,7 +68,7 @@ struct reada_extent { u32 blocksize; int err; struct list_head extctl; - struct kref refcnt; + int refcnt; spinlock_t lock; struct reada_zone *zones[BTRFS_MAX_MIRRORS]; int nzones; @@ -126,7 +126,7 @@ static int __readahead_hook(struct btrfs_root *root, struct extent_buffer *eb, spin_lock(&fs_info->reada_lock); re = radix_tree_lookup(&fs_info->reada_tree, index); if (re) - kref_get(&re->refcnt); + re->refcnt++; spin_unlock(&fs_info->reada_lock); if (!re) @@ -336,7 +336,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_root *root, spin_lock(&fs_info->reada_lock); re = radix_tree_lookup(&fs_info->reada_tree, index); if (re) - kref_get(&re->refcnt); + re->refcnt++; spin_unlock(&fs_info->reada_lock); if (re) @@ -352,7 +352,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_root *root, re->top = *top; INIT_LIST_HEAD(&re->extctl); spin_lock_init(&re->lock); - kref_init(&re->refcnt); + re->refcnt = 1; /* * map block @@ -398,7 +398,7 @@ static struct reada_extent *reada_find_extent(struct btrfs_root *root, if (ret == -EEXIST) { re_exist = radix_tree_lookup(&fs_info->reada_tree, index); BUG_ON(!re_exist); - kref_get(&re_exist->refcnt); + re_exist->refcnt++; spin_unlock(&fs_info->reada_lock); goto error; } @@ -465,10 +465,6 @@ error: return re_exist; } -static void reada_kref_dummy(struct kref *kr) -{ -} - static void reada_extent_put(struct btrfs_fs_info *fs_info, struct reada_extent *re) { @@ -476,7 +472,7 @@ static void reada_extent_put(struct btrfs_fs_info *fs_info, unsigned long index = re->logical >> PAGE_CACHE_SHIFT; spin_lock(&fs_info->reada_lock); - if (!kref_put(&re->refcnt, reada_kref_dummy)) { + if (--re->refcnt) { spin_unlock(&fs_info->reada_lock); return; } @@ -671,7 +667,7 @@ static int reada_start_machine_dev(struct btrfs_fs_info *fs_info, return 0; } dev->reada_next = re->logical + re->blocksize; - kref_get(&re->refcnt); + re->refcnt++; spin_unlock(&fs_info->reada_lock); diff --git a/fs/ceph/inode.c b/fs/ceph/inode.c index 4b5762e..ba95eea 100644 --- a/fs/ceph/inode.c +++ b/fs/ceph/inode.c @@ -1104,7 +1104,7 @@ int ceph_fill_trace(struct super_block *sb, struct ceph_mds_request *req, pr_err("fill_trace bad get_inode " "%llx.%llx\n", vino.ino, vino.snap); err = PTR_ERR(in); - d_delete(dn); + d_drop(dn); goto done; } dn = splice_dentry(dn, in, &have_lease, true); @@ -1277,7 +1277,7 @@ retry_lookup: in = ceph_get_inode(parent->d_sb, vino); if (IS_ERR(in)) { dout("new_inode badness\n"); - d_delete(dn); + d_drop(dn); dput(dn); err = PTR_ERR(in); goto out; diff --git a/fs/ceph/super.c b/fs/ceph/super.c index b982239..3a42d93 100644 --- a/fs/ceph/super.c +++ b/fs/ceph/super.c @@ -603,6 +603,11 @@ bad_cap: static void destroy_caches(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ceph_inode_cachep); kmem_cache_destroy(ceph_cap_cachep); kmem_cache_destroy(ceph_dentry_cachep); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index a41044a..e7931cc 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -968,6 +968,11 @@ cifs_init_inodecache(void) static void cifs_destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(cifs_inode_cachep); } diff --git a/fs/coda/inode.c b/fs/coda/inode.c index f181312..be2aa49 100644 --- a/fs/coda/inode.c +++ b/fs/coda/inode.c @@ -85,6 +85,11 @@ int coda_init_inodecache(void) void coda_destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(coda_inode_cachep); } @@ -107,43 +112,41 @@ static const struct super_operations coda_super_operations = static int get_device_index(struct coda_mount_data *data) { - struct file *file; + struct fd f; struct inode *inode; int idx; - if(data == NULL) { + if (data == NULL) { printk("coda_read_super: Bad mount data\n"); return -1; } - if(data->version != CODA_MOUNT_VERSION) { + if (data->version != CODA_MOUNT_VERSION) { printk("coda_read_super: Bad mount version\n"); return -1; } - file = fget(data->fd); - inode = NULL; - if(file) - inode = file->f_path.dentry->d_inode; - - if(!inode || !S_ISCHR(inode->i_mode) || - imajor(inode) != CODA_PSDEV_MAJOR) { - if(file) - fput(file); - - printk("coda_read_super: Bad file\n"); - return -1; + f = fdget(data->fd); + if (!f.file) + goto Ebadf; + inode = f.file->f_path.dentry->d_inode; + if (!S_ISCHR(inode->i_mode) || imajor(inode) != CODA_PSDEV_MAJOR) { + fdput(f); + goto Ebadf; } idx = iminor(inode); - fput(file); + fdput(f); - if(idx < 0 || idx >= MAX_CODADEVS) { + if (idx < 0 || idx >= MAX_CODADEVS) { printk("coda_read_super: Bad minor number\n"); return -1; } return idx; +Ebadf: + printk("coda_read_super: Bad file\n"); + return -1; } static int coda_fill_super(struct super_block *sb, void *data, int silent) diff --git a/fs/compat.c b/fs/compat.c index 1bdb350..b7a24d0 100644 --- a/fs/compat.c +++ b/fs/compat.c @@ -870,22 +870,20 @@ asmlinkage long compat_sys_old_readdir(unsigned int fd, struct compat_old_linux_dirent __user *dirent, unsigned int count) { int error; - struct file *file; - int fput_needed; + struct fd f = fdget(fd); struct compat_readdir_callback buf; - file = fget_light(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADF; buf.result = 0; buf.dirent = dirent; - error = vfs_readdir(file, compat_fillonedir, &buf); + error = vfs_readdir(f.file, compat_fillonedir, &buf); if (buf.result) error = buf.result; - fput_light(file, fput_needed); + fdput(f); return error; } @@ -949,17 +947,16 @@ efault: asmlinkage long compat_sys_getdents(unsigned int fd, struct compat_linux_dirent __user *dirent, unsigned int count) { - struct file * file; + struct fd f; struct compat_linux_dirent __user * lastdirent; struct compat_getdents_callback buf; - int fput_needed; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) return -EFAULT; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return -EBADF; buf.current_dir = dirent; @@ -967,17 +964,17 @@ asmlinkage long compat_sys_getdents(unsigned int fd, buf.count = count; buf.error = 0; - error = vfs_readdir(file, compat_filldir, &buf); + error = vfs_readdir(f.file, compat_filldir, &buf); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(file->f_pos, &lastdirent->d_off)) + if (put_user(f.file->f_pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; } - fput_light(file, fput_needed); + fdput(f); return error; } @@ -1035,17 +1032,16 @@ efault: asmlinkage long compat_sys_getdents64(unsigned int fd, struct linux_dirent64 __user * dirent, unsigned int count) { - struct file * file; + struct fd f; struct linux_dirent64 __user * lastdirent; struct compat_getdents_callback64 buf; - int fput_needed; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) return -EFAULT; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return -EBADF; buf.current_dir = dirent; @@ -1053,18 +1049,18 @@ asmlinkage long compat_sys_getdents64(unsigned int fd, buf.count = count; buf.error = 0; - error = vfs_readdir(file, compat_filldir64, &buf); + error = vfs_readdir(f.file, compat_filldir64, &buf); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - typeof(lastdirent->d_off) d_off = file->f_pos; + typeof(lastdirent->d_off) d_off = f.file->f_pos; if (__put_user_unaligned(d_off, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; } - fput_light(file, fput_needed); + fdput(f); return error; } #endif /* ! __ARCH_OMIT_COMPAT_SYS_GETDENTS64 */ @@ -1152,18 +1148,16 @@ asmlinkage ssize_t compat_sys_readv(unsigned long fd, const struct compat_iovec __user *vec, unsigned long vlen) { - struct file *file; - int fput_needed; + struct fd f = fdget(fd); ssize_t ret; loff_t pos; - file = fget_light(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADF; - pos = file->f_pos; - ret = compat_readv(file, vec, vlen, &pos); - file->f_pos = pos; - fput_light(file, fput_needed); + pos = f.file->f_pos; + ret = compat_readv(f.file, vec, vlen, &pos); + f.file->f_pos = pos; + fdput(f); return ret; } @@ -1171,19 +1165,18 @@ asmlinkage ssize_t compat_sys_preadv64(unsigned long fd, const struct compat_iovec __user *vec, unsigned long vlen, loff_t pos) { - struct file *file; - int fput_needed; + struct fd f; ssize_t ret; if (pos < 0) return -EINVAL; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return -EBADF; ret = -ESPIPE; - if (file->f_mode & FMODE_PREAD) - ret = compat_readv(file, vec, vlen, &pos); - fput_light(file, fput_needed); + if (f.file->f_mode & FMODE_PREAD) + ret = compat_readv(f.file, vec, vlen, &pos); + fdput(f); return ret; } @@ -1221,18 +1214,16 @@ asmlinkage ssize_t compat_sys_writev(unsigned long fd, const struct compat_iovec __user *vec, unsigned long vlen) { - struct file *file; - int fput_needed; + struct fd f = fdget(fd); ssize_t ret; loff_t pos; - file = fget_light(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADF; - pos = file->f_pos; - ret = compat_writev(file, vec, vlen, &pos); - file->f_pos = pos; - fput_light(file, fput_needed); + pos = f.file->f_pos; + ret = compat_writev(f.file, vec, vlen, &pos); + f.file->f_pos = pos; + fdput(f); return ret; } @@ -1240,19 +1231,18 @@ asmlinkage ssize_t compat_sys_pwritev64(unsigned long fd, const struct compat_iovec __user *vec, unsigned long vlen, loff_t pos) { - struct file *file; - int fput_needed; + struct fd f; ssize_t ret; if (pos < 0) return -EINVAL; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return -EBADF; ret = -ESPIPE; - if (file->f_mode & FMODE_PWRITE) - ret = compat_writev(file, vec, vlen, &pos); - fput_light(file, fput_needed); + if (f.file->f_mode & FMODE_PWRITE) + ret = compat_writev(f.file, vec, vlen, &pos); + fdput(f); return ret; } @@ -1802,3 +1792,25 @@ compat_sys_open_by_handle_at(int mountdirfd, return do_handle_open(mountdirfd, handle, flags); } #endif + +#ifdef __ARCH_WANT_COMPAT_SYS_SENDFILE +asmlinkage long compat_sys_sendfile(int out_fd, int in_fd, + compat_off_t __user *offset, compat_size_t count) +{ + loff_t pos; + off_t off; + ssize_t ret; + + if (offset) { + if (unlikely(get_user(off, offset))) + return -EFAULT; + pos = off; + ret = do_sendfile(out_fd, in_fd, &pos, count, MAX_NON_LFS); + if (unlikely(put_user(pos, offset))) + return -EFAULT; + return ret; + } + + return do_sendfile(out_fd, in_fd, NULL, count, 0); +} +#endif /* __ARCH_WANT_COMPAT_SYS_SENDFILE */ diff --git a/fs/compat_ioctl.c b/fs/compat_ioctl.c index 9c03a3a..f505402 100644 --- a/fs/compat_ioctl.c +++ b/fs/compat_ioctl.c @@ -1539,16 +1539,13 @@ static int compat_ioctl_check_table(unsigned int xcmd) asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, unsigned long arg) { - struct file *filp; + struct fd f = fdget(fd); int error = -EBADF; - int fput_needed; - - filp = fget_light(fd, &fput_needed); - if (!filp) + if (!f.file) goto out; /* RED-PEN how should LSM module know it's handling 32bit? */ - error = security_file_ioctl(filp, cmd, arg); + error = security_file_ioctl(f.file, cmd, arg); if (error) goto out_fput; @@ -1568,30 +1565,30 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, #if defined(CONFIG_IA64) || defined(CONFIG_X86_64) case FS_IOC_RESVSP_32: case FS_IOC_RESVSP64_32: - error = compat_ioctl_preallocate(filp, compat_ptr(arg)); + error = compat_ioctl_preallocate(f.file, compat_ptr(arg)); goto out_fput; #else case FS_IOC_RESVSP: case FS_IOC_RESVSP64: - error = ioctl_preallocate(filp, compat_ptr(arg)); + error = ioctl_preallocate(f.file, compat_ptr(arg)); goto out_fput; #endif case FIBMAP: case FIGETBSZ: case FIONREAD: - if (S_ISREG(filp->f_path.dentry->d_inode->i_mode)) + if (S_ISREG(f.file->f_path.dentry->d_inode->i_mode)) break; /*FALL THROUGH*/ default: - if (filp->f_op && filp->f_op->compat_ioctl) { - error = filp->f_op->compat_ioctl(filp, cmd, arg); + if (f.file->f_op && f.file->f_op->compat_ioctl) { + error = f.file->f_op->compat_ioctl(f.file, cmd, arg); if (error != -ENOIOCTLCMD) goto out_fput; } - if (!filp->f_op || !filp->f_op->unlocked_ioctl) + if (!f.file->f_op || !f.file->f_op->unlocked_ioctl) goto do_ioctl; break; } @@ -1599,7 +1596,7 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, if (compat_ioctl_check_table(XFORM(cmd))) goto found_handler; - error = do_ioctl_trans(fd, cmd, arg, filp); + error = do_ioctl_trans(fd, cmd, arg, f.file); if (error == -ENOIOCTLCMD) error = -ENOTTY; @@ -1608,9 +1605,9 @@ asmlinkage long compat_sys_ioctl(unsigned int fd, unsigned int cmd, found_handler: arg = (unsigned long)compat_ptr(arg); do_ioctl: - error = do_vfs_ioctl(filp, fd, cmd, arg); + error = do_vfs_ioctl(f.file, fd, cmd, arg); out_fput: - fput_light(filp, fput_needed); + fdput(f); out: return error; } diff --git a/fs/coredump.c b/fs/coredump.c new file mode 100644 index 0000000..f045bba --- /dev/null +++ b/fs/coredump.c @@ -0,0 +1,686 @@ +#include <linux/slab.h> +#include <linux/file.h> +#include <linux/fdtable.h> +#include <linux/mm.h> +#include <linux/stat.h> +#include <linux/fcntl.h> +#include <linux/swap.h> +#include <linux/string.h> +#include <linux/init.h> +#include <linux/pagemap.h> +#include <linux/perf_event.h> +#include <linux/highmem.h> +#include <linux/spinlock.h> +#include <linux/key.h> +#include <linux/personality.h> +#include <linux/binfmts.h> +#include <linux/utsname.h> +#include <linux/pid_namespace.h> +#include <linux/module.h> +#include <linux/namei.h> +#include <linux/mount.h> +#include <linux/security.h> +#include <linux/syscalls.h> +#include <linux/tsacct_kern.h> +#include <linux/cn_proc.h> +#include <linux/audit.h> +#include <linux/tracehook.h> +#include <linux/kmod.h> +#include <linux/fsnotify.h> +#include <linux/fs_struct.h> +#include <linux/pipe_fs_i.h> +#include <linux/oom.h> +#include <linux/compat.h> + +#include <asm/uaccess.h> +#include <asm/mmu_context.h> +#include <asm/tlb.h> +#include <asm/exec.h> + +#include <trace/events/task.h> +#include "internal.h" + +#include <trace/events/sched.h> + +int core_uses_pid; +char core_pattern[CORENAME_MAX_SIZE] = "core"; +unsigned int core_pipe_limit; + +struct core_name { + char *corename; + int used, size; +}; +static atomic_t call_count = ATOMIC_INIT(1); + +/* The maximal length of core_pattern is also specified in sysctl.c */ + +static int expand_corename(struct core_name *cn) +{ + char *old_corename = cn->corename; + + cn->size = CORENAME_MAX_SIZE * atomic_inc_return(&call_count); + cn->corename = krealloc(old_corename, cn->size, GFP_KERNEL); + + if (!cn->corename) { + kfree(old_corename); + return -ENOMEM; + } + + return 0; +} + +static int cn_printf(struct core_name *cn, const char *fmt, ...) +{ + char *cur; + int need; + int ret; + va_list arg; + + va_start(arg, fmt); + need = vsnprintf(NULL, 0, fmt, arg); + va_end(arg); + + if (likely(need < cn->size - cn->used - 1)) + goto out_printf; + + ret = expand_corename(cn); + if (ret) + goto expand_fail; + +out_printf: + cur = cn->corename + cn->used; + va_start(arg, fmt); + vsnprintf(cur, need + 1, fmt, arg); + va_end(arg); + cn->used += need; + return 0; + +expand_fail: + return ret; +} + +static void cn_escape(char *str) +{ + for (; *str; str++) + if (*str == '/') + *str = '!'; +} + +static int cn_print_exe_file(struct core_name *cn) +{ + struct file *exe_file; + char *pathbuf, *path; + int ret; + + exe_file = get_mm_exe_file(current->mm); + if (!exe_file) { + char *commstart = cn->corename + cn->used; + ret = cn_printf(cn, "%s (path unknown)", current->comm); + cn_escape(commstart); + return ret; + } + + pathbuf = kmalloc(PATH_MAX, GFP_TEMPORARY); + if (!pathbuf) { + ret = -ENOMEM; + goto put_exe_file; + } + + path = d_path(&exe_file->f_path, pathbuf, PATH_MAX); + if (IS_ERR(path)) { + ret = PTR_ERR(path); + goto free_buf; + } + + cn_escape(path); + + ret = cn_printf(cn, "%s", path); + +free_buf: + kfree(pathbuf); +put_exe_file: + fput(exe_file); + return ret; +} + +/* format_corename will inspect the pattern parameter, and output a + * name into corename, which must have space for at least + * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator. + */ +static int format_corename(struct core_name *cn, long signr) +{ + const struct cred *cred = current_cred(); + const char *pat_ptr = core_pattern; + int ispipe = (*pat_ptr == '|'); + int pid_in_pattern = 0; + int err = 0; + + cn->size = CORENAME_MAX_SIZE * atomic_read(&call_count); + cn->corename = kmalloc(cn->size, GFP_KERNEL); + cn->used = 0; + + if (!cn->corename) + return -ENOMEM; + + /* Repeat as long as we have more pattern to process and more output + space */ + while (*pat_ptr) { + if (*pat_ptr != '%') { + if (*pat_ptr == 0) + goto out; + err = cn_printf(cn, "%c", *pat_ptr++); + } else { + switch (*++pat_ptr) { + /* single % at the end, drop that */ + case 0: + goto out; + /* Double percent, output one percent */ + case '%': + err = cn_printf(cn, "%c", '%'); + break; + /* pid */ + case 'p': + pid_in_pattern = 1; + err = cn_printf(cn, "%d", + task_tgid_vnr(current)); + break; + /* uid */ + case 'u': + err = cn_printf(cn, "%d", cred->uid); + break; + /* gid */ + case 'g': + err = cn_printf(cn, "%d", cred->gid); + break; + /* signal that caused the coredump */ + case 's': + err = cn_printf(cn, "%ld", signr); + break; + /* UNIX time of coredump */ + case 't': { + struct timeval tv; + do_gettimeofday(&tv); + err = cn_printf(cn, "%lu", tv.tv_sec); + break; + } + /* hostname */ + case 'h': { + char *namestart = cn->corename + cn->used; + down_read(&uts_sem); + err = cn_printf(cn, "%s", + utsname()->nodename); + up_read(&uts_sem); + cn_escape(namestart); + break; + } + /* executable */ + case 'e': { + char *commstart = cn->corename + cn->used; + err = cn_printf(cn, "%s", current->comm); + cn_escape(commstart); + break; + } + case 'E': + err = cn_print_exe_file(cn); + break; + /* core limit size */ + case 'c': + err = cn_printf(cn, "%lu", + rlimit(RLIMIT_CORE)); + break; + default: + break; + } + ++pat_ptr; + } + + if (err) + return err; + } + + /* Backward compatibility with core_uses_pid: + * + * If core_pattern does not include a %p (as is the default) + * and core_uses_pid is set, then .%pid will be appended to + * the filename. Do not do this for piped commands. */ + if (!ispipe && !pid_in_pattern && core_uses_pid) { + err = cn_printf(cn, ".%d", task_tgid_vnr(current)); + if (err) + return err; + } +out: + return ispipe; +} + +static int zap_process(struct task_struct *start, int exit_code) +{ + struct task_struct *t; + int nr = 0; + + start->signal->flags = SIGNAL_GROUP_EXIT; + start->signal->group_exit_code = exit_code; + start->signal->group_stop_count = 0; + + t = start; + do { + task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK); + if (t != current && t->mm) { + sigaddset(&t->pending.signal, SIGKILL); + signal_wake_up(t, 1); + nr++; + } + } while_each_thread(start, t); + + return nr; +} + +static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, + struct core_state *core_state, int exit_code) +{ + struct task_struct *g, *p; + unsigned long flags; + int nr = -EAGAIN; + + spin_lock_irq(&tsk->sighand->siglock); + if (!signal_group_exit(tsk->signal)) { + mm->core_state = core_state; + nr = zap_process(tsk, exit_code); + } + spin_unlock_irq(&tsk->sighand->siglock); + if (unlikely(nr < 0)) + return nr; + + if (atomic_read(&mm->mm_users) == nr + 1) + goto done; + /* + * We should find and kill all tasks which use this mm, and we should + * count them correctly into ->nr_threads. We don't take tasklist + * lock, but this is safe wrt: + * + * fork: + * None of sub-threads can fork after zap_process(leader). All + * processes which were created before this point should be + * visible to zap_threads() because copy_process() adds the new + * process to the tail of init_task.tasks list, and lock/unlock + * of ->siglock provides a memory barrier. + * + * do_exit: + * The caller holds mm->mmap_sem. This means that the task which + * uses this mm can't pass exit_mm(), so it can't exit or clear + * its ->mm. + * + * de_thread: + * It does list_replace_rcu(&leader->tasks, ¤t->tasks), + * we must see either old or new leader, this does not matter. + * However, it can change p->sighand, so lock_task_sighand(p) + * must be used. Since p->mm != NULL and we hold ->mmap_sem + * it can't fail. + * + * Note also that "g" can be the old leader with ->mm == NULL + * and already unhashed and thus removed from ->thread_group. + * This is OK, __unhash_process()->list_del_rcu() does not + * clear the ->next pointer, we will find the new leader via + * next_thread(). + */ + rcu_read_lock(); + for_each_process(g) { + if (g == tsk->group_leader) + continue; + if (g->flags & PF_KTHREAD) + continue; + p = g; + do { + if (p->mm) { + if (unlikely(p->mm == mm)) { + lock_task_sighand(p, &flags); + nr += zap_process(p, exit_code); + unlock_task_sighand(p, &flags); + } + break; + } + } while_each_thread(g, p); + } + rcu_read_unlock(); +done: + atomic_set(&core_state->nr_threads, nr); + return nr; +} + +static int coredump_wait(int exit_code, struct core_state *core_state) +{ + struct task_struct *tsk = current; + struct mm_struct *mm = tsk->mm; + int core_waiters = -EBUSY; + + init_completion(&core_state->startup); + core_state->dumper.task = tsk; + core_state->dumper.next = NULL; + + down_write(&mm->mmap_sem); + if (!mm->core_state) + core_waiters = zap_threads(tsk, mm, core_state, exit_code); + up_write(&mm->mmap_sem); + + if (core_waiters > 0) { + struct core_thread *ptr; + + wait_for_completion(&core_state->startup); + /* + * Wait for all the threads to become inactive, so that + * all the thread context (extended register state, like + * fpu etc) gets copied to the memory. + */ + ptr = core_state->dumper.next; + while (ptr != NULL) { + wait_task_inactive(ptr->task, 0); + ptr = ptr->next; + } + } + + return core_waiters; +} + +static void coredump_finish(struct mm_struct *mm) +{ + struct core_thread *curr, *next; + struct task_struct *task; + + next = mm->core_state->dumper.next; + while ((curr = next) != NULL) { + next = curr->next; + task = curr->task; + /* + * see exit_mm(), curr->task must not see + * ->task == NULL before we read ->next. + */ + smp_mb(); + curr->task = NULL; + wake_up_process(task); + } + + mm->core_state = NULL; +} + +static void wait_for_dump_helpers(struct file *file) +{ + struct pipe_inode_info *pipe; + + pipe = file->f_path.dentry->d_inode->i_pipe; + + pipe_lock(pipe); + pipe->readers++; + pipe->writers--; + + while ((pipe->readers > 1) && (!signal_pending(current))) { + wake_up_interruptible_sync(&pipe->wait); + kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); + pipe_wait(pipe); + } + + pipe->readers--; + pipe->writers++; + pipe_unlock(pipe); + +} + +/* + * umh_pipe_setup + * helper function to customize the process used + * to collect the core in userspace. Specifically + * it sets up a pipe and installs it as fd 0 (stdin) + * for the process. Returns 0 on success, or + * PTR_ERR on failure. + * Note that it also sets the core limit to 1. This + * is a special value that we use to trap recursive + * core dumps + */ +static int umh_pipe_setup(struct subprocess_info *info, struct cred *new) +{ + struct file *files[2]; + struct coredump_params *cp = (struct coredump_params *)info->data; + int err = create_pipe_files(files, 0); + if (err) + return err; + + cp->file = files[1]; + + replace_fd(0, files[0], 0); + /* and disallow core files too */ + current->signal->rlim[RLIMIT_CORE] = (struct rlimit){1, 1}; + + return 0; +} + +void do_coredump(long signr, int exit_code, struct pt_regs *regs) +{ + struct core_state core_state; + struct core_name cn; + struct mm_struct *mm = current->mm; + struct linux_binfmt * binfmt; + const struct cred *old_cred; + struct cred *cred; + int retval = 0; + int flag = 0; + int ispipe; + struct files_struct *displaced; + bool need_nonrelative = false; + static atomic_t core_dump_count = ATOMIC_INIT(0); + struct coredump_params cprm = { + .signr = signr, + .regs = regs, + .limit = rlimit(RLIMIT_CORE), + /* + * We must use the same mm->flags while dumping core to avoid + * inconsistency of bit flags, since this flag is not protected + * by any locks. + */ + .mm_flags = mm->flags, + }; + + audit_core_dumps(signr); + + binfmt = mm->binfmt; + if (!binfmt || !binfmt->core_dump) + goto fail; + if (!__get_dumpable(cprm.mm_flags)) + goto fail; + + cred = prepare_creds(); + if (!cred) + goto fail; + /* + * We cannot trust fsuid as being the "true" uid of the process + * nor do we know its entire history. We only know it was tainted + * so we dump it as root in mode 2, and only into a controlled + * environment (pipe handler or fully qualified path). + */ + if (__get_dumpable(cprm.mm_flags) == SUID_DUMPABLE_SAFE) { + /* Setuid core dump mode */ + flag = O_EXCL; /* Stop rewrite attacks */ + cred->fsuid = GLOBAL_ROOT_UID; /* Dump root private */ + need_nonrelative = true; + } + + retval = coredump_wait(exit_code, &core_state); + if (retval < 0) + goto fail_creds; + + old_cred = override_creds(cred); + + /* + * Clear any false indication of pending signals that might + * be seen by the filesystem code called to write the core file. + */ + clear_thread_flag(TIF_SIGPENDING); + + ispipe = format_corename(&cn, signr); + + if (ispipe) { + int dump_count; + char **helper_argv; + + if (ispipe < 0) { + printk(KERN_WARNING "format_corename failed\n"); + printk(KERN_WARNING "Aborting core\n"); + goto fail_corename; + } + + if (cprm.limit == 1) { + /* See umh_pipe_setup() which sets RLIMIT_CORE = 1. + * + * Normally core limits are irrelevant to pipes, since + * we're not writing to the file system, but we use + * cprm.limit of 1 here as a speacial value, this is a + * consistent way to catch recursive crashes. + * We can still crash if the core_pattern binary sets + * RLIM_CORE = !1, but it runs as root, and can do + * lots of stupid things. + * + * Note that we use task_tgid_vnr here to grab the pid + * of the process group leader. That way we get the + * right pid if a thread in a multi-threaded + * core_pattern process dies. + */ + printk(KERN_WARNING + "Process %d(%s) has RLIMIT_CORE set to 1\n", + task_tgid_vnr(current), current->comm); + printk(KERN_WARNING "Aborting core\n"); + goto fail_unlock; + } + cprm.limit = RLIM_INFINITY; + + dump_count = atomic_inc_return(&core_dump_count); + if (core_pipe_limit && (core_pipe_limit < dump_count)) { + printk(KERN_WARNING "Pid %d(%s) over core_pipe_limit\n", + task_tgid_vnr(current), current->comm); + printk(KERN_WARNING "Skipping core dump\n"); + goto fail_dropcount; + } + + helper_argv = argv_split(GFP_KERNEL, cn.corename+1, NULL); + if (!helper_argv) { + printk(KERN_WARNING "%s failed to allocate memory\n", + __func__); + goto fail_dropcount; + } + + retval = call_usermodehelper_fns(helper_argv[0], helper_argv, + NULL, UMH_WAIT_EXEC, umh_pipe_setup, + NULL, &cprm); + argv_free(helper_argv); + if (retval) { + printk(KERN_INFO "Core dump to %s pipe failed\n", + cn.corename); + goto close_fail; + } + } else { + struct inode *inode; + + if (cprm.limit < binfmt->min_coredump) + goto fail_unlock; + + if (need_nonrelative && cn.corename[0] != '/') { + printk(KERN_WARNING "Pid %d(%s) can only dump core "\ + "to fully qualified path!\n", + task_tgid_vnr(current), current->comm); + printk(KERN_WARNING "Skipping core dump\n"); + goto fail_unlock; + } + + cprm.file = filp_open(cn.corename, + O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, + 0600); + if (IS_ERR(cprm.file)) + goto fail_unlock; + + inode = cprm.file->f_path.dentry->d_inode; + if (inode->i_nlink > 1) + goto close_fail; + if (d_unhashed(cprm.file->f_path.dentry)) + goto close_fail; + /* + * AK: actually i see no reason to not allow this for named + * pipes etc, but keep the previous behaviour for now. + */ + if (!S_ISREG(inode->i_mode)) + goto close_fail; + /* + * Dont allow local users get cute and trick others to coredump + * into their pre-created files. + */ + if (!uid_eq(inode->i_uid, current_fsuid())) + goto close_fail; + if (!cprm.file->f_op || !cprm.file->f_op->write) + goto close_fail; + if (do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file)) + goto close_fail; + } + + /* get us an unshared descriptor table; almost always a no-op */ + retval = unshare_files(&displaced); + if (retval) + goto close_fail; + if (displaced) + put_files_struct(displaced); + retval = binfmt->core_dump(&cprm); + if (retval) + current->signal->group_exit_code |= 0x80; + + if (ispipe && core_pipe_limit) + wait_for_dump_helpers(cprm.file); +close_fail: + if (cprm.file) + filp_close(cprm.file, NULL); +fail_dropcount: + if (ispipe) + atomic_dec(&core_dump_count); +fail_unlock: + kfree(cn.corename); +fail_corename: + coredump_finish(mm); + revert_creds(old_cred); +fail_creds: + put_cred(cred); +fail: + return; +} + +/* + * Core dumping helper functions. These are the only things you should + * do on a core-file: use only these functions to write out all the + * necessary info. + */ +int dump_write(struct file *file, const void *addr, int nr) +{ + return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr; +} +EXPORT_SYMBOL(dump_write); + +int dump_seek(struct file *file, loff_t off) +{ + int ret = 1; + + if (file->f_op->llseek && file->f_op->llseek != no_llseek) { + if (file->f_op->llseek(file, off, SEEK_CUR) < 0) + return 0; + } else { + char *buf = (char *)get_zeroed_page(GFP_KERNEL); + + if (!buf) + return 0; + while (off > 0) { + unsigned long n = off; + + if (n > PAGE_SIZE) + n = PAGE_SIZE; + if (!dump_write(file, buf, n)) { + ret = 0; + break; + } + off -= n; + } + free_page((unsigned long)buf); + } + return ret; +} +EXPORT_SYMBOL(dump_seek); diff --git a/fs/dcache.c b/fs/dcache.c index 693f95b..3a463d0 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -2113,7 +2113,7 @@ again: inode = dentry->d_inode; isdir = S_ISDIR(inode->i_mode); if (dentry->d_count == 1) { - if (inode && !spin_trylock(&inode->i_lock)) { + if (!spin_trylock(&inode->i_lock)) { spin_unlock(&dentry->d_lock); cpu_relax(); goto again; diff --git a/fs/ecryptfs/main.c b/fs/ecryptfs/main.c index 24bb043..4e0886c 100644 --- a/fs/ecryptfs/main.c +++ b/fs/ecryptfs/main.c @@ -711,6 +711,12 @@ static void ecryptfs_free_kmem_caches(void) { int i; + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); + for (i = 0; i < ARRAY_SIZE(ecryptfs_cache_infos); i++) { struct ecryptfs_cache_info *info; diff --git a/fs/efs/super.c b/fs/efs/super.c index e755ec7..2002431 100644 --- a/fs/efs/super.c +++ b/fs/efs/super.c @@ -96,6 +96,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(efs_inode_cachep); } diff --git a/fs/eventpoll.c b/fs/eventpoll.c index eedec84..cd96649 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -1810,7 +1810,7 @@ SYSCALL_DEFINE4(epoll_wait, int, epfd, struct epoll_event __user *, events, int, maxevents, int, timeout) { int error; - struct file *file; + struct fd f; struct eventpoll *ep; /* The maximum number of event must be greater than zero */ @@ -1818,38 +1818,33 @@ SYSCALL_DEFINE4(epoll_wait, int, epfd, struct epoll_event __user *, events, return -EINVAL; /* Verify that the area passed by the user is writeable */ - if (!access_ok(VERIFY_WRITE, events, maxevents * sizeof(struct epoll_event))) { - error = -EFAULT; - goto error_return; - } + if (!access_ok(VERIFY_WRITE, events, maxevents * sizeof(struct epoll_event))) + return -EFAULT; /* Get the "struct file *" for the eventpoll file */ - error = -EBADF; - file = fget(epfd); - if (!file) - goto error_return; + f = fdget(epfd); + if (!f.file) + return -EBADF; /* * We have to check that the file structure underneath the fd * the user passed to us _is_ an eventpoll file. */ error = -EINVAL; - if (!is_file_epoll(file)) + if (!is_file_epoll(f.file)) goto error_fput; /* * At this point it is safe to assume that the "private_data" contains * our own data structure. */ - ep = file->private_data; + ep = f.file->private_data; /* Time to fish for events ... */ error = ep_poll(ep, events, maxevents, timeout); error_fput: - fput(file); -error_return: - + fdput(f); return error; } @@ -66,19 +66,8 @@ #include <trace/events/sched.h> -int core_uses_pid; -char core_pattern[CORENAME_MAX_SIZE] = "core"; -unsigned int core_pipe_limit; int suid_dumpable = 0; -struct core_name { - char *corename; - int used, size; -}; -static atomic_t call_count = ATOMIC_INIT(1); - -/* The maximal length of core_pattern is also specified in sysctl.c */ - static LIST_HEAD(formats); static DEFINE_RWLOCK(binfmt_lock); @@ -1006,40 +995,6 @@ no_thread_group: return 0; } -/* - * These functions flushes out all traces of the currently running executable - * so that a new one can be started - */ -static void flush_old_files(struct files_struct * files) -{ - long j = -1; - struct fdtable *fdt; - - spin_lock(&files->file_lock); - for (;;) { - unsigned long set, i; - - j++; - i = j * BITS_PER_LONG; - fdt = files_fdtable(files); - if (i >= fdt->max_fds) - break; - set = fdt->close_on_exec[j]; - if (!set) - continue; - fdt->close_on_exec[j] = 0; - spin_unlock(&files->file_lock); - for ( ; set ; i++,set >>= 1) { - if (set & 1) { - sys_close(i); - } - } - spin_lock(&files->file_lock); - - } - spin_unlock(&files->file_lock); -} - char *get_task_comm(char *buf, struct task_struct *tsk) { /* buf must be at least sizeof(tsk->comm) in size */ @@ -1050,6 +1005,11 @@ char *get_task_comm(char *buf, struct task_struct *tsk) } EXPORT_SYMBOL_GPL(get_task_comm); +/* + * These functions flushes out all traces of the currently running executable + * so that a new one can be started + */ + void set_task_comm(struct task_struct *tsk, char *buf) { task_lock(tsk); @@ -1171,7 +1131,7 @@ void setup_new_exec(struct linux_binprm * bprm) current->self_exec_id++; flush_signal_handlers(current, 0); - flush_old_files(current->files); + do_close_on_exec(current->files); } EXPORT_SYMBOL(setup_new_exec); @@ -1632,353 +1592,6 @@ void set_binfmt(struct linux_binfmt *new) EXPORT_SYMBOL(set_binfmt); -static int expand_corename(struct core_name *cn) -{ - char *old_corename = cn->corename; - - cn->size = CORENAME_MAX_SIZE * atomic_inc_return(&call_count); - cn->corename = krealloc(old_corename, cn->size, GFP_KERNEL); - - if (!cn->corename) { - kfree(old_corename); - return -ENOMEM; - } - - return 0; -} - -static int cn_printf(struct core_name *cn, const char *fmt, ...) -{ - char *cur; - int need; - int ret; - va_list arg; - - va_start(arg, fmt); - need = vsnprintf(NULL, 0, fmt, arg); - va_end(arg); - - if (likely(need < cn->size - cn->used - 1)) - goto out_printf; - - ret = expand_corename(cn); - if (ret) - goto expand_fail; - -out_printf: - cur = cn->corename + cn->used; - va_start(arg, fmt); - vsnprintf(cur, need + 1, fmt, arg); - va_end(arg); - cn->used += need; - return 0; - -expand_fail: - return ret; -} - -static void cn_escape(char *str) -{ - for (; *str; str++) - if (*str == '/') - *str = '!'; -} - -static int cn_print_exe_file(struct core_name *cn) -{ - struct file *exe_file; - char *pathbuf, *path; - int ret; - - exe_file = get_mm_exe_file(current->mm); - if (!exe_file) { - char *commstart = cn->corename + cn->used; - ret = cn_printf(cn, "%s (path unknown)", current->comm); - cn_escape(commstart); - return ret; - } - - pathbuf = kmalloc(PATH_MAX, GFP_TEMPORARY); - if (!pathbuf) { - ret = -ENOMEM; - goto put_exe_file; - } - - path = d_path(&exe_file->f_path, pathbuf, PATH_MAX); - if (IS_ERR(path)) { - ret = PTR_ERR(path); - goto free_buf; - } - - cn_escape(path); - - ret = cn_printf(cn, "%s", path); - -free_buf: - kfree(pathbuf); -put_exe_file: - fput(exe_file); - return ret; -} - -/* format_corename will inspect the pattern parameter, and output a - * name into corename, which must have space for at least - * CORENAME_MAX_SIZE bytes plus one byte for the zero terminator. - */ -static int format_corename(struct core_name *cn, long signr) -{ - const struct cred *cred = current_cred(); - const char *pat_ptr = core_pattern; - int ispipe = (*pat_ptr == '|'); - int pid_in_pattern = 0; - int err = 0; - - cn->size = CORENAME_MAX_SIZE * atomic_read(&call_count); - cn->corename = kmalloc(cn->size, GFP_KERNEL); - cn->used = 0; - - if (!cn->corename) - return -ENOMEM; - - /* Repeat as long as we have more pattern to process and more output - space */ - while (*pat_ptr) { - if (*pat_ptr != '%') { - if (*pat_ptr == 0) - goto out; - err = cn_printf(cn, "%c", *pat_ptr++); - } else { - switch (*++pat_ptr) { - /* single % at the end, drop that */ - case 0: - goto out; - /* Double percent, output one percent */ - case '%': - err = cn_printf(cn, "%c", '%'); - break; - /* pid */ - case 'p': - pid_in_pattern = 1; - err = cn_printf(cn, "%d", - task_tgid_vnr(current)); - break; - /* uid */ - case 'u': - err = cn_printf(cn, "%d", cred->uid); - break; - /* gid */ - case 'g': - err = cn_printf(cn, "%d", cred->gid); - break; - /* signal that caused the coredump */ - case 's': - err = cn_printf(cn, "%ld", signr); - break; - /* UNIX time of coredump */ - case 't': { - struct timeval tv; - do_gettimeofday(&tv); - err = cn_printf(cn, "%lu", tv.tv_sec); - break; - } - /* hostname */ - case 'h': { - char *namestart = cn->corename + cn->used; - down_read(&uts_sem); - err = cn_printf(cn, "%s", - utsname()->nodename); - up_read(&uts_sem); - cn_escape(namestart); - break; - } - /* executable */ - case 'e': { - char *commstart = cn->corename + cn->used; - err = cn_printf(cn, "%s", current->comm); - cn_escape(commstart); - break; - } - case 'E': - err = cn_print_exe_file(cn); - break; - /* core limit size */ - case 'c': - err = cn_printf(cn, "%lu", - rlimit(RLIMIT_CORE)); - break; - default: - break; - } - ++pat_ptr; - } - - if (err) - return err; - } - - /* Backward compatibility with core_uses_pid: - * - * If core_pattern does not include a %p (as is the default) - * and core_uses_pid is set, then .%pid will be appended to - * the filename. Do not do this for piped commands. */ - if (!ispipe && !pid_in_pattern && core_uses_pid) { - err = cn_printf(cn, ".%d", task_tgid_vnr(current)); - if (err) - return err; - } -out: - return ispipe; -} - -static int zap_process(struct task_struct *start, int exit_code) -{ - struct task_struct *t; - int nr = 0; - - start->signal->flags = SIGNAL_GROUP_EXIT; - start->signal->group_exit_code = exit_code; - start->signal->group_stop_count = 0; - - t = start; - do { - task_clear_jobctl_pending(t, JOBCTL_PENDING_MASK); - if (t != current && t->mm) { - sigaddset(&t->pending.signal, SIGKILL); - signal_wake_up(t, 1); - nr++; - } - } while_each_thread(start, t); - - return nr; -} - -static inline int zap_threads(struct task_struct *tsk, struct mm_struct *mm, - struct core_state *core_state, int exit_code) -{ - struct task_struct *g, *p; - unsigned long flags; - int nr = -EAGAIN; - - spin_lock_irq(&tsk->sighand->siglock); - if (!signal_group_exit(tsk->signal)) { - mm->core_state = core_state; - nr = zap_process(tsk, exit_code); - } - spin_unlock_irq(&tsk->sighand->siglock); - if (unlikely(nr < 0)) - return nr; - - if (atomic_read(&mm->mm_users) == nr + 1) - goto done; - /* - * We should find and kill all tasks which use this mm, and we should - * count them correctly into ->nr_threads. We don't take tasklist - * lock, but this is safe wrt: - * - * fork: - * None of sub-threads can fork after zap_process(leader). All - * processes which were created before this point should be - * visible to zap_threads() because copy_process() adds the new - * process to the tail of init_task.tasks list, and lock/unlock - * of ->siglock provides a memory barrier. - * - * do_exit: - * The caller holds mm->mmap_sem. This means that the task which - * uses this mm can't pass exit_mm(), so it can't exit or clear - * its ->mm. - * - * de_thread: - * It does list_replace_rcu(&leader->tasks, ¤t->tasks), - * we must see either old or new leader, this does not matter. - * However, it can change p->sighand, so lock_task_sighand(p) - * must be used. Since p->mm != NULL and we hold ->mmap_sem - * it can't fail. - * - * Note also that "g" can be the old leader with ->mm == NULL - * and already unhashed and thus removed from ->thread_group. - * This is OK, __unhash_process()->list_del_rcu() does not - * clear the ->next pointer, we will find the new leader via - * next_thread(). - */ - rcu_read_lock(); - for_each_process(g) { - if (g == tsk->group_leader) - continue; - if (g->flags & PF_KTHREAD) - continue; - p = g; - do { - if (p->mm) { - if (unlikely(p->mm == mm)) { - lock_task_sighand(p, &flags); - nr += zap_process(p, exit_code); - unlock_task_sighand(p, &flags); - } - break; - } - } while_each_thread(g, p); - } - rcu_read_unlock(); -done: - atomic_set(&core_state->nr_threads, nr); - return nr; -} - -static int coredump_wait(int exit_code, struct core_state *core_state) -{ - struct task_struct *tsk = current; - struct mm_struct *mm = tsk->mm; - int core_waiters = -EBUSY; - - init_completion(&core_state->startup); - core_state->dumper.task = tsk; - core_state->dumper.next = NULL; - - down_write(&mm->mmap_sem); - if (!mm->core_state) - core_waiters = zap_threads(tsk, mm, core_state, exit_code); - up_write(&mm->mmap_sem); - - if (core_waiters > 0) { - struct core_thread *ptr; - - wait_for_completion(&core_state->startup); - /* - * Wait for all the threads to become inactive, so that - * all the thread context (extended register state, like - * fpu etc) gets copied to the memory. - */ - ptr = core_state->dumper.next; - while (ptr != NULL) { - wait_task_inactive(ptr->task, 0); - ptr = ptr->next; - } - } - - return core_waiters; -} - -static void coredump_finish(struct mm_struct *mm) -{ - struct core_thread *curr, *next; - struct task_struct *task; - - next = mm->core_state->dumper.next; - while ((curr = next) != NULL) { - next = curr->next; - task = curr->task; - /* - * see exit_mm(), curr->task must not see - * ->task == NULL before we read ->next. - */ - smp_mb(); - curr->task = NULL; - wake_up_process(task); - } - - mm->core_state = NULL; -} - /* * set_dumpable converts traditional three-value dumpable to two flags and * stores them into mm->flags. It modifies lower two bits of mm->flags, but @@ -2020,7 +1633,7 @@ void set_dumpable(struct mm_struct *mm, int value) } } -static int __get_dumpable(unsigned long mm_flags) +int __get_dumpable(unsigned long mm_flags) { int ret; @@ -2032,290 +1645,3 @@ int get_dumpable(struct mm_struct *mm) { return __get_dumpable(mm->flags); } - -static void wait_for_dump_helpers(struct file *file) -{ - struct pipe_inode_info *pipe; - - pipe = file->f_path.dentry->d_inode->i_pipe; - - pipe_lock(pipe); - pipe->readers++; - pipe->writers--; - - while ((pipe->readers > 1) && (!signal_pending(current))) { - wake_up_interruptible_sync(&pipe->wait); - kill_fasync(&pipe->fasync_readers, SIGIO, POLL_IN); - pipe_wait(pipe); - } - - pipe->readers--; - pipe->writers++; - pipe_unlock(pipe); - -} - - -/* - * umh_pipe_setup - * helper function to customize the process used - * to collect the core in userspace. Specifically - * it sets up a pipe and installs it as fd 0 (stdin) - * for the process. Returns 0 on success, or - * PTR_ERR on failure. - * Note that it also sets the core limit to 1. This - * is a special value that we use to trap recursive - * core dumps - */ -static int umh_pipe_setup(struct subprocess_info *info, struct cred *new) -{ - struct file *files[2]; - struct fdtable *fdt; - struct coredump_params *cp = (struct coredump_params *)info->data; - struct files_struct *cf = current->files; - int err = create_pipe_files(files, 0); - if (err) - return err; - - cp->file = files[1]; - - sys_close(0); - fd_install(0, files[0]); - spin_lock(&cf->file_lock); - fdt = files_fdtable(cf); - __set_open_fd(0, fdt); - __clear_close_on_exec(0, fdt); - spin_unlock(&cf->file_lock); - - /* and disallow core files too */ - current->signal->rlim[RLIMIT_CORE] = (struct rlimit){1, 1}; - - return 0; -} - -void do_coredump(long signr, int exit_code, struct pt_regs *regs) -{ - struct core_state core_state; - struct core_name cn; - struct mm_struct *mm = current->mm; - struct linux_binfmt * binfmt; - const struct cred *old_cred; - struct cred *cred; - int retval = 0; - int flag = 0; - int ispipe; - bool need_nonrelative = false; - static atomic_t core_dump_count = ATOMIC_INIT(0); - struct coredump_params cprm = { - .signr = signr, - .regs = regs, - .limit = rlimit(RLIMIT_CORE), - /* - * We must use the same mm->flags while dumping core to avoid - * inconsistency of bit flags, since this flag is not protected - * by any locks. - */ - .mm_flags = mm->flags, - }; - - audit_core_dumps(signr); - - binfmt = mm->binfmt; - if (!binfmt || !binfmt->core_dump) - goto fail; - if (!__get_dumpable(cprm.mm_flags)) - goto fail; - - cred = prepare_creds(); - if (!cred) - goto fail; - /* - * We cannot trust fsuid as being the "true" uid of the process - * nor do we know its entire history. We only know it was tainted - * so we dump it as root in mode 2, and only into a controlled - * environment (pipe handler or fully qualified path). - */ - if (__get_dumpable(cprm.mm_flags) == SUID_DUMPABLE_SAFE) { - /* Setuid core dump mode */ - flag = O_EXCL; /* Stop rewrite attacks */ - cred->fsuid = GLOBAL_ROOT_UID; /* Dump root private */ - need_nonrelative = true; - } - - retval = coredump_wait(exit_code, &core_state); - if (retval < 0) - goto fail_creds; - - old_cred = override_creds(cred); - - /* - * Clear any false indication of pending signals that might - * be seen by the filesystem code called to write the core file. - */ - clear_thread_flag(TIF_SIGPENDING); - - ispipe = format_corename(&cn, signr); - - if (ispipe) { - int dump_count; - char **helper_argv; - - if (ispipe < 0) { - printk(KERN_WARNING "format_corename failed\n"); - printk(KERN_WARNING "Aborting core\n"); - goto fail_corename; - } - - if (cprm.limit == 1) { - /* See umh_pipe_setup() which sets RLIMIT_CORE = 1. - * - * Normally core limits are irrelevant to pipes, since - * we're not writing to the file system, but we use - * cprm.limit of 1 here as a speacial value, this is a - * consistent way to catch recursive crashes. - * We can still crash if the core_pattern binary sets - * RLIM_CORE = !1, but it runs as root, and can do - * lots of stupid things. - * - * Note that we use task_tgid_vnr here to grab the pid - * of the process group leader. That way we get the - * right pid if a thread in a multi-threaded - * core_pattern process dies. - */ - printk(KERN_WARNING - "Process %d(%s) has RLIMIT_CORE set to 1\n", - task_tgid_vnr(current), current->comm); - printk(KERN_WARNING "Aborting core\n"); - goto fail_unlock; - } - cprm.limit = RLIM_INFINITY; - - dump_count = atomic_inc_return(&core_dump_count); - if (core_pipe_limit && (core_pipe_limit < dump_count)) { - printk(KERN_WARNING "Pid %d(%s) over core_pipe_limit\n", - task_tgid_vnr(current), current->comm); - printk(KERN_WARNING "Skipping core dump\n"); - goto fail_dropcount; - } - - helper_argv = argv_split(GFP_KERNEL, cn.corename+1, NULL); - if (!helper_argv) { - printk(KERN_WARNING "%s failed to allocate memory\n", - __func__); - goto fail_dropcount; - } - - retval = call_usermodehelper_fns(helper_argv[0], helper_argv, - NULL, UMH_WAIT_EXEC, umh_pipe_setup, - NULL, &cprm); - argv_free(helper_argv); - if (retval) { - printk(KERN_INFO "Core dump to %s pipe failed\n", - cn.corename); - goto close_fail; - } - } else { - struct inode *inode; - - if (cprm.limit < binfmt->min_coredump) - goto fail_unlock; - - if (need_nonrelative && cn.corename[0] != '/') { - printk(KERN_WARNING "Pid %d(%s) can only dump core "\ - "to fully qualified path!\n", - task_tgid_vnr(current), current->comm); - printk(KERN_WARNING "Skipping core dump\n"); - goto fail_unlock; - } - - cprm.file = filp_open(cn.corename, - O_CREAT | 2 | O_NOFOLLOW | O_LARGEFILE | flag, - 0600); - if (IS_ERR(cprm.file)) - goto fail_unlock; - - inode = cprm.file->f_path.dentry->d_inode; - if (inode->i_nlink > 1) - goto close_fail; - if (d_unhashed(cprm.file->f_path.dentry)) - goto close_fail; - /* - * AK: actually i see no reason to not allow this for named - * pipes etc, but keep the previous behaviour for now. - */ - if (!S_ISREG(inode->i_mode)) - goto close_fail; - /* - * Dont allow local users get cute and trick others to coredump - * into their pre-created files. - */ - if (!uid_eq(inode->i_uid, current_fsuid())) - goto close_fail; - if (!cprm.file->f_op || !cprm.file->f_op->write) - goto close_fail; - if (do_truncate(cprm.file->f_path.dentry, 0, 0, cprm.file)) - goto close_fail; - } - - retval = binfmt->core_dump(&cprm); - if (retval) - current->signal->group_exit_code |= 0x80; - - if (ispipe && core_pipe_limit) - wait_for_dump_helpers(cprm.file); -close_fail: - if (cprm.file) - filp_close(cprm.file, NULL); -fail_dropcount: - if (ispipe) - atomic_dec(&core_dump_count); -fail_unlock: - kfree(cn.corename); -fail_corename: - coredump_finish(mm); - revert_creds(old_cred); -fail_creds: - put_cred(cred); -fail: - return; -} - -/* - * Core dumping helper functions. These are the only things you should - * do on a core-file: use only these functions to write out all the - * necessary info. - */ -int dump_write(struct file *file, const void *addr, int nr) -{ - return access_ok(VERIFY_READ, addr, nr) && file->f_op->write(file, addr, nr, &file->f_pos) == nr; -} -EXPORT_SYMBOL(dump_write); - -int dump_seek(struct file *file, loff_t off) -{ - int ret = 1; - - if (file->f_op->llseek && file->f_op->llseek != no_llseek) { - if (file->f_op->llseek(file, off, SEEK_CUR) < 0) - return 0; - } else { - char *buf = (char *)get_zeroed_page(GFP_KERNEL); - - if (!buf) - return 0; - while (off > 0) { - unsigned long n = off; - - if (n > PAGE_SIZE) - n = PAGE_SIZE; - if (!dump_write(file, buf, n)) { - ret = 0; - break; - } - off -= n; - } - free_page((unsigned long)buf); - } - return ret; -} -EXPORT_SYMBOL(dump_seek); diff --git a/fs/exofs/super.c b/fs/exofs/super.c index dde41a7..59e3bbf 100644 --- a/fs/exofs/super.c +++ b/fs/exofs/super.c @@ -206,6 +206,11 @@ static int init_inodecache(void) */ static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(exofs_inode_cachep); } diff --git a/fs/ext2/super.c b/fs/ext2/super.c index af74d9e..6c205d0 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -206,6 +206,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ext2_inode_cachep); } diff --git a/fs/ext3/super.c b/fs/ext3/super.c index 09b8455..bd29894 100644 --- a/fs/ext3/super.c +++ b/fs/ext3/super.c @@ -532,6 +532,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ext3_inode_cachep); } diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 7f7dad7..5439d6a 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -233,7 +233,7 @@ group_extend_out: case EXT4_IOC_MOVE_EXT: { struct move_extent me; - struct file *donor_filp; + struct fd donor; int err; if (!(filp->f_mode & FMODE_READ) || @@ -245,11 +245,11 @@ group_extend_out: return -EFAULT; me.moved_len = 0; - donor_filp = fget(me.donor_fd); - if (!donor_filp) + donor = fdget(me.donor_fd); + if (!donor.file) return -EBADF; - if (!(donor_filp->f_mode & FMODE_WRITE)) { + if (!(donor.file->f_mode & FMODE_WRITE)) { err = -EBADF; goto mext_out; } @@ -258,14 +258,15 @@ group_extend_out: EXT4_FEATURE_RO_COMPAT_BIGALLOC)) { ext4_msg(sb, KERN_ERR, "Online defrag not supported with bigalloc"); - return -EOPNOTSUPP; + err = -EOPNOTSUPP; + goto mext_out; } err = mnt_want_write_file(filp); if (err) goto mext_out; - err = ext4_move_extents(filp, donor_filp, me.orig_start, + err = ext4_move_extents(filp, donor.file, me.orig_start, me.donor_start, me.len, &me.moved_len); mnt_drop_write_file(filp); @@ -273,7 +274,7 @@ group_extend_out: &me, sizeof(me))) err = -EFAULT; mext_out: - fput(donor_filp); + fdput(donor); return err; } diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 1f15cc8..69c55d4 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1019,6 +1019,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ext4_inode_cachep); } diff --git a/fs/fat/inode.c b/fs/fat/inode.c index 47d9eb0..4e5a6ac 100644 --- a/fs/fat/inode.c +++ b/fs/fat/inode.c @@ -521,6 +521,11 @@ static int __init fat_init_inodecache(void) static void __exit fat_destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(fat_inode_cachep); } @@ -26,124 +26,6 @@ #include <asm/siginfo.h> #include <asm/uaccess.h> -void set_close_on_exec(unsigned int fd, int flag) -{ - struct files_struct *files = current->files; - struct fdtable *fdt; - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - if (flag) - __set_close_on_exec(fd, fdt); - else - __clear_close_on_exec(fd, fdt); - spin_unlock(&files->file_lock); -} - -static bool get_close_on_exec(unsigned int fd) -{ - struct files_struct *files = current->files; - struct fdtable *fdt; - bool res; - rcu_read_lock(); - fdt = files_fdtable(files); - res = close_on_exec(fd, fdt); - rcu_read_unlock(); - return res; -} - -SYSCALL_DEFINE3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags) -{ - int err = -EBADF; - struct file * file, *tofree; - struct files_struct * files = current->files; - struct fdtable *fdt; - - if ((flags & ~O_CLOEXEC) != 0) - return -EINVAL; - - if (unlikely(oldfd == newfd)) - return -EINVAL; - - spin_lock(&files->file_lock); - err = expand_files(files, newfd); - file = fcheck(oldfd); - if (unlikely(!file)) - goto Ebadf; - if (unlikely(err < 0)) { - if (err == -EMFILE) - goto Ebadf; - goto out_unlock; - } - /* - * We need to detect attempts to do dup2() over allocated but still - * not finished descriptor. NB: OpenBSD avoids that at the price of - * extra work in their equivalent of fget() - they insert struct - * file immediately after grabbing descriptor, mark it larval if - * more work (e.g. actual opening) is needed and make sure that - * fget() treats larval files as absent. Potentially interesting, - * but while extra work in fget() is trivial, locking implications - * and amount of surgery on open()-related paths in VFS are not. - * FreeBSD fails with -EBADF in the same situation, NetBSD "solution" - * deadlocks in rather amusing ways, AFAICS. All of that is out of - * scope of POSIX or SUS, since neither considers shared descriptor - * tables and this condition does not arise without those. - */ - err = -EBUSY; - fdt = files_fdtable(files); - tofree = fdt->fd[newfd]; - if (!tofree && fd_is_open(newfd, fdt)) - goto out_unlock; - get_file(file); - rcu_assign_pointer(fdt->fd[newfd], file); - __set_open_fd(newfd, fdt); - if (flags & O_CLOEXEC) - __set_close_on_exec(newfd, fdt); - else - __clear_close_on_exec(newfd, fdt); - spin_unlock(&files->file_lock); - - if (tofree) - filp_close(tofree, files); - - return newfd; - -Ebadf: - err = -EBADF; -out_unlock: - spin_unlock(&files->file_lock); - return err; -} - -SYSCALL_DEFINE2(dup2, unsigned int, oldfd, unsigned int, newfd) -{ - if (unlikely(newfd == oldfd)) { /* corner case */ - struct files_struct *files = current->files; - int retval = oldfd; - - rcu_read_lock(); - if (!fcheck_files(files, oldfd)) - retval = -EBADF; - rcu_read_unlock(); - return retval; - } - return sys_dup3(oldfd, newfd, 0); -} - -SYSCALL_DEFINE1(dup, unsigned int, fildes) -{ - int ret = -EBADF; - struct file *file = fget_raw(fildes); - - if (file) { - ret = get_unused_fd(); - if (ret >= 0) - fd_install(ret, file); - else - fput(file); - } - return ret; -} - #define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME) static int setfl(int fd, struct file * filp, unsigned long arg) @@ -267,7 +149,7 @@ pid_t f_getown(struct file *filp) static int f_setown_ex(struct file *filp, unsigned long arg) { - struct f_owner_ex * __user owner_p = (void * __user)arg; + struct f_owner_ex __user *owner_p = (void __user *)arg; struct f_owner_ex owner; struct pid *pid; int type; @@ -307,7 +189,7 @@ static int f_setown_ex(struct file *filp, unsigned long arg) static int f_getown_ex(struct file *filp, unsigned long arg) { - struct f_owner_ex * __user owner_p = (void * __user)arg; + struct f_owner_ex __user *owner_p = (void __user *)arg; struct f_owner_ex owner; int ret = 0; @@ -345,7 +227,7 @@ static int f_getown_ex(struct file *filp, unsigned long arg) static int f_getowner_uids(struct file *filp, unsigned long arg) { struct user_namespace *user_ns = current_user_ns(); - uid_t * __user dst = (void * __user)arg; + uid_t __user *dst = (void __user *)arg; uid_t src[2]; int err; @@ -373,14 +255,10 @@ static long do_fcntl(int fd, unsigned int cmd, unsigned long arg, switch (cmd) { case F_DUPFD: + err = f_dupfd(arg, filp, 0); + break; case F_DUPFD_CLOEXEC: - if (arg >= rlimit(RLIMIT_NOFILE)) - break; - err = alloc_fd(arg, cmd == F_DUPFD_CLOEXEC ? O_CLOEXEC : 0); - if (err >= 0) { - get_file(filp); - fd_install(err, filp); - } + err = f_dupfd(arg, filp, FD_CLOEXEC); break; case F_GETFD: err = get_close_on_exec(fd) ? FD_CLOEXEC : 0; @@ -470,25 +348,23 @@ static int check_fcntl_cmd(unsigned cmd) SYSCALL_DEFINE3(fcntl, unsigned int, fd, unsigned int, cmd, unsigned long, arg) { - struct file *filp; - int fput_needed; + struct fd f = fdget_raw(fd); long err = -EBADF; - filp = fget_raw_light(fd, &fput_needed); - if (!filp) + if (!f.file) goto out; - if (unlikely(filp->f_mode & FMODE_PATH)) { + if (unlikely(f.file->f_mode & FMODE_PATH)) { if (!check_fcntl_cmd(cmd)) goto out1; } - err = security_file_fcntl(filp, cmd, arg); + err = security_file_fcntl(f.file, cmd, arg); if (!err) - err = do_fcntl(fd, cmd, arg, filp); + err = do_fcntl(fd, cmd, arg, f.file); out1: - fput_light(filp, fput_needed); + fdput(f); out: return err; } @@ -497,38 +373,36 @@ out: SYSCALL_DEFINE3(fcntl64, unsigned int, fd, unsigned int, cmd, unsigned long, arg) { - struct file * filp; + struct fd f = fdget_raw(fd); long err = -EBADF; - int fput_needed; - filp = fget_raw_light(fd, &fput_needed); - if (!filp) + if (!f.file) goto out; - if (unlikely(filp->f_mode & FMODE_PATH)) { + if (unlikely(f.file->f_mode & FMODE_PATH)) { if (!check_fcntl_cmd(cmd)) goto out1; } - err = security_file_fcntl(filp, cmd, arg); + err = security_file_fcntl(f.file, cmd, arg); if (err) goto out1; switch (cmd) { case F_GETLK64: - err = fcntl_getlk64(filp, (struct flock64 __user *) arg); + err = fcntl_getlk64(f.file, (struct flock64 __user *) arg); break; case F_SETLK64: case F_SETLKW64: - err = fcntl_setlk64(fd, filp, cmd, + err = fcntl_setlk64(fd, f.file, cmd, (struct flock64 __user *) arg); break; default: - err = do_fcntl(fd, cmd, arg, filp); + err = do_fcntl(fd, cmd, arg, f.file); break; } out1: - fput_light(filp, fput_needed); + fdput(f); out: return err; } diff --git a/fs/fhandle.c b/fs/fhandle.c index a48e4a1..f775bfd 100644 --- a/fs/fhandle.c +++ b/fs/fhandle.c @@ -113,24 +113,21 @@ SYSCALL_DEFINE5(name_to_handle_at, int, dfd, const char __user *, name, static struct vfsmount *get_vfsmount_from_fd(int fd) { - struct path path; + struct vfsmount *mnt; if (fd == AT_FDCWD) { struct fs_struct *fs = current->fs; spin_lock(&fs->lock); - path = fs->pwd; - mntget(path.mnt); + mnt = mntget(fs->pwd.mnt); spin_unlock(&fs->lock); } else { - int fput_needed; - struct file *file = fget_light(fd, &fput_needed); - if (!file) + struct fd f = fdget(fd); + if (!f.file) return ERR_PTR(-EBADF); - path = file->f_path; - mntget(path.mnt); - fput_light(file, fput_needed); + mnt = mntget(f.file->f_path.mnt); + fdput(f); } - return path.mnt; + return mnt; } static int vfs_dentry_acceptable(void *context, struct dentry *dentry) @@ -6,6 +6,7 @@ * Manage the dynamic fd arrays in the process files_struct. */ +#include <linux/syscalls.h> #include <linux/export.h> #include <linux/fs.h> #include <linux/mm.h> @@ -84,22 +85,14 @@ static void free_fdtable_work(struct work_struct *work) } } -void free_fdtable_rcu(struct rcu_head *rcu) +static void free_fdtable_rcu(struct rcu_head *rcu) { struct fdtable *fdt = container_of(rcu, struct fdtable, rcu); struct fdtable_defer *fddef; BUG_ON(!fdt); + BUG_ON(fdt->max_fds <= NR_OPEN_DEFAULT); - if (fdt->max_fds <= NR_OPEN_DEFAULT) { - /* - * This fdtable is embedded in the files structure and that - * structure itself is getting destroyed. - */ - kmem_cache_free(files_cachep, - container_of(fdt, struct files_struct, fdtab)); - return; - } if (!is_vmalloc_addr(fdt->fd) && !is_vmalloc_addr(fdt->open_fds)) { kfree(fdt->fd); kfree(fdt->open_fds); @@ -229,7 +222,7 @@ static int expand_fdtable(struct files_struct *files, int nr) copy_fdtable(new_fdt, cur_fdt); rcu_assign_pointer(files->fdt, new_fdt); if (cur_fdt->max_fds > NR_OPEN_DEFAULT) - free_fdtable(cur_fdt); + call_rcu(&cur_fdt->rcu, free_fdtable_rcu); } else { /* Somebody else expanded, so undo our attempt */ __free_fdtable(new_fdt); @@ -245,19 +238,12 @@ static int expand_fdtable(struct files_struct *files, int nr) * expanded and execution may have blocked. * The files->file_lock should be held on entry, and will be held on exit. */ -int expand_files(struct files_struct *files, int nr) +static int expand_files(struct files_struct *files, int nr) { struct fdtable *fdt; fdt = files_fdtable(files); - /* - * N.B. For clone tasks sharing a files structure, this test - * will limit the total number of files that can be opened. - */ - if (nr >= rlimit(RLIMIT_NOFILE)) - return -EMFILE; - /* Do we need to expand? */ if (nr < fdt->max_fds) return 0; @@ -270,6 +256,26 @@ int expand_files(struct files_struct *files, int nr) return expand_fdtable(files, nr); } +static inline void __set_close_on_exec(int fd, struct fdtable *fdt) +{ + __set_bit(fd, fdt->close_on_exec); +} + +static inline void __clear_close_on_exec(int fd, struct fdtable *fdt) +{ + __clear_bit(fd, fdt->close_on_exec); +} + +static inline void __set_open_fd(int fd, struct fdtable *fdt) +{ + __set_bit(fd, fdt->open_fds); +} + +static inline void __clear_open_fd(int fd, struct fdtable *fdt) +{ + __clear_bit(fd, fdt->open_fds); +} + static int count_open_files(struct fdtable *fdt) { int size = fdt->max_fds; @@ -395,6 +401,95 @@ out: return NULL; } +static void close_files(struct files_struct * files) +{ + int i, j; + struct fdtable *fdt; + + j = 0; + + /* + * It is safe to dereference the fd table without RCU or + * ->file_lock because this is the last reference to the + * files structure. But use RCU to shut RCU-lockdep up. + */ + rcu_read_lock(); + fdt = files_fdtable(files); + rcu_read_unlock(); + for (;;) { + unsigned long set; + i = j * BITS_PER_LONG; + if (i >= fdt->max_fds) + break; + set = fdt->open_fds[j++]; + while (set) { + if (set & 1) { + struct file * file = xchg(&fdt->fd[i], NULL); + if (file) { + filp_close(file, files); + cond_resched(); + } + } + i++; + set >>= 1; + } + } +} + +struct files_struct *get_files_struct(struct task_struct *task) +{ + struct files_struct *files; + + task_lock(task); + files = task->files; + if (files) + atomic_inc(&files->count); + task_unlock(task); + + return files; +} + +void put_files_struct(struct files_struct *files) +{ + struct fdtable *fdt; + + if (atomic_dec_and_test(&files->count)) { + close_files(files); + /* not really needed, since nobody can see us */ + rcu_read_lock(); + fdt = files_fdtable(files); + rcu_read_unlock(); + /* free the arrays if they are not embedded */ + if (fdt != &files->fdtab) + __free_fdtable(fdt); + kmem_cache_free(files_cachep, files); + } +} + +void reset_files_struct(struct files_struct *files) +{ + struct task_struct *tsk = current; + struct files_struct *old; + + old = tsk->files; + task_lock(tsk); + tsk->files = files; + task_unlock(tsk); + put_files_struct(old); +} + +void exit_files(struct task_struct *tsk) +{ + struct files_struct * files = tsk->files; + + if (files) { + task_lock(tsk); + tsk->files = NULL; + task_unlock(tsk); + put_files_struct(files); + } +} + static void __devinit fdtable_defer_list_init(int cpu) { struct fdtable_defer *fddef = &per_cpu(fdtable_defer_list, cpu); @@ -424,12 +519,18 @@ struct files_struct init_files = { .file_lock = __SPIN_LOCK_UNLOCKED(init_task.file_lock), }; +void daemonize_descriptors(void) +{ + atomic_inc(&init_files.count); + reset_files_struct(&init_files); +} + /* * allocate a file descriptor, mark it busy. */ -int alloc_fd(unsigned start, unsigned flags) +int __alloc_fd(struct files_struct *files, + unsigned start, unsigned end, unsigned flags) { - struct files_struct *files = current->files; unsigned int fd; int error; struct fdtable *fdt; @@ -444,6 +545,14 @@ repeat: if (fd < fdt->max_fds) fd = find_next_zero_bit(fdt->open_fds, fdt->max_fds, fd); + /* + * N.B. For clone tasks sharing a files structure, this test + * will limit the total number of files that can be opened. + */ + error = -EMFILE; + if (fd >= end) + goto out; + error = expand_files(files, fd); if (error < 0) goto out; @@ -477,8 +586,424 @@ out: return error; } -int get_unused_fd(void) +static int alloc_fd(unsigned start, unsigned flags) +{ + return __alloc_fd(current->files, start, rlimit(RLIMIT_NOFILE), flags); +} + +int get_unused_fd_flags(unsigned flags) +{ + return __alloc_fd(current->files, 0, rlimit(RLIMIT_NOFILE), flags); +} +EXPORT_SYMBOL(get_unused_fd_flags); + +static void __put_unused_fd(struct files_struct *files, unsigned int fd) +{ + struct fdtable *fdt = files_fdtable(files); + __clear_open_fd(fd, fdt); + if (fd < files->next_fd) + files->next_fd = fd; +} + +void put_unused_fd(unsigned int fd) +{ + struct files_struct *files = current->files; + spin_lock(&files->file_lock); + __put_unused_fd(files, fd); + spin_unlock(&files->file_lock); +} + +EXPORT_SYMBOL(put_unused_fd); + +/* + * Install a file pointer in the fd array. + * + * The VFS is full of places where we drop the files lock between + * setting the open_fds bitmap and installing the file in the file + * array. At any such point, we are vulnerable to a dup2() race + * installing a file in the array before us. We need to detect this and + * fput() the struct file we are about to overwrite in this case. + * + * It should never happen - if we allow dup2() do it, _really_ bad things + * will follow. + * + * NOTE: __fd_install() variant is really, really low-level; don't + * use it unless you are forced to by truly lousy API shoved down + * your throat. 'files' *MUST* be either current->files or obtained + * by get_files_struct(current) done by whoever had given it to you, + * or really bad things will happen. Normally you want to use + * fd_install() instead. + */ + +void __fd_install(struct files_struct *files, unsigned int fd, + struct file *file) +{ + struct fdtable *fdt; + spin_lock(&files->file_lock); + fdt = files_fdtable(files); + BUG_ON(fdt->fd[fd] != NULL); + rcu_assign_pointer(fdt->fd[fd], file); + spin_unlock(&files->file_lock); +} + +void fd_install(unsigned int fd, struct file *file) { - return alloc_fd(0, 0); + __fd_install(current->files, fd, file); +} + +EXPORT_SYMBOL(fd_install); + +/* + * The same warnings as for __alloc_fd()/__fd_install() apply here... + */ +int __close_fd(struct files_struct *files, unsigned fd) +{ + struct file *file; + struct fdtable *fdt; + + spin_lock(&files->file_lock); + fdt = files_fdtable(files); + if (fd >= fdt->max_fds) + goto out_unlock; + file = fdt->fd[fd]; + if (!file) + goto out_unlock; + rcu_assign_pointer(fdt->fd[fd], NULL); + __clear_close_on_exec(fd, fdt); + __put_unused_fd(files, fd); + spin_unlock(&files->file_lock); + return filp_close(file, files); + +out_unlock: + spin_unlock(&files->file_lock); + return -EBADF; +} + +void do_close_on_exec(struct files_struct *files) +{ + unsigned i; + struct fdtable *fdt; + + /* exec unshares first */ + BUG_ON(atomic_read(&files->count) != 1); + spin_lock(&files->file_lock); + for (i = 0; ; i++) { + unsigned long set; + unsigned fd = i * BITS_PER_LONG; + fdt = files_fdtable(files); + if (fd >= fdt->max_fds) + break; + set = fdt->close_on_exec[i]; + if (!set) + continue; + fdt->close_on_exec[i] = 0; + for ( ; set ; fd++, set >>= 1) { + struct file *file; + if (!(set & 1)) + continue; + file = fdt->fd[fd]; + if (!file) + continue; + rcu_assign_pointer(fdt->fd[fd], NULL); + __put_unused_fd(files, fd); + spin_unlock(&files->file_lock); + filp_close(file, files); + cond_resched(); + spin_lock(&files->file_lock); + } + + } + spin_unlock(&files->file_lock); +} + +struct file *fget(unsigned int fd) +{ + struct file *file; + struct files_struct *files = current->files; + + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + /* File object ref couldn't be taken */ + if (file->f_mode & FMODE_PATH || + !atomic_long_inc_not_zero(&file->f_count)) + file = NULL; + } + rcu_read_unlock(); + + return file; +} + +EXPORT_SYMBOL(fget); + +struct file *fget_raw(unsigned int fd) +{ + struct file *file; + struct files_struct *files = current->files; + + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + /* File object ref couldn't be taken */ + if (!atomic_long_inc_not_zero(&file->f_count)) + file = NULL; + } + rcu_read_unlock(); + + return file; +} + +EXPORT_SYMBOL(fget_raw); + +/* + * Lightweight file lookup - no refcnt increment if fd table isn't shared. + * + * You can use this instead of fget if you satisfy all of the following + * conditions: + * 1) You must call fput_light before exiting the syscall and returning control + * to userspace (i.e. you cannot remember the returned struct file * after + * returning to userspace). + * 2) You must not call filp_close on the returned struct file * in between + * calls to fget_light and fput_light. + * 3) You must not clone the current task in between the calls to fget_light + * and fput_light. + * + * The fput_needed flag returned by fget_light should be passed to the + * corresponding fput_light. + */ +struct file *fget_light(unsigned int fd, int *fput_needed) +{ + struct file *file; + struct files_struct *files = current->files; + + *fput_needed = 0; + if (atomic_read(&files->count) == 1) { + file = fcheck_files(files, fd); + if (file && (file->f_mode & FMODE_PATH)) + file = NULL; + } else { + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + if (!(file->f_mode & FMODE_PATH) && + atomic_long_inc_not_zero(&file->f_count)) + *fput_needed = 1; + else + /* Didn't get the reference, someone's freed */ + file = NULL; + } + rcu_read_unlock(); + } + + return file; +} +EXPORT_SYMBOL(fget_light); + +struct file *fget_raw_light(unsigned int fd, int *fput_needed) +{ + struct file *file; + struct files_struct *files = current->files; + + *fput_needed = 0; + if (atomic_read(&files->count) == 1) { + file = fcheck_files(files, fd); + } else { + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + if (atomic_long_inc_not_zero(&file->f_count)) + *fput_needed = 1; + else + /* Didn't get the reference, someone's freed */ + file = NULL; + } + rcu_read_unlock(); + } + + return file; +} + +void set_close_on_exec(unsigned int fd, int flag) +{ + struct files_struct *files = current->files; + struct fdtable *fdt; + spin_lock(&files->file_lock); + fdt = files_fdtable(files); + if (flag) + __set_close_on_exec(fd, fdt); + else + __clear_close_on_exec(fd, fdt); + spin_unlock(&files->file_lock); +} + +bool get_close_on_exec(unsigned int fd) +{ + struct files_struct *files = current->files; + struct fdtable *fdt; + bool res; + rcu_read_lock(); + fdt = files_fdtable(files); + res = close_on_exec(fd, fdt); + rcu_read_unlock(); + return res; +} + +static int do_dup2(struct files_struct *files, + struct file *file, unsigned fd, unsigned flags) +{ + struct file *tofree; + struct fdtable *fdt; + + /* + * We need to detect attempts to do dup2() over allocated but still + * not finished descriptor. NB: OpenBSD avoids that at the price of + * extra work in their equivalent of fget() - they insert struct + * file immediately after grabbing descriptor, mark it larval if + * more work (e.g. actual opening) is needed and make sure that + * fget() treats larval files as absent. Potentially interesting, + * but while extra work in fget() is trivial, locking implications + * and amount of surgery on open()-related paths in VFS are not. + * FreeBSD fails with -EBADF in the same situation, NetBSD "solution" + * deadlocks in rather amusing ways, AFAICS. All of that is out of + * scope of POSIX or SUS, since neither considers shared descriptor + * tables and this condition does not arise without those. + */ + fdt = files_fdtable(files); + tofree = fdt->fd[fd]; + if (!tofree && fd_is_open(fd, fdt)) + goto Ebusy; + get_file(file); + rcu_assign_pointer(fdt->fd[fd], file); + __set_open_fd(fd, fdt); + if (flags & O_CLOEXEC) + __set_close_on_exec(fd, fdt); + else + __clear_close_on_exec(fd, fdt); + spin_unlock(&files->file_lock); + + if (tofree) + filp_close(tofree, files); + + return fd; + +Ebusy: + spin_unlock(&files->file_lock); + return -EBUSY; +} + +int replace_fd(unsigned fd, struct file *file, unsigned flags) +{ + int err; + struct files_struct *files = current->files; + + if (!file) + return __close_fd(files, fd); + + if (fd >= rlimit(RLIMIT_NOFILE)) + return -EMFILE; + + spin_lock(&files->file_lock); + err = expand_files(files, fd); + if (unlikely(err < 0)) + goto out_unlock; + return do_dup2(files, file, fd, flags); + +out_unlock: + spin_unlock(&files->file_lock); + return err; +} + +SYSCALL_DEFINE3(dup3, unsigned int, oldfd, unsigned int, newfd, int, flags) +{ + int err = -EBADF; + struct file *file; + struct files_struct *files = current->files; + + if ((flags & ~O_CLOEXEC) != 0) + return -EINVAL; + + if (newfd >= rlimit(RLIMIT_NOFILE)) + return -EMFILE; + + spin_lock(&files->file_lock); + err = expand_files(files, newfd); + file = fcheck(oldfd); + if (unlikely(!file)) + goto Ebadf; + if (unlikely(err < 0)) { + if (err == -EMFILE) + goto Ebadf; + goto out_unlock; + } + return do_dup2(files, file, newfd, flags); + +Ebadf: + err = -EBADF; +out_unlock: + spin_unlock(&files->file_lock); + return err; +} + +SYSCALL_DEFINE2(dup2, unsigned int, oldfd, unsigned int, newfd) +{ + if (unlikely(newfd == oldfd)) { /* corner case */ + struct files_struct *files = current->files; + int retval = oldfd; + + rcu_read_lock(); + if (!fcheck_files(files, oldfd)) + retval = -EBADF; + rcu_read_unlock(); + return retval; + } + return sys_dup3(oldfd, newfd, 0); +} + +SYSCALL_DEFINE1(dup, unsigned int, fildes) +{ + int ret = -EBADF; + struct file *file = fget_raw(fildes); + + if (file) { + ret = get_unused_fd(); + if (ret >= 0) + fd_install(ret, file); + else + fput(file); + } + return ret; +} + +int f_dupfd(unsigned int from, struct file *file, unsigned flags) +{ + int err; + if (from >= rlimit(RLIMIT_NOFILE)) + return -EINVAL; + err = alloc_fd(from, flags); + if (err >= 0) { + get_file(file); + fd_install(err, file); + } + return err; +} + +int iterate_fd(struct files_struct *files, unsigned n, + int (*f)(const void *, struct file *, unsigned), + const void *p) +{ + struct fdtable *fdt; + struct file *file; + int res = 0; + if (!files) + return 0; + spin_lock(&files->file_lock); + fdt = files_fdtable(files); + while (!res && n < fdt->max_fds) { + file = rcu_dereference_check_fdtable(files, fdt->fd[n++]); + if (file) + res = f(p, file, n); + } + spin_unlock(&files->file_lock); + return res; } -EXPORT_SYMBOL(get_unused_fd); +EXPORT_SYMBOL(iterate_fd); diff --git a/fs/file_table.c b/fs/file_table.c index 701985e..dac6792 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -243,10 +243,10 @@ static void __fput(struct file *file) if (file->f_op && file->f_op->fasync) file->f_op->fasync(-1, file, 0); } + ima_file_free(file); if (file->f_op && file->f_op->release) file->f_op->release(inode, file); security_file_free(file); - ima_file_free(file); if (unlikely(S_ISCHR(inode->i_mode) && inode->i_cdev != NULL && !(file->f_mode & FMODE_PATH))) { cdev_put(inode->i_cdev); @@ -339,112 +339,6 @@ void __fput_sync(struct file *file) EXPORT_SYMBOL(fput); -struct file *fget(unsigned int fd) -{ - struct file *file; - struct files_struct *files = current->files; - - rcu_read_lock(); - file = fcheck_files(files, fd); - if (file) { - /* File object ref couldn't be taken */ - if (file->f_mode & FMODE_PATH || - !atomic_long_inc_not_zero(&file->f_count)) - file = NULL; - } - rcu_read_unlock(); - - return file; -} - -EXPORT_SYMBOL(fget); - -struct file *fget_raw(unsigned int fd) -{ - struct file *file; - struct files_struct *files = current->files; - - rcu_read_lock(); - file = fcheck_files(files, fd); - if (file) { - /* File object ref couldn't be taken */ - if (!atomic_long_inc_not_zero(&file->f_count)) - file = NULL; - } - rcu_read_unlock(); - - return file; -} - -EXPORT_SYMBOL(fget_raw); - -/* - * Lightweight file lookup - no refcnt increment if fd table isn't shared. - * - * You can use this instead of fget if you satisfy all of the following - * conditions: - * 1) You must call fput_light before exiting the syscall and returning control - * to userspace (i.e. you cannot remember the returned struct file * after - * returning to userspace). - * 2) You must not call filp_close on the returned struct file * in between - * calls to fget_light and fput_light. - * 3) You must not clone the current task in between the calls to fget_light - * and fput_light. - * - * The fput_needed flag returned by fget_light should be passed to the - * corresponding fput_light. - */ -struct file *fget_light(unsigned int fd, int *fput_needed) -{ - struct file *file; - struct files_struct *files = current->files; - - *fput_needed = 0; - if (atomic_read(&files->count) == 1) { - file = fcheck_files(files, fd); - if (file && (file->f_mode & FMODE_PATH)) - file = NULL; - } else { - rcu_read_lock(); - file = fcheck_files(files, fd); - if (file) { - if (!(file->f_mode & FMODE_PATH) && - atomic_long_inc_not_zero(&file->f_count)) - *fput_needed = 1; - else - /* Didn't get the reference, someone's freed */ - file = NULL; - } - rcu_read_unlock(); - } - - return file; -} - -struct file *fget_raw_light(unsigned int fd, int *fput_needed) -{ - struct file *file; - struct files_struct *files = current->files; - - *fput_needed = 0; - if (atomic_read(&files->count) == 1) { - file = fcheck_files(files, fd); - } else { - rcu_read_lock(); - file = fcheck_files(files, fd); - if (file) { - if (atomic_long_inc_not_zero(&file->f_count)) - *fput_needed = 1; - else - /* Didn't get the reference, someone's freed */ - file = NULL; - } - rcu_read_unlock(); - } - - return file; -} - void put_filp(struct file *file) { if (atomic_long_dec_and_test(&file->f_count)) { diff --git a/fs/freevxfs/vxfs_super.c b/fs/freevxfs/vxfs_super.c index d4fabd2..fed2c8a 100644 --- a/fs/freevxfs/vxfs_super.c +++ b/fs/freevxfs/vxfs_super.c @@ -279,6 +279,11 @@ static void __exit vxfs_cleanup(void) { unregister_filesystem(&vxfs_fs_type); + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(vxfs_inode_cachep); } diff --git a/fs/fuse/dev.c b/fs/fuse/dev.c index f4246cf..8c23fa7 100644 --- a/fs/fuse/dev.c +++ b/fs/fuse/dev.c @@ -148,8 +148,7 @@ static struct fuse_req *get_reserved_req(struct fuse_conn *fc, if (ff->reserved_req) { req = ff->reserved_req; ff->reserved_req = NULL; - get_file(file); - req->stolen_file = file; + req->stolen_file = get_file(file); } spin_unlock(&fc->lock); } while (!req); diff --git a/fs/fuse/inode.c b/fs/fuse/inode.c index fca222d..f0eda12 100644 --- a/fs/fuse/inode.c +++ b/fs/fuse/inode.c @@ -1197,6 +1197,12 @@ static void fuse_fs_cleanup(void) { unregister_filesystem(&fuse_fs_type); unregister_fuseblk(); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(fuse_inode_cachep); } diff --git a/fs/hfs/super.c b/fs/hfs/super.c index 0b63d13..e93ddaa 100644 --- a/fs/hfs/super.c +++ b/fs/hfs/super.c @@ -492,6 +492,12 @@ static int __init init_hfs_fs(void) static void __exit exit_hfs_fs(void) { unregister_filesystem(&hfs_fs_type); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(hfs_inode_cachep); } diff --git a/fs/hfsplus/super.c b/fs/hfsplus/super.c index fdafb2d..811a84d 100644 --- a/fs/hfsplus/super.c +++ b/fs/hfsplus/super.c @@ -635,6 +635,12 @@ static int __init init_hfsplus_fs(void) static void __exit exit_hfsplus_fs(void) { unregister_filesystem(&hfsplus_fs_type); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(hfsplus_inode_cachep); } diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c index a152783..bc28bf07 100644 --- a/fs/hpfs/super.c +++ b/fs/hpfs/super.c @@ -210,6 +210,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(hpfs_inode_cachep); } diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 6e572c4..9460120 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -1048,6 +1048,11 @@ static int __init init_hugetlbfs_fs(void) static void __exit exit_hugetlbfs_fs(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(hugetlbfs_inode_cachep); kern_unmount(hugetlbfs_vfsmount); unregister_filesystem(&hugetlbfs_fs_type); @@ -603,21 +603,14 @@ int do_vfs_ioctl(struct file *filp, unsigned int fd, unsigned int cmd, SYSCALL_DEFINE3(ioctl, unsigned int, fd, unsigned int, cmd, unsigned long, arg) { - struct file *filp; - int error = -EBADF; - int fput_needed; - - filp = fget_light(fd, &fput_needed); - if (!filp) - goto out; - - error = security_file_ioctl(filp, cmd, arg); - if (error) - goto out_fput; - - error = do_vfs_ioctl(filp, fd, cmd, arg); - out_fput: - fput_light(filp, fput_needed); - out: + int error; + struct fd f = fdget(fd); + + if (!f.file) + return -EBADF; + error = security_file_ioctl(f.file, cmd, arg); + if (!error) + error = do_vfs_ioctl(f.file, fd, cmd, arg); + fdput(f); return error; } diff --git a/fs/isofs/inode.c b/fs/isofs/inode.c index a7d8e6c..67ce52507 100644 --- a/fs/isofs/inode.c +++ b/fs/isofs/inode.c @@ -115,6 +115,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(isofs_inode_cachep); } diff --git a/fs/jffs2/super.c b/fs/jffs2/super.c index 61ea413..ff48795 100644 --- a/fs/jffs2/super.c +++ b/fs/jffs2/super.c @@ -418,6 +418,12 @@ static void __exit exit_jffs2_fs(void) unregister_filesystem(&jffs2_fs_type); jffs2_destroy_slab_caches(); jffs2_compressors_exit(); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(jffs2_inode_cachep); } diff --git a/fs/jfs/super.c b/fs/jfs/super.c index 706692f..efdf883 100644 --- a/fs/jfs/super.c +++ b/fs/jfs/super.c @@ -911,6 +911,12 @@ static void __exit exit_jfs_fs(void) jfs_proc_clean(); #endif unregister_filesystem(&jfs_fs_type); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(jfs_inode_cachep); } @@ -1625,15 +1625,13 @@ EXPORT_SYMBOL(flock_lock_file_wait); */ SYSCALL_DEFINE2(flock, unsigned int, fd, unsigned int, cmd) { - struct file *filp; - int fput_needed; + struct fd f = fdget(fd); struct file_lock *lock; int can_sleep, unlock; int error; error = -EBADF; - filp = fget_light(fd, &fput_needed); - if (!filp) + if (!f.file) goto out; can_sleep = !(cmd & LOCK_NB); @@ -1641,31 +1639,31 @@ SYSCALL_DEFINE2(flock, unsigned int, fd, unsigned int, cmd) unlock = (cmd == LOCK_UN); if (!unlock && !(cmd & LOCK_MAND) && - !(filp->f_mode & (FMODE_READ|FMODE_WRITE))) + !(f.file->f_mode & (FMODE_READ|FMODE_WRITE))) goto out_putf; - error = flock_make_lock(filp, &lock, cmd); + error = flock_make_lock(f.file, &lock, cmd); if (error) goto out_putf; if (can_sleep) lock->fl_flags |= FL_SLEEP; - error = security_file_lock(filp, lock->fl_type); + error = security_file_lock(f.file, lock->fl_type); if (error) goto out_free; - if (filp->f_op && filp->f_op->flock) - error = filp->f_op->flock(filp, + if (f.file->f_op && f.file->f_op->flock) + error = f.file->f_op->flock(f.file, (can_sleep) ? F_SETLKW : F_SETLK, lock); else - error = flock_lock_file_wait(filp, lock); + error = flock_lock_file_wait(f.file, lock); out_free: locks_free_lock(lock); out_putf: - fput_light(filp, fput_needed); + fdput(f); out: return error; } diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c index bda3908..adb9011 100644 --- a/fs/logfs/inode.c +++ b/fs/logfs/inode.c @@ -417,5 +417,10 @@ int logfs_init_inode_cache(void) void logfs_destroy_inode_cache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(logfs_inode_cache); } diff --git a/fs/minix/inode.c b/fs/minix/inode.c index d0e42c6..4fc5f8a 100644 --- a/fs/minix/inode.c +++ b/fs/minix/inode.c @@ -100,6 +100,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(minix_inode_cachep); } @@ -1797,8 +1797,6 @@ static int path_init(int dfd, const char *name, unsigned int flags, struct nameidata *nd, struct file **fp) { int retval = 0; - int fput_needed; - struct file *file; nd->last_type = LAST_ROOT; /* if there are only slashes... */ nd->flags = flags | LOOKUP_JUMPED; @@ -1850,44 +1848,41 @@ static int path_init(int dfd, const char *name, unsigned int flags, get_fs_pwd(current->fs, &nd->path); } } else { + struct fd f = fdget_raw(dfd); struct dentry *dentry; - file = fget_raw_light(dfd, &fput_needed); - retval = -EBADF; - if (!file) - goto out_fail; + if (!f.file) + return -EBADF; - dentry = file->f_path.dentry; + dentry = f.file->f_path.dentry; if (*name) { - retval = -ENOTDIR; - if (!S_ISDIR(dentry->d_inode->i_mode)) - goto fput_fail; + if (!S_ISDIR(dentry->d_inode->i_mode)) { + fdput(f); + return -ENOTDIR; + } retval = inode_permission(dentry->d_inode, MAY_EXEC); - if (retval) - goto fput_fail; + if (retval) { + fdput(f); + return retval; + } } - nd->path = file->f_path; + nd->path = f.file->f_path; if (flags & LOOKUP_RCU) { - if (fput_needed) - *fp = file; + if (f.need_put) + *fp = f.file; nd->seq = __read_seqcount_begin(&nd->path.dentry->d_seq); lock_rcu_walk(); } else { - path_get(&file->f_path); - fput_light(file, fput_needed); + path_get(&nd->path); + fdput(f); } } nd->inode = nd->path.dentry->d_inode; return 0; - -fput_fail: - fput_light(file, fput_needed); -out_fail: - return retval; } static inline int lookup_last(struct nameidata *nd, struct path *path) @@ -3971,7 +3966,7 @@ EXPORT_SYMBOL(user_path_at); EXPORT_SYMBOL(follow_down_one); EXPORT_SYMBOL(follow_down); EXPORT_SYMBOL(follow_up); -EXPORT_SYMBOL(get_write_access); /* binfmt_aout */ +EXPORT_SYMBOL(get_write_access); /* nfsd */ EXPORT_SYMBOL(getname); EXPORT_SYMBOL(lock_rename); EXPORT_SYMBOL(lookup_one_len); diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c index eaa7432..d7e9fe7 100644 --- a/fs/ncpfs/inode.c +++ b/fs/ncpfs/inode.c @@ -89,6 +89,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ncp_inode_cachep); } diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c index 9b47610..e4c716d 100644 --- a/fs/nfs/inode.c +++ b/fs/nfs/inode.c @@ -1571,6 +1571,11 @@ static int __init nfs_init_inodecache(void) static void nfs_destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(nfs_inode_cachep); } diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index cc894ed..48a1bad 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -2837,8 +2837,7 @@ static int nfs4_setlease(struct nfs4_delegation *dp, int flag) return -ENOMEM; } fp->fi_lease = fl; - fp->fi_deleg_file = fl->fl_file; - get_file(fp->fi_deleg_file); + fp->fi_deleg_file = get_file(fl->fl_file); atomic_set(&fp->fi_delegees, 1); list_add(&dp->dl_perfile, &fp->fi_delegations); return 0; diff --git a/fs/nilfs2/super.c b/fs/nilfs2/super.c index 6a10812..3c991dc 100644 --- a/fs/nilfs2/super.c +++ b/fs/nilfs2/super.c @@ -1382,6 +1382,12 @@ static void nilfs_segbuf_init_once(void *obj) static void nilfs_destroy_cachep(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); + if (nilfs_inode_cachep) kmem_cache_destroy(nilfs_inode_cachep); if (nilfs_transaction_cachep) diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index d438036..721d692 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -58,7 +58,9 @@ static struct fsnotify_event *get_one_event(struct fsnotify_group *group, return fsnotify_remove_notify_event(group); } -static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event) +static int create_fd(struct fsnotify_group *group, + struct fsnotify_event *event, + struct file **file) { int client_fd; struct file *new_file; @@ -98,7 +100,7 @@ static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event) put_unused_fd(client_fd); client_fd = PTR_ERR(new_file); } else { - fd_install(client_fd, new_file); + *file = new_file; } return client_fd; @@ -106,13 +108,15 @@ static int create_fd(struct fsnotify_group *group, struct fsnotify_event *event) static int fill_event_metadata(struct fsnotify_group *group, struct fanotify_event_metadata *metadata, - struct fsnotify_event *event) + struct fsnotify_event *event, + struct file **file) { int ret = 0; pr_debug("%s: group=%p metadata=%p event=%p\n", __func__, group, metadata, event); + *file = NULL; metadata->event_len = FAN_EVENT_METADATA_LEN; metadata->metadata_len = FAN_EVENT_METADATA_LEN; metadata->vers = FANOTIFY_METADATA_VERSION; @@ -121,7 +125,7 @@ static int fill_event_metadata(struct fsnotify_group *group, if (unlikely(event->mask & FAN_Q_OVERFLOW)) metadata->fd = FAN_NOFD; else { - metadata->fd = create_fd(group, event); + metadata->fd = create_fd(group, event, file); if (metadata->fd < 0) ret = metadata->fd; } @@ -220,25 +224,6 @@ static int prepare_for_access_response(struct fsnotify_group *group, return 0; } -static void remove_access_response(struct fsnotify_group *group, - struct fsnotify_event *event, - __s32 fd) -{ - struct fanotify_response_event *re; - - if (!(event->mask & FAN_ALL_PERM_EVENTS)) - return; - - re = dequeue_re(group, fd); - if (!re) - return; - - BUG_ON(re->event != event); - - kmem_cache_free(fanotify_response_event_cache, re); - - return; -} #else static int prepare_for_access_response(struct fsnotify_group *group, struct fsnotify_event *event, @@ -247,12 +232,6 @@ static int prepare_for_access_response(struct fsnotify_group *group, return 0; } -static void remove_access_response(struct fsnotify_group *group, - struct fsnotify_event *event, - __s32 fd) -{ - return; -} #endif static ssize_t copy_event_to_user(struct fsnotify_group *group, @@ -260,31 +239,33 @@ static ssize_t copy_event_to_user(struct fsnotify_group *group, char __user *buf) { struct fanotify_event_metadata fanotify_event_metadata; + struct file *f; int fd, ret; pr_debug("%s: group=%p event=%p\n", __func__, group, event); - ret = fill_event_metadata(group, &fanotify_event_metadata, event); + ret = fill_event_metadata(group, &fanotify_event_metadata, event, &f); if (ret < 0) goto out; fd = fanotify_event_metadata.fd; - ret = prepare_for_access_response(group, event, fd); - if (ret) - goto out_close_fd; - ret = -EFAULT; if (copy_to_user(buf, &fanotify_event_metadata, fanotify_event_metadata.event_len)) - goto out_kill_access_response; + goto out_close_fd; + ret = prepare_for_access_response(group, event, fd); + if (ret) + goto out_close_fd; + + fd_install(fd, f); return fanotify_event_metadata.event_len; -out_kill_access_response: - remove_access_response(group, event, fd); out_close_fd: - if (fd != FAN_NOFD) - sys_close(fd); + if (fd != FAN_NOFD) { + put_unused_fd(fd); + fput(f); + } out: #ifdef CONFIG_FANOTIFY_ACCESS_PERMISSIONS if (event->mask & FAN_ALL_PERM_EVENTS) { @@ -470,24 +451,22 @@ static int fanotify_find_path(int dfd, const char __user *filename, dfd, filename, flags); if (filename == NULL) { - struct file *file; - int fput_needed; + struct fd f = fdget(dfd); ret = -EBADF; - file = fget_light(dfd, &fput_needed); - if (!file) + if (!f.file) goto out; ret = -ENOTDIR; if ((flags & FAN_MARK_ONLYDIR) && - !(S_ISDIR(file->f_path.dentry->d_inode->i_mode))) { - fput_light(file, fput_needed); + !(S_ISDIR(f.file->f_path.dentry->d_inode->i_mode))) { + fdput(f); goto out; } - *path = file->f_path; + *path = f.file->f_path; path_get(path); - fput_light(file, fput_needed); + fdput(f); } else { unsigned int lookup_flags = 0; @@ -767,9 +746,9 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, struct inode *inode = NULL; struct vfsmount *mnt = NULL; struct fsnotify_group *group; - struct file *filp; + struct fd f; struct path path; - int ret, fput_needed; + int ret; pr_debug("%s: fanotify_fd=%d flags=%x dfd=%d pathname=%p mask=%llx\n", __func__, fanotify_fd, flags, dfd, pathname, mask); @@ -803,15 +782,15 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, #endif return -EINVAL; - filp = fget_light(fanotify_fd, &fput_needed); - if (unlikely(!filp)) + f = fdget(fanotify_fd); + if (unlikely(!f.file)) return -EBADF; /* verify that this is indeed an fanotify instance */ ret = -EINVAL; - if (unlikely(filp->f_op != &fanotify_fops)) + if (unlikely(f.file->f_op != &fanotify_fops)) goto fput_and_out; - group = filp->private_data; + group = f.file->private_data; /* * group->priority == FS_PRIO_0 == FAN_CLASS_NOTIF. These are not @@ -858,7 +837,7 @@ SYSCALL_DEFINE(fanotify_mark)(int fanotify_fd, unsigned int flags, path_put(&path); fput_and_out: - fput_light(filp, fput_needed); + fdput(f); return ret; } diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 8445fbc..c311dda 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -757,16 +757,16 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname, struct fsnotify_group *group; struct inode *inode; struct path path; - struct file *filp; - int ret, fput_needed; + struct fd f; + int ret; unsigned flags = 0; - filp = fget_light(fd, &fput_needed); - if (unlikely(!filp)) + f = fdget(fd); + if (unlikely(!f.file)) return -EBADF; /* verify that this is indeed an inotify instance */ - if (unlikely(filp->f_op != &inotify_fops)) { + if (unlikely(f.file->f_op != &inotify_fops)) { ret = -EINVAL; goto fput_and_out; } @@ -782,13 +782,13 @@ SYSCALL_DEFINE3(inotify_add_watch, int, fd, const char __user *, pathname, /* inode held in place by reference to path; group by fget on fd */ inode = path.dentry->d_inode; - group = filp->private_data; + group = f.file->private_data; /* create/update an inode mark */ ret = inotify_update_watch(group, inode, mask); path_put(&path); fput_and_out: - fput_light(filp, fput_needed); + fdput(f); return ret; } @@ -796,19 +796,19 @@ SYSCALL_DEFINE2(inotify_rm_watch, int, fd, __s32, wd) { struct fsnotify_group *group; struct inotify_inode_mark *i_mark; - struct file *filp; - int ret = 0, fput_needed; + struct fd f; + int ret = 0; - filp = fget_light(fd, &fput_needed); - if (unlikely(!filp)) + f = fdget(fd); + if (unlikely(!f.file)) return -EBADF; /* verify that this is indeed an inotify instance */ ret = -EINVAL; - if (unlikely(filp->f_op != &inotify_fops)) + if (unlikely(f.file->f_op != &inotify_fops)) goto out; - group = filp->private_data; + group = f.file->private_data; ret = -EINVAL; i_mark = inotify_idr_find(group, wd); @@ -823,7 +823,7 @@ SYSCALL_DEFINE2(inotify_rm_watch, int, fd, __s32, wd) fsnotify_put_mark(&i_mark->fsn_mark); out: - fput_light(filp, fput_needed); + fdput(f); return ret; } diff --git a/fs/ntfs/super.c b/fs/ntfs/super.c index da01c16..4a8289f8 100644 --- a/fs/ntfs/super.c +++ b/fs/ntfs/super.c @@ -3193,6 +3193,12 @@ static void __exit exit_ntfs_fs(void) ntfs_debug("Unregistering NTFS driver."); unregister_filesystem(&ntfs_fs_type); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ntfs_big_inode_cache); kmem_cache_destroy(ntfs_inode_cache); kmem_cache_destroy(ntfs_name_cache); diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index a4e855e..f7c648d 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -1746,8 +1746,8 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, long fd; int sectsize; char *p = (char *)page; - struct file *filp = NULL; - struct inode *inode = NULL; + struct fd f; + struct inode *inode; ssize_t ret = -EINVAL; int live_threshold; @@ -1766,26 +1766,26 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, if (fd < 0 || fd >= INT_MAX) goto out; - filp = fget(fd); - if (filp == NULL) + f = fdget(fd); + if (f.file == NULL) goto out; if (reg->hr_blocks == 0 || reg->hr_start_block == 0 || reg->hr_block_bytes == 0) - goto out; + goto out2; - inode = igrab(filp->f_mapping->host); + inode = igrab(f.file->f_mapping->host); if (inode == NULL) - goto out; + goto out2; if (!S_ISBLK(inode->i_mode)) - goto out; + goto out3; - reg->hr_bdev = I_BDEV(filp->f_mapping->host); + reg->hr_bdev = I_BDEV(f.file->f_mapping->host); ret = blkdev_get(reg->hr_bdev, FMODE_WRITE | FMODE_READ, NULL); if (ret) { reg->hr_bdev = NULL; - goto out; + goto out3; } inode = NULL; @@ -1797,7 +1797,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, "blocksize %u incorrect for device, expected %d", reg->hr_block_bytes, sectsize); ret = -EINVAL; - goto out; + goto out3; } o2hb_init_region_params(reg); @@ -1811,13 +1811,13 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, ret = o2hb_map_slot_data(reg); if (ret) { mlog_errno(ret); - goto out; + goto out3; } ret = o2hb_populate_slot_data(reg); if (ret) { mlog_errno(ret); - goto out; + goto out3; } INIT_DELAYED_WORK(®->hr_write_timeout_work, o2hb_write_timeout); @@ -1847,7 +1847,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, if (IS_ERR(hb_task)) { ret = PTR_ERR(hb_task); mlog_errno(ret); - goto out; + goto out3; } spin_lock(&o2hb_live_lock); @@ -1863,7 +1863,7 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, if (reg->hr_aborted_start) { ret = -EIO; - goto out; + goto out3; } /* Ok, we were woken. Make sure it wasn't by drop_item() */ @@ -1882,11 +1882,11 @@ static ssize_t o2hb_region_dev_write(struct o2hb_region *reg, printk(KERN_NOTICE "o2hb: Heartbeat started on region %s (%s)\n", config_item_name(®->hr_item), reg->hr_dev_name); +out3: + iput(inode); +out2: + fdput(f); out: - if (filp) - fput(filp); - if (inode) - iput(inode); if (ret < 0) { if (reg->hr_bdev) { blkdev_put(reg->hr_bdev, FMODE_READ|FMODE_WRITE); diff --git a/fs/ocfs2/dlmfs/dlmfs.c b/fs/ocfs2/dlmfs/dlmfs.c index 83b6f98..16b712d 100644 --- a/fs/ocfs2/dlmfs/dlmfs.c +++ b/fs/ocfs2/dlmfs/dlmfs.c @@ -691,6 +691,11 @@ static void __exit exit_dlmfs_fs(void) flush_workqueue(user_dlm_worker); destroy_workqueue(user_dlm_worker); + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(dlmfs_inode_cache); bdi_destroy(&dlmfs_backing_dev_info); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 68f4541..0e91ec2 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -1818,6 +1818,11 @@ static int ocfs2_initialize_mem_caches(void) static void ocfs2_free_mem_caches(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); if (ocfs2_inode_cachep) kmem_cache_destroy(ocfs2_inode_cachep); ocfs2_inode_cachep = NULL; @@ -132,27 +132,27 @@ SYSCALL_DEFINE2(truncate, const char __user *, path, long, length) static long do_sys_ftruncate(unsigned int fd, loff_t length, int small) { - struct inode * inode; + struct inode *inode; struct dentry *dentry; - struct file * file; + struct fd f; int error; error = -EINVAL; if (length < 0) goto out; error = -EBADF; - file = fget(fd); - if (!file) + f = fdget(fd); + if (!f.file) goto out; /* explicitly opened as large or we are on 64-bit box */ - if (file->f_flags & O_LARGEFILE) + if (f.file->f_flags & O_LARGEFILE) small = 0; - dentry = file->f_path.dentry; + dentry = f.file->f_path.dentry; inode = dentry->d_inode; error = -EINVAL; - if (!S_ISREG(inode->i_mode) || !(file->f_mode & FMODE_WRITE)) + if (!S_ISREG(inode->i_mode) || !(f.file->f_mode & FMODE_WRITE)) goto out_putf; error = -EINVAL; @@ -165,14 +165,14 @@ static long do_sys_ftruncate(unsigned int fd, loff_t length, int small) goto out_putf; sb_start_write(inode->i_sb); - error = locks_verify_truncate(inode, file, length); + error = locks_verify_truncate(inode, f.file, length); if (!error) - error = security_path_truncate(&file->f_path); + error = security_path_truncate(&f.file->f_path); if (!error) - error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, file); + error = do_truncate(dentry, length, ATTR_MTIME|ATTR_CTIME, f.file); sb_end_write(inode->i_sb); out_putf: - fput(file); + fdput(f); out: return error; } @@ -276,15 +276,13 @@ int do_fallocate(struct file *file, int mode, loff_t offset, loff_t len) SYSCALL_DEFINE(fallocate)(int fd, int mode, loff_t offset, loff_t len) { - struct file *file; + struct fd f = fdget(fd); int error = -EBADF; - file = fget(fd); - if (file) { - error = do_fallocate(file, mode, offset, len); - fput(file); + if (f.file) { + error = do_fallocate(f.file, mode, offset, len); + fdput(f); } - return error; } @@ -400,16 +398,15 @@ out: SYSCALL_DEFINE1(fchdir, unsigned int, fd) { - struct file *file; + struct fd f = fdget_raw(fd); struct inode *inode; - int error, fput_needed; + int error = -EBADF; error = -EBADF; - file = fget_raw_light(fd, &fput_needed); - if (!file) + if (!f.file) goto out; - inode = file->f_path.dentry->d_inode; + inode = f.file->f_path.dentry->d_inode; error = -ENOTDIR; if (!S_ISDIR(inode->i_mode)) @@ -417,9 +414,9 @@ SYSCALL_DEFINE1(fchdir, unsigned int, fd) error = inode_permission(inode, MAY_EXEC | MAY_CHDIR); if (!error) - set_fs_pwd(current->fs, &file->f_path); + set_fs_pwd(current->fs, &f.file->f_path); out_putf: - fput_light(file, fput_needed); + fdput(f); out: return error; } @@ -582,23 +579,20 @@ SYSCALL_DEFINE3(lchown, const char __user *, filename, uid_t, user, gid_t, group SYSCALL_DEFINE3(fchown, unsigned int, fd, uid_t, user, gid_t, group) { - struct file * file; + struct fd f = fdget(fd); int error = -EBADF; - struct dentry * dentry; - file = fget(fd); - if (!file) + if (!f.file) goto out; - error = mnt_want_write_file(file); + error = mnt_want_write_file(f.file); if (error) goto out_fput; - dentry = file->f_path.dentry; - audit_inode(NULL, dentry); - error = chown_common(&file->f_path, user, group); - mnt_drop_write_file(file); + audit_inode(NULL, f.file->f_path.dentry); + error = chown_common(&f.file->f_path, user, group); + mnt_drop_write_file(f.file); out_fput: - fput(file); + fdput(f); out: return error; } @@ -803,50 +797,6 @@ struct file *dentry_open(const struct path *path, int flags, } EXPORT_SYMBOL(dentry_open); -static void __put_unused_fd(struct files_struct *files, unsigned int fd) -{ - struct fdtable *fdt = files_fdtable(files); - __clear_open_fd(fd, fdt); - if (fd < files->next_fd) - files->next_fd = fd; -} - -void put_unused_fd(unsigned int fd) -{ - struct files_struct *files = current->files; - spin_lock(&files->file_lock); - __put_unused_fd(files, fd); - spin_unlock(&files->file_lock); -} - -EXPORT_SYMBOL(put_unused_fd); - -/* - * Install a file pointer in the fd array. - * - * The VFS is full of places where we drop the files lock between - * setting the open_fds bitmap and installing the file in the file - * array. At any such point, we are vulnerable to a dup2() race - * installing a file in the array before us. We need to detect this and - * fput() the struct file we are about to overwrite in this case. - * - * It should never happen - if we allow dup2() do it, _really_ bad things - * will follow. - */ - -void fd_install(unsigned int fd, struct file *file) -{ - struct files_struct *files = current->files; - struct fdtable *fdt; - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - BUG_ON(fdt->fd[fd] != NULL); - rcu_assign_pointer(fdt->fd[fd], file); - spin_unlock(&files->file_lock); -} - -EXPORT_SYMBOL(fd_install); - static inline int build_open_flags(int flags, umode_t mode, struct open_flags *op) { int lookup_flags = 0; @@ -858,7 +808,7 @@ static inline int build_open_flags(int flags, umode_t mode, struct open_flags *o op->mode = 0; /* Must never be set by userspace */ - flags &= ~FMODE_NONOTIFY; + flags &= ~FMODE_NONOTIFY & ~O_CLOEXEC; /* * O_SYNC is implemented as __O_SYNC|O_DSYNC. As many places only @@ -1038,23 +988,7 @@ EXPORT_SYMBOL(filp_close); */ SYSCALL_DEFINE1(close, unsigned int, fd) { - struct file * filp; - struct files_struct *files = current->files; - struct fdtable *fdt; - int retval; - - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - if (fd >= fdt->max_fds) - goto out_unlock; - filp = fdt->fd[fd]; - if (!filp) - goto out_unlock; - rcu_assign_pointer(fdt->fd[fd], NULL); - __clear_close_on_exec(fd, fdt); - __put_unused_fd(files, fd); - spin_unlock(&files->file_lock); - retval = filp_close(filp, files); + int retval = __close_fd(current->files, fd); /* can't restart close syscall because file table entry was cleared */ if (unlikely(retval == -ERESTARTSYS || @@ -1064,10 +998,6 @@ SYSCALL_DEFINE1(close, unsigned int, fd) retval = -EINTR; return retval; - -out_unlock: - spin_unlock(&files->file_lock); - return -EBADF; } EXPORT_SYMBOL(sys_close); diff --git a/fs/openpromfs/inode.c b/fs/openpromfs/inode.c index 4a34779..2ad080f 100644 --- a/fs/openpromfs/inode.c +++ b/fs/openpromfs/inode.c @@ -463,6 +463,11 @@ static int __init init_openprom_fs(void) static void __exit exit_openprom_fs(void) { unregister_filesystem(&openprom_fs_type); + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(op_inode_cachep); } @@ -1064,9 +1064,8 @@ err_inode: return err; } -int do_pipe_flags(int *fd, int flags) +static int __do_pipe_flags(int *fd, struct file **files, int flags) { - struct file *files[2]; int error; int fdw, fdr; @@ -1088,11 +1087,8 @@ int do_pipe_flags(int *fd, int flags) fdw = error; audit_fd_pair(fdr, fdw); - fd_install(fdr, files[0]); - fd_install(fdw, files[1]); fd[0] = fdr; fd[1] = fdw; - return 0; err_fdr: @@ -1103,21 +1099,38 @@ int do_pipe_flags(int *fd, int flags) return error; } +int do_pipe_flags(int *fd, int flags) +{ + struct file *files[2]; + int error = __do_pipe_flags(fd, files, flags); + if (!error) { + fd_install(fd[0], files[0]); + fd_install(fd[1], files[1]); + } + return error; +} + /* * sys_pipe() is the normal C calling standard for creating * a pipe. It's not the way Unix traditionally does this, though. */ SYSCALL_DEFINE2(pipe2, int __user *, fildes, int, flags) { + struct file *files[2]; int fd[2]; int error; - error = do_pipe_flags(fd, flags); + error = __do_pipe_flags(fd, files, flags); if (!error) { - if (copy_to_user(fildes, fd, sizeof(fd))) { - sys_close(fd[0]); - sys_close(fd[1]); + if (unlikely(copy_to_user(fildes, fd, sizeof(fd)))) { + fput(files[0]); + fput(files[1]); + put_unused_fd(fd[0]); + put_unused_fd(fd[1]); error = -EFAULT; + } else { + fd_install(fd[0], files[0]); + fd_install(fd[1], files[1]); } } return error; diff --git a/fs/proc/Makefile b/fs/proc/Makefile index c1c7293..99349ef 100644 --- a/fs/proc/Makefile +++ b/fs/proc/Makefile @@ -8,7 +8,7 @@ proc-y := nommu.o task_nommu.o proc-$(CONFIG_MMU) := mmu.o task_mmu.o proc-y += inode.o root.o base.o generic.o array.o \ - proc_tty.o + proc_tty.o fd.o proc-y += cmdline.o proc-y += consoles.o proc-y += cpuinfo.o diff --git a/fs/proc/base.c b/fs/proc/base.c index acd1960..d295af99 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -90,6 +90,7 @@ #endif #include <trace/events/oom.h> #include "internal.h" +#include "fd.h" /* NOTE: * Implementing inode permission operations in /proc is almost @@ -136,8 +137,6 @@ struct pid_entry { NULL, &proc_single_file_operations, \ { .proc_show = show } ) -static int proc_fd_permission(struct inode *inode, int mask); - /* * Count the number of hardlinks for the pid_entry table, excluding the . * and .. links. @@ -1500,7 +1499,7 @@ out: return error; } -static const struct inode_operations proc_pid_link_inode_operations = { +const struct inode_operations proc_pid_link_inode_operations = { .readlink = proc_pid_readlink, .follow_link = proc_pid_follow_link, .setattr = proc_setattr, @@ -1509,21 +1508,6 @@ static const struct inode_operations proc_pid_link_inode_operations = { /* building an inode */ -static int task_dumpable(struct task_struct *task) -{ - int dumpable = 0; - struct mm_struct *mm; - - task_lock(task); - mm = task->mm; - if (mm) - dumpable = get_dumpable(mm); - task_unlock(task); - if(dumpable == 1) - return 1; - return 0; -} - struct inode *proc_pid_make_inode(struct super_block * sb, struct task_struct *task) { struct inode * inode; @@ -1649,15 +1633,6 @@ int pid_revalidate(struct dentry *dentry, unsigned int flags) return 0; } -static int pid_delete_dentry(const struct dentry * dentry) -{ - /* Is the task we represent dead? - * If so, then don't put the dentry on the lru list, - * kill it immediately. - */ - return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first; -} - const struct dentry_operations pid_dentry_operations = { .d_revalidate = pid_revalidate, @@ -1720,289 +1695,6 @@ end_instantiate: return filldir(dirent, name, len, filp->f_pos, ino, type); } -static unsigned name_to_int(struct dentry *dentry) -{ - const char *name = dentry->d_name.name; - int len = dentry->d_name.len; - unsigned n = 0; - - if (len > 1 && *name == '0') - goto out; - while (len-- > 0) { - unsigned c = *name++ - '0'; - if (c > 9) - goto out; - if (n >= (~0U-9)/10) - goto out; - n *= 10; - n += c; - } - return n; -out: - return ~0U; -} - -#define PROC_FDINFO_MAX 64 - -static int proc_fd_info(struct inode *inode, struct path *path, char *info) -{ - struct task_struct *task = get_proc_task(inode); - struct files_struct *files = NULL; - struct file *file; - int fd = proc_fd(inode); - - if (task) { - files = get_files_struct(task); - put_task_struct(task); - } - if (files) { - /* - * We are not taking a ref to the file structure, so we must - * hold ->file_lock. - */ - spin_lock(&files->file_lock); - file = fcheck_files(files, fd); - if (file) { - unsigned int f_flags; - struct fdtable *fdt; - - fdt = files_fdtable(files); - f_flags = file->f_flags & ~O_CLOEXEC; - if (close_on_exec(fd, fdt)) - f_flags |= O_CLOEXEC; - - if (path) { - *path = file->f_path; - path_get(&file->f_path); - } - if (info) - snprintf(info, PROC_FDINFO_MAX, - "pos:\t%lli\n" - "flags:\t0%o\n", - (long long) file->f_pos, - f_flags); - spin_unlock(&files->file_lock); - put_files_struct(files); - return 0; - } - spin_unlock(&files->file_lock); - put_files_struct(files); - } - return -ENOENT; -} - -static int proc_fd_link(struct dentry *dentry, struct path *path) -{ - return proc_fd_info(dentry->d_inode, path, NULL); -} - -static int tid_fd_revalidate(struct dentry *dentry, unsigned int flags) -{ - struct inode *inode; - struct task_struct *task; - int fd; - struct files_struct *files; - const struct cred *cred; - - if (flags & LOOKUP_RCU) - return -ECHILD; - - inode = dentry->d_inode; - task = get_proc_task(inode); - fd = proc_fd(inode); - - if (task) { - files = get_files_struct(task); - if (files) { - struct file *file; - rcu_read_lock(); - file = fcheck_files(files, fd); - if (file) { - unsigned f_mode = file->f_mode; - - rcu_read_unlock(); - put_files_struct(files); - - if (task_dumpable(task)) { - rcu_read_lock(); - cred = __task_cred(task); - inode->i_uid = cred->euid; - inode->i_gid = cred->egid; - rcu_read_unlock(); - } else { - inode->i_uid = GLOBAL_ROOT_UID; - inode->i_gid = GLOBAL_ROOT_GID; - } - - if (S_ISLNK(inode->i_mode)) { - unsigned i_mode = S_IFLNK; - if (f_mode & FMODE_READ) - i_mode |= S_IRUSR | S_IXUSR; - if (f_mode & FMODE_WRITE) - i_mode |= S_IWUSR | S_IXUSR; - inode->i_mode = i_mode; - } - - security_task_to_inode(task, inode); - put_task_struct(task); - return 1; - } - rcu_read_unlock(); - put_files_struct(files); - } - put_task_struct(task); - } - d_drop(dentry); - return 0; -} - -static const struct dentry_operations tid_fd_dentry_operations = -{ - .d_revalidate = tid_fd_revalidate, - .d_delete = pid_delete_dentry, -}; - -static struct dentry *proc_fd_instantiate(struct inode *dir, - struct dentry *dentry, struct task_struct *task, const void *ptr) -{ - unsigned fd = (unsigned long)ptr; - struct inode *inode; - struct proc_inode *ei; - struct dentry *error = ERR_PTR(-ENOENT); - - inode = proc_pid_make_inode(dir->i_sb, task); - if (!inode) - goto out; - ei = PROC_I(inode); - ei->fd = fd; - - inode->i_mode = S_IFLNK; - inode->i_op = &proc_pid_link_inode_operations; - inode->i_size = 64; - ei->op.proc_get_link = proc_fd_link; - d_set_d_op(dentry, &tid_fd_dentry_operations); - d_add(dentry, inode); - /* Close the race of the process dying before we return the dentry */ - if (tid_fd_revalidate(dentry, 0)) - error = NULL; - - out: - return error; -} - -static struct dentry *proc_lookupfd_common(struct inode *dir, - struct dentry *dentry, - instantiate_t instantiate) -{ - struct task_struct *task = get_proc_task(dir); - unsigned fd = name_to_int(dentry); - struct dentry *result = ERR_PTR(-ENOENT); - - if (!task) - goto out_no_task; - if (fd == ~0U) - goto out; - - result = instantiate(dir, dentry, task, (void *)(unsigned long)fd); -out: - put_task_struct(task); -out_no_task: - return result; -} - -static int proc_readfd_common(struct file * filp, void * dirent, - filldir_t filldir, instantiate_t instantiate) -{ - struct dentry *dentry = filp->f_path.dentry; - struct inode *inode = dentry->d_inode; - struct task_struct *p = get_proc_task(inode); - unsigned int fd, ino; - int retval; - struct files_struct * files; - - retval = -ENOENT; - if (!p) - goto out_no_task; - retval = 0; - - fd = filp->f_pos; - switch (fd) { - case 0: - if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - case 1: - ino = parent_ino(dentry); - if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) - goto out; - filp->f_pos++; - default: - files = get_files_struct(p); - if (!files) - goto out; - rcu_read_lock(); - for (fd = filp->f_pos-2; - fd < files_fdtable(files)->max_fds; - fd++, filp->f_pos++) { - char name[PROC_NUMBUF]; - int len; - int rv; - - if (!fcheck_files(files, fd)) - continue; - rcu_read_unlock(); - - len = snprintf(name, sizeof(name), "%d", fd); - rv = proc_fill_cache(filp, dirent, filldir, - name, len, instantiate, p, - (void *)(unsigned long)fd); - if (rv < 0) - goto out_fd_loop; - rcu_read_lock(); - } - rcu_read_unlock(); -out_fd_loop: - put_files_struct(files); - } -out: - put_task_struct(p); -out_no_task: - return retval; -} - -static struct dentry *proc_lookupfd(struct inode *dir, struct dentry *dentry, - unsigned int flags) -{ - return proc_lookupfd_common(dir, dentry, proc_fd_instantiate); -} - -static int proc_readfd(struct file *filp, void *dirent, filldir_t filldir) -{ - return proc_readfd_common(filp, dirent, filldir, proc_fd_instantiate); -} - -static ssize_t proc_fdinfo_read(struct file *file, char __user *buf, - size_t len, loff_t *ppos) -{ - char tmp[PROC_FDINFO_MAX]; - int err = proc_fd_info(file->f_path.dentry->d_inode, NULL, tmp); - if (!err) - err = simple_read_from_buffer(buf, len, ppos, tmp, strlen(tmp)); - return err; -} - -static const struct file_operations proc_fdinfo_file_operations = { - .open = nonseekable_open, - .read = proc_fdinfo_read, - .llseek = no_llseek, -}; - -static const struct file_operations proc_fd_operations = { - .read = generic_read_dir, - .readdir = proc_readfd, - .llseek = default_llseek, -}; - #ifdef CONFIG_CHECKPOINT_RESTORE /* @@ -2121,7 +1813,7 @@ out: } struct map_files_info { - struct file *file; + fmode_t mode; unsigned long len; unsigned char name[4*sizeof(long)+2]; /* max: %lx-%lx\0 */ }; @@ -2130,13 +1822,10 @@ static struct dentry * proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { - const struct file *file = ptr; + fmode_t mode = (fmode_t)(unsigned long)ptr; struct proc_inode *ei; struct inode *inode; - if (!file) - return ERR_PTR(-ENOENT); - inode = proc_pid_make_inode(dir->i_sb, task); if (!inode) return ERR_PTR(-ENOENT); @@ -2148,9 +1837,9 @@ proc_map_files_instantiate(struct inode *dir, struct dentry *dentry, inode->i_size = 64; inode->i_mode = S_IFLNK; - if (file->f_mode & FMODE_READ) + if (mode & FMODE_READ) inode->i_mode |= S_IRUSR; - if (file->f_mode & FMODE_WRITE) + if (mode & FMODE_WRITE) inode->i_mode |= S_IWUSR; d_set_d_op(dentry, &tid_map_files_dentry_operations); @@ -2194,7 +1883,8 @@ static struct dentry *proc_map_files_lookup(struct inode *dir, if (!vma) goto out_no_vma; - result = proc_map_files_instantiate(dir, dentry, task, vma->vm_file); + result = proc_map_files_instantiate(dir, dentry, task, + (void *)(unsigned long)vma->vm_file->f_mode); out_no_vma: up_read(&mm->mmap_sem); @@ -2295,8 +1985,7 @@ proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) if (++pos <= filp->f_pos) continue; - get_file(vma->vm_file); - info.file = vma->vm_file; + info.mode = vma->vm_file->f_mode; info.len = snprintf(info.name, sizeof(info.name), "%lx-%lx", vma->vm_start, vma->vm_end); @@ -2311,19 +2000,11 @@ proc_map_files_readdir(struct file *filp, void *dirent, filldir_t filldir) ret = proc_fill_cache(filp, dirent, filldir, p->name, p->len, proc_map_files_instantiate, - task, p->file); + task, + (void *)(unsigned long)p->mode); if (ret) break; filp->f_pos++; - fput(p->file); - } - for (; i < nr_files; i++) { - /* - * In case of error don't forget - * to put rest of file refs. - */ - p = flex_array_get(fa, i); - fput(p->file); } if (fa) flex_array_free(fa); @@ -2345,82 +2026,6 @@ static const struct file_operations proc_map_files_operations = { #endif /* CONFIG_CHECKPOINT_RESTORE */ -/* - * /proc/pid/fd needs a special permission handler so that a process can still - * access /proc/self/fd after it has executed a setuid(). - */ -static int proc_fd_permission(struct inode *inode, int mask) -{ - int rv = generic_permission(inode, mask); - if (rv == 0) - return 0; - if (task_pid(current) == proc_pid(inode)) - rv = 0; - return rv; -} - -/* - * proc directories can do almost nothing.. - */ -static const struct inode_operations proc_fd_inode_operations = { - .lookup = proc_lookupfd, - .permission = proc_fd_permission, - .setattr = proc_setattr, -}; - -static struct dentry *proc_fdinfo_instantiate(struct inode *dir, - struct dentry *dentry, struct task_struct *task, const void *ptr) -{ - unsigned fd = (unsigned long)ptr; - struct inode *inode; - struct proc_inode *ei; - struct dentry *error = ERR_PTR(-ENOENT); - - inode = proc_pid_make_inode(dir->i_sb, task); - if (!inode) - goto out; - ei = PROC_I(inode); - ei->fd = fd; - inode->i_mode = S_IFREG | S_IRUSR; - inode->i_fop = &proc_fdinfo_file_operations; - d_set_d_op(dentry, &tid_fd_dentry_operations); - d_add(dentry, inode); - /* Close the race of the process dying before we return the dentry */ - if (tid_fd_revalidate(dentry, 0)) - error = NULL; - - out: - return error; -} - -static struct dentry *proc_lookupfdinfo(struct inode *dir, - struct dentry *dentry, - unsigned int flags) -{ - return proc_lookupfd_common(dir, dentry, proc_fdinfo_instantiate); -} - -static int proc_readfdinfo(struct file *filp, void *dirent, filldir_t filldir) -{ - return proc_readfd_common(filp, dirent, filldir, - proc_fdinfo_instantiate); -} - -static const struct file_operations proc_fdinfo_operations = { - .read = generic_read_dir, - .readdir = proc_readfdinfo, - .llseek = default_llseek, -}; - -/* - * proc directories can do almost nothing.. - */ -static const struct inode_operations proc_fdinfo_inode_operations = { - .lookup = proc_lookupfdinfo, - .setattr = proc_setattr, -}; - - static struct dentry *proc_pident_instantiate(struct inode *dir, struct dentry *dentry, struct task_struct *task, const void *ptr) { diff --git a/fs/proc/fd.c b/fs/proc/fd.c new file mode 100644 index 0000000..f28a875 --- /dev/null +++ b/fs/proc/fd.c @@ -0,0 +1,367 @@ +#include <linux/sched.h> +#include <linux/errno.h> +#include <linux/dcache.h> +#include <linux/path.h> +#include <linux/fdtable.h> +#include <linux/namei.h> +#include <linux/pid.h> +#include <linux/security.h> +#include <linux/file.h> +#include <linux/seq_file.h> + +#include <linux/proc_fs.h> + +#include "internal.h" +#include "fd.h" + +static int seq_show(struct seq_file *m, void *v) +{ + struct files_struct *files = NULL; + int f_flags = 0, ret = -ENOENT; + struct file *file = NULL; + struct task_struct *task; + + task = get_proc_task(m->private); + if (!task) + return -ENOENT; + + files = get_files_struct(task); + put_task_struct(task); + + if (files) { + int fd = proc_fd(m->private); + + spin_lock(&files->file_lock); + file = fcheck_files(files, fd); + if (file) { + struct fdtable *fdt = files_fdtable(files); + + f_flags = file->f_flags; + if (close_on_exec(fd, fdt)) + f_flags |= O_CLOEXEC; + + get_file(file); + ret = 0; + } + spin_unlock(&files->file_lock); + put_files_struct(files); + } + + if (!ret) { + seq_printf(m, "pos:\t%lli\nflags:\t0%o\n", + (long long)file->f_pos, f_flags); + fput(file); + } + + return ret; +} + +static int seq_fdinfo_open(struct inode *inode, struct file *file) +{ + return single_open(file, seq_show, inode); +} + +static const struct file_operations proc_fdinfo_file_operations = { + .open = seq_fdinfo_open, + .read = seq_read, + .llseek = seq_lseek, + .release = single_release, +}; + +static int tid_fd_revalidate(struct dentry *dentry, unsigned int flags) +{ + struct files_struct *files; + struct task_struct *task; + const struct cred *cred; + struct inode *inode; + int fd; + + if (flags & LOOKUP_RCU) + return -ECHILD; + + inode = dentry->d_inode; + task = get_proc_task(inode); + fd = proc_fd(inode); + + if (task) { + files = get_files_struct(task); + if (files) { + struct file *file; + + rcu_read_lock(); + file = fcheck_files(files, fd); + if (file) { + unsigned f_mode = file->f_mode; + + rcu_read_unlock(); + put_files_struct(files); + + if (task_dumpable(task)) { + rcu_read_lock(); + cred = __task_cred(task); + inode->i_uid = cred->euid; + inode->i_gid = cred->egid; + rcu_read_unlock(); + } else { + inode->i_uid = GLOBAL_ROOT_UID; + inode->i_gid = GLOBAL_ROOT_GID; + } + + if (S_ISLNK(inode->i_mode)) { + unsigned i_mode = S_IFLNK; + if (f_mode & FMODE_READ) + i_mode |= S_IRUSR | S_IXUSR; + if (f_mode & FMODE_WRITE) + i_mode |= S_IWUSR | S_IXUSR; + inode->i_mode = i_mode; + } + + security_task_to_inode(task, inode); + put_task_struct(task); + return 1; + } + rcu_read_unlock(); + put_files_struct(files); + } + put_task_struct(task); + } + + d_drop(dentry); + return 0; +} + +static const struct dentry_operations tid_fd_dentry_operations = { + .d_revalidate = tid_fd_revalidate, + .d_delete = pid_delete_dentry, +}; + +static int proc_fd_link(struct dentry *dentry, struct path *path) +{ + struct files_struct *files = NULL; + struct task_struct *task; + int ret = -ENOENT; + + task = get_proc_task(dentry->d_inode); + if (task) { + files = get_files_struct(task); + put_task_struct(task); + } + + if (files) { + int fd = proc_fd(dentry->d_inode); + struct file *fd_file; + + spin_lock(&files->file_lock); + fd_file = fcheck_files(files, fd); + if (fd_file) { + *path = fd_file->f_path; + path_get(&fd_file->f_path); + ret = 0; + } + spin_unlock(&files->file_lock); + put_files_struct(files); + } + + return ret; +} + +static struct dentry * +proc_fd_instantiate(struct inode *dir, struct dentry *dentry, + struct task_struct *task, const void *ptr) +{ + struct dentry *error = ERR_PTR(-ENOENT); + unsigned fd = (unsigned long)ptr; + struct proc_inode *ei; + struct inode *inode; + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + goto out; + + ei = PROC_I(inode); + ei->fd = fd; + + inode->i_mode = S_IFLNK; + inode->i_op = &proc_pid_link_inode_operations; + inode->i_size = 64; + + ei->op.proc_get_link = proc_fd_link; + + d_set_d_op(dentry, &tid_fd_dentry_operations); + d_add(dentry, inode); + + /* Close the race of the process dying before we return the dentry */ + if (tid_fd_revalidate(dentry, 0)) + error = NULL; + out: + return error; +} + +static struct dentry *proc_lookupfd_common(struct inode *dir, + struct dentry *dentry, + instantiate_t instantiate) +{ + struct task_struct *task = get_proc_task(dir); + struct dentry *result = ERR_PTR(-ENOENT); + unsigned fd = name_to_int(dentry); + + if (!task) + goto out_no_task; + if (fd == ~0U) + goto out; + + result = instantiate(dir, dentry, task, (void *)(unsigned long)fd); +out: + put_task_struct(task); +out_no_task: + return result; +} + +static int proc_readfd_common(struct file * filp, void * dirent, + filldir_t filldir, instantiate_t instantiate) +{ + struct dentry *dentry = filp->f_path.dentry; + struct inode *inode = dentry->d_inode; + struct task_struct *p = get_proc_task(inode); + struct files_struct *files; + unsigned int fd, ino; + int retval; + + retval = -ENOENT; + if (!p) + goto out_no_task; + retval = 0; + + fd = filp->f_pos; + switch (fd) { + case 0: + if (filldir(dirent, ".", 1, 0, inode->i_ino, DT_DIR) < 0) + goto out; + filp->f_pos++; + case 1: + ino = parent_ino(dentry); + if (filldir(dirent, "..", 2, 1, ino, DT_DIR) < 0) + goto out; + filp->f_pos++; + default: + files = get_files_struct(p); + if (!files) + goto out; + rcu_read_lock(); + for (fd = filp->f_pos - 2; + fd < files_fdtable(files)->max_fds; + fd++, filp->f_pos++) { + char name[PROC_NUMBUF]; + int len; + int rv; + + if (!fcheck_files(files, fd)) + continue; + rcu_read_unlock(); + + len = snprintf(name, sizeof(name), "%d", fd); + rv = proc_fill_cache(filp, dirent, filldir, + name, len, instantiate, p, + (void *)(unsigned long)fd); + if (rv < 0) + goto out_fd_loop; + rcu_read_lock(); + } + rcu_read_unlock(); +out_fd_loop: + put_files_struct(files); + } +out: + put_task_struct(p); +out_no_task: + return retval; +} + +static int proc_readfd(struct file *filp, void *dirent, filldir_t filldir) +{ + return proc_readfd_common(filp, dirent, filldir, proc_fd_instantiate); +} + +const struct file_operations proc_fd_operations = { + .read = generic_read_dir, + .readdir = proc_readfd, + .llseek = default_llseek, +}; + +static struct dentry *proc_lookupfd(struct inode *dir, struct dentry *dentry, + unsigned int flags) +{ + return proc_lookupfd_common(dir, dentry, proc_fd_instantiate); +} + +/* + * /proc/pid/fd needs a special permission handler so that a process can still + * access /proc/self/fd after it has executed a setuid(). + */ +int proc_fd_permission(struct inode *inode, int mask) +{ + int rv = generic_permission(inode, mask); + if (rv == 0) + return 0; + if (task_pid(current) == proc_pid(inode)) + rv = 0; + return rv; +} + +const struct inode_operations proc_fd_inode_operations = { + .lookup = proc_lookupfd, + .permission = proc_fd_permission, + .setattr = proc_setattr, +}; + +static struct dentry * +proc_fdinfo_instantiate(struct inode *dir, struct dentry *dentry, + struct task_struct *task, const void *ptr) +{ + struct dentry *error = ERR_PTR(-ENOENT); + unsigned fd = (unsigned long)ptr; + struct proc_inode *ei; + struct inode *inode; + + inode = proc_pid_make_inode(dir->i_sb, task); + if (!inode) + goto out; + + ei = PROC_I(inode); + ei->fd = fd; + + inode->i_mode = S_IFREG | S_IRUSR; + inode->i_fop = &proc_fdinfo_file_operations; + + d_set_d_op(dentry, &tid_fd_dentry_operations); + d_add(dentry, inode); + + /* Close the race of the process dying before we return the dentry */ + if (tid_fd_revalidate(dentry, 0)) + error = NULL; + out: + return error; +} + +static struct dentry * +proc_lookupfdinfo(struct inode *dir, struct dentry *dentry, unsigned int flags) +{ + return proc_lookupfd_common(dir, dentry, proc_fdinfo_instantiate); +} + +static int proc_readfdinfo(struct file *filp, void *dirent, filldir_t filldir) +{ + return proc_readfd_common(filp, dirent, filldir, + proc_fdinfo_instantiate); +} + +const struct inode_operations proc_fdinfo_inode_operations = { + .lookup = proc_lookupfdinfo, + .setattr = proc_setattr, +}; + +const struct file_operations proc_fdinfo_operations = { + .read = generic_read_dir, + .readdir = proc_readfdinfo, + .llseek = default_llseek, +}; diff --git a/fs/proc/fd.h b/fs/proc/fd.h new file mode 100644 index 0000000..cbb1d47 --- /dev/null +++ b/fs/proc/fd.h @@ -0,0 +1,14 @@ +#ifndef __PROCFS_FD_H__ +#define __PROCFS_FD_H__ + +#include <linux/fs.h> + +extern const struct file_operations proc_fd_operations; +extern const struct inode_operations proc_fd_inode_operations; + +extern const struct file_operations proc_fdinfo_operations; +extern const struct inode_operations proc_fdinfo_inode_operations; + +extern int proc_fd_permission(struct inode *inode, int mask); + +#endif /* __PROCFS_FD_H__ */ diff --git a/fs/proc/internal.h b/fs/proc/internal.h index e1167a1..67925a7 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -9,6 +9,7 @@ * 2 of the License, or (at your option) any later version. */ +#include <linux/sched.h> #include <linux/proc_fs.h> struct ctl_table_header; @@ -65,6 +66,7 @@ extern const struct file_operations proc_clear_refs_operations; extern const struct file_operations proc_pagemap_operations; extern const struct file_operations proc_net_operations; extern const struct inode_operations proc_net_inode_operations; +extern const struct inode_operations proc_pid_link_inode_operations; struct proc_maps_private { struct pid *pid; @@ -91,6 +93,52 @@ static inline int proc_fd(struct inode *inode) return PROC_I(inode)->fd; } +static inline int task_dumpable(struct task_struct *task) +{ + int dumpable = 0; + struct mm_struct *mm; + + task_lock(task); + mm = task->mm; + if (mm) + dumpable = get_dumpable(mm); + task_unlock(task); + if(dumpable == 1) + return 1; + return 0; +} + +static inline int pid_delete_dentry(const struct dentry * dentry) +{ + /* Is the task we represent dead? + * If so, then don't put the dentry on the lru list, + * kill it immediately. + */ + return !proc_pid(dentry->d_inode)->tasks[PIDTYPE_PID].first; +} + +static inline unsigned name_to_int(struct dentry *dentry) +{ + const char *name = dentry->d_name.name; + int len = dentry->d_name.len; + unsigned n = 0; + + if (len > 1 && *name == '0') + goto out; + while (len-- > 0) { + unsigned c = *name++ - '0'; + if (c > 9) + goto out; + if (n >= (~0U-9)/10) + goto out; + n *= 10; + n += c; + } + return n; +out: + return ~0U; +} + struct dentry *proc_lookup_de(struct proc_dir_entry *de, struct inode *ino, struct dentry *dentry); int proc_readdir_de(struct proc_dir_entry *de, struct file *filp, void *dirent, diff --git a/fs/qnx4/inode.c b/fs/qnx4/inode.c index 5c3c7b0..43098bb 100644 --- a/fs/qnx4/inode.c +++ b/fs/qnx4/inode.c @@ -391,6 +391,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(qnx4_inode_cachep); } diff --git a/fs/qnx6/inode.c b/fs/qnx6/inode.c index f4eef0b..b6addf5 100644 --- a/fs/qnx6/inode.c +++ b/fs/qnx6/inode.c @@ -651,6 +651,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(qnx6_inode_cachep); } diff --git a/fs/read_write.c b/fs/read_write.c index 1adfb69..d065348 100644 --- a/fs/read_write.c +++ b/fs/read_write.c @@ -232,23 +232,18 @@ EXPORT_SYMBOL(vfs_llseek); SYSCALL_DEFINE3(lseek, unsigned int, fd, off_t, offset, unsigned int, origin) { off_t retval; - struct file * file; - int fput_needed; - - retval = -EBADF; - file = fget_light(fd, &fput_needed); - if (!file) - goto bad; + struct fd f = fdget(fd); + if (!f.file) + return -EBADF; retval = -EINVAL; if (origin <= SEEK_MAX) { - loff_t res = vfs_llseek(file, offset, origin); + loff_t res = vfs_llseek(f.file, offset, origin); retval = res; if (res != (loff_t)retval) retval = -EOVERFLOW; /* LFS: should only happen on 32 bit platforms */ } - fput_light(file, fput_needed); -bad: + fdput(f); return retval; } @@ -258,20 +253,17 @@ SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high, unsigned int, origin) { int retval; - struct file * file; + struct fd f = fdget(fd); loff_t offset; - int fput_needed; - retval = -EBADF; - file = fget_light(fd, &fput_needed); - if (!file) - goto bad; + if (!f.file) + return -EBADF; retval = -EINVAL; if (origin > SEEK_MAX) goto out_putf; - offset = vfs_llseek(file, ((loff_t) offset_high << 32) | offset_low, + offset = vfs_llseek(f.file, ((loff_t) offset_high << 32) | offset_low, origin); retval = (int)offset; @@ -281,8 +273,7 @@ SYSCALL_DEFINE5(llseek, unsigned int, fd, unsigned long, offset_high, retval = 0; } out_putf: - fput_light(file, fput_needed); -bad: + fdput(f); return retval; } #endif @@ -461,34 +452,29 @@ static inline void file_pos_write(struct file *file, loff_t pos) SYSCALL_DEFINE3(read, unsigned int, fd, char __user *, buf, size_t, count) { - struct file *file; + struct fd f = fdget(fd); ssize_t ret = -EBADF; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (file) { - loff_t pos = file_pos_read(file); - ret = vfs_read(file, buf, count, &pos); - file_pos_write(file, pos); - fput_light(file, fput_needed); + if (f.file) { + loff_t pos = file_pos_read(f.file); + ret = vfs_read(f.file, buf, count, &pos); + file_pos_write(f.file, pos); + fdput(f); } - return ret; } SYSCALL_DEFINE3(write, unsigned int, fd, const char __user *, buf, size_t, count) { - struct file *file; + struct fd f = fdget(fd); ssize_t ret = -EBADF; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (file) { - loff_t pos = file_pos_read(file); - ret = vfs_write(file, buf, count, &pos); - file_pos_write(file, pos); - fput_light(file, fput_needed); + if (f.file) { + loff_t pos = file_pos_read(f.file); + ret = vfs_write(f.file, buf, count, &pos); + file_pos_write(f.file, pos); + fdput(f); } return ret; @@ -497,19 +483,18 @@ SYSCALL_DEFINE3(write, unsigned int, fd, const char __user *, buf, SYSCALL_DEFINE(pread64)(unsigned int fd, char __user *buf, size_t count, loff_t pos) { - struct file *file; + struct fd f; ssize_t ret = -EBADF; - int fput_needed; if (pos < 0) return -EINVAL; - file = fget_light(fd, &fput_needed); - if (file) { + f = fdget(fd); + if (f.file) { ret = -ESPIPE; - if (file->f_mode & FMODE_PREAD) - ret = vfs_read(file, buf, count, &pos); - fput_light(file, fput_needed); + if (f.file->f_mode & FMODE_PREAD) + ret = vfs_read(f.file, buf, count, &pos); + fdput(f); } return ret; @@ -526,19 +511,18 @@ SYSCALL_ALIAS(sys_pread64, SyS_pread64); SYSCALL_DEFINE(pwrite64)(unsigned int fd, const char __user *buf, size_t count, loff_t pos) { - struct file *file; + struct fd f; ssize_t ret = -EBADF; - int fput_needed; if (pos < 0) return -EINVAL; - file = fget_light(fd, &fput_needed); - if (file) { + f = fdget(fd); + if (f.file) { ret = -ESPIPE; - if (file->f_mode & FMODE_PWRITE) - ret = vfs_write(file, buf, count, &pos); - fput_light(file, fput_needed); + if (f.file->f_mode & FMODE_PWRITE) + ret = vfs_write(f.file, buf, count, &pos); + fdput(f); } return ret; @@ -789,16 +773,14 @@ EXPORT_SYMBOL(vfs_writev); SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec, unsigned long, vlen) { - struct file *file; + struct fd f = fdget(fd); ssize_t ret = -EBADF; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (file) { - loff_t pos = file_pos_read(file); - ret = vfs_readv(file, vec, vlen, &pos); - file_pos_write(file, pos); - fput_light(file, fput_needed); + if (f.file) { + loff_t pos = file_pos_read(f.file); + ret = vfs_readv(f.file, vec, vlen, &pos); + file_pos_write(f.file, pos); + fdput(f); } if (ret > 0) @@ -810,16 +792,14 @@ SYSCALL_DEFINE3(readv, unsigned long, fd, const struct iovec __user *, vec, SYSCALL_DEFINE3(writev, unsigned long, fd, const struct iovec __user *, vec, unsigned long, vlen) { - struct file *file; + struct fd f = fdget(fd); ssize_t ret = -EBADF; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (file) { - loff_t pos = file_pos_read(file); - ret = vfs_writev(file, vec, vlen, &pos); - file_pos_write(file, pos); - fput_light(file, fput_needed); + if (f.file) { + loff_t pos = file_pos_read(f.file); + ret = vfs_writev(f.file, vec, vlen, &pos); + file_pos_write(f.file, pos); + fdput(f); } if (ret > 0) @@ -838,19 +818,18 @@ SYSCALL_DEFINE5(preadv, unsigned long, fd, const struct iovec __user *, vec, unsigned long, vlen, unsigned long, pos_l, unsigned long, pos_h) { loff_t pos = pos_from_hilo(pos_h, pos_l); - struct file *file; + struct fd f; ssize_t ret = -EBADF; - int fput_needed; if (pos < 0) return -EINVAL; - file = fget_light(fd, &fput_needed); - if (file) { + f = fdget(fd); + if (f.file) { ret = -ESPIPE; - if (file->f_mode & FMODE_PREAD) - ret = vfs_readv(file, vec, vlen, &pos); - fput_light(file, fput_needed); + if (f.file->f_mode & FMODE_PREAD) + ret = vfs_readv(f.file, vec, vlen, &pos); + fdput(f); } if (ret > 0) @@ -863,19 +842,18 @@ SYSCALL_DEFINE5(pwritev, unsigned long, fd, const struct iovec __user *, vec, unsigned long, vlen, unsigned long, pos_l, unsigned long, pos_h) { loff_t pos = pos_from_hilo(pos_h, pos_l); - struct file *file; + struct fd f; ssize_t ret = -EBADF; - int fput_needed; if (pos < 0) return -EINVAL; - file = fget_light(fd, &fput_needed); - if (file) { + f = fdget(fd); + if (f.file) { ret = -ESPIPE; - if (file->f_mode & FMODE_PWRITE) - ret = vfs_writev(file, vec, vlen, &pos); - fput_light(file, fput_needed); + if (f.file->f_mode & FMODE_PWRITE) + ret = vfs_writev(f.file, vec, vlen, &pos); + fdput(f); } if (ret > 0) @@ -884,31 +862,31 @@ SYSCALL_DEFINE5(pwritev, unsigned long, fd, const struct iovec __user *, vec, return ret; } -static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, - size_t count, loff_t max) +ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, size_t count, + loff_t max) { - struct file * in_file, * out_file; - struct inode * in_inode, * out_inode; + struct fd in, out; + struct inode *in_inode, *out_inode; loff_t pos; ssize_t retval; - int fput_needed_in, fput_needed_out, fl; + int fl; /* * Get input file, and verify that it is ok.. */ retval = -EBADF; - in_file = fget_light(in_fd, &fput_needed_in); - if (!in_file) + in = fdget(in_fd); + if (!in.file) goto out; - if (!(in_file->f_mode & FMODE_READ)) + if (!(in.file->f_mode & FMODE_READ)) goto fput_in; retval = -ESPIPE; if (!ppos) - ppos = &in_file->f_pos; + ppos = &in.file->f_pos; else - if (!(in_file->f_mode & FMODE_PREAD)) + if (!(in.file->f_mode & FMODE_PREAD)) goto fput_in; - retval = rw_verify_area(READ, in_file, ppos, count); + retval = rw_verify_area(READ, in.file, ppos, count); if (retval < 0) goto fput_in; count = retval; @@ -917,15 +895,15 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, * Get output file, and verify that it is ok.. */ retval = -EBADF; - out_file = fget_light(out_fd, &fput_needed_out); - if (!out_file) + out = fdget(out_fd); + if (!out.file) goto fput_in; - if (!(out_file->f_mode & FMODE_WRITE)) + if (!(out.file->f_mode & FMODE_WRITE)) goto fput_out; retval = -EINVAL; - in_inode = in_file->f_path.dentry->d_inode; - out_inode = out_file->f_path.dentry->d_inode; - retval = rw_verify_area(WRITE, out_file, &out_file->f_pos, count); + in_inode = in.file->f_path.dentry->d_inode; + out_inode = out.file->f_path.dentry->d_inode; + retval = rw_verify_area(WRITE, out.file, &out.file->f_pos, count); if (retval < 0) goto fput_out; count = retval; @@ -949,10 +927,10 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, * and the application is arguably buggy if it doesn't expect * EAGAIN on a non-blocking file descriptor. */ - if (in_file->f_flags & O_NONBLOCK) + if (in.file->f_flags & O_NONBLOCK) fl = SPLICE_F_NONBLOCK; #endif - retval = do_splice_direct(in_file, ppos, out_file, count, fl); + retval = do_splice_direct(in.file, ppos, out.file, count, fl); if (retval > 0) { add_rchar(current, retval); @@ -965,9 +943,9 @@ static ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, retval = -EOVERFLOW; fput_out: - fput_light(out_file, fput_needed_out); + fdput(out); fput_in: - fput_light(in_file, fput_needed_in); + fdput(in); out: return retval; } diff --git a/fs/read_write.h b/fs/read_write.h index d07b954..d3e00ef 100644 --- a/fs/read_write.h +++ b/fs/read_write.h @@ -12,3 +12,5 @@ ssize_t do_sync_readv_writev(struct file *filp, const struct iovec *iov, unsigned long nr_segs, size_t len, loff_t *ppos, iov_fn_t fn); ssize_t do_loop_readv_writev(struct file *filp, struct iovec *iov, unsigned long nr_segs, loff_t *ppos, io_fn_t fn); +ssize_t do_sendfile(int out_fd, int in_fd, loff_t *ppos, size_t count, + loff_t max); diff --git a/fs/readdir.c b/fs/readdir.c index 39e3370..5e69ef5 100644 --- a/fs/readdir.c +++ b/fs/readdir.c @@ -106,22 +106,20 @@ SYSCALL_DEFINE3(old_readdir, unsigned int, fd, struct old_linux_dirent __user *, dirent, unsigned int, count) { int error; - struct file * file; + struct fd f = fdget(fd); struct readdir_callback buf; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADF; buf.result = 0; buf.dirent = dirent; - error = vfs_readdir(file, fillonedir, &buf); + error = vfs_readdir(f.file, fillonedir, &buf); if (buf.result) error = buf.result; - fput_light(file, fput_needed); + fdput(f); return error; } @@ -191,17 +189,16 @@ efault: SYSCALL_DEFINE3(getdents, unsigned int, fd, struct linux_dirent __user *, dirent, unsigned int, count) { - struct file * file; + struct fd f; struct linux_dirent __user * lastdirent; struct getdents_callback buf; - int fput_needed; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) return -EFAULT; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return -EBADF; buf.current_dir = dirent; @@ -209,17 +206,17 @@ SYSCALL_DEFINE3(getdents, unsigned int, fd, buf.count = count; buf.error = 0; - error = vfs_readdir(file, filldir, &buf); + error = vfs_readdir(f.file, filldir, &buf); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - if (put_user(file->f_pos, &lastdirent->d_off)) + if (put_user(f.file->f_pos, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; } - fput_light(file, fput_needed); + fdput(f); return error; } @@ -272,17 +269,16 @@ efault: SYSCALL_DEFINE3(getdents64, unsigned int, fd, struct linux_dirent64 __user *, dirent, unsigned int, count) { - struct file * file; + struct fd f; struct linux_dirent64 __user * lastdirent; struct getdents_callback64 buf; - int fput_needed; int error; if (!access_ok(VERIFY_WRITE, dirent, count)) return -EFAULT; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return -EBADF; buf.current_dir = dirent; @@ -290,17 +286,17 @@ SYSCALL_DEFINE3(getdents64, unsigned int, fd, buf.count = count; buf.error = 0; - error = vfs_readdir(file, filldir64, &buf); + error = vfs_readdir(f.file, filldir64, &buf); if (error >= 0) error = buf.error; lastdirent = buf.previous; if (lastdirent) { - typeof(lastdirent->d_off) d_off = file->f_pos; + typeof(lastdirent->d_off) d_off = f.file->f_pos; if (__put_user(d_off, &lastdirent->d_off)) error = -EFAULT; else error = count - buf.count; } - fput_light(file, fput_needed); + fdput(f); return error; } diff --git a/fs/reiserfs/super.c b/fs/reiserfs/super.c index 7a37dab..1078ae1 100644 --- a/fs/reiserfs/super.c +++ b/fs/reiserfs/super.c @@ -608,6 +608,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(reiserfs_inode_cachep); } diff --git a/fs/romfs/super.c b/fs/romfs/super.c index 77c5f21..fd7c5f6 100644 --- a/fs/romfs/super.c +++ b/fs/romfs/super.c @@ -648,6 +648,11 @@ error_register: static void __exit exit_romfs_fs(void) { unregister_filesystem(&romfs_fs_type); + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(romfs_inode_cachep); } diff --git a/fs/select.c b/fs/select.c index db14c78..2ef72d9 100644 --- a/fs/select.c +++ b/fs/select.c @@ -220,8 +220,7 @@ static void __pollwait(struct file *filp, wait_queue_head_t *wait_address, struct poll_table_entry *entry = poll_get_entry(pwq); if (!entry) return; - get_file(filp); - entry->filp = filp; + entry->filp = get_file(filp); entry->wait_address = wait_address; entry->key = p->_key; init_waitqueue_func_entry(&entry->wait, pollwake); @@ -429,8 +428,6 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time) for (i = 0; i < n; ++rinp, ++routp, ++rexp) { unsigned long in, out, ex, all_bits, bit = 1, mask, j; unsigned long res_in = 0, res_out = 0, res_ex = 0; - const struct file_operations *f_op = NULL; - struct file *file = NULL; in = *inp++; out = *outp++; ex = *exp++; all_bits = in | out | ex; @@ -440,20 +437,21 @@ int do_select(int n, fd_set_bits *fds, struct timespec *end_time) } for (j = 0; j < BITS_PER_LONG; ++j, ++i, bit <<= 1) { - int fput_needed; + struct fd f; if (i >= n) break; if (!(bit & all_bits)) continue; - file = fget_light(i, &fput_needed); - if (file) { - f_op = file->f_op; + f = fdget(i); + if (f.file) { + const struct file_operations *f_op; + f_op = f.file->f_op; mask = DEFAULT_POLLMASK; if (f_op && f_op->poll) { wait_key_set(wait, in, out, bit); - mask = (*f_op->poll)(file, wait); + mask = (*f_op->poll)(f.file, wait); } - fput_light(file, fput_needed); + fdput(f); if ((mask & POLLIN_SET) && (in & bit)) { res_in |= bit; retval++; @@ -726,20 +724,17 @@ static inline unsigned int do_pollfd(struct pollfd *pollfd, poll_table *pwait) mask = 0; fd = pollfd->fd; if (fd >= 0) { - int fput_needed; - struct file * file; - - file = fget_light(fd, &fput_needed); + struct fd f = fdget(fd); mask = POLLNVAL; - if (file != NULL) { + if (f.file) { mask = DEFAULT_POLLMASK; - if (file->f_op && file->f_op->poll) { + if (f.file->f_op && f.file->f_op->poll) { pwait->_key = pollfd->events|POLLERR|POLLHUP; - mask = file->f_op->poll(file, pwait); + mask = f.file->f_op->poll(f.file, pwait); } /* Mask out unneeded events. */ mask &= pollfd->events | POLLERR | POLLHUP; - fput_light(file, fput_needed); + fdput(f); } } pollfd->revents = mask; diff --git a/fs/signalfd.c b/fs/signalfd.c index 9f35a37..8bee4e5 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -269,13 +269,12 @@ SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask, if (ufd < 0) kfree(ctx); } else { - int fput_needed; - struct file *file = fget_light(ufd, &fput_needed); - if (!file) + struct fd f = fdget(ufd); + if (!f.file) return -EBADF; - ctx = file->private_data; - if (file->f_op != &signalfd_fops) { - fput_light(file, fput_needed); + ctx = f.file->private_data; + if (f.file->f_op != &signalfd_fops) { + fdput(f); return -EINVAL; } spin_lock_irq(¤t->sighand->siglock); @@ -283,7 +282,7 @@ SYSCALL_DEFINE4(signalfd4, int, ufd, sigset_t __user *, user_mask, spin_unlock_irq(¤t->sighand->siglock); wake_up(¤t->sighand->signalfd_wqh); - fput_light(file, fput_needed); + fdput(f); } return ufd; diff --git a/fs/splice.c b/fs/splice.c index 41514dd..13e5b47 100644 --- a/fs/splice.c +++ b/fs/splice.c @@ -1666,9 +1666,8 @@ static long vmsplice_to_pipe(struct file *file, const struct iovec __user *iov, SYSCALL_DEFINE4(vmsplice, int, fd, const struct iovec __user *, iov, unsigned long, nr_segs, unsigned int, flags) { - struct file *file; + struct fd f; long error; - int fput; if (unlikely(nr_segs > UIO_MAXIOV)) return -EINVAL; @@ -1676,14 +1675,14 @@ SYSCALL_DEFINE4(vmsplice, int, fd, const struct iovec __user *, iov, return 0; error = -EBADF; - file = fget_light(fd, &fput); - if (file) { - if (file->f_mode & FMODE_WRITE) - error = vmsplice_to_pipe(file, iov, nr_segs, flags); - else if (file->f_mode & FMODE_READ) - error = vmsplice_to_user(file, iov, nr_segs, flags); - - fput_light(file, fput); + f = fdget(fd); + if (f.file) { + if (f.file->f_mode & FMODE_WRITE) + error = vmsplice_to_pipe(f.file, iov, nr_segs, flags); + else if (f.file->f_mode & FMODE_READ) + error = vmsplice_to_user(f.file, iov, nr_segs, flags); + + fdput(f); } return error; @@ -1693,30 +1692,27 @@ SYSCALL_DEFINE6(splice, int, fd_in, loff_t __user *, off_in, int, fd_out, loff_t __user *, off_out, size_t, len, unsigned int, flags) { + struct fd in, out; long error; - struct file *in, *out; - int fput_in, fput_out; if (unlikely(!len)) return 0; error = -EBADF; - in = fget_light(fd_in, &fput_in); - if (in) { - if (in->f_mode & FMODE_READ) { - out = fget_light(fd_out, &fput_out); - if (out) { - if (out->f_mode & FMODE_WRITE) - error = do_splice(in, off_in, - out, off_out, + in = fdget(fd_in); + if (in.file) { + if (in.file->f_mode & FMODE_READ) { + out = fdget(fd_out); + if (out.file) { + if (out.file->f_mode & FMODE_WRITE) + error = do_splice(in.file, off_in, + out.file, off_out, len, flags); - fput_light(out, fput_out); + fdput(out); } } - - fput_light(in, fput_in); + fdput(in); } - return error; } @@ -2027,26 +2023,25 @@ static long do_tee(struct file *in, struct file *out, size_t len, SYSCALL_DEFINE4(tee, int, fdin, int, fdout, size_t, len, unsigned int, flags) { - struct file *in; - int error, fput_in; + struct fd in; + int error; if (unlikely(!len)) return 0; error = -EBADF; - in = fget_light(fdin, &fput_in); - if (in) { - if (in->f_mode & FMODE_READ) { - int fput_out; - struct file *out = fget_light(fdout, &fput_out); - - if (out) { - if (out->f_mode & FMODE_WRITE) - error = do_tee(in, out, len, flags); - fput_light(out, fput_out); + in = fdget(fdin); + if (in.file) { + if (in.file->f_mode & FMODE_READ) { + struct fd out = fdget(fdout); + if (out.file) { + if (out.file->f_mode & FMODE_WRITE) + error = do_tee(in.file, out.file, + len, flags); + fdput(out); } } - fput_light(in, fput_in); + fdput(in); } return error; diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index 29cd014..260e392 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -425,6 +425,11 @@ static int __init init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(squashfs_inode_cachep); } @@ -57,13 +57,13 @@ EXPORT_SYMBOL(vfs_getattr); int vfs_fstat(unsigned int fd, struct kstat *stat) { - int fput_needed; - struct file *f = fget_raw_light(fd, &fput_needed); + struct fd f = fdget_raw(fd); int error = -EBADF; - if (f) { - error = vfs_getattr(f->f_path.mnt, f->f_path.dentry, stat); - fput_light(f, fput_needed); + if (f.file) { + error = vfs_getattr(f.file->f_path.mnt, f.file->f_path.dentry, + stat); + fdput(f); } return error; } diff --git a/fs/statfs.c b/fs/statfs.c index 95ad5c0..f8e832e 100644 --- a/fs/statfs.c +++ b/fs/statfs.c @@ -87,12 +87,11 @@ int user_statfs(const char __user *pathname, struct kstatfs *st) int fd_statfs(int fd, struct kstatfs *st) { - int fput_needed; - struct file *file = fget_light(fd, &fput_needed); + struct fd f = fdget(fd); int error = -EBADF; - if (file) { - error = vfs_statfs(&file->f_path, st); - fput_light(file, fput_needed); + if (f.file) { + error = vfs_statfs(&f.file->f_path, st); + fdput(f); } return error; } @@ -307,12 +307,6 @@ void deactivate_locked_super(struct super_block *s) /* caches are now gone, we can safely kill the shrinker now */ unregister_shrinker(&s->s_shrink); - - /* - * We need to call rcu_barrier so all the delayed rcu free - * inodes are flushed before we release the fs module. - */ - rcu_barrier(); put_filesystem(fs); put_super(s); } else { @@ -148,21 +148,19 @@ void emergency_sync(void) */ SYSCALL_DEFINE1(syncfs, int, fd) { - struct file *file; + struct fd f = fdget(fd); struct super_block *sb; int ret; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADF; - sb = file->f_dentry->d_sb; + sb = f.file->f_dentry->d_sb; down_read(&sb->s_umount); ret = sync_filesystem(sb); up_read(&sb->s_umount); - fput_light(file, fput_needed); + fdput(f); return ret; } @@ -201,14 +199,12 @@ EXPORT_SYMBOL(vfs_fsync); static int do_fsync(unsigned int fd, int datasync) { - struct file *file; + struct fd f = fdget(fd); int ret = -EBADF; - int fput_needed; - file = fget_light(fd, &fput_needed); - if (file) { - ret = vfs_fsync(file, datasync); - fput_light(file, fput_needed); + if (f.file) { + ret = vfs_fsync(f.file, datasync); + fdput(f); } return ret; } @@ -291,10 +287,9 @@ SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes, unsigned int flags) { int ret; - struct file *file; + struct fd f; struct address_space *mapping; loff_t endbyte; /* inclusive */ - int fput_needed; umode_t i_mode; ret = -EINVAL; @@ -333,17 +328,17 @@ SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes, endbyte--; /* inclusive */ ret = -EBADF; - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) goto out; - i_mode = file->f_path.dentry->d_inode->i_mode; + i_mode = f.file->f_path.dentry->d_inode->i_mode; ret = -ESPIPE; if (!S_ISREG(i_mode) && !S_ISBLK(i_mode) && !S_ISDIR(i_mode) && !S_ISLNK(i_mode)) goto out_put; - mapping = file->f_mapping; + mapping = f.file->f_mapping; if (!mapping) { ret = -EINVAL; goto out_put; @@ -366,7 +361,7 @@ SYSCALL_DEFINE(sync_file_range)(int fd, loff_t offset, loff_t nbytes, ret = filemap_fdatawait_range(mapping, offset, endbyte); out_put: - fput_light(file, fput_needed); + fdput(f); out: return ret; } diff --git a/fs/sysv/inode.c b/fs/sysv/inode.c index b23ab73..d33e506 100644 --- a/fs/sysv/inode.c +++ b/fs/sysv/inode.c @@ -360,5 +360,10 @@ int __init sysv_init_icache(void) void sysv_destroy_icache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(sysv_inode_cachep); } diff --git a/fs/timerfd.c b/fs/timerfd.c index dffeb37..d03822b 100644 --- a/fs/timerfd.c +++ b/fs/timerfd.c @@ -234,19 +234,17 @@ static const struct file_operations timerfd_fops = { .llseek = noop_llseek, }; -static struct file *timerfd_fget(int fd) +static int timerfd_fget(int fd, struct fd *p) { - struct file *file; - - file = fget(fd); - if (!file) - return ERR_PTR(-EBADF); - if (file->f_op != &timerfd_fops) { - fput(file); - return ERR_PTR(-EINVAL); + struct fd f = fdget(fd); + if (!f.file) + return -EBADF; + if (f.file->f_op != &timerfd_fops) { + fdput(f); + return -EINVAL; } - - return file; + *p = f; + return 0; } SYSCALL_DEFINE2(timerfd_create, int, clockid, int, flags) @@ -284,7 +282,7 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, const struct itimerspec __user *, utmr, struct itimerspec __user *, otmr) { - struct file *file; + struct fd f; struct timerfd_ctx *ctx; struct itimerspec ktmr, kotmr; int ret; @@ -297,10 +295,10 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, !timespec_valid(&ktmr.it_interval)) return -EINVAL; - file = timerfd_fget(ufd); - if (IS_ERR(file)) - return PTR_ERR(file); - ctx = file->private_data; + ret = timerfd_fget(ufd, &f); + if (ret) + return ret; + ctx = f.file->private_data; timerfd_setup_cancel(ctx, flags); @@ -334,7 +332,7 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, ret = timerfd_setup(ctx, flags, &ktmr); spin_unlock_irq(&ctx->wqh.lock); - fput(file); + fdput(f); if (otmr && copy_to_user(otmr, &kotmr, sizeof(kotmr))) return -EFAULT; @@ -343,14 +341,13 @@ SYSCALL_DEFINE4(timerfd_settime, int, ufd, int, flags, SYSCALL_DEFINE2(timerfd_gettime, int, ufd, struct itimerspec __user *, otmr) { - struct file *file; + struct fd f; struct timerfd_ctx *ctx; struct itimerspec kotmr; - - file = timerfd_fget(ufd); - if (IS_ERR(file)) - return PTR_ERR(file); - ctx = file->private_data; + int ret = timerfd_fget(ufd, &f); + if (ret) + return ret; + ctx = f.file->private_data; spin_lock_irq(&ctx->wqh.lock); if (ctx->expired && ctx->tintv.tv64) { @@ -362,7 +359,7 @@ SYSCALL_DEFINE2(timerfd_gettime, int, ufd, struct itimerspec __user *, otmr) kotmr.it_value = ktime_to_timespec(timerfd_get_remaining(ctx)); kotmr.it_interval = ktime_to_timespec(ctx->tintv); spin_unlock_irq(&ctx->wqh.lock); - fput(file); + fdput(f); return copy_to_user(otmr, &kotmr, sizeof(kotmr)) ? -EFAULT: 0; } diff --git a/fs/ubifs/budget.c b/fs/ubifs/budget.c index 969489e..e8e01d7 100644 --- a/fs/ubifs/budget.c +++ b/fs/ubifs/budget.c @@ -342,9 +342,8 @@ static int do_budget_space(struct ubifs_info *c) lebs = c->lst.empty_lebs + c->freeable_cnt + c->idx_gc_cnt - c->lst.taken_empty_lebs; if (unlikely(rsvd_idx_lebs > lebs)) { - dbg_budg("out of indexing space: min_idx_lebs %d (old %d), " - "rsvd_idx_lebs %d", min_idx_lebs, c->bi.min_idx_lebs, - rsvd_idx_lebs); + dbg_budg("out of indexing space: min_idx_lebs %d (old %d), rsvd_idx_lebs %d", + min_idx_lebs, c->bi.min_idx_lebs, rsvd_idx_lebs); return -ENOSPC; } diff --git a/fs/ubifs/commit.c b/fs/ubifs/commit.c index 8eda717..ff822934 100644 --- a/fs/ubifs/commit.c +++ b/fs/ubifs/commit.c @@ -293,8 +293,8 @@ int ubifs_bg_thread(void *info) int err; struct ubifs_info *c = info; - dbg_msg("background thread \"%s\" started, PID %d", - c->bgt_name, current->pid); + ubifs_msg("background thread \"%s\" started, PID %d", + c->bgt_name, current->pid); set_freezable(); while (1) { @@ -328,7 +328,7 @@ int ubifs_bg_thread(void *info) cond_resched(); } - dbg_msg("background thread \"%s\" stops", c->bgt_name); + ubifs_msg("background thread \"%s\" stops", c->bgt_name); return 0; } @@ -514,7 +514,7 @@ struct idx_node { struct list_head list; int iip; union ubifs_key upper_key; - struct ubifs_idx_node idx __attribute__((aligned(8))); + struct ubifs_idx_node idx __aligned(8); }; /** diff --git a/fs/ubifs/compress.c b/fs/ubifs/compress.c index 11e4132..2bfa095 100644 --- a/fs/ubifs/compress.c +++ b/fs/ubifs/compress.c @@ -112,8 +112,7 @@ void ubifs_compress(const void *in_buf, int in_len, void *out_buf, int *out_len, if (compr->comp_mutex) mutex_unlock(compr->comp_mutex); if (unlikely(err)) { - ubifs_warn("cannot compress %d bytes, compressor %s, " - "error %d, leave data uncompressed", + ubifs_warn("cannot compress %d bytes, compressor %s, error %d, leave data uncompressed", in_len, compr->name, err); goto no_compr; } @@ -176,8 +175,8 @@ int ubifs_decompress(const void *in_buf, int in_len, void *out_buf, if (compr->decomp_mutex) mutex_unlock(compr->decomp_mutex); if (err) - ubifs_err("cannot decompress %d bytes, compressor %s, " - "error %d", in_len, compr->name, err); + ubifs_err("cannot decompress %d bytes, compressor %s, error %d", + in_len, compr->name, err); return err; } diff --git a/fs/ubifs/debug.c b/fs/ubifs/debug.c index 340d1af..6291163 100644 --- a/fs/ubifs/debug.c +++ b/fs/ubifs/debug.c @@ -219,15 +219,15 @@ const char *dbg_jhead(int jhead) static void dump_ch(const struct ubifs_ch *ch) { - printk(KERN_ERR "\tmagic %#x\n", le32_to_cpu(ch->magic)); - printk(KERN_ERR "\tcrc %#x\n", le32_to_cpu(ch->crc)); - printk(KERN_ERR "\tnode_type %d (%s)\n", ch->node_type, + pr_err("\tmagic %#x\n", le32_to_cpu(ch->magic)); + pr_err("\tcrc %#x\n", le32_to_cpu(ch->crc)); + pr_err("\tnode_type %d (%s)\n", ch->node_type, dbg_ntype(ch->node_type)); - printk(KERN_ERR "\tgroup_type %d (%s)\n", ch->group_type, + pr_err("\tgroup_type %d (%s)\n", ch->group_type, dbg_gtype(ch->group_type)); - printk(KERN_ERR "\tsqnum %llu\n", + pr_err("\tsqnum %llu\n", (unsigned long long)le64_to_cpu(ch->sqnum)); - printk(KERN_ERR "\tlen %u\n", le32_to_cpu(ch->len)); + pr_err("\tlen %u\n", le32_to_cpu(ch->len)); } void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode) @@ -238,43 +238,43 @@ void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode) struct ubifs_dent_node *dent, *pdent = NULL; int count = 2; - printk(KERN_ERR "Dump in-memory inode:"); - printk(KERN_ERR "\tinode %lu\n", inode->i_ino); - printk(KERN_ERR "\tsize %llu\n", + pr_err("Dump in-memory inode:"); + pr_err("\tinode %lu\n", inode->i_ino); + pr_err("\tsize %llu\n", (unsigned long long)i_size_read(inode)); - printk(KERN_ERR "\tnlink %u\n", inode->i_nlink); - printk(KERN_ERR "\tuid %u\n", (unsigned int)i_uid_read(inode)); - printk(KERN_ERR "\tgid %u\n", (unsigned int)i_gid_read(inode)); - printk(KERN_ERR "\tatime %u.%u\n", + pr_err("\tnlink %u\n", inode->i_nlink); + pr_err("\tuid %u\n", (unsigned int)i_uid_read(inode)); + pr_err("\tgid %u\n", (unsigned int)i_gid_read(inode)); + pr_err("\tatime %u.%u\n", (unsigned int)inode->i_atime.tv_sec, (unsigned int)inode->i_atime.tv_nsec); - printk(KERN_ERR "\tmtime %u.%u\n", + pr_err("\tmtime %u.%u\n", (unsigned int)inode->i_mtime.tv_sec, (unsigned int)inode->i_mtime.tv_nsec); - printk(KERN_ERR "\tctime %u.%u\n", + pr_err("\tctime %u.%u\n", (unsigned int)inode->i_ctime.tv_sec, (unsigned int)inode->i_ctime.tv_nsec); - printk(KERN_ERR "\tcreat_sqnum %llu\n", ui->creat_sqnum); - printk(KERN_ERR "\txattr_size %u\n", ui->xattr_size); - printk(KERN_ERR "\txattr_cnt %u\n", ui->xattr_cnt); - printk(KERN_ERR "\txattr_names %u\n", ui->xattr_names); - printk(KERN_ERR "\tdirty %u\n", ui->dirty); - printk(KERN_ERR "\txattr %u\n", ui->xattr); - printk(KERN_ERR "\tbulk_read %u\n", ui->xattr); - printk(KERN_ERR "\tsynced_i_size %llu\n", + pr_err("\tcreat_sqnum %llu\n", ui->creat_sqnum); + pr_err("\txattr_size %u\n", ui->xattr_size); + pr_err("\txattr_cnt %u\n", ui->xattr_cnt); + pr_err("\txattr_names %u\n", ui->xattr_names); + pr_err("\tdirty %u\n", ui->dirty); + pr_err("\txattr %u\n", ui->xattr); + pr_err("\tbulk_read %u\n", ui->xattr); + pr_err("\tsynced_i_size %llu\n", (unsigned long long)ui->synced_i_size); - printk(KERN_ERR "\tui_size %llu\n", + pr_err("\tui_size %llu\n", (unsigned long long)ui->ui_size); - printk(KERN_ERR "\tflags %d\n", ui->flags); - printk(KERN_ERR "\tcompr_type %d\n", ui->compr_type); - printk(KERN_ERR "\tlast_page_read %lu\n", ui->last_page_read); - printk(KERN_ERR "\tread_in_a_row %lu\n", ui->read_in_a_row); - printk(KERN_ERR "\tdata_len %d\n", ui->data_len); + pr_err("\tflags %d\n", ui->flags); + pr_err("\tcompr_type %d\n", ui->compr_type); + pr_err("\tlast_page_read %lu\n", ui->last_page_read); + pr_err("\tread_in_a_row %lu\n", ui->read_in_a_row); + pr_err("\tdata_len %d\n", ui->data_len); if (!S_ISDIR(inode->i_mode)) return; - printk(KERN_ERR "List of directory entries:\n"); + pr_err("List of directory entries:\n"); ubifs_assert(!mutex_is_locked(&c->tnc_mutex)); lowest_dent_key(c, &key, inode->i_ino); @@ -282,11 +282,11 @@ void ubifs_dump_inode(struct ubifs_info *c, const struct inode *inode) dent = ubifs_tnc_next_ent(c, &key, &nm); if (IS_ERR(dent)) { if (PTR_ERR(dent) != -ENOENT) - printk(KERN_ERR "error %ld\n", PTR_ERR(dent)); + pr_err("error %ld\n", PTR_ERR(dent)); break; } - printk(KERN_ERR "\t%d: %s (%s)\n", + pr_err("\t%d: %s (%s)\n", count++, dent->name, get_dent_type(dent->type)); nm.name = dent->name; @@ -305,12 +305,9 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) const struct ubifs_ch *ch = node; char key_buf[DBG_KEY_BUF_LEN]; - if (dbg_is_tst_rcvry(c)) - return; - /* If the magic is incorrect, just hexdump the first bytes */ if (le32_to_cpu(ch->magic) != UBIFS_NODE_MAGIC) { - printk(KERN_ERR "Not a node, first %zu bytes:", UBIFS_CH_SZ); + pr_err("Not a node, first %zu bytes:", UBIFS_CH_SZ); print_hex_dump(KERN_ERR, "", DUMP_PREFIX_OFFSET, 32, 1, (void *)node, UBIFS_CH_SZ, 1); return; @@ -324,8 +321,7 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) { const struct ubifs_pad_node *pad = node; - printk(KERN_ERR "\tpad_len %u\n", - le32_to_cpu(pad->pad_len)); + pr_err("\tpad_len %u\n", le32_to_cpu(pad->pad_len)); break; } case UBIFS_SB_NODE: @@ -333,112 +329,77 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) const struct ubifs_sb_node *sup = node; unsigned int sup_flags = le32_to_cpu(sup->flags); - printk(KERN_ERR "\tkey_hash %d (%s)\n", + pr_err("\tkey_hash %d (%s)\n", (int)sup->key_hash, get_key_hash(sup->key_hash)); - printk(KERN_ERR "\tkey_fmt %d (%s)\n", + pr_err("\tkey_fmt %d (%s)\n", (int)sup->key_fmt, get_key_fmt(sup->key_fmt)); - printk(KERN_ERR "\tflags %#x\n", sup_flags); - printk(KERN_ERR "\t big_lpt %u\n", + pr_err("\tflags %#x\n", sup_flags); + pr_err("\t big_lpt %u\n", !!(sup_flags & UBIFS_FLG_BIGLPT)); - printk(KERN_ERR "\t space_fixup %u\n", + pr_err("\t space_fixup %u\n", !!(sup_flags & UBIFS_FLG_SPACE_FIXUP)); - printk(KERN_ERR "\tmin_io_size %u\n", - le32_to_cpu(sup->min_io_size)); - printk(KERN_ERR "\tleb_size %u\n", - le32_to_cpu(sup->leb_size)); - printk(KERN_ERR "\tleb_cnt %u\n", - le32_to_cpu(sup->leb_cnt)); - printk(KERN_ERR "\tmax_leb_cnt %u\n", - le32_to_cpu(sup->max_leb_cnt)); - printk(KERN_ERR "\tmax_bud_bytes %llu\n", + pr_err("\tmin_io_size %u\n", le32_to_cpu(sup->min_io_size)); + pr_err("\tleb_size %u\n", le32_to_cpu(sup->leb_size)); + pr_err("\tleb_cnt %u\n", le32_to_cpu(sup->leb_cnt)); + pr_err("\tmax_leb_cnt %u\n", le32_to_cpu(sup->max_leb_cnt)); + pr_err("\tmax_bud_bytes %llu\n", (unsigned long long)le64_to_cpu(sup->max_bud_bytes)); - printk(KERN_ERR "\tlog_lebs %u\n", - le32_to_cpu(sup->log_lebs)); - printk(KERN_ERR "\tlpt_lebs %u\n", - le32_to_cpu(sup->lpt_lebs)); - printk(KERN_ERR "\torph_lebs %u\n", - le32_to_cpu(sup->orph_lebs)); - printk(KERN_ERR "\tjhead_cnt %u\n", - le32_to_cpu(sup->jhead_cnt)); - printk(KERN_ERR "\tfanout %u\n", - le32_to_cpu(sup->fanout)); - printk(KERN_ERR "\tlsave_cnt %u\n", - le32_to_cpu(sup->lsave_cnt)); - printk(KERN_ERR "\tdefault_compr %u\n", + pr_err("\tlog_lebs %u\n", le32_to_cpu(sup->log_lebs)); + pr_err("\tlpt_lebs %u\n", le32_to_cpu(sup->lpt_lebs)); + pr_err("\torph_lebs %u\n", le32_to_cpu(sup->orph_lebs)); + pr_err("\tjhead_cnt %u\n", le32_to_cpu(sup->jhead_cnt)); + pr_err("\tfanout %u\n", le32_to_cpu(sup->fanout)); + pr_err("\tlsave_cnt %u\n", le32_to_cpu(sup->lsave_cnt)); + pr_err("\tdefault_compr %u\n", (int)le16_to_cpu(sup->default_compr)); - printk(KERN_ERR "\trp_size %llu\n", + pr_err("\trp_size %llu\n", (unsigned long long)le64_to_cpu(sup->rp_size)); - printk(KERN_ERR "\trp_uid %u\n", - le32_to_cpu(sup->rp_uid)); - printk(KERN_ERR "\trp_gid %u\n", - le32_to_cpu(sup->rp_gid)); - printk(KERN_ERR "\tfmt_version %u\n", - le32_to_cpu(sup->fmt_version)); - printk(KERN_ERR "\ttime_gran %u\n", - le32_to_cpu(sup->time_gran)); - printk(KERN_ERR "\tUUID %pUB\n", - sup->uuid); + pr_err("\trp_uid %u\n", le32_to_cpu(sup->rp_uid)); + pr_err("\trp_gid %u\n", le32_to_cpu(sup->rp_gid)); + pr_err("\tfmt_version %u\n", le32_to_cpu(sup->fmt_version)); + pr_err("\ttime_gran %u\n", le32_to_cpu(sup->time_gran)); + pr_err("\tUUID %pUB\n", sup->uuid); break; } case UBIFS_MST_NODE: { const struct ubifs_mst_node *mst = node; - printk(KERN_ERR "\thighest_inum %llu\n", + pr_err("\thighest_inum %llu\n", (unsigned long long)le64_to_cpu(mst->highest_inum)); - printk(KERN_ERR "\tcommit number %llu\n", + pr_err("\tcommit number %llu\n", (unsigned long long)le64_to_cpu(mst->cmt_no)); - printk(KERN_ERR "\tflags %#x\n", - le32_to_cpu(mst->flags)); - printk(KERN_ERR "\tlog_lnum %u\n", - le32_to_cpu(mst->log_lnum)); - printk(KERN_ERR "\troot_lnum %u\n", - le32_to_cpu(mst->root_lnum)); - printk(KERN_ERR "\troot_offs %u\n", - le32_to_cpu(mst->root_offs)); - printk(KERN_ERR "\troot_len %u\n", - le32_to_cpu(mst->root_len)); - printk(KERN_ERR "\tgc_lnum %u\n", - le32_to_cpu(mst->gc_lnum)); - printk(KERN_ERR "\tihead_lnum %u\n", - le32_to_cpu(mst->ihead_lnum)); - printk(KERN_ERR "\tihead_offs %u\n", - le32_to_cpu(mst->ihead_offs)); - printk(KERN_ERR "\tindex_size %llu\n", + pr_err("\tflags %#x\n", le32_to_cpu(mst->flags)); + pr_err("\tlog_lnum %u\n", le32_to_cpu(mst->log_lnum)); + pr_err("\troot_lnum %u\n", le32_to_cpu(mst->root_lnum)); + pr_err("\troot_offs %u\n", le32_to_cpu(mst->root_offs)); + pr_err("\troot_len %u\n", le32_to_cpu(mst->root_len)); + pr_err("\tgc_lnum %u\n", le32_to_cpu(mst->gc_lnum)); + pr_err("\tihead_lnum %u\n", le32_to_cpu(mst->ihead_lnum)); + pr_err("\tihead_offs %u\n", le32_to_cpu(mst->ihead_offs)); + pr_err("\tindex_size %llu\n", (unsigned long long)le64_to_cpu(mst->index_size)); - printk(KERN_ERR "\tlpt_lnum %u\n", - le32_to_cpu(mst->lpt_lnum)); - printk(KERN_ERR "\tlpt_offs %u\n", - le32_to_cpu(mst->lpt_offs)); - printk(KERN_ERR "\tnhead_lnum %u\n", - le32_to_cpu(mst->nhead_lnum)); - printk(KERN_ERR "\tnhead_offs %u\n", - le32_to_cpu(mst->nhead_offs)); - printk(KERN_ERR "\tltab_lnum %u\n", - le32_to_cpu(mst->ltab_lnum)); - printk(KERN_ERR "\tltab_offs %u\n", - le32_to_cpu(mst->ltab_offs)); - printk(KERN_ERR "\tlsave_lnum %u\n", - le32_to_cpu(mst->lsave_lnum)); - printk(KERN_ERR "\tlsave_offs %u\n", - le32_to_cpu(mst->lsave_offs)); - printk(KERN_ERR "\tlscan_lnum %u\n", - le32_to_cpu(mst->lscan_lnum)); - printk(KERN_ERR "\tleb_cnt %u\n", - le32_to_cpu(mst->leb_cnt)); - printk(KERN_ERR "\tempty_lebs %u\n", - le32_to_cpu(mst->empty_lebs)); - printk(KERN_ERR "\tidx_lebs %u\n", - le32_to_cpu(mst->idx_lebs)); - printk(KERN_ERR "\ttotal_free %llu\n", + pr_err("\tlpt_lnum %u\n", le32_to_cpu(mst->lpt_lnum)); + pr_err("\tlpt_offs %u\n", le32_to_cpu(mst->lpt_offs)); + pr_err("\tnhead_lnum %u\n", le32_to_cpu(mst->nhead_lnum)); + pr_err("\tnhead_offs %u\n", le32_to_cpu(mst->nhead_offs)); + pr_err("\tltab_lnum %u\n", le32_to_cpu(mst->ltab_lnum)); + pr_err("\tltab_offs %u\n", le32_to_cpu(mst->ltab_offs)); + pr_err("\tlsave_lnum %u\n", le32_to_cpu(mst->lsave_lnum)); + pr_err("\tlsave_offs %u\n", le32_to_cpu(mst->lsave_offs)); + pr_err("\tlscan_lnum %u\n", le32_to_cpu(mst->lscan_lnum)); + pr_err("\tleb_cnt %u\n", le32_to_cpu(mst->leb_cnt)); + pr_err("\tempty_lebs %u\n", le32_to_cpu(mst->empty_lebs)); + pr_err("\tidx_lebs %u\n", le32_to_cpu(mst->idx_lebs)); + pr_err("\ttotal_free %llu\n", (unsigned long long)le64_to_cpu(mst->total_free)); - printk(KERN_ERR "\ttotal_dirty %llu\n", + pr_err("\ttotal_dirty %llu\n", (unsigned long long)le64_to_cpu(mst->total_dirty)); - printk(KERN_ERR "\ttotal_used %llu\n", + pr_err("\ttotal_used %llu\n", (unsigned long long)le64_to_cpu(mst->total_used)); - printk(KERN_ERR "\ttotal_dead %llu\n", + pr_err("\ttotal_dead %llu\n", (unsigned long long)le64_to_cpu(mst->total_dead)); - printk(KERN_ERR "\ttotal_dark %llu\n", + pr_err("\ttotal_dark %llu\n", (unsigned long long)le64_to_cpu(mst->total_dark)); break; } @@ -446,12 +407,9 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) { const struct ubifs_ref_node *ref = node; - printk(KERN_ERR "\tlnum %u\n", - le32_to_cpu(ref->lnum)); - printk(KERN_ERR "\toffs %u\n", - le32_to_cpu(ref->offs)); - printk(KERN_ERR "\tjhead %u\n", - le32_to_cpu(ref->jhead)); + pr_err("\tlnum %u\n", le32_to_cpu(ref->lnum)); + pr_err("\toffs %u\n", le32_to_cpu(ref->offs)); + pr_err("\tjhead %u\n", le32_to_cpu(ref->jhead)); break; } case UBIFS_INO_NODE: @@ -459,41 +417,32 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) const struct ubifs_ino_node *ino = node; key_read(c, &ino->key, &key); - printk(KERN_ERR "\tkey %s\n", + pr_err("\tkey %s\n", dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); - printk(KERN_ERR "\tcreat_sqnum %llu\n", + pr_err("\tcreat_sqnum %llu\n", (unsigned long long)le64_to_cpu(ino->creat_sqnum)); - printk(KERN_ERR "\tsize %llu\n", + pr_err("\tsize %llu\n", (unsigned long long)le64_to_cpu(ino->size)); - printk(KERN_ERR "\tnlink %u\n", - le32_to_cpu(ino->nlink)); - printk(KERN_ERR "\tatime %lld.%u\n", + pr_err("\tnlink %u\n", le32_to_cpu(ino->nlink)); + pr_err("\tatime %lld.%u\n", (long long)le64_to_cpu(ino->atime_sec), le32_to_cpu(ino->atime_nsec)); - printk(KERN_ERR "\tmtime %lld.%u\n", + pr_err("\tmtime %lld.%u\n", (long long)le64_to_cpu(ino->mtime_sec), le32_to_cpu(ino->mtime_nsec)); - printk(KERN_ERR "\tctime %lld.%u\n", + pr_err("\tctime %lld.%u\n", (long long)le64_to_cpu(ino->ctime_sec), le32_to_cpu(ino->ctime_nsec)); - printk(KERN_ERR "\tuid %u\n", - le32_to_cpu(ino->uid)); - printk(KERN_ERR "\tgid %u\n", - le32_to_cpu(ino->gid)); - printk(KERN_ERR "\tmode %u\n", - le32_to_cpu(ino->mode)); - printk(KERN_ERR "\tflags %#x\n", - le32_to_cpu(ino->flags)); - printk(KERN_ERR "\txattr_cnt %u\n", - le32_to_cpu(ino->xattr_cnt)); - printk(KERN_ERR "\txattr_size %u\n", - le32_to_cpu(ino->xattr_size)); - printk(KERN_ERR "\txattr_names %u\n", - le32_to_cpu(ino->xattr_names)); - printk(KERN_ERR "\tcompr_type %#x\n", + pr_err("\tuid %u\n", le32_to_cpu(ino->uid)); + pr_err("\tgid %u\n", le32_to_cpu(ino->gid)); + pr_err("\tmode %u\n", le32_to_cpu(ino->mode)); + pr_err("\tflags %#x\n", le32_to_cpu(ino->flags)); + pr_err("\txattr_cnt %u\n", le32_to_cpu(ino->xattr_cnt)); + pr_err("\txattr_size %u\n", le32_to_cpu(ino->xattr_size)); + pr_err("\txattr_names %u\n", le32_to_cpu(ino->xattr_names)); + pr_err("\tcompr_type %#x\n", (int)le16_to_cpu(ino->compr_type)); - printk(KERN_ERR "\tdata len %u\n", - le32_to_cpu(ino->data_len)); + pr_err("\tdata len %u\n", le32_to_cpu(ino->data_len)); break; } case UBIFS_DENT_NODE: @@ -503,22 +452,21 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) int nlen = le16_to_cpu(dent->nlen); key_read(c, &dent->key, &key); - printk(KERN_ERR "\tkey %s\n", + pr_err("\tkey %s\n", dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); - printk(KERN_ERR "\tinum %llu\n", + pr_err("\tinum %llu\n", (unsigned long long)le64_to_cpu(dent->inum)); - printk(KERN_ERR "\ttype %d\n", (int)dent->type); - printk(KERN_ERR "\tnlen %d\n", nlen); - printk(KERN_ERR "\tname "); + pr_err("\ttype %d\n", (int)dent->type); + pr_err("\tnlen %d\n", nlen); + pr_err("\tname "); if (nlen > UBIFS_MAX_NLEN) - printk(KERN_ERR "(bad name length, not printing, " - "bad or corrupted node)"); + pr_err("(bad name length, not printing, bad or corrupted node)"); else { for (i = 0; i < nlen && dent->name[i]; i++) - printk(KERN_CONT "%c", dent->name[i]); + pr_cont("%c", dent->name[i]); } - printk(KERN_CONT "\n"); + pr_cont("\n"); break; } @@ -528,15 +476,13 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) int dlen = le32_to_cpu(ch->len) - UBIFS_DATA_NODE_SZ; key_read(c, &dn->key, &key); - printk(KERN_ERR "\tkey %s\n", + pr_err("\tkey %s\n", dbg_snprintf_key(c, &key, key_buf, DBG_KEY_BUF_LEN)); - printk(KERN_ERR "\tsize %u\n", - le32_to_cpu(dn->size)); - printk(KERN_ERR "\tcompr_typ %d\n", + pr_err("\tsize %u\n", le32_to_cpu(dn->size)); + pr_err("\tcompr_typ %d\n", (int)le16_to_cpu(dn->compr_type)); - printk(KERN_ERR "\tdata size %d\n", - dlen); - printk(KERN_ERR "\tdata:\n"); + pr_err("\tdata size %d\n", dlen); + pr_err("\tdata:\n"); print_hex_dump(KERN_ERR, "\t", DUMP_PREFIX_OFFSET, 32, 1, (void *)&dn->data, dlen, 0); break; @@ -545,11 +491,10 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) { const struct ubifs_trun_node *trun = node; - printk(KERN_ERR "\tinum %u\n", - le32_to_cpu(trun->inum)); - printk(KERN_ERR "\told_size %llu\n", + pr_err("\tinum %u\n", le32_to_cpu(trun->inum)); + pr_err("\told_size %llu\n", (unsigned long long)le64_to_cpu(trun->old_size)); - printk(KERN_ERR "\tnew_size %llu\n", + pr_err("\tnew_size %llu\n", (unsigned long long)le64_to_cpu(trun->new_size)); break; } @@ -558,17 +503,16 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) const struct ubifs_idx_node *idx = node; n = le16_to_cpu(idx->child_cnt); - printk(KERN_ERR "\tchild_cnt %d\n", n); - printk(KERN_ERR "\tlevel %d\n", - (int)le16_to_cpu(idx->level)); - printk(KERN_ERR "\tBranches:\n"); + pr_err("\tchild_cnt %d\n", n); + pr_err("\tlevel %d\n", (int)le16_to_cpu(idx->level)); + pr_err("\tBranches:\n"); for (i = 0; i < n && i < c->fanout - 1; i++) { const struct ubifs_branch *br; br = ubifs_idx_branch(c, idx, i); key_read(c, &br->key, &key); - printk(KERN_ERR "\t%d: LEB %d:%d len %d key %s\n", + pr_err("\t%d: LEB %d:%d len %d key %s\n", i, le32_to_cpu(br->lnum), le32_to_cpu(br->offs), le32_to_cpu(br->len), dbg_snprintf_key(c, &key, key_buf, @@ -582,20 +526,20 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) { const struct ubifs_orph_node *orph = node; - printk(KERN_ERR "\tcommit number %llu\n", + pr_err("\tcommit number %llu\n", (unsigned long long) le64_to_cpu(orph->cmt_no) & LLONG_MAX); - printk(KERN_ERR "\tlast node flag %llu\n", + pr_err("\tlast node flag %llu\n", (unsigned long long)(le64_to_cpu(orph->cmt_no)) >> 63); n = (le32_to_cpu(ch->len) - UBIFS_ORPH_NODE_SZ) >> 3; - printk(KERN_ERR "\t%d orphan inode numbers:\n", n); + pr_err("\t%d orphan inode numbers:\n", n); for (i = 0; i < n; i++) - printk(KERN_ERR "\t ino %llu\n", + pr_err("\t ino %llu\n", (unsigned long long)le64_to_cpu(orph->inos[i])); break; } default: - printk(KERN_ERR "node type %d was not recognized\n", + pr_err("node type %d was not recognized\n", (int)ch->node_type); } spin_unlock(&dbg_lock); @@ -604,16 +548,16 @@ void ubifs_dump_node(const struct ubifs_info *c, const void *node) void ubifs_dump_budget_req(const struct ubifs_budget_req *req) { spin_lock(&dbg_lock); - printk(KERN_ERR "Budgeting request: new_ino %d, dirtied_ino %d\n", + pr_err("Budgeting request: new_ino %d, dirtied_ino %d\n", req->new_ino, req->dirtied_ino); - printk(KERN_ERR "\tnew_ino_d %d, dirtied_ino_d %d\n", + pr_err("\tnew_ino_d %d, dirtied_ino_d %d\n", req->new_ino_d, req->dirtied_ino_d); - printk(KERN_ERR "\tnew_page %d, dirtied_page %d\n", + pr_err("\tnew_page %d, dirtied_page %d\n", req->new_page, req->dirtied_page); - printk(KERN_ERR "\tnew_dent %d, mod_dent %d\n", + pr_err("\tnew_dent %d, mod_dent %d\n", req->new_dent, req->mod_dent); - printk(KERN_ERR "\tidx_growth %d\n", req->idx_growth); - printk(KERN_ERR "\tdata_growth %d dd_growth %d\n", + pr_err("\tidx_growth %d\n", req->idx_growth); + pr_err("\tdata_growth %d dd_growth %d\n", req->data_growth, req->dd_growth); spin_unlock(&dbg_lock); } @@ -621,14 +565,12 @@ void ubifs_dump_budget_req(const struct ubifs_budget_req *req) void ubifs_dump_lstats(const struct ubifs_lp_stats *lst) { spin_lock(&dbg_lock); - printk(KERN_ERR "(pid %d) Lprops statistics: empty_lebs %d, " - "idx_lebs %d\n", current->pid, lst->empty_lebs, lst->idx_lebs); - printk(KERN_ERR "\ttaken_empty_lebs %d, total_free %lld, " - "total_dirty %lld\n", lst->taken_empty_lebs, lst->total_free, - lst->total_dirty); - printk(KERN_ERR "\ttotal_used %lld, total_dark %lld, " - "total_dead %lld\n", lst->total_used, lst->total_dark, - lst->total_dead); + pr_err("(pid %d) Lprops statistics: empty_lebs %d, idx_lebs %d\n", + current->pid, lst->empty_lebs, lst->idx_lebs); + pr_err("\ttaken_empty_lebs %d, total_free %lld, total_dirty %lld\n", + lst->taken_empty_lebs, lst->total_free, lst->total_dirty); + pr_err("\ttotal_used %lld, total_dark %lld, total_dead %lld\n", + lst->total_used, lst->total_dark, lst->total_dead); spin_unlock(&dbg_lock); } @@ -642,21 +584,17 @@ void ubifs_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi) spin_lock(&c->space_lock); spin_lock(&dbg_lock); - printk(KERN_ERR "(pid %d) Budgeting info: data budget sum %lld, " - "total budget sum %lld\n", current->pid, - bi->data_growth + bi->dd_growth, + pr_err("(pid %d) Budgeting info: data budget sum %lld, total budget sum %lld\n", + current->pid, bi->data_growth + bi->dd_growth, bi->data_growth + bi->dd_growth + bi->idx_growth); - printk(KERN_ERR "\tbudg_data_growth %lld, budg_dd_growth %lld, " - "budg_idx_growth %lld\n", bi->data_growth, bi->dd_growth, - bi->idx_growth); - printk(KERN_ERR "\tmin_idx_lebs %d, old_idx_sz %llu, " - "uncommitted_idx %lld\n", bi->min_idx_lebs, bi->old_idx_sz, - bi->uncommitted_idx); - printk(KERN_ERR "\tpage_budget %d, inode_budget %d, dent_budget %d\n", + pr_err("\tbudg_data_growth %lld, budg_dd_growth %lld, budg_idx_growth %lld\n", + bi->data_growth, bi->dd_growth, bi->idx_growth); + pr_err("\tmin_idx_lebs %d, old_idx_sz %llu, uncommitted_idx %lld\n", + bi->min_idx_lebs, bi->old_idx_sz, bi->uncommitted_idx); + pr_err("\tpage_budget %d, inode_budget %d, dent_budget %d\n", bi->page_budget, bi->inode_budget, bi->dent_budget); - printk(KERN_ERR "\tnospace %u, nospace_rp %u\n", - bi->nospace, bi->nospace_rp); - printk(KERN_ERR "\tdark_wm %d, dead_wm %d, max_idx_node_sz %d\n", + pr_err("\tnospace %u, nospace_rp %u\n", bi->nospace, bi->nospace_rp); + pr_err("\tdark_wm %d, dead_wm %d, max_idx_node_sz %d\n", c->dark_wm, c->dead_wm, c->max_idx_node_sz); if (bi != &c->bi) @@ -667,38 +605,37 @@ void ubifs_dump_budg(struct ubifs_info *c, const struct ubifs_budg_info *bi) */ goto out_unlock; - printk(KERN_ERR "\tfreeable_cnt %d, calc_idx_sz %lld, idx_gc_cnt %d\n", + pr_err("\tfreeable_cnt %d, calc_idx_sz %lld, idx_gc_cnt %d\n", c->freeable_cnt, c->calc_idx_sz, c->idx_gc_cnt); - printk(KERN_ERR "\tdirty_pg_cnt %ld, dirty_zn_cnt %ld, " - "clean_zn_cnt %ld\n", atomic_long_read(&c->dirty_pg_cnt), + pr_err("\tdirty_pg_cnt %ld, dirty_zn_cnt %ld, clean_zn_cnt %ld\n", + atomic_long_read(&c->dirty_pg_cnt), atomic_long_read(&c->dirty_zn_cnt), atomic_long_read(&c->clean_zn_cnt)); - printk(KERN_ERR "\tgc_lnum %d, ihead_lnum %d\n", - c->gc_lnum, c->ihead_lnum); + pr_err("\tgc_lnum %d, ihead_lnum %d\n", c->gc_lnum, c->ihead_lnum); /* If we are in R/O mode, journal heads do not exist */ if (c->jheads) for (i = 0; i < c->jhead_cnt; i++) - printk(KERN_ERR "\tjhead %s\t LEB %d\n", + pr_err("\tjhead %s\t LEB %d\n", dbg_jhead(c->jheads[i].wbuf.jhead), c->jheads[i].wbuf.lnum); for (rb = rb_first(&c->buds); rb; rb = rb_next(rb)) { bud = rb_entry(rb, struct ubifs_bud, rb); - printk(KERN_ERR "\tbud LEB %d\n", bud->lnum); + pr_err("\tbud LEB %d\n", bud->lnum); } list_for_each_entry(bud, &c->old_buds, list) - printk(KERN_ERR "\told bud LEB %d\n", bud->lnum); + pr_err("\told bud LEB %d\n", bud->lnum); list_for_each_entry(idx_gc, &c->idx_gc, list) - printk(KERN_ERR "\tGC'ed idx LEB %d unmap %d\n", + pr_err("\tGC'ed idx LEB %d unmap %d\n", idx_gc->lnum, idx_gc->unmap); - printk(KERN_ERR "\tcommit state %d\n", c->cmt_state); + pr_err("\tcommit state %d\n", c->cmt_state); /* Print budgeting predictions */ available = ubifs_calc_available(c, c->bi.min_idx_lebs); outstanding = c->bi.data_growth + c->bi.dd_growth; free = ubifs_get_free_space_nolock(c); - printk(KERN_ERR "Budgeting predictions:\n"); - printk(KERN_ERR "\tavailable: %lld, outstanding %lld, free %lld\n", + pr_err("Budgeting predictions:\n"); + pr_err("\tavailable: %lld, outstanding %lld, free %lld\n", available, outstanding, free); out_unlock: spin_unlock(&dbg_lock); @@ -718,21 +655,19 @@ void ubifs_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp) dark = ubifs_calc_dark(c, spc); if (lp->flags & LPROPS_INDEX) - printk(KERN_ERR "LEB %-7d free %-8d dirty %-8d used %-8d " - "free + dirty %-8d flags %#x (", lp->lnum, lp->free, - lp->dirty, c->leb_size - spc, spc, lp->flags); + pr_err("LEB %-7d free %-8d dirty %-8d used %-8d free + dirty %-8d flags %#x (", + lp->lnum, lp->free, lp->dirty, c->leb_size - spc, spc, + lp->flags); else - printk(KERN_ERR "LEB %-7d free %-8d dirty %-8d used %-8d " - "free + dirty %-8d dark %-4d dead %-4d nodes fit %-3d " - "flags %#-4x (", lp->lnum, lp->free, lp->dirty, - c->leb_size - spc, spc, dark, dead, - (int)(spc / UBIFS_MAX_NODE_SZ), lp->flags); + pr_err("LEB %-7d free %-8d dirty %-8d used %-8d free + dirty %-8d dark %-4d dead %-4d nodes fit %-3d flags %#-4x (", + lp->lnum, lp->free, lp->dirty, c->leb_size - spc, spc, + dark, dead, (int)(spc / UBIFS_MAX_NODE_SZ), lp->flags); if (lp->flags & LPROPS_TAKEN) { if (lp->flags & LPROPS_INDEX) - printk(KERN_CONT "index, taken"); + pr_cont("index, taken"); else - printk(KERN_CONT "taken"); + pr_cont("taken"); } else { const char *s; @@ -769,7 +704,7 @@ void ubifs_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp) break; } } - printk(KERN_CONT "%s", s); + pr_cont("%s", s); } for (rb = rb_first((struct rb_root *)&c->buds); rb; rb = rb_next(rb)) { @@ -784,19 +719,18 @@ void ubifs_dump_lprop(const struct ubifs_info *c, const struct ubifs_lprops *lp) */ if (c->jheads && lp->lnum == c->jheads[i].wbuf.lnum) { - printk(KERN_CONT ", jhead %s", - dbg_jhead(i)); + pr_cont(", jhead %s", dbg_jhead(i)); head = 1; } } if (!head) - printk(KERN_CONT ", bud of jhead %s", + pr_cont(", bud of jhead %s", dbg_jhead(bud->jhead)); } } if (lp->lnum == c->gc_lnum) - printk(KERN_CONT ", GC LEB"); - printk(KERN_CONT ")\n"); + pr_cont(", GC LEB"); + pr_cont(")\n"); } void ubifs_dump_lprops(struct ubifs_info *c) @@ -805,8 +739,7 @@ void ubifs_dump_lprops(struct ubifs_info *c) struct ubifs_lprops lp; struct ubifs_lp_stats lst; - printk(KERN_ERR "(pid %d) start dumping LEB properties\n", - current->pid); + pr_err("(pid %d) start dumping LEB properties\n", current->pid); ubifs_get_lp_stats(c, &lst); ubifs_dump_lstats(&lst); @@ -817,8 +750,7 @@ void ubifs_dump_lprops(struct ubifs_info *c) ubifs_dump_lprop(c, &lp); } - printk(KERN_ERR "(pid %d) finish dumping LEB properties\n", - current->pid); + pr_err("(pid %d) finish dumping LEB properties\n", current->pid); } void ubifs_dump_lpt_info(struct ubifs_info *c) @@ -826,37 +758,36 @@ void ubifs_dump_lpt_info(struct ubifs_info *c) int i; spin_lock(&dbg_lock); - printk(KERN_ERR "(pid %d) dumping LPT information\n", current->pid); - printk(KERN_ERR "\tlpt_sz: %lld\n", c->lpt_sz); - printk(KERN_ERR "\tpnode_sz: %d\n", c->pnode_sz); - printk(KERN_ERR "\tnnode_sz: %d\n", c->nnode_sz); - printk(KERN_ERR "\tltab_sz: %d\n", c->ltab_sz); - printk(KERN_ERR "\tlsave_sz: %d\n", c->lsave_sz); - printk(KERN_ERR "\tbig_lpt: %d\n", c->big_lpt); - printk(KERN_ERR "\tlpt_hght: %d\n", c->lpt_hght); - printk(KERN_ERR "\tpnode_cnt: %d\n", c->pnode_cnt); - printk(KERN_ERR "\tnnode_cnt: %d\n", c->nnode_cnt); - printk(KERN_ERR "\tdirty_pn_cnt: %d\n", c->dirty_pn_cnt); - printk(KERN_ERR "\tdirty_nn_cnt: %d\n", c->dirty_nn_cnt); - printk(KERN_ERR "\tlsave_cnt: %d\n", c->lsave_cnt); - printk(KERN_ERR "\tspace_bits: %d\n", c->space_bits); - printk(KERN_ERR "\tlpt_lnum_bits: %d\n", c->lpt_lnum_bits); - printk(KERN_ERR "\tlpt_offs_bits: %d\n", c->lpt_offs_bits); - printk(KERN_ERR "\tlpt_spc_bits: %d\n", c->lpt_spc_bits); - printk(KERN_ERR "\tpcnt_bits: %d\n", c->pcnt_bits); - printk(KERN_ERR "\tlnum_bits: %d\n", c->lnum_bits); - printk(KERN_ERR "\tLPT root is at %d:%d\n", c->lpt_lnum, c->lpt_offs); - printk(KERN_ERR "\tLPT head is at %d:%d\n", + pr_err("(pid %d) dumping LPT information\n", current->pid); + pr_err("\tlpt_sz: %lld\n", c->lpt_sz); + pr_err("\tpnode_sz: %d\n", c->pnode_sz); + pr_err("\tnnode_sz: %d\n", c->nnode_sz); + pr_err("\tltab_sz: %d\n", c->ltab_sz); + pr_err("\tlsave_sz: %d\n", c->lsave_sz); + pr_err("\tbig_lpt: %d\n", c->big_lpt); + pr_err("\tlpt_hght: %d\n", c->lpt_hght); + pr_err("\tpnode_cnt: %d\n", c->pnode_cnt); + pr_err("\tnnode_cnt: %d\n", c->nnode_cnt); + pr_err("\tdirty_pn_cnt: %d\n", c->dirty_pn_cnt); + pr_err("\tdirty_nn_cnt: %d\n", c->dirty_nn_cnt); + pr_err("\tlsave_cnt: %d\n", c->lsave_cnt); + pr_err("\tspace_bits: %d\n", c->space_bits); + pr_err("\tlpt_lnum_bits: %d\n", c->lpt_lnum_bits); + pr_err("\tlpt_offs_bits: %d\n", c->lpt_offs_bits); + pr_err("\tlpt_spc_bits: %d\n", c->lpt_spc_bits); + pr_err("\tpcnt_bits: %d\n", c->pcnt_bits); + pr_err("\tlnum_bits: %d\n", c->lnum_bits); + pr_err("\tLPT root is at %d:%d\n", c->lpt_lnum, c->lpt_offs); + pr_err("\tLPT head is at %d:%d\n", c->nhead_lnum, c->nhead_offs); - printk(KERN_ERR "\tLPT ltab is at %d:%d\n", - c->ltab_lnum, c->ltab_offs); + pr_err("\tLPT ltab is at %d:%d\n", c->ltab_lnum, c->ltab_offs); if (c->big_lpt) - printk(KERN_ERR "\tLPT lsave is at %d:%d\n", + pr_err("\tLPT lsave is at %d:%d\n", c->lsave_lnum, c->lsave_offs); for (i = 0; i < c->lpt_lebs; i++) - printk(KERN_ERR "\tLPT LEB %d free %d dirty %d tgc %d " - "cmt %d\n", i + c->lpt_first, c->ltab[i].free, - c->ltab[i].dirty, c->ltab[i].tgc, c->ltab[i].cmt); + pr_err("\tLPT LEB %d free %d dirty %d tgc %d cmt %d\n", + i + c->lpt_first, c->ltab[i].free, c->ltab[i].dirty, + c->ltab[i].tgc, c->ltab[i].cmt); spin_unlock(&dbg_lock); } @@ -865,13 +796,13 @@ void ubifs_dump_sleb(const struct ubifs_info *c, { struct ubifs_scan_node *snod; - printk(KERN_ERR "(pid %d) start dumping scanned data from LEB %d:%d\n", + pr_err("(pid %d) start dumping scanned data from LEB %d:%d\n", current->pid, sleb->lnum, offs); list_for_each_entry(snod, &sleb->nodes, list) { cond_resched(); - printk(KERN_ERR "Dumping node at LEB %d:%d len %d\n", sleb->lnum, - snod->offs, snod->len); + pr_err("Dumping node at LEB %d:%d len %d\n", + sleb->lnum, snod->offs, snod->len); ubifs_dump_node(c, snod->node); } } @@ -882,11 +813,7 @@ void ubifs_dump_leb(const struct ubifs_info *c, int lnum) struct ubifs_scan_node *snod; void *buf; - if (dbg_is_tst_rcvry(c)) - return; - - printk(KERN_ERR "(pid %d) start dumping LEB %d\n", - current->pid, lnum); + pr_err("(pid %d) start dumping LEB %d\n", current->pid, lnum); buf = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL); if (!buf) { @@ -900,18 +827,17 @@ void ubifs_dump_leb(const struct ubifs_info *c, int lnum) goto out; } - printk(KERN_ERR "LEB %d has %d nodes ending at %d\n", lnum, + pr_err("LEB %d has %d nodes ending at %d\n", lnum, sleb->nodes_cnt, sleb->endpt); list_for_each_entry(snod, &sleb->nodes, list) { cond_resched(); - printk(KERN_ERR "Dumping node at LEB %d:%d len %d\n", lnum, + pr_err("Dumping node at LEB %d:%d len %d\n", lnum, snod->offs, snod->len); ubifs_dump_node(c, snod->node); } - printk(KERN_ERR "(pid %d) finish dumping LEB %d\n", - current->pid, lnum); + pr_err("(pid %d) finish dumping LEB %d\n", current->pid, lnum); ubifs_scan_destroy(sleb); out: @@ -932,33 +858,28 @@ void ubifs_dump_znode(const struct ubifs_info *c, else zbr = &c->zroot; - printk(KERN_ERR "znode %p, LEB %d:%d len %d parent %p iip %d level %d" - " child_cnt %d flags %lx\n", znode, zbr->lnum, zbr->offs, - zbr->len, znode->parent, znode->iip, znode->level, - znode->child_cnt, znode->flags); + pr_err("znode %p, LEB %d:%d len %d parent %p iip %d level %d child_cnt %d flags %lx\n", + znode, zbr->lnum, zbr->offs, zbr->len, znode->parent, znode->iip, + znode->level, znode->child_cnt, znode->flags); if (znode->child_cnt <= 0 || znode->child_cnt > c->fanout) { spin_unlock(&dbg_lock); return; } - printk(KERN_ERR "zbranches:\n"); + pr_err("zbranches:\n"); for (n = 0; n < znode->child_cnt; n++) { zbr = &znode->zbranch[n]; if (znode->level > 0) - printk(KERN_ERR "\t%d: znode %p LEB %d:%d len %d key " - "%s\n", n, zbr->znode, zbr->lnum, - zbr->offs, zbr->len, - dbg_snprintf_key(c, &zbr->key, - key_buf, - DBG_KEY_BUF_LEN)); + pr_err("\t%d: znode %p LEB %d:%d len %d key %s\n", + n, zbr->znode, zbr->lnum, zbr->offs, zbr->len, + dbg_snprintf_key(c, &zbr->key, key_buf, + DBG_KEY_BUF_LEN)); else - printk(KERN_ERR "\t%d: LNC %p LEB %d:%d len %d key " - "%s\n", n, zbr->znode, zbr->lnum, - zbr->offs, zbr->len, - dbg_snprintf_key(c, &zbr->key, - key_buf, - DBG_KEY_BUF_LEN)); + pr_err("\t%d: LNC %p LEB %d:%d len %d key %s\n", + n, zbr->znode, zbr->lnum, zbr->offs, zbr->len, + dbg_snprintf_key(c, &zbr->key, key_buf, + DBG_KEY_BUF_LEN)); } spin_unlock(&dbg_lock); } @@ -967,16 +888,16 @@ void ubifs_dump_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat) { int i; - printk(KERN_ERR "(pid %d) start dumping heap cat %d (%d elements)\n", + pr_err("(pid %d) start dumping heap cat %d (%d elements)\n", current->pid, cat, heap->cnt); for (i = 0; i < heap->cnt; i++) { struct ubifs_lprops *lprops = heap->arr[i]; - printk(KERN_ERR "\t%d. LEB %d hpos %d free %d dirty %d " - "flags %d\n", i, lprops->lnum, lprops->hpos, - lprops->free, lprops->dirty, lprops->flags); + pr_err("\t%d. LEB %d hpos %d free %d dirty %d flags %d\n", + i, lprops->lnum, lprops->hpos, lprops->free, + lprops->dirty, lprops->flags); } - printk(KERN_ERR "(pid %d) finish dumping heap\n", current->pid); + pr_err("(pid %d) finish dumping heap\n", current->pid); } void ubifs_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode, @@ -984,15 +905,15 @@ void ubifs_dump_pnode(struct ubifs_info *c, struct ubifs_pnode *pnode, { int i; - printk(KERN_ERR "(pid %d) dumping pnode:\n", current->pid); - printk(KERN_ERR "\taddress %zx parent %zx cnext %zx\n", + pr_err("(pid %d) dumping pnode:\n", current->pid); + pr_err("\taddress %zx parent %zx cnext %zx\n", (size_t)pnode, (size_t)parent, (size_t)pnode->cnext); - printk(KERN_ERR "\tflags %lu iip %d level %d num %d\n", + pr_err("\tflags %lu iip %d level %d num %d\n", pnode->flags, iip, pnode->level, pnode->num); for (i = 0; i < UBIFS_LPT_FANOUT; i++) { struct ubifs_lprops *lp = &pnode->lprops[i]; - printk(KERN_ERR "\t%d: free %d dirty %d flags %d lnum %d\n", + pr_err("\t%d: free %d dirty %d flags %d lnum %d\n", i, lp->free, lp->dirty, lp->flags, lp->lnum); } } @@ -1002,20 +923,20 @@ void ubifs_dump_tnc(struct ubifs_info *c) struct ubifs_znode *znode; int level; - printk(KERN_ERR "\n"); - printk(KERN_ERR "(pid %d) start dumping TNC tree\n", current->pid); + pr_err("\n"); + pr_err("(pid %d) start dumping TNC tree\n", current->pid); znode = ubifs_tnc_levelorder_next(c->zroot.znode, NULL); level = znode->level; - printk(KERN_ERR "== Level %d ==\n", level); + pr_err("== Level %d ==\n", level); while (znode) { if (level != znode->level) { level = znode->level; - printk(KERN_ERR "== Level %d ==\n", level); + pr_err("== Level %d ==\n", level); } ubifs_dump_znode(c, znode); znode = ubifs_tnc_levelorder_next(c->zroot.znode, znode); } - printk(KERN_ERR "(pid %d) finish dumping TNC tree\n", current->pid); + pr_err("(pid %d) finish dumping TNC tree\n", current->pid); } static int dump_znode(struct ubifs_info *c, struct ubifs_znode *znode, @@ -1154,8 +1075,8 @@ int dbg_check_synced_i_size(const struct ubifs_info *c, struct inode *inode) mutex_lock(&ui->ui_mutex); spin_lock(&ui->ui_lock); if (ui->ui_size != ui->synced_i_size && !ui->dirty) { - ubifs_err("ui_size is %lld, synced_i_size is %lld, but inode " - "is clean", ui->ui_size, ui->synced_i_size); + ubifs_err("ui_size is %lld, synced_i_size is %lld, but inode is clean", + ui->ui_size, ui->synced_i_size); ubifs_err("i_ino %lu, i_mode %#x, i_size %lld", inode->i_ino, inode->i_mode, i_size_read(inode)); dump_stack(); @@ -1217,17 +1138,16 @@ int dbg_check_dir(struct ubifs_info *c, const struct inode *dir) kfree(pdent); if (i_size_read(dir) != size) { - ubifs_err("directory inode %lu has size %llu, " - "but calculated size is %llu", dir->i_ino, - (unsigned long long)i_size_read(dir), + ubifs_err("directory inode %lu has size %llu, but calculated size is %llu", + dir->i_ino, (unsigned long long)i_size_read(dir), (unsigned long long)size); ubifs_dump_inode(c, dir); dump_stack(); return -EINVAL; } if (dir->i_nlink != nlink) { - ubifs_err("directory inode %lu has nlink %u, but calculated " - "nlink is %u", dir->i_ino, dir->i_nlink, nlink); + ubifs_err("directory inode %lu has nlink %u, but calculated nlink is %u", + dir->i_ino, dir->i_nlink, nlink); ubifs_dump_inode(c, dir); dump_stack(); return -EINVAL; @@ -1686,8 +1606,8 @@ int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb, if (znode_cb) { err = znode_cb(c, znode, priv); if (err) { - ubifs_err("znode checking function returned " - "error %d", err); + ubifs_err("znode checking function returned error %d", + err); ubifs_dump_znode(c, znode); goto out_dump; } @@ -1697,9 +1617,7 @@ int dbg_walk_index(struct ubifs_info *c, dbg_leaf_callback leaf_cb, zbr = &znode->zbranch[idx]; err = leaf_cb(c, zbr, priv); if (err) { - ubifs_err("leaf checking function " - "returned error %d, for leaf " - "at LEB %d:%d", + ubifs_err("leaf checking function returned error %d, for leaf at LEB %d:%d", err, zbr->lnum, zbr->offs); goto out_dump; } @@ -1807,8 +1725,8 @@ int dbg_check_idx_size(struct ubifs_info *c, long long idx_size) } if (calc != idx_size) { - ubifs_err("index size check failed: calculated size is %lld, " - "should be %lld", calc, idx_size); + ubifs_err("index size check failed: calculated size is %lld, should be %lld", + calc, idx_size); dump_stack(); return -EINVAL; } @@ -2120,8 +2038,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr, fscki = read_add_inode(c, priv, inum); if (IS_ERR(fscki)) { err = PTR_ERR(fscki); - ubifs_err("error %d while processing data node and " - "trying to find inode node %lu", + ubifs_err("error %d while processing data node and trying to find inode node %lu", err, (unsigned long)inum); goto out_dump; } @@ -2131,9 +2048,8 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr, blk_offs <<= UBIFS_BLOCK_SHIFT; blk_offs += le32_to_cpu(dn->size); if (blk_offs > fscki->size) { - ubifs_err("data node at LEB %d:%d is not within inode " - "size %lld", zbr->lnum, zbr->offs, - fscki->size); + ubifs_err("data node at LEB %d:%d is not within inode size %lld", + zbr->lnum, zbr->offs, fscki->size); err = -EINVAL; goto out_dump; } @@ -2154,8 +2070,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr, fscki = read_add_inode(c, priv, inum); if (IS_ERR(fscki)) { err = PTR_ERR(fscki); - ubifs_err("error %d while processing entry node and " - "trying to find inode node %lu", + ubifs_err("error %d while processing entry node and trying to find inode node %lu", err, (unsigned long)inum); goto out_dump; } @@ -2167,8 +2082,7 @@ static int check_leaf(struct ubifs_info *c, struct ubifs_zbranch *zbr, fscki1 = read_add_inode(c, priv, inum); if (IS_ERR(fscki1)) { err = PTR_ERR(fscki1); - ubifs_err("error %d while processing entry node and " - "trying to find parent inode node %lu", + ubifs_err("error %d while processing entry node and trying to find parent inode node %lu", err, (unsigned long)inum); goto out_dump; } @@ -2258,61 +2172,52 @@ static int check_inodes(struct ubifs_info *c, struct fsck_data *fsckd) */ if (fscki->inum != UBIFS_ROOT_INO && fscki->references != 1) { - ubifs_err("directory inode %lu has %d " - "direntries which refer it, but " - "should be 1", + ubifs_err("directory inode %lu has %d direntries which refer it, but should be 1", (unsigned long)fscki->inum, fscki->references); goto out_dump; } if (fscki->inum == UBIFS_ROOT_INO && fscki->references != 0) { - ubifs_err("root inode %lu has non-zero (%d) " - "direntries which refer it", + ubifs_err("root inode %lu has non-zero (%d) direntries which refer it", (unsigned long)fscki->inum, fscki->references); goto out_dump; } if (fscki->calc_sz != fscki->size) { - ubifs_err("directory inode %lu size is %lld, " - "but calculated size is %lld", + ubifs_err("directory inode %lu size is %lld, but calculated size is %lld", (unsigned long)fscki->inum, fscki->size, fscki->calc_sz); goto out_dump; } if (fscki->calc_cnt != fscki->nlink) { - ubifs_err("directory inode %lu nlink is %d, " - "but calculated nlink is %d", + ubifs_err("directory inode %lu nlink is %d, but calculated nlink is %d", (unsigned long)fscki->inum, fscki->nlink, fscki->calc_cnt); goto out_dump; } } else { if (fscki->references != fscki->nlink) { - ubifs_err("inode %lu nlink is %d, but " - "calculated nlink is %d", + ubifs_err("inode %lu nlink is %d, but calculated nlink is %d", (unsigned long)fscki->inum, fscki->nlink, fscki->references); goto out_dump; } } if (fscki->xattr_sz != fscki->calc_xsz) { - ubifs_err("inode %lu has xattr size %u, but " - "calculated size is %lld", + ubifs_err("inode %lu has xattr size %u, but calculated size is %lld", (unsigned long)fscki->inum, fscki->xattr_sz, fscki->calc_xsz); goto out_dump; } if (fscki->xattr_cnt != fscki->calc_xcnt) { - ubifs_err("inode %lu has %u xattrs, but " - "calculated count is %lld", + ubifs_err("inode %lu has %u xattrs, but calculated count is %lld", (unsigned long)fscki->inum, fscki->xattr_cnt, fscki->calc_xcnt); goto out_dump; } if (fscki->xattr_nms != fscki->calc_xnms) { - ubifs_err("inode %lu has xattr names' size %u, but " - "calculated names' size is %lld", + ubifs_err("inode %lu has xattr names' size %u, but calculated names' size is %lld", (unsigned long)fscki->inum, fscki->xattr_nms, fscki->calc_xnms); goto out_dump; @@ -2652,20 +2557,18 @@ static int power_cut_emulated(struct ubifs_info *c, int lnum, int write) return 1; } -static void cut_data(const void *buf, unsigned int len) +static int corrupt_data(const struct ubifs_info *c, const void *buf, + unsigned int len) { unsigned int from, to, i, ffs = chance(1, 2); unsigned char *p = (void *)buf; from = random32() % (len + 1); - if (chance(1, 2)) - to = random32() % (len - from + 1); - else - to = len; + /* Corruption may only span one max. write unit */ + to = min(len, ALIGN(from, c->max_write_size)); - if (from < to) - ubifs_warn("filled bytes %u-%u with %s", from, to - 1, - ffs ? "0xFFs" : "random data"); + ubifs_warn("filled bytes %u-%u with %s", from, to - 1, + ffs ? "0xFFs" : "random data"); if (ffs) for (i = from; i < to; i++) @@ -2673,6 +2576,8 @@ static void cut_data(const void *buf, unsigned int len) else for (i = from; i < to; i++) p[i] = random32() % 0x100; + + return to; } int dbg_leb_write(struct ubifs_info *c, int lnum, const void *buf, @@ -2685,7 +2590,9 @@ int dbg_leb_write(struct ubifs_info *c, int lnum, const void *buf, failing = power_cut_emulated(c, lnum, 1); if (failing) - cut_data(buf, len); + len = corrupt_data(c, buf, len); + ubifs_warn("actually write %d bytes to LEB %d:%d (the buffer was corrupted)", + len, lnum, offs); err = ubi_leb_write(c->ubi, lnum, buf, offs, len); if (err) return err; diff --git a/fs/ubifs/debug.h b/fs/ubifs/debug.h index 760de72..e03d517 100644 --- a/fs/ubifs/debug.h +++ b/fs/ubifs/debug.h @@ -150,7 +150,7 @@ struct ubifs_global_debug_info { #define ubifs_assert(expr) do { \ if (unlikely(!(expr))) { \ - printk(KERN_CRIT "UBIFS assert failed in %s at %u (pid %d)\n", \ + pr_crit("UBIFS assert failed in %s at %u (pid %d)\n", \ __func__, __LINE__, current->pid); \ dump_stack(); \ } \ @@ -159,26 +159,23 @@ struct ubifs_global_debug_info { #define ubifs_assert_cmt_locked(c) do { \ if (unlikely(down_write_trylock(&(c)->commit_sem))) { \ up_write(&(c)->commit_sem); \ - printk(KERN_CRIT "commit lock is not locked!\n"); \ + pr_crit("commit lock is not locked!\n"); \ ubifs_assert(0); \ } \ } while (0) #define ubifs_dbg_msg(type, fmt, ...) \ - pr_debug("UBIFS DBG " type ": " fmt "\n", ##__VA_ARGS__) + pr_debug("UBIFS DBG " type " (pid %d): " fmt "\n", current->pid, \ + ##__VA_ARGS__) #define DBG_KEY_BUF_LEN 48 #define ubifs_dbg_msg_key(type, key, fmt, ...) do { \ char __tmp_key_buf[DBG_KEY_BUF_LEN]; \ - pr_debug("UBIFS DBG " type ": " fmt "%s\n", ##__VA_ARGS__, \ + pr_debug("UBIFS DBG " type " (pid %d): " fmt "%s\n", current->pid, \ + ##__VA_ARGS__, \ dbg_snprintf_key(c, key, __tmp_key_buf, DBG_KEY_BUF_LEN)); \ } while (0) -/* Just a debugging messages not related to any specific UBIFS subsystem */ -#define dbg_msg(fmt, ...) \ - printk(KERN_DEBUG "UBIFS DBG (pid %d): %s: " fmt "\n", current->pid, \ - __func__, ##__VA_ARGS__) - /* General messages */ #define dbg_gen(fmt, ...) ubifs_dbg_msg("gen", fmt, ##__VA_ARGS__) /* Additional journal messages */ diff --git a/fs/ubifs/dir.c b/fs/ubifs/dir.c index c95681c..e271fba 100644 --- a/fs/ubifs/dir.c +++ b/fs/ubifs/dir.c @@ -980,8 +980,8 @@ static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry, * separately. */ - dbg_gen("dent '%.*s' ino %lu in dir ino %lu to dent '%.*s' in " - "dir ino %lu", old_dentry->d_name.len, old_dentry->d_name.name, + dbg_gen("dent '%.*s' ino %lu in dir ino %lu to dent '%.*s' in dir ino %lu", + old_dentry->d_name.len, old_dentry->d_name.name, old_inode->i_ino, old_dir->i_ino, new_dentry->d_name.len, new_dentry->d_name.name, new_dir->i_ino); ubifs_assert(mutex_is_locked(&old_dir->i_mutex)); diff --git a/fs/ubifs/file.c b/fs/ubifs/file.c index 7bd6e72..ff48c5a 100644 --- a/fs/ubifs/file.c +++ b/fs/ubifs/file.c @@ -1486,8 +1486,8 @@ static int ubifs_vm_page_mkwrite(struct vm_area_struct *vma, err = ubifs_budget_space(c, &req); if (unlikely(err)) { if (err == -ENOSPC) - ubifs_warn("out of space for mmapped file " - "(inode number %lu)", inode->i_ino); + ubifs_warn("out of space for mmapped file (inode number %lu)", + inode->i_ino); return VM_FAULT_SIGBUS; } diff --git a/fs/ubifs/gc.c b/fs/ubifs/gc.c index 04dd6f4..76ca53c 100644 --- a/fs/ubifs/gc.c +++ b/fs/ubifs/gc.c @@ -714,9 +714,9 @@ int ubifs_garbage_collect(struct ubifs_info *c, int anyway) break; } - dbg_gc("found LEB %d: free %d, dirty %d, sum %d " - "(min. space %d)", lp.lnum, lp.free, lp.dirty, - lp.free + lp.dirty, min_space); + dbg_gc("found LEB %d: free %d, dirty %d, sum %d (min. space %d)", + lp.lnum, lp.free, lp.dirty, lp.free + lp.dirty, + min_space); space_before = c->leb_size - wbuf->offs - wbuf->used; if (wbuf->lnum == -1) diff --git a/fs/ubifs/log.c b/fs/ubifs/log.c index c80b15d..36bd4ef 100644 --- a/fs/ubifs/log.c +++ b/fs/ubifs/log.c @@ -315,17 +315,15 @@ static void remove_buds(struct ubifs_info *c) * heads (non-closed buds). */ c->cmt_bud_bytes += wbuf->offs - bud->start; - dbg_log("preserve %d:%d, jhead %s, bud bytes %d, " - "cmt_bud_bytes %lld", bud->lnum, bud->start, - dbg_jhead(bud->jhead), wbuf->offs - bud->start, - c->cmt_bud_bytes); + dbg_log("preserve %d:%d, jhead %s, bud bytes %d, cmt_bud_bytes %lld", + bud->lnum, bud->start, dbg_jhead(bud->jhead), + wbuf->offs - bud->start, c->cmt_bud_bytes); bud->start = wbuf->offs; } else { c->cmt_bud_bytes += c->leb_size - bud->start; - dbg_log("remove %d:%d, jhead %s, bud bytes %d, " - "cmt_bud_bytes %lld", bud->lnum, bud->start, - dbg_jhead(bud->jhead), c->leb_size - bud->start, - c->cmt_bud_bytes); + dbg_log("remove %d:%d, jhead %s, bud bytes %d, cmt_bud_bytes %lld", + bud->lnum, bud->start, dbg_jhead(bud->jhead), + c->leb_size - bud->start, c->cmt_bud_bytes); rb_erase(p1, &c->buds); /* * If the commit does not finish, the recovery will need diff --git a/fs/ubifs/lprops.c b/fs/ubifs/lprops.c index 86eb8e5..e5a2a35 100644 --- a/fs/ubifs/lprops.c +++ b/fs/ubifs/lprops.c @@ -867,15 +867,15 @@ int dbg_check_cats(struct ubifs_info *c) list_for_each_entry(lprops, &c->empty_list, list) { if (lprops->free != c->leb_size) { - ubifs_err("non-empty LEB %d on empty list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("non-empty LEB %d on empty list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } if (lprops->flags & LPROPS_TAKEN) { - ubifs_err("taken LEB %d on empty list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("taken LEB %d on empty list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } } @@ -883,15 +883,15 @@ int dbg_check_cats(struct ubifs_info *c) i = 0; list_for_each_entry(lprops, &c->freeable_list, list) { if (lprops->free + lprops->dirty != c->leb_size) { - ubifs_err("non-freeable LEB %d on freeable list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("non-freeable LEB %d on freeable list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } if (lprops->flags & LPROPS_TAKEN) { - ubifs_err("taken LEB %d on freeable list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("taken LEB %d on freeable list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } i += 1; @@ -913,21 +913,21 @@ int dbg_check_cats(struct ubifs_info *c) list_for_each_entry(lprops, &c->frdi_idx_list, list) { if (lprops->free + lprops->dirty != c->leb_size) { - ubifs_err("non-freeable LEB %d on frdi_idx list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("non-freeable LEB %d on frdi_idx list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } if (lprops->flags & LPROPS_TAKEN) { - ubifs_err("taken LEB %d on frdi_idx list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("taken LEB %d on frdi_idx list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } if (!(lprops->flags & LPROPS_INDEX)) { - ubifs_err("non-index LEB %d on frdi_idx list " - "(free %d dirty %d flags %d)", lprops->lnum, - lprops->free, lprops->dirty, lprops->flags); + ubifs_err("non-index LEB %d on frdi_idx list (free %d dirty %d flags %d)", + lprops->lnum, lprops->free, lprops->dirty, + lprops->flags); return -EINVAL; } } @@ -982,9 +982,9 @@ void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat, goto out; } if (lprops != lp) { - dbg_msg("lprops %zx lp %zx lprops->lnum %d lp->lnum %d", - (size_t)lprops, (size_t)lp, lprops->lnum, - lp->lnum); + ubifs_err("lprops %zx lp %zx lprops->lnum %d lp->lnum %d", + (size_t)lprops, (size_t)lp, lprops->lnum, + lp->lnum); err = 4; goto out; } @@ -1002,7 +1002,7 @@ void dbg_check_heap(struct ubifs_info *c, struct ubifs_lpt_heap *heap, int cat, } out: if (err) { - dbg_msg("failed cat %d hpos %d err %d", cat, i, err); + ubifs_err("failed cat %d hpos %d err %d", cat, i, err); dump_stack(); ubifs_dump_heap(c, heap, cat); } @@ -1153,8 +1153,8 @@ static int scan_check_cb(struct ubifs_info *c, if (free > c->leb_size || free < 0 || dirty > c->leb_size || dirty < 0) { - ubifs_err("bad calculated accounting for LEB %d: " - "free %d, dirty %d", lnum, free, dirty); + ubifs_err("bad calculated accounting for LEB %d: free %d, dirty %d", + lnum, free, dirty); goto out_destroy; } @@ -1200,8 +1200,7 @@ static int scan_check_cb(struct ubifs_info *c, /* Free but not unmapped LEB, it's fine */ is_idx = 0; else { - ubifs_err("indexing node without indexing " - "flag"); + ubifs_err("indexing node without indexing flag"); goto out_print; } } @@ -1236,8 +1235,7 @@ static int scan_check_cb(struct ubifs_info *c, return LPT_SCAN_CONTINUE; out_print: - ubifs_err("bad accounting of LEB %d: free %d, dirty %d flags %#x, " - "should be free %d, dirty %d", + ubifs_err("bad accounting of LEB %d: free %d, dirty %d flags %#x, should be free %d, dirty %d", lnum, lp->free, lp->dirty, lp->flags, free, dirty); ubifs_dump_leb(c, lnum); out_destroy: @@ -1290,12 +1288,10 @@ int dbg_check_lprops(struct ubifs_info *c) lst.total_dirty != c->lst.total_dirty || lst.total_used != c->lst.total_used) { ubifs_err("bad overall accounting"); - ubifs_err("calculated: empty_lebs %d, idx_lebs %d, " - "total_free %lld, total_dirty %lld, total_used %lld", + ubifs_err("calculated: empty_lebs %d, idx_lebs %d, total_free %lld, total_dirty %lld, total_used %lld", lst.empty_lebs, lst.idx_lebs, lst.total_free, lst.total_dirty, lst.total_used); - ubifs_err("read from lprops: empty_lebs %d, idx_lebs %d, " - "total_free %lld, total_dirty %lld, total_used %lld", + ubifs_err("read from lprops: empty_lebs %d, idx_lebs %d, total_free %lld, total_dirty %lld, total_used %lld", c->lst.empty_lebs, c->lst.idx_lebs, c->lst.total_free, c->lst.total_dirty, c->lst.total_used); err = -EINVAL; diff --git a/fs/ubifs/lpt.c b/fs/ubifs/lpt.c index 8640920..d46b19e 100644 --- a/fs/ubifs/lpt.c +++ b/fs/ubifs/lpt.c @@ -1311,7 +1311,7 @@ out: ubifs_err("error %d reading pnode at %d:%d", err, lnum, offs); ubifs_dump_pnode(c, pnode, parent, iip); dump_stack(); - dbg_msg("calc num: %d", calc_pnode_num_from_parent(c, parent, iip)); + ubifs_err("calc num: %d", calc_pnode_num_from_parent(c, parent, iip)); kfree(pnode); return err; } @@ -2237,8 +2237,7 @@ int dbg_check_lpt_nodes(struct ubifs_info *c, struct ubifs_cnode *cnode, /* cnode is a nnode */ num = calc_nnode_num(row, col); if (cnode->num != num) { - ubifs_err("nnode num %d expected %d " - "parent num %d iip %d", + ubifs_err("nnode num %d expected %d parent num %d iip %d", cnode->num, num, (nnode ? nnode->num : 0), cnode->iip); return -EINVAL; diff --git a/fs/ubifs/lpt_commit.c b/fs/ubifs/lpt_commit.c index 4fa7073..9daaeef 100644 --- a/fs/ubifs/lpt_commit.c +++ b/fs/ubifs/lpt_commit.c @@ -320,8 +320,8 @@ static int layout_cnodes(struct ubifs_info *c) return 0; no_space: - ubifs_err("LPT out of space at LEB %d:%d needing %d, done_ltab %d, " - "done_lsave %d", lnum, offs, len, done_ltab, done_lsave); + ubifs_err("LPT out of space at LEB %d:%d needing %d, done_ltab %d, done_lsave %d", + lnum, offs, len, done_ltab, done_lsave); ubifs_dump_lpt_info(c); ubifs_dump_lpt_lebs(c); dump_stack(); @@ -545,8 +545,8 @@ static int write_cnodes(struct ubifs_info *c) return 0; no_space: - ubifs_err("LPT out of space mismatch at LEB %d:%d needing %d, done_ltab " - "%d, done_lsave %d", lnum, offs, len, done_ltab, done_lsave); + ubifs_err("LPT out of space mismatch at LEB %d:%d needing %d, done_ltab %d, done_lsave %d", + lnum, offs, len, done_ltab, done_lsave); ubifs_dump_lpt_info(c); ubifs_dump_lpt_lebs(c); dump_stack(); @@ -1662,21 +1662,19 @@ static int dbg_check_ltab_lnum(struct ubifs_info *c, int lnum) continue; } if (!dbg_is_all_ff(p, len)) { - dbg_msg("invalid empty space in LEB %d at %d", - lnum, c->leb_size - len); + ubifs_err("invalid empty space in LEB %d at %d", + lnum, c->leb_size - len); err = -EINVAL; } i = lnum - c->lpt_first; if (len != c->ltab[i].free) { - dbg_msg("invalid free space in LEB %d " - "(free %d, expected %d)", - lnum, len, c->ltab[i].free); + ubifs_err("invalid free space in LEB %d (free %d, expected %d)", + lnum, len, c->ltab[i].free); err = -EINVAL; } if (dirty != c->ltab[i].dirty) { - dbg_msg("invalid dirty space in LEB %d " - "(dirty %d, expected %d)", - lnum, dirty, c->ltab[i].dirty); + ubifs_err("invalid dirty space in LEB %d (dirty %d, expected %d)", + lnum, dirty, c->ltab[i].dirty); err = -EINVAL; } goto out; @@ -1888,8 +1886,7 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum) int err, len = c->leb_size, node_type, node_num, node_len, offs; void *buf, *p; - printk(KERN_DEBUG "(pid %d) start dumping LEB %d\n", - current->pid, lnum); + pr_err("(pid %d) start dumping LEB %d\n", current->pid, lnum); buf = p = __vmalloc(c->leb_size, GFP_NOFS, PAGE_KERNEL); if (!buf) { ubifs_err("cannot allocate memory to dump LPT"); @@ -1907,14 +1904,14 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum) pad_len = get_pad_len(c, p, len); if (pad_len) { - printk(KERN_DEBUG "LEB %d:%d, pad %d bytes\n", + pr_err("LEB %d:%d, pad %d bytes\n", lnum, offs, pad_len); p += pad_len; len -= pad_len; continue; } if (len) - printk(KERN_DEBUG "LEB %d:%d, free %d bytes\n", + pr_err("LEB %d:%d, free %d bytes\n", lnum, offs, len); break; } @@ -1925,11 +1922,10 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum) { node_len = c->pnode_sz; if (c->big_lpt) - printk(KERN_DEBUG "LEB %d:%d, pnode num %d\n", + pr_err("LEB %d:%d, pnode num %d\n", lnum, offs, node_num); else - printk(KERN_DEBUG "LEB %d:%d, pnode\n", - lnum, offs); + pr_err("LEB %d:%d, pnode\n", lnum, offs); break; } case UBIFS_LPT_NNODE: @@ -1939,29 +1935,28 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum) node_len = c->nnode_sz; if (c->big_lpt) - printk(KERN_DEBUG "LEB %d:%d, nnode num %d, ", + pr_err("LEB %d:%d, nnode num %d, ", lnum, offs, node_num); else - printk(KERN_DEBUG "LEB %d:%d, nnode, ", + pr_err("LEB %d:%d, nnode, ", lnum, offs); err = ubifs_unpack_nnode(c, p, &nnode); for (i = 0; i < UBIFS_LPT_FANOUT; i++) { - printk(KERN_CONT "%d:%d", nnode.nbranch[i].lnum, + pr_cont("%d:%d", nnode.nbranch[i].lnum, nnode.nbranch[i].offs); if (i != UBIFS_LPT_FANOUT - 1) - printk(KERN_CONT ", "); + pr_cont(", "); } - printk(KERN_CONT "\n"); + pr_cont("\n"); break; } case UBIFS_LPT_LTAB: node_len = c->ltab_sz; - printk(KERN_DEBUG "LEB %d:%d, ltab\n", - lnum, offs); + pr_err("LEB %d:%d, ltab\n", lnum, offs); break; case UBIFS_LPT_LSAVE: node_len = c->lsave_sz; - printk(KERN_DEBUG "LEB %d:%d, lsave len\n", lnum, offs); + pr_err("LEB %d:%d, lsave len\n", lnum, offs); break; default: ubifs_err("LPT node type %d not recognized", node_type); @@ -1972,8 +1967,7 @@ static void dump_lpt_leb(const struct ubifs_info *c, int lnum) len -= node_len; } - printk(KERN_DEBUG "(pid %d) finish dumping LEB %d\n", - current->pid, lnum); + pr_err("(pid %d) finish dumping LEB %d\n", current->pid, lnum); out: vfree(buf); return; @@ -1990,12 +1984,10 @@ void ubifs_dump_lpt_lebs(const struct ubifs_info *c) { int i; - printk(KERN_DEBUG "(pid %d) start dumping all LPT LEBs\n", - current->pid); + pr_err("(pid %d) start dumping all LPT LEBs\n", current->pid); for (i = 0; i < c->lpt_lebs; i++) dump_lpt_leb(c, i + c->lpt_first); - printk(KERN_DEBUG "(pid %d) finish dumping all LPT LEBs\n", - current->pid); + pr_err("(pid %d) finish dumping all LPT LEBs\n", current->pid); } /** diff --git a/fs/ubifs/orphan.c b/fs/ubifs/orphan.c index cebf17e..769701c 100644 --- a/fs/ubifs/orphan.c +++ b/fs/ubifs/orphan.c @@ -562,8 +562,8 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb, list_for_each_entry(snod, &sleb->nodes, list) { if (snod->type != UBIFS_ORPH_NODE) { - ubifs_err("invalid node type %d in orphan area at " - "%d:%d", snod->type, sleb->lnum, snod->offs); + ubifs_err("invalid node type %d in orphan area at %d:%d", + snod->type, sleb->lnum, snod->offs); ubifs_dump_node(c, snod->node); return -EINVAL; } @@ -589,8 +589,7 @@ static int do_kill_orphans(struct ubifs_info *c, struct ubifs_scan_leb *sleb, * number. That makes this orphan node, out of date. */ if (!first) { - ubifs_err("out of order commit number %llu in " - "orphan node at %d:%d", + ubifs_err("out of order commit number %llu in orphan node at %d:%d", cmt_no, sleb->lnum, snod->offs); ubifs_dump_node(c, snod->node); return -EINVAL; diff --git a/fs/ubifs/recovery.c b/fs/ubifs/recovery.c index edeec49..065096e 100644 --- a/fs/ubifs/recovery.c +++ b/fs/ubifs/recovery.c @@ -609,7 +609,8 @@ static void drop_last_node(struct ubifs_scan_leb *sleb, int *offs) snod = list_entry(sleb->nodes.prev, struct ubifs_scan_node, list); - dbg_rcvry("dropping last node at %d:%d", sleb->lnum, snod->offs); + dbg_rcvry("dropping last node at %d:%d", + sleb->lnum, snod->offs); *offs = snod->offs; list_del(&snod->list); kfree(snod); @@ -702,8 +703,8 @@ struct ubifs_scan_leb *ubifs_recover_leb(struct ubifs_info *c, int lnum, * See header comment for this file for more * explanations about the reasons we have this check. */ - ubifs_err("corrupt empty space LEB %d:%d, corruption " - "starts at %d", lnum, offs, corruption); + ubifs_err("corrupt empty space LEB %d:%d, corruption starts at %d", + lnum, offs, corruption); /* Make sure we dump interesting non-0xFF data */ offs += corruption; buf += corruption; @@ -899,8 +900,8 @@ struct ubifs_scan_leb *ubifs_recover_log_leb(struct ubifs_info *c, int lnum, } } if (snod->sqnum > cs_sqnum) { - ubifs_err("unrecoverable log corruption " - "in LEB %d", lnum); + ubifs_err("unrecoverable log corruption in LEB %d", + lnum); ubifs_scan_destroy(sleb); return ERR_PTR(-EUCLEAN); } diff --git a/fs/ubifs/replay.c b/fs/ubifs/replay.c index 94d78fc..3187925 100644 --- a/fs/ubifs/replay.c +++ b/fs/ubifs/replay.c @@ -141,9 +141,9 @@ static int set_bud_lprops(struct ubifs_info *c, struct bud_entry *b) * during the replay. */ if (dirty != 0) - dbg_msg("LEB %d lp: %d free %d dirty " - "replay: %d free %d dirty", b->bud->lnum, - lp->free, lp->dirty, b->free, b->dirty); + dbg_mnt("LEB %d lp: %d free %d dirty replay: %d free %d dirty", + b->bud->lnum, lp->free, lp->dirty, b->free, + b->dirty); } lp = ubifs_change_lp(c, lp, b->free, dirty + b->dirty, lp->flags | LPROPS_TAKEN, 0); @@ -677,7 +677,8 @@ static int replay_bud(struct ubifs_info *c, struct bud_entry *b) b->dirty = sleb->endpt - offs - used; b->free = c->leb_size - sleb->endpt; - dbg_mnt("bud LEB %d replied: dirty %d, free %d", lnum, b->dirty, b->free); + dbg_mnt("bud LEB %d replied: dirty %d, free %d", + lnum, b->dirty, b->free); out: ubifs_scan_destroy(sleb); @@ -865,8 +866,7 @@ static int replay_log_leb(struct ubifs_info *c, int lnum, int offs, void *sbuf) goto out_dump; } if (le64_to_cpu(node->cmt_no) != c->cmt_no) { - ubifs_err("first CS node at LEB %d:%d has wrong " - "commit number %llu expected %llu", + ubifs_err("first CS node at LEB %d:%d has wrong commit number %llu expected %llu", lnum, offs, (unsigned long long)le64_to_cpu(node->cmt_no), c->cmt_no); @@ -1058,8 +1058,8 @@ int ubifs_replay_journal(struct ubifs_info *c) c->bi.uncommitted_idx *= c->max_idx_node_sz; ubifs_assert(c->bud_bytes <= c->max_bud_bytes || c->need_recovery); - dbg_mnt("finished, log head LEB %d:%d, max_sqnum %llu, " - "highest_inum %lu", c->lhead_lnum, c->lhead_offs, c->max_sqnum, + dbg_mnt("finished, log head LEB %d:%d, max_sqnum %llu, highest_inum %lu", + c->lhead_lnum, c->lhead_offs, c->max_sqnum, (unsigned long)c->highest_inum); out: destroy_replay_list(c); diff --git a/fs/ubifs/sb.c b/fs/ubifs/sb.c index 52c21f4..4c37607 100644 --- a/fs/ubifs/sb.c +++ b/fs/ubifs/sb.c @@ -391,9 +391,8 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup) min_leb_cnt += c->lpt_lebs + c->orph_lebs + c->jhead_cnt + 6; if (c->leb_cnt < min_leb_cnt || c->leb_cnt > c->vi.size) { - ubifs_err("bad LEB count: %d in superblock, %d on UBI volume, " - "%d minimum required", c->leb_cnt, c->vi.size, - min_leb_cnt); + ubifs_err("bad LEB count: %d in superblock, %d on UBI volume, %d minimum required", + c->leb_cnt, c->vi.size, min_leb_cnt); goto failed; } @@ -411,15 +410,14 @@ static int validate_sb(struct ubifs_info *c, struct ubifs_sb_node *sup) max_bytes = (long long)c->leb_size * UBIFS_MIN_BUD_LEBS; if (c->max_bud_bytes < max_bytes) { - ubifs_err("too small journal (%lld bytes), must be at least " - "%lld bytes", c->max_bud_bytes, max_bytes); + ubifs_err("too small journal (%lld bytes), must be at least %lld bytes", + c->max_bud_bytes, max_bytes); goto failed; } max_bytes = (long long)c->leb_size * c->main_lebs; if (c->max_bud_bytes > max_bytes) { - ubifs_err("too large journal size (%lld bytes), only %lld bytes" - "available in the main area", + ubifs_err("too large journal size (%lld bytes), only %lld bytes available in the main area", c->max_bud_bytes, max_bytes); goto failed; } @@ -549,10 +547,9 @@ int ubifs_read_superblock(struct ubifs_info *c) ubifs_assert(!c->ro_media || c->ro_mount); if (!c->ro_mount || c->ro_compat_version > UBIFS_RO_COMPAT_VERSION) { - ubifs_err("on-flash format version is w%d/r%d, but " - "software only supports up to version " - "w%d/r%d", c->fmt_version, - c->ro_compat_version, UBIFS_FORMAT_VERSION, + ubifs_err("on-flash format version is w%d/r%d, but software only supports up to version w%d/r%d", + c->fmt_version, c->ro_compat_version, + UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION); if (c->ro_compat_version <= UBIFS_RO_COMPAT_VERSION) { ubifs_msg("only R/O mounting is possible"); diff --git a/fs/ubifs/scan.c b/fs/ubifs/scan.c index 7c40e60..58aa05d 100644 --- a/fs/ubifs/scan.c +++ b/fs/ubifs/scan.c @@ -75,7 +75,7 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum, magic = le32_to_cpu(ch->magic); if (magic == 0xFFFFFFFF) { - dbg_scan("hit empty space"); + dbg_scan("hit empty space at LEB %d:%d", lnum, offs); return SCANNED_EMPTY_SPACE; } @@ -85,7 +85,8 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum, if (len < UBIFS_CH_SZ) return SCANNED_GARBAGE; - dbg_scan("scanning %s", dbg_ntype(ch->node_type)); + dbg_scan("scanning %s at LEB %d:%d", + dbg_ntype(ch->node_type), lnum, offs); if (ubifs_check_node(c, buf, lnum, offs, quiet, 1)) return SCANNED_A_CORRUPT_NODE; @@ -114,8 +115,8 @@ int ubifs_scan_a_node(const struct ubifs_info *c, void *buf, int len, int lnum, return SCANNED_A_BAD_PAD_NODE; } - dbg_scan("%d bytes padded, offset now %d", - pad_len, ALIGN(offs + node_len + pad_len, 8)); + dbg_scan("%d bytes padded at LEB %d:%d, offset now %d", pad_len, + lnum, offs, ALIGN(offs + node_len + pad_len, 8)); return node_len + pad_len; } @@ -150,8 +151,8 @@ struct ubifs_scan_leb *ubifs_start_scan(const struct ubifs_info *c, int lnum, err = ubifs_leb_read(c, lnum, sbuf + offs, offs, c->leb_size - offs, 0); if (err && err != -EBADMSG) { - ubifs_err("cannot read %d bytes from LEB %d:%d," - " error %d", c->leb_size - offs, lnum, offs, err); + ubifs_err("cannot read %d bytes from LEB %d:%d, error %d", + c->leb_size - offs, lnum, offs, err); kfree(sleb); return ERR_PTR(err); } @@ -240,8 +241,6 @@ void ubifs_scanned_corruption(const struct ubifs_info *c, int lnum, int offs, int len; ubifs_err("corruption at LEB %d:%d", lnum, offs); - if (dbg_is_tst_rcvry(c)) - return; len = c->leb_size - offs; if (len > 8192) len = 8192; diff --git a/fs/ubifs/super.c b/fs/ubifs/super.c index 681f3a9..ddc0f6a 100644 --- a/fs/ubifs/super.c +++ b/fs/ubifs/super.c @@ -89,9 +89,8 @@ static int validate_inode(struct ubifs_info *c, const struct inode *inode) return 5; if (!ubifs_compr_present(ui->compr_type)) { - ubifs_warn("inode %lu uses '%s' compression, but it was not " - "compiled in", inode->i_ino, - ubifs_compr_name(ui->compr_type)); + ubifs_warn("inode %lu uses '%s' compression, but it was not compiled in", + inode->i_ino, ubifs_compr_name(ui->compr_type)); } err = dbg_check_dir(c, inode); @@ -1061,8 +1060,8 @@ static int ubifs_parse_options(struct ubifs_info *c, char *options, flag = parse_standard_option(p); if (!flag) { - ubifs_err("unrecognized mount option \"%s\" " - "or missing value", p); + ubifs_err("unrecognized mount option \"%s\" or missing value", + p); return -EINVAL; } sb->s_flags |= flag; @@ -1124,8 +1123,8 @@ again: } /* Just disable bulk-read */ - ubifs_warn("Cannot allocate %d bytes of memory for bulk-read, " - "disabling it", c->max_bu_buf_len); + ubifs_warn("cannot allocate %d bytes of memory for bulk-read, disabling it", + c->max_bu_buf_len); c->mount_opts.bulk_read = 1; c->bulk_read = 0; return; @@ -1161,7 +1160,7 @@ static int check_free_space(struct ubifs_info *c) static int mount_ubifs(struct ubifs_info *c) { int err; - long long x; + long long x, y; size_t sz; c->ro_mount = !!(c->vfs_sb->s_flags & MS_RDONLY); @@ -1411,75 +1410,69 @@ static int mount_ubifs(struct ubifs_info *c) c->mounting = 0; - ubifs_msg("mounted UBI device %d, volume %d, name \"%s\"", - c->vi.ubi_num, c->vi.vol_id, c->vi.name); - if (c->ro_mount) - ubifs_msg("mounted read-only"); + ubifs_msg("mounted UBI device %d, volume %d, name \"%s\"%s", + c->vi.ubi_num, c->vi.vol_id, c->vi.name, + c->ro_mount ? ", R/O mode" : NULL); x = (long long)c->main_lebs * c->leb_size; - ubifs_msg("file system size: %lld bytes (%lld KiB, %lld MiB, %d " - "LEBs)", x, x >> 10, x >> 20, c->main_lebs); - x = (long long)c->log_lebs * c->leb_size + c->max_bud_bytes; - ubifs_msg("journal size: %lld bytes (%lld KiB, %lld MiB, %d " - "LEBs)", x, x >> 10, x >> 20, c->log_lebs + c->max_bud_cnt); - ubifs_msg("media format: w%d/r%d (latest is w%d/r%d)", + y = (long long)c->log_lebs * c->leb_size + c->max_bud_bytes; + ubifs_msg("LEB size: %d bytes (%d KiB), min./max. I/O unit sizes: %d bytes/%d bytes", + c->leb_size, c->leb_size >> 10, c->min_io_size, + c->max_write_size); + ubifs_msg("FS size: %lld bytes (%lld MiB, %d LEBs), journal size %lld bytes (%lld MiB, %d LEBs)", + x, x >> 20, c->main_lebs, + y, y >> 20, c->log_lebs + c->max_bud_cnt); + ubifs_msg("reserved for root: %llu bytes (%llu KiB)", + c->report_rp_size, c->report_rp_size >> 10); + ubifs_msg("media format: w%d/r%d (latest is w%d/r%d), UUID %pUB%s", c->fmt_version, c->ro_compat_version, - UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION); - ubifs_msg("default compressor: %s", ubifs_compr_name(c->default_compr)); - ubifs_msg("reserved for root: %llu bytes (%llu KiB)", - c->report_rp_size, c->report_rp_size >> 10); - - dbg_msg("compiled on: " __DATE__ " at " __TIME__); - dbg_msg("min. I/O unit size: %d bytes", c->min_io_size); - dbg_msg("max. write size: %d bytes", c->max_write_size); - dbg_msg("LEB size: %d bytes (%d KiB)", - c->leb_size, c->leb_size >> 10); - dbg_msg("data journal heads: %d", + UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION, c->uuid, + c->big_lpt ? ", big LPT model" : ", small LPT model"); + + dbg_gen("default compressor: %s", ubifs_compr_name(c->default_compr)); + dbg_gen("data journal heads: %d", c->jhead_cnt - NONDATA_JHEADS_CNT); - dbg_msg("UUID: %pUB", c->uuid); - dbg_msg("big_lpt %d", c->big_lpt); - dbg_msg("log LEBs: %d (%d - %d)", + dbg_gen("log LEBs: %d (%d - %d)", c->log_lebs, UBIFS_LOG_LNUM, c->log_last); - dbg_msg("LPT area LEBs: %d (%d - %d)", + dbg_gen("LPT area LEBs: %d (%d - %d)", c->lpt_lebs, c->lpt_first, c->lpt_last); - dbg_msg("orphan area LEBs: %d (%d - %d)", + dbg_gen("orphan area LEBs: %d (%d - %d)", c->orph_lebs, c->orph_first, c->orph_last); - dbg_msg("main area LEBs: %d (%d - %d)", + dbg_gen("main area LEBs: %d (%d - %d)", c->main_lebs, c->main_first, c->leb_cnt - 1); - dbg_msg("index LEBs: %d", c->lst.idx_lebs); - dbg_msg("total index bytes: %lld (%lld KiB, %lld MiB)", + dbg_gen("index LEBs: %d", c->lst.idx_lebs); + dbg_gen("total index bytes: %lld (%lld KiB, %lld MiB)", c->bi.old_idx_sz, c->bi.old_idx_sz >> 10, c->bi.old_idx_sz >> 20); - dbg_msg("key hash type: %d", c->key_hash_type); - dbg_msg("tree fanout: %d", c->fanout); - dbg_msg("reserved GC LEB: %d", c->gc_lnum); - dbg_msg("first main LEB: %d", c->main_first); - dbg_msg("max. znode size %d", c->max_znode_sz); - dbg_msg("max. index node size %d", c->max_idx_node_sz); - dbg_msg("node sizes: data %zu, inode %zu, dentry %zu", + dbg_gen("key hash type: %d", c->key_hash_type); + dbg_gen("tree fanout: %d", c->fanout); + dbg_gen("reserved GC LEB: %d", c->gc_lnum); + dbg_gen("max. znode size %d", c->max_znode_sz); + dbg_gen("max. index node size %d", c->max_idx_node_sz); + dbg_gen("node sizes: data %zu, inode %zu, dentry %zu", UBIFS_DATA_NODE_SZ, UBIFS_INO_NODE_SZ, UBIFS_DENT_NODE_SZ); - dbg_msg("node sizes: trun %zu, sb %zu, master %zu", + dbg_gen("node sizes: trun %zu, sb %zu, master %zu", UBIFS_TRUN_NODE_SZ, UBIFS_SB_NODE_SZ, UBIFS_MST_NODE_SZ); - dbg_msg("node sizes: ref %zu, cmt. start %zu, orph %zu", + dbg_gen("node sizes: ref %zu, cmt. start %zu, orph %zu", UBIFS_REF_NODE_SZ, UBIFS_CS_NODE_SZ, UBIFS_ORPH_NODE_SZ); - dbg_msg("max. node sizes: data %zu, inode %zu dentry %zu, idx %d", + dbg_gen("max. node sizes: data %zu, inode %zu dentry %zu, idx %d", UBIFS_MAX_DATA_NODE_SZ, UBIFS_MAX_INO_NODE_SZ, UBIFS_MAX_DENT_NODE_SZ, ubifs_idx_node_sz(c, c->fanout)); - dbg_msg("dead watermark: %d", c->dead_wm); - dbg_msg("dark watermark: %d", c->dark_wm); - dbg_msg("LEB overhead: %d", c->leb_overhead); + dbg_gen("dead watermark: %d", c->dead_wm); + dbg_gen("dark watermark: %d", c->dark_wm); + dbg_gen("LEB overhead: %d", c->leb_overhead); x = (long long)c->main_lebs * c->dark_wm; - dbg_msg("max. dark space: %lld (%lld KiB, %lld MiB)", + dbg_gen("max. dark space: %lld (%lld KiB, %lld MiB)", x, x >> 10, x >> 20); - dbg_msg("maximum bud bytes: %lld (%lld KiB, %lld MiB)", + dbg_gen("maximum bud bytes: %lld (%lld KiB, %lld MiB)", c->max_bud_bytes, c->max_bud_bytes >> 10, c->max_bud_bytes >> 20); - dbg_msg("BG commit bud bytes: %lld (%lld KiB, %lld MiB)", + dbg_gen("BG commit bud bytes: %lld (%lld KiB, %lld MiB)", c->bg_bud_bytes, c->bg_bud_bytes >> 10, c->bg_bud_bytes >> 20); - dbg_msg("current bud bytes %lld (%lld KiB, %lld MiB)", + dbg_gen("current bud bytes %lld (%lld KiB, %lld MiB)", c->bud_bytes, c->bud_bytes >> 10, c->bud_bytes >> 20); - dbg_msg("max. seq. number: %llu", c->max_sqnum); - dbg_msg("commit number: %llu", c->cmt_no); + dbg_gen("max. seq. number: %llu", c->max_sqnum); + dbg_gen("commit number: %llu", c->cmt_no); return 0; @@ -1564,10 +1557,9 @@ static int ubifs_remount_rw(struct ubifs_info *c) if (c->rw_incompat) { ubifs_err("the file-system is not R/W-compatible"); - ubifs_msg("on-flash format version is w%d/r%d, but software " - "only supports up to version w%d/r%d", c->fmt_version, - c->ro_compat_version, UBIFS_FORMAT_VERSION, - UBIFS_RO_COMPAT_VERSION); + ubifs_msg("on-flash format version is w%d/r%d, but software only supports up to version w%d/r%d", + c->fmt_version, c->ro_compat_version, + UBIFS_FORMAT_VERSION, UBIFS_RO_COMPAT_VERSION); return -EROFS; } @@ -1828,8 +1820,8 @@ static void ubifs_put_super(struct super_block *sb) * next mount, so we just print a message and * continue to unmount normally. */ - ubifs_err("failed to write master node, " - "error %d", err); + ubifs_err("failed to write master node, error %d", + err); } else { for (i = 0; i < c->jhead_cnt; i++) /* Make sure write-buffer timers are canceled */ @@ -2248,8 +2240,7 @@ static int __init ubifs_init(void) * UBIFS_BLOCK_SIZE. It is assumed that both are powers of 2. */ if (PAGE_CACHE_SIZE < UBIFS_BLOCK_SIZE) { - ubifs_err("VFS page cache size is %u bytes, but UBIFS requires" - " at least 4096 bytes", + ubifs_err("VFS page cache size is %u bytes, but UBIFS requires at least 4096 bytes", (unsigned int)PAGE_CACHE_SIZE); return -EINVAL; } @@ -2298,6 +2289,12 @@ static void __exit ubifs_exit(void) dbg_debugfs_exit(); ubifs_compressors_exit(); unregister_shrinker(&ubifs_shrinker_info); + + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ubifs_inode_slab); unregister_filesystem(&ubifs_fs_type); } diff --git a/fs/ubifs/tnc_misc.c b/fs/ubifs/tnc_misc.c index d38ac7f..f6bf899 100644 --- a/fs/ubifs/tnc_misc.c +++ b/fs/ubifs/tnc_misc.c @@ -328,8 +328,8 @@ static int read_znode(struct ubifs_info *c, int lnum, int offs, int len, case UBIFS_XENT_KEY: break; default: - dbg_msg("bad key type at slot %d: %d", - i, key_type(c, &zbr->key)); + ubifs_err("bad key type at slot %d: %d", + i, key_type(c, &zbr->key)); err = 3; goto out_dump; } diff --git a/fs/ubifs/ubifs.h b/fs/ubifs/ubifs.h index 64f2367..5486346 100644 --- a/fs/ubifs/ubifs.h +++ b/fs/ubifs/ubifs.h @@ -42,16 +42,15 @@ #define UBIFS_VERSION 1 /* Normal UBIFS messages */ -#define ubifs_msg(fmt, ...) \ - printk(KERN_NOTICE "UBIFS: " fmt "\n", ##__VA_ARGS__) +#define ubifs_msg(fmt, ...) pr_notice("UBIFS: " fmt "\n", ##__VA_ARGS__) /* UBIFS error messages */ -#define ubifs_err(fmt, ...) \ - printk(KERN_ERR "UBIFS error (pid %d): %s: " fmt "\n", current->pid, \ +#define ubifs_err(fmt, ...) \ + pr_err("UBIFS error (pid %d): %s: " fmt "\n", current->pid, \ __func__, ##__VA_ARGS__) /* UBIFS warning messages */ -#define ubifs_warn(fmt, ...) \ - printk(KERN_WARNING "UBIFS warning (pid %d): %s: " fmt "\n", \ - current->pid, __func__, ##__VA_ARGS__) +#define ubifs_warn(fmt, ...) \ + pr_warn("UBIFS warning (pid %d): %s: " fmt "\n", \ + current->pid, __func__, ##__VA_ARGS__) /* UBIFS file system VFS magic number */ #define UBIFS_SUPER_MAGIC 0x24051905 diff --git a/fs/udf/super.c b/fs/udf/super.c index 862741d..d44fb56 100644 --- a/fs/udf/super.c +++ b/fs/udf/super.c @@ -171,6 +171,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(udf_inode_cachep); } diff --git a/fs/ufs/super.c b/fs/ufs/super.c index 444927e..f7cfecf 100644 --- a/fs/ufs/super.c +++ b/fs/ufs/super.c @@ -1466,6 +1466,11 @@ static int init_inodecache(void) static void destroy_inodecache(void) { + /* + * Make sure all delayed rcu free inodes are flushed before we + * destroy cache. + */ + rcu_barrier(); kmem_cache_destroy(ufs_inode_cachep); } diff --git a/fs/utimes.c b/fs/utimes.c index fa4dbe4..bb0696a 100644 --- a/fs/utimes.c +++ b/fs/utimes.c @@ -140,19 +140,18 @@ long do_utimes(int dfd, const char __user *filename, struct timespec *times, goto out; if (filename == NULL && dfd != AT_FDCWD) { - int fput_needed; - struct file *file; + struct fd f; if (flags & AT_SYMLINK_NOFOLLOW) goto out; - file = fget_light(dfd, &fput_needed); + f = fdget(dfd); error = -EBADF; - if (!file) + if (!f.file) goto out; - error = utimes_common(&file->f_path, times); - fput_light(file, fput_needed); + error = utimes_common(&f.file->f_path, times); + fdput(f); } else { struct path path; int lookup_flags = 0; @@ -296,11 +296,13 @@ vfs_removexattr(struct dentry *dentry, const char *name) if (error) return error; + mutex_lock(&inode->i_mutex); error = security_inode_removexattr(dentry, name); - if (error) + if (error) { + mutex_unlock(&inode->i_mutex); return error; + } - mutex_lock(&inode->i_mutex); error = inode->i_op->removexattr(dentry, name); mutex_unlock(&inode->i_mutex); @@ -403,22 +405,20 @@ SYSCALL_DEFINE5(lsetxattr, const char __user *, pathname, SYSCALL_DEFINE5(fsetxattr, int, fd, const char __user *, name, const void __user *,value, size_t, size, int, flags) { - int fput_needed; - struct file *f; + struct fd f = fdget(fd); struct dentry *dentry; int error = -EBADF; - f = fget_light(fd, &fput_needed); - if (!f) + if (!f.file) return error; - dentry = f->f_path.dentry; + dentry = f.file->f_path.dentry; audit_inode(NULL, dentry); - error = mnt_want_write_file(f); + error = mnt_want_write_file(f.file); if (!error) { error = setxattr(dentry, name, value, size, flags); - mnt_drop_write_file(f); + mnt_drop_write_file(f.file); } - fput_light(f, fput_needed); + fdput(f); return error; } @@ -502,16 +502,14 @@ SYSCALL_DEFINE4(lgetxattr, const char __user *, pathname, SYSCALL_DEFINE4(fgetxattr, int, fd, const char __user *, name, void __user *, value, size_t, size) { - int fput_needed; - struct file *f; + struct fd f = fdget(fd); ssize_t error = -EBADF; - f = fget_light(fd, &fput_needed); - if (!f) + if (!f.file) return error; - audit_inode(NULL, f->f_path.dentry); - error = getxattr(f->f_path.dentry, name, value, size); - fput_light(f, fput_needed); + audit_inode(NULL, f.file->f_path.dentry); + error = getxattr(f.file->f_path.dentry, name, value, size); + fdput(f); return error; } @@ -583,16 +581,14 @@ SYSCALL_DEFINE3(llistxattr, const char __user *, pathname, char __user *, list, SYSCALL_DEFINE3(flistxattr, int, fd, char __user *, list, size_t, size) { - int fput_needed; - struct file *f; + struct fd f = fdget(fd); ssize_t error = -EBADF; - f = fget_light(fd, &fput_needed); - if (!f) + if (!f.file) return error; - audit_inode(NULL, f->f_path.dentry); - error = listxattr(f->f_path.dentry, list, size); - fput_light(f, fput_needed); + audit_inode(NULL, f.file->f_path.dentry); + error = listxattr(f.file->f_path.dentry, list, size); + fdput(f); return error; } @@ -652,22 +648,20 @@ SYSCALL_DEFINE2(lremovexattr, const char __user *, pathname, SYSCALL_DEFINE2(fremovexattr, int, fd, const char __user *, name) { - int fput_needed; - struct file *f; + struct fd f = fdget(fd); struct dentry *dentry; int error = -EBADF; - f = fget_light(fd, &fput_needed); - if (!f) + if (!f.file) return error; - dentry = f->f_path.dentry; + dentry = f.file->f_path.dentry; audit_inode(NULL, dentry); - error = mnt_want_write_file(f); + error = mnt_want_write_file(f.file); if (!error) { error = removexattr(dentry, name); - mnt_drop_write_file(f); + mnt_drop_write_file(f.file); } - fput_light(f, fput_needed); + fdput(f); return error; } diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c index e00de08..b9b8646 100644 --- a/fs/xfs/xfs_dfrag.c +++ b/fs/xfs/xfs_dfrag.c @@ -48,44 +48,44 @@ xfs_swapext( xfs_swapext_t *sxp) { xfs_inode_t *ip, *tip; - struct file *file, *tmp_file; + struct fd f, tmp; int error = 0; /* Pull information for the target fd */ - file = fget((int)sxp->sx_fdtarget); - if (!file) { + f = fdget((int)sxp->sx_fdtarget); + if (!f.file) { error = XFS_ERROR(EINVAL); goto out; } - if (!(file->f_mode & FMODE_WRITE) || - !(file->f_mode & FMODE_READ) || - (file->f_flags & O_APPEND)) { + if (!(f.file->f_mode & FMODE_WRITE) || + !(f.file->f_mode & FMODE_READ) || + (f.file->f_flags & O_APPEND)) { error = XFS_ERROR(EBADF); goto out_put_file; } - tmp_file = fget((int)sxp->sx_fdtmp); - if (!tmp_file) { + tmp = fdget((int)sxp->sx_fdtmp); + if (!tmp.file) { error = XFS_ERROR(EINVAL); goto out_put_file; } - if (!(tmp_file->f_mode & FMODE_WRITE) || - !(tmp_file->f_mode & FMODE_READ) || - (tmp_file->f_flags & O_APPEND)) { + if (!(tmp.file->f_mode & FMODE_WRITE) || + !(tmp.file->f_mode & FMODE_READ) || + (tmp.file->f_flags & O_APPEND)) { error = XFS_ERROR(EBADF); goto out_put_tmp_file; } - if (IS_SWAPFILE(file->f_path.dentry->d_inode) || - IS_SWAPFILE(tmp_file->f_path.dentry->d_inode)) { + if (IS_SWAPFILE(f.file->f_path.dentry->d_inode) || + IS_SWAPFILE(tmp.file->f_path.dentry->d_inode)) { error = XFS_ERROR(EINVAL); goto out_put_tmp_file; } - ip = XFS_I(file->f_path.dentry->d_inode); - tip = XFS_I(tmp_file->f_path.dentry->d_inode); + ip = XFS_I(f.file->f_path.dentry->d_inode); + tip = XFS_I(tmp.file->f_path.dentry->d_inode); if (ip->i_mount != tip->i_mount) { error = XFS_ERROR(EINVAL); @@ -105,9 +105,9 @@ xfs_swapext( error = xfs_swap_extents(ip, tip, sxp); out_put_tmp_file: - fput(tmp_file); + fdput(tmp); out_put_file: - fput(file); + fdput(f); out: return error; } diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 56afcdb..1eaeb8b 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -36,6 +36,7 @@ #include <linux/dcache.h> #include <linux/falloc.h> +#include <linux/pagevec.h> static const struct vm_operations_struct xfs_file_vm_ops; @@ -959,17 +960,232 @@ xfs_vm_page_mkwrite( return block_page_mkwrite(vma, vmf, xfs_get_blocks); } +/* + * This type is designed to indicate the type of offset we would like + * to search from page cache for either xfs_seek_data() or xfs_seek_hole(). + */ +enum { + HOLE_OFF = 0, + DATA_OFF, +}; + +/* + * Lookup the desired type of offset from the given page. + * + * On success, return true and the offset argument will point to the + * start of the region that was found. Otherwise this function will + * return false and keep the offset argument unchanged. + */ +STATIC bool +xfs_lookup_buffer_offset( + struct page *page, + loff_t *offset, + unsigned int type) +{ + loff_t lastoff = page_offset(page); + bool found = false; + struct buffer_head *bh, *head; + + bh = head = page_buffers(page); + do { + /* + * Unwritten extents that have data in the page + * cache covering them can be identified by the + * BH_Unwritten state flag. Pages with multiple + * buffers might have a mix of holes, data and + * unwritten extents - any buffer with valid + * data in it should have BH_Uptodate flag set + * on it. + */ + if (buffer_unwritten(bh) || + buffer_uptodate(bh)) { + if (type == DATA_OFF) + found = true; + } else { + if (type == HOLE_OFF) + found = true; + } + + if (found) { + *offset = lastoff; + break; + } + lastoff += bh->b_size; + } while ((bh = bh->b_this_page) != head); + + return found; +} + +/* + * This routine is called to find out and return a data or hole offset + * from the page cache for unwritten extents according to the desired + * type for xfs_seek_data() or xfs_seek_hole(). + * + * The argument offset is used to tell where we start to search from the + * page cache. Map is used to figure out the end points of the range to + * lookup pages. + * + * Return true if the desired type of offset was found, and the argument + * offset is filled with that address. Otherwise, return false and keep + * offset unchanged. + */ +STATIC bool +xfs_find_get_desired_pgoff( + struct inode *inode, + struct xfs_bmbt_irec *map, + unsigned int type, + loff_t *offset) +{ + struct xfs_inode *ip = XFS_I(inode); + struct xfs_mount *mp = ip->i_mount; + struct pagevec pvec; + pgoff_t index; + pgoff_t end; + loff_t endoff; + loff_t startoff = *offset; + loff_t lastoff = startoff; + bool found = false; + + pagevec_init(&pvec, 0); + + index = startoff >> PAGE_CACHE_SHIFT; + endoff = XFS_FSB_TO_B(mp, map->br_startoff + map->br_blockcount); + end = endoff >> PAGE_CACHE_SHIFT; + do { + int want; + unsigned nr_pages; + unsigned int i; + + want = min_t(pgoff_t, end - index, PAGEVEC_SIZE); + nr_pages = pagevec_lookup(&pvec, inode->i_mapping, index, + want); + /* + * No page mapped into given range. If we are searching holes + * and if this is the first time we got into the loop, it means + * that the given offset is landed in a hole, return it. + * + * If we have already stepped through some block buffers to find + * holes but they all contains data. In this case, the last + * offset is already updated and pointed to the end of the last + * mapped page, if it does not reach the endpoint to search, + * that means there should be a hole between them. + */ + if (nr_pages == 0) { + /* Data search found nothing */ + if (type == DATA_OFF) + break; + + ASSERT(type == HOLE_OFF); + if (lastoff == startoff || lastoff < endoff) { + found = true; + *offset = lastoff; + } + break; + } + + /* + * At lease we found one page. If this is the first time we + * step into the loop, and if the first page index offset is + * greater than the given search offset, a hole was found. + */ + if (type == HOLE_OFF && lastoff == startoff && + lastoff < page_offset(pvec.pages[0])) { + found = true; + break; + } + + for (i = 0; i < nr_pages; i++) { + struct page *page = pvec.pages[i]; + loff_t b_offset; + + /* + * At this point, the page may be truncated or + * invalidated (changing page->mapping to NULL), + * or even swizzled back from swapper_space to tmpfs + * file mapping. However, page->index will not change + * because we have a reference on the page. + * + * Searching done if the page index is out of range. + * If the current offset is not reaches the end of + * the specified search range, there should be a hole + * between them. + */ + if (page->index > end) { + if (type == HOLE_OFF && lastoff < endoff) { + *offset = lastoff; + found = true; + } + goto out; + } + + lock_page(page); + /* + * Page truncated or invalidated(page->mapping == NULL). + * We can freely skip it and proceed to check the next + * page. + */ + if (unlikely(page->mapping != inode->i_mapping)) { + unlock_page(page); + continue; + } + + if (!page_has_buffers(page)) { + unlock_page(page); + continue; + } + + found = xfs_lookup_buffer_offset(page, &b_offset, type); + if (found) { + /* + * The found offset may be less than the start + * point to search if this is the first time to + * come here. + */ + *offset = max_t(loff_t, startoff, b_offset); + unlock_page(page); + goto out; + } + + /* + * We either searching data but nothing was found, or + * searching hole but found a data buffer. In either + * case, probably the next page contains the desired + * things, update the last offset to it so. + */ + lastoff = page_offset(page) + PAGE_SIZE; + unlock_page(page); + } + + /* + * The number of returned pages less than our desired, search + * done. In this case, nothing was found for searching data, + * but we found a hole behind the last offset. + */ + if (nr_pages < want) { + if (type == HOLE_OFF) { + *offset = lastoff; + found = true; + } + break; + } + + index = pvec.pages[i - 1]->index + 1; + pagevec_release(&pvec); + } while (index <= end); + +out: + pagevec_release(&pvec); + return found; +} + STATIC loff_t xfs_seek_data( struct file *file, - loff_t start, - u32 type) + loff_t start) { struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; - struct xfs_bmbt_irec map[2]; - int nmap = 2; loff_t uninitialized_var(offset); xfs_fsize_t isize; xfs_fileoff_t fsbno; @@ -985,36 +1201,74 @@ xfs_seek_data( goto out_unlock; } - fsbno = XFS_B_TO_FSBT(mp, start); - /* * Try to read extents from the first block indicated * by fsbno to the end block of the file. */ + fsbno = XFS_B_TO_FSBT(mp, start); end = XFS_B_TO_FSB(mp, isize); + for (;;) { + struct xfs_bmbt_irec map[2]; + int nmap = 2; + unsigned int i; - error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap, - XFS_BMAPI_ENTIRE); - if (error) - goto out_unlock; + error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap, + XFS_BMAPI_ENTIRE); + if (error) + goto out_unlock; - /* - * Treat unwritten extent as data extent since it might - * contains dirty data in page cache. - */ - if (map[0].br_startblock != HOLESTARTBLOCK) { - offset = max_t(loff_t, start, - XFS_FSB_TO_B(mp, map[0].br_startoff)); - } else { + /* No extents at given offset, must be beyond EOF */ + if (nmap == 0) { + error = ENXIO; + goto out_unlock; + } + + for (i = 0; i < nmap; i++) { + offset = max_t(loff_t, start, + XFS_FSB_TO_B(mp, map[i].br_startoff)); + + /* Landed in a data extent */ + if (map[i].br_startblock == DELAYSTARTBLOCK || + (map[i].br_state == XFS_EXT_NORM && + !isnullstartblock(map[i].br_startblock))) + goto out; + + /* + * Landed in an unwritten extent, try to search data + * from page cache. + */ + if (map[i].br_state == XFS_EXT_UNWRITTEN) { + if (xfs_find_get_desired_pgoff(inode, &map[i], + DATA_OFF, &offset)) + goto out; + } + } + + /* + * map[0] is hole or its an unwritten extent but + * without data in page cache. Probably means that + * we are reading after EOF if nothing in map[1]. + */ if (nmap == 1) { error = ENXIO; goto out_unlock; } - offset = max_t(loff_t, start, - XFS_FSB_TO_B(mp, map[1].br_startoff)); + ASSERT(i > 1); + + /* + * Nothing was found, proceed to the next round of search + * if reading offset not beyond or hit EOF. + */ + fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount; + start = XFS_FSB_TO_B(mp, fsbno); + if (start >= isize) { + error = ENXIO; + goto out_unlock; + } } +out: if (offset != file->f_pos) file->f_pos = offset; @@ -1029,16 +1283,15 @@ out_unlock: STATIC loff_t xfs_seek_hole( struct file *file, - loff_t start, - u32 type) + loff_t start) { struct inode *inode = file->f_mapping->host; struct xfs_inode *ip = XFS_I(inode); struct xfs_mount *mp = ip->i_mount; loff_t uninitialized_var(offset); - loff_t holeoff; xfs_fsize_t isize; xfs_fileoff_t fsbno; + xfs_filblks_t end; uint lock; int error; @@ -1054,21 +1307,77 @@ xfs_seek_hole( } fsbno = XFS_B_TO_FSBT(mp, start); - error = xfs_bmap_first_unused(NULL, ip, 1, &fsbno, XFS_DATA_FORK); - if (error) - goto out_unlock; + end = XFS_B_TO_FSB(mp, isize); + + for (;;) { + struct xfs_bmbt_irec map[2]; + int nmap = 2; + unsigned int i; + + error = xfs_bmapi_read(ip, fsbno, end - fsbno, map, &nmap, + XFS_BMAPI_ENTIRE); + if (error) + goto out_unlock; + + /* No extents at given offset, must be beyond EOF */ + if (nmap == 0) { + error = ENXIO; + goto out_unlock; + } + + for (i = 0; i < nmap; i++) { + offset = max_t(loff_t, start, + XFS_FSB_TO_B(mp, map[i].br_startoff)); + + /* Landed in a hole */ + if (map[i].br_startblock == HOLESTARTBLOCK) + goto out; + + /* + * Landed in an unwritten extent, try to search hole + * from page cache. + */ + if (map[i].br_state == XFS_EXT_UNWRITTEN) { + if (xfs_find_get_desired_pgoff(inode, &map[i], + HOLE_OFF, &offset)) + goto out; + } + } + + /* + * map[0] contains data or its unwritten but contains + * data in page cache, probably means that we are + * reading after EOF. We should fix offset to point + * to the end of the file(i.e., there is an implicit + * hole at the end of any file). + */ + if (nmap == 1) { + offset = isize; + break; + } + + ASSERT(i > 1); - holeoff = XFS_FSB_TO_B(mp, fsbno); - if (holeoff <= start) - offset = start; - else { /* - * xfs_bmap_first_unused() could return a value bigger than - * isize if there are no more holes past the supplied offset. + * Both mappings contains data, proceed to the next round of + * search if the current reading offset not beyond or hit EOF. */ - offset = min_t(loff_t, holeoff, isize); + fsbno = map[i - 1].br_startoff + map[i - 1].br_blockcount; + start = XFS_FSB_TO_B(mp, fsbno); + if (start >= isize) { + offset = isize; + break; + } } +out: + /* + * At this point, we must have found a hole. However, the returned + * offset may be bigger than the file size as it may be aligned to + * page boundary for unwritten extents, we need to deal with this + * situation in particular. + */ + offset = min_t(loff_t, offset, isize); if (offset != file->f_pos) file->f_pos = offset; @@ -1092,9 +1401,9 @@ xfs_file_llseek( case SEEK_SET: return generic_file_llseek(file, offset, origin); case SEEK_DATA: - return xfs_seek_data(file, offset, origin); + return xfs_seek_data(file, offset); case SEEK_HOLE: - return xfs_seek_hole(file, offset, origin); + return xfs_seek_hole(file, offset); default: return -EINVAL; } diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c index 5aceb3f..445bf1a 100644 --- a/fs/xfs/xfs_ialloc.c +++ b/fs/xfs/xfs_ialloc.c @@ -431,7 +431,7 @@ xfs_ialloc_next_ag( spin_lock(&mp->m_agirotor_lock); agno = mp->m_agirotor; - if (++mp->m_agirotor == mp->m_maxagi) + if (++mp->m_agirotor >= mp->m_maxagi) mp->m_agirotor = 0; spin_unlock(&mp->m_agirotor_lock); diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 0e0232c..8305f2a 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -70,16 +70,16 @@ xfs_find_handle( int hsize; xfs_handle_t handle; struct inode *inode; - struct file *file = NULL; + struct fd f; struct path path; int error; struct xfs_inode *ip; if (cmd == XFS_IOC_FD_TO_HANDLE) { - file = fget(hreq->fd); - if (!file) + f = fdget(hreq->fd); + if (!f.file) return -EBADF; - inode = file->f_path.dentry->d_inode; + inode = f.file->f_path.dentry->d_inode; } else { error = user_lpath((const char __user *)hreq->path, &path); if (error) @@ -134,7 +134,7 @@ xfs_find_handle( out_put: if (cmd == XFS_IOC_FD_TO_HANDLE) - fput(file); + fdput(f); else path_put(&path); return error; diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c index 29c2f83..b2bd3a0 100644 --- a/fs/xfs/xfs_mount.c +++ b/fs/xfs/xfs_mount.c @@ -440,7 +440,7 @@ xfs_initialize_perag( xfs_agnumber_t agcount, xfs_agnumber_t *maxagi) { - xfs_agnumber_t index, max_metadata; + xfs_agnumber_t index; xfs_agnumber_t first_initialised = 0; xfs_perag_t *pag; xfs_agino_t agino; @@ -500,43 +500,10 @@ xfs_initialize_perag( else mp->m_flags &= ~XFS_MOUNT_32BITINODES; - if (mp->m_flags & XFS_MOUNT_32BITINODES) { - /* - * Calculate how much should be reserved for inodes to meet - * the max inode percentage. - */ - if (mp->m_maxicount) { - __uint64_t icount; - - icount = sbp->sb_dblocks * sbp->sb_imax_pct; - do_div(icount, 100); - icount += sbp->sb_agblocks - 1; - do_div(icount, sbp->sb_agblocks); - max_metadata = icount; - } else { - max_metadata = agcount; - } - - for (index = 0; index < agcount; index++) { - ino = XFS_AGINO_TO_INO(mp, index, agino); - if (ino > XFS_MAXINUMBER_32) { - index++; - break; - } - - pag = xfs_perag_get(mp, index); - pag->pagi_inodeok = 1; - if (index < max_metadata) - pag->pagf_metadata = 1; - xfs_perag_put(pag); - } - } else { - for (index = 0; index < agcount; index++) { - pag = xfs_perag_get(mp, index); - pag->pagi_inodeok = 1; - xfs_perag_put(pag); - } - } + if (mp->m_flags & XFS_MOUNT_32BITINODES) + index = xfs_set_inode32(mp); + else + index = xfs_set_inode64(mp); if (maxagi) *maxagi = index; diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h index 05a05a7..deee09e 100644 --- a/fs/xfs/xfs_mount.h +++ b/fs/xfs/xfs_mount.h @@ -54,12 +54,7 @@ typedef struct xfs_trans_reservations { #include "xfs_sync.h" struct xlog; -struct xfs_mount_args; struct xfs_inode; -struct xfs_bmbt_irec; -struct xfs_bmap_free; -struct xfs_extdelta; -struct xfs_swapext; struct xfs_mru_cache; struct xfs_nameops; struct xfs_ail; diff --git a/fs/xfs/xfs_super.c b/fs/xfs/xfs_super.c index 001537f..26a09bd 100644 --- a/fs/xfs/xfs_super.c +++ b/fs/xfs/xfs_super.c @@ -88,6 +88,8 @@ mempool_t *xfs_ioend_pool; * unwritten extent conversion */ #define MNTOPT_NOBARRIER "nobarrier" /* .. disable */ #define MNTOPT_64BITINODE "inode64" /* inodes can be allocated anywhere */ +#define MNTOPT_32BITINODE "inode32" /* inode allocation limited to + * XFS_MAXINUMBER_32 */ #define MNTOPT_IKEEP "ikeep" /* do not free empty inode clusters */ #define MNTOPT_NOIKEEP "noikeep" /* free empty inode clusters */ #define MNTOPT_LARGEIO "largeio" /* report large I/O sizes in stat() */ @@ -120,12 +122,18 @@ mempool_t *xfs_ioend_pool; * in the future, too. */ enum { - Opt_barrier, Opt_nobarrier, Opt_err + Opt_barrier, + Opt_nobarrier, + Opt_inode64, + Opt_inode32, + Opt_err }; static const match_table_t tokens = { {Opt_barrier, "barrier"}, {Opt_nobarrier, "nobarrier"}, + {Opt_inode64, "inode64"}, + {Opt_inode32, "inode32"}, {Opt_err, NULL} }; @@ -197,7 +205,9 @@ xfs_parseargs( */ mp->m_flags |= XFS_MOUNT_BARRIER; mp->m_flags |= XFS_MOUNT_COMPAT_IOSIZE; +#if !XFS_BIG_INUMS mp->m_flags |= XFS_MOUNT_SMALL_INUMS; +#endif /* * These can be overridden by the mount option parsing. @@ -294,6 +304,8 @@ xfs_parseargs( return EINVAL; } dswidth = simple_strtoul(value, &eov, 10); + } else if (!strcmp(this_char, MNTOPT_32BITINODE)) { + mp->m_flags |= XFS_MOUNT_SMALL_INUMS; } else if (!strcmp(this_char, MNTOPT_64BITINODE)) { mp->m_flags &= ~XFS_MOUNT_SMALL_INUMS; #if !XFS_BIG_INUMS @@ -492,6 +504,7 @@ xfs_showargs( { XFS_MOUNT_FILESTREAMS, "," MNTOPT_FILESTREAM }, { XFS_MOUNT_GRPID, "," MNTOPT_GRPID }, { XFS_MOUNT_DISCARD, "," MNTOPT_DISCARD }, + { XFS_MOUNT_SMALL_INUMS, "," MNTOPT_32BITINODE }, { 0, NULL } }; static struct proc_xfs_info xfs_info_unset[] = { @@ -591,6 +604,80 @@ xfs_max_file_offset( return (((__uint64_t)pagefactor) << bitshift) - 1; } +xfs_agnumber_t +xfs_set_inode32(struct xfs_mount *mp) +{ + xfs_agnumber_t index = 0; + xfs_agnumber_t maxagi = 0; + xfs_sb_t *sbp = &mp->m_sb; + xfs_agnumber_t max_metadata; + xfs_agino_t agino = XFS_OFFBNO_TO_AGINO(mp, sbp->sb_agblocks -1, 0); + xfs_ino_t ino = XFS_AGINO_TO_INO(mp, sbp->sb_agcount -1, agino); + xfs_perag_t *pag; + + /* Calculate how much should be reserved for inodes to meet + * the max inode percentage. + */ + if (mp->m_maxicount) { + __uint64_t icount; + + icount = sbp->sb_dblocks * sbp->sb_imax_pct; + do_div(icount, 100); + icount += sbp->sb_agblocks - 1; + do_div(icount, sbp->sb_agblocks); + max_metadata = icount; + } else { + max_metadata = sbp->sb_agcount; + } + + for (index = 0; index < sbp->sb_agcount; index++) { + ino = XFS_AGINO_TO_INO(mp, index, agino); + + if (ino > XFS_MAXINUMBER_32) { + pag = xfs_perag_get(mp, index); + pag->pagi_inodeok = 0; + pag->pagf_metadata = 0; + xfs_perag_put(pag); + continue; + } + + pag = xfs_perag_get(mp, index); + pag->pagi_inodeok = 1; + maxagi++; + if (index < max_metadata) + pag->pagf_metadata = 1; + xfs_perag_put(pag); + } + mp->m_flags |= (XFS_MOUNT_32BITINODES | + XFS_MOUNT_SMALL_INUMS); + + return maxagi; +} + +xfs_agnumber_t +xfs_set_inode64(struct xfs_mount *mp) +{ + xfs_agnumber_t index = 0; + + for (index = 0; index < mp->m_sb.sb_agcount; index++) { + struct xfs_perag *pag; + + pag = xfs_perag_get(mp, index); + pag->pagi_inodeok = 1; + pag->pagf_metadata = 0; + xfs_perag_put(pag); + } + + /* There is no need for lock protection on m_flags, + * the rw_semaphore of the VFS superblock is locked + * during mount/umount/remount operations, so this is + * enough to avoid concurency on the m_flags field + */ + mp->m_flags &= ~(XFS_MOUNT_32BITINODES | + XFS_MOUNT_SMALL_INUMS); + return index; +} + STATIC int xfs_blkdev_get( xfs_mount_t *mp, @@ -1056,6 +1143,12 @@ xfs_fs_remount( case Opt_nobarrier: mp->m_flags &= ~XFS_MOUNT_BARRIER; break; + case Opt_inode64: + mp->m_maxagi = xfs_set_inode64(mp); + break; + case Opt_inode32: + mp->m_maxagi = xfs_set_inode32(mp); + break; default: /* * Logically we would return an error here to prevent @@ -1506,6 +1599,11 @@ xfs_init_zones(void) STATIC void xfs_destroy_zones(void) { + /* + * Make sure all delayed rcu free are flushed before we + * destroy caches. + */ + rcu_barrier(); kmem_zone_destroy(xfs_ili_zone); kmem_zone_destroy(xfs_inode_zone); kmem_zone_destroy(xfs_efi_zone); diff --git a/fs/xfs/xfs_super.h b/fs/xfs/xfs_super.h index 09b0c26..9de4a92 100644 --- a/fs/xfs/xfs_super.h +++ b/fs/xfs/xfs_super.h @@ -75,6 +75,8 @@ struct block_device; extern __uint64_t xfs_max_file_offset(unsigned int); extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); +extern xfs_agnumber_t xfs_set_inode32(struct xfs_mount *); +extern xfs_agnumber_t xfs_set_inode64(struct xfs_mount *); extern const struct export_operations xfs_export_operations; extern const struct xattr_handler *xfs_xattr_handlers[]; diff --git a/fs/xfs/xfs_trace.h b/fs/xfs/xfs_trace.h index e5795dd..7d36ccf 100644 --- a/fs/xfs/xfs_trace.h +++ b/fs/xfs/xfs_trace.h @@ -37,6 +37,7 @@ struct xlog_recover; struct xlog_recover_item; struct xfs_buf_log_format; struct xfs_inode_log_format; +struct xfs_bmbt_irec; DECLARE_EVENT_CLASS(xfs_attr_list_class, TP_PROTO(struct xfs_attr_list_context *ctx), diff --git a/include/acpi/processor.h b/include/acpi/processor.h index 64ec644..555d033 100644 --- a/include/acpi/processor.h +++ b/include/acpi/processor.h @@ -3,7 +3,6 @@ #include <linux/kernel.h> #include <linux/cpu.h> -#include <linux/cpuidle.h> #include <linux/thermal.h> #include <asm/acpi.h> @@ -59,13 +58,11 @@ struct acpi_processor_cx { u8 entry_method; u8 index; u32 latency; - u32 power; u8 bm_sts_skip; char desc[ACPI_CX_DESC_LEN]; }; struct acpi_processor_power { - struct cpuidle_device dev; struct acpi_processor_cx *state; unsigned long bm_check_timestamp; u32 default_state; @@ -325,12 +322,10 @@ extern void acpi_processor_reevaluate_tstate(struct acpi_processor *pr, extern const struct file_operations acpi_processor_throttling_fops; extern void acpi_processor_throttling_init(void); /* in processor_idle.c */ -int acpi_processor_power_init(struct acpi_processor *pr, - struct acpi_device *device); +int acpi_processor_power_init(struct acpi_processor *pr); +int acpi_processor_power_exit(struct acpi_processor *pr); int acpi_processor_cst_has_changed(struct acpi_processor *pr); int acpi_processor_hotplug(struct acpi_processor *pr); -int acpi_processor_power_exit(struct acpi_processor *pr, - struct acpi_device *device); int acpi_processor_suspend(struct device *dev); int acpi_processor_resume(struct device *dev); extern struct cpuidle_driver acpi_idle_driver; diff --git a/include/linux/ata.h b/include/linux/ata.h index 5713d3a..408da95 100644 --- a/include/linux/ata.h +++ b/include/linux/ata.h @@ -77,6 +77,9 @@ enum { ATA_ID_EIDE_PIO_IORDY = 68, ATA_ID_ADDITIONAL_SUPP = 69, ATA_ID_QUEUE_DEPTH = 75, + ATA_ID_SATA_CAPABILITY = 76, + ATA_ID_SATA_CAPABILITY_2 = 77, + ATA_ID_FEATURE_SUPP = 78, ATA_ID_MAJOR_VER = 80, ATA_ID_COMMAND_SET_1 = 82, ATA_ID_COMMAND_SET_2 = 83, @@ -292,6 +295,13 @@ enum { /* READ_LOG_EXT pages */ ATA_LOG_SATA_NCQ = 0x10, + ATA_LOG_SATA_ID_DEV_DATA = 0x30, + ATA_LOG_SATA_SETTINGS = 0x08, + ATA_LOG_DEVSLP_MDAT = 0x30, + ATA_LOG_DEVSLP_MDAT_MASK = 0x1F, + ATA_LOG_DEVSLP_DETO = 0x31, + ATA_LOG_DEVSLP_VALID = 0x37, + ATA_LOG_DEVSLP_VALID_MASK = 0x80, /* READ/WRITE LONG (obsolete) */ ATA_CMD_READ_LONG = 0x22, @@ -345,6 +355,7 @@ enum { SATA_FPDMA_IN_ORDER = 0x04, /* FPDMA in-order data delivery */ SATA_AN = 0x05, /* Asynchronous Notification */ SATA_SSP = 0x06, /* Software Settings Preservation */ + SATA_DEVSLP = 0x09, /* Device Sleep */ /* feature values for SET_MAX */ ATA_SET_MAX_ADDR = 0x00, @@ -558,15 +569,17 @@ static inline int ata_is_data(u8 prot) #define ata_id_is_ata(id) (((id)[ATA_ID_CONFIG] & (1 << 15)) == 0) #define ata_id_has_lba(id) ((id)[ATA_ID_CAPABILITY] & (1 << 9)) #define ata_id_has_dma(id) ((id)[ATA_ID_CAPABILITY] & (1 << 8)) -#define ata_id_has_ncq(id) ((id)[76] & (1 << 8)) +#define ata_id_has_ncq(id) ((id)[ATA_ID_SATA_CAPABILITY] & (1 << 8)) #define ata_id_queue_depth(id) (((id)[ATA_ID_QUEUE_DEPTH] & 0x1f) + 1) #define ata_id_removeable(id) ((id)[ATA_ID_CONFIG] & (1 << 7)) #define ata_id_has_atapi_AN(id) \ - ( (((id)[76] != 0x0000) && ((id)[76] != 0xffff)) && \ - ((id)[78] & (1 << 5)) ) + ((((id)[ATA_ID_SATA_CAPABILITY] != 0x0000) && \ + ((id)[ATA_ID_SATA_CAPABILITY] != 0xffff)) && \ + ((id)[ATA_ID_FEATURE_SUPP] & (1 << 5))) #define ata_id_has_fpdma_aa(id) \ - ( (((id)[76] != 0x0000) && ((id)[76] != 0xffff)) && \ - ((id)[78] & (1 << 2)) ) + ((((id)[ATA_ID_SATA_CAPABILITY] != 0x0000) && \ + ((id)[ATA_ID_SATA_CAPABILITY] != 0xffff)) && \ + ((id)[ATA_ID_FEATURE_SUPP] & (1 << 2))) #define ata_id_iordy_disable(id) ((id)[ATA_ID_CAPABILITY] & (1 << 10)) #define ata_id_has_iordy(id) ((id)[ATA_ID_CAPABILITY] & (1 << 11)) #define ata_id_u32(id,n) \ @@ -578,11 +591,12 @@ static inline int ata_is_data(u8 prot) ((u64) (id)[(n) + 0]) ) #define ata_id_cdb_intr(id) (((id)[ATA_ID_CONFIG] & 0x60) == 0x20) -#define ata_id_has_da(id) ((id)[77] & (1 << 4)) +#define ata_id_has_da(id) ((id)[ATA_ID_SATA_CAPABILITY_2] & (1 << 4)) +#define ata_id_has_devslp(id) ((id)[ATA_ID_FEATURE_SUPP] & (1 << 8)) static inline bool ata_id_has_hipm(const u16 *id) { - u16 val = id[76]; + u16 val = id[ATA_ID_SATA_CAPABILITY]; if (val == 0 || val == 0xffff) return false; @@ -592,7 +606,7 @@ static inline bool ata_id_has_hipm(const u16 *id) static inline bool ata_id_has_dipm(const u16 *id) { - u16 val = id[78]; + u16 val = id[ATA_ID_FEATURE_SUPP]; if (val == 0 || val == 0xffff) return false; diff --git a/include/linux/audit.h b/include/linux/audit.h index 12367cb..e7c836d 100644 --- a/include/linux/audit.h +++ b/include/linux/audit.h @@ -531,6 +531,7 @@ extern int audit_set_loginuid(kuid_t loginuid); #define audit_get_loginuid(t) ((t)->loginuid) #define audit_get_sessionid(t) ((t)->sessionid) extern void audit_log_task_context(struct audit_buffer *ab); +extern void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk); extern void __audit_ipc_obj(struct kern_ipc_perm *ipcp); extern void __audit_ipc_set_perm(unsigned long qbytes, uid_t uid, gid_t gid, umode_t mode); extern int __audit_bprm(struct linux_binprm *bprm); @@ -642,6 +643,7 @@ extern int audit_signals; #define audit_get_loginuid(t) (INVALID_UID) #define audit_get_sessionid(t) (-1) #define audit_log_task_context(b) do { ; } while (0) +#define audit_log_task_info(b, t) do { ; } while (0) #define audit_ipc_obj(i) ((void)0) #define audit_ipc_set_perm(q,u,g,m) ((void)0) #define audit_bprm(p) ({ 0; }) diff --git a/include/linux/clockchips.h b/include/linux/clockchips.h index acba8943..8a7096f 100644 --- a/include/linux/clockchips.h +++ b/include/linux/clockchips.h @@ -97,6 +97,8 @@ struct clock_event_device { void (*broadcast)(const struct cpumask *mask); void (*set_mode)(enum clock_event_mode mode, struct clock_event_device *); + void (*suspend)(struct clock_event_device *); + void (*resume)(struct clock_event_device *); unsigned long min_delta_ticks; unsigned long max_delta_ticks; @@ -156,6 +158,9 @@ clockevents_calc_mult_shift(struct clock_event_device *ce, u32 freq, u32 minsec) freq, minsec); } +extern void clockevents_suspend(void); +extern void clockevents_resume(void); + #ifdef CONFIG_GENERIC_CLOCKEVENTS extern void clockevents_notify(unsigned long reason, void *arg); #else @@ -164,6 +169,9 @@ extern void clockevents_notify(unsigned long reason, void *arg); #else /* CONFIG_GENERIC_CLOCKEVENTS_BUILD */ +static inline void clockevents_suspend(void) {} +static inline void clockevents_resume(void) {} + #define clockevents_notify(reason, arg) do { } while (0) #endif diff --git a/include/linux/compat.h b/include/linux/compat.h index 09b28b7..fd4e2995 100644 --- a/include/linux/compat.h +++ b/include/linux/compat.h @@ -590,6 +590,9 @@ asmlinkage ssize_t compat_sys_process_vm_writev(compat_pid_t pid, unsigned long liovcnt, const struct compat_iovec __user *rvec, unsigned long riovcnt, unsigned long flags); +asmlinkage long compat_sys_sendfile(int out_fd, int in_fd, + compat_off_t __user *offset, compat_size_t count); + #else #define is_compat_task() (0) diff --git a/include/linux/device.h b/include/linux/device.h index af92883..86ef6ab 100644 --- a/include/linux/device.h +++ b/include/linux/device.h @@ -776,6 +776,13 @@ static inline void pm_suspend_ignore_children(struct device *dev, bool enable) dev->power.ignore_children = enable; } +static inline void dev_pm_syscore_device(struct device *dev, bool val) +{ +#ifdef CONFIG_PM_SLEEP + dev->power.syscore = val; +#endif +} + static inline void device_lock(struct device *dev) { mutex_lock(&dev->mutex); diff --git a/include/linux/fdtable.h b/include/linux/fdtable.h index 158a41e..45052aa 100644 --- a/include/linux/fdtable.h +++ b/include/linux/fdtable.h @@ -30,31 +30,11 @@ struct fdtable { struct fdtable *next; }; -static inline void __set_close_on_exec(int fd, struct fdtable *fdt) -{ - __set_bit(fd, fdt->close_on_exec); -} - -static inline void __clear_close_on_exec(int fd, struct fdtable *fdt) -{ - __clear_bit(fd, fdt->close_on_exec); -} - static inline bool close_on_exec(int fd, const struct fdtable *fdt) { return test_bit(fd, fdt->close_on_exec); } -static inline void __set_open_fd(int fd, struct fdtable *fdt) -{ - __set_bit(fd, fdt->open_fds); -} - -static inline void __clear_open_fd(int fd, struct fdtable *fdt) -{ - __clear_bit(fd, fdt->open_fds); -} - static inline bool fd_is_open(int fd, const struct fdtable *fdt) { return test_bit(fd, fdt->open_fds); @@ -93,15 +73,8 @@ struct file_operations; struct vfsmount; struct dentry; -extern int expand_files(struct files_struct *, int nr); -extern void free_fdtable_rcu(struct rcu_head *rcu); extern void __init files_defer_init(void); -static inline void free_fdtable(struct fdtable *fdt) -{ - call_rcu(&fdt->rcu, free_fdtable_rcu); -} - static inline struct file * fcheck_files(struct files_struct *files, unsigned int fd) { struct file * file = NULL; @@ -122,8 +95,20 @@ struct task_struct; struct files_struct *get_files_struct(struct task_struct *); void put_files_struct(struct files_struct *fs); void reset_files_struct(struct files_struct *); +void daemonize_descriptors(void); int unshare_files(struct files_struct **); struct files_struct *dup_fd(struct files_struct *, int *); +void do_close_on_exec(struct files_struct *); +int iterate_fd(struct files_struct *, unsigned, + int (*)(const void *, struct file *, unsigned), + const void *); + +extern int __alloc_fd(struct files_struct *files, + unsigned start, unsigned end, unsigned flags); +extern void __fd_install(struct files_struct *files, + unsigned int fd, struct file *file); +extern int __close_fd(struct files_struct *files, + unsigned int fd); extern struct kmem_cache *files_cachep; diff --git a/include/linux/file.h b/include/linux/file.h index a22408b..cbacf4f 100644 --- a/include/linux/file.h +++ b/include/linux/file.h @@ -26,15 +26,44 @@ static inline void fput_light(struct file *file, int fput_needed) fput(file); } +struct fd { + struct file *file; + int need_put; +}; + +static inline void fdput(struct fd fd) +{ + if (fd.need_put) + fput(fd.file); +} + extern struct file *fget(unsigned int fd); extern struct file *fget_light(unsigned int fd, int *fput_needed); + +static inline struct fd fdget(unsigned int fd) +{ + int b; + struct file *f = fget_light(fd, &b); + return (struct fd){f,b}; +} + extern struct file *fget_raw(unsigned int fd); extern struct file *fget_raw_light(unsigned int fd, int *fput_needed); + +static inline struct fd fdget_raw(unsigned int fd) +{ + int b; + struct file *f = fget_raw_light(fd, &b); + return (struct fd){f,b}; +} + +extern int f_dupfd(unsigned int from, struct file *file, unsigned flags); +extern int replace_fd(unsigned fd, struct file *file, unsigned flags); extern void set_close_on_exec(unsigned int fd, int flag); +extern bool get_close_on_exec(unsigned int fd); extern void put_filp(struct file *); -extern int alloc_fd(unsigned start, unsigned flags); -extern int get_unused_fd(void); -#define get_unused_fd_flags(flags) alloc_fd(0, (flags)) +extern int get_unused_fd_flags(unsigned flags); +#define get_unused_fd() get_unused_fd_flags(0) extern void put_unused_fd(unsigned int fd); extern void fd_install(unsigned int fd, struct file *file); diff --git a/include/linux/frontswap.h b/include/linux/frontswap.h index 0e4e2ee..3044254 100644 --- a/include/linux/frontswap.h +++ b/include/linux/frontswap.h @@ -19,6 +19,8 @@ extern struct frontswap_ops extern void frontswap_shrink(unsigned long); extern unsigned long frontswap_curr_pages(void); extern void frontswap_writethrough(bool); +#define FRONTSWAP_HAS_EXCLUSIVE_GETS +extern void frontswap_tmem_exclusive_gets(bool); extern void __frontswap_init(unsigned type); extern int __frontswap_store(struct page *page); diff --git a/include/linux/fs.h b/include/linux/fs.h index aa11047..ca6d8c8 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -1074,7 +1074,11 @@ struct file_handle { unsigned char f_handle[0]; }; -#define get_file(x) atomic_long_inc(&(x)->f_count) +static inline struct file *get_file(struct file *f) +{ + atomic_long_inc(&f->f_count); + return f; +} #define fput_atomic(x) atomic_long_add_unless(&(x)->f_count, -1, 1) #define file_count(x) atomic_long_read(&(x)->f_count) @@ -1126,9 +1130,9 @@ static inline int file_check_writeable(struct file *filp) /* Page cache limit. The filesystems should put that into their s_maxbytes limits, otherwise bad things can happen in VM. */ #if BITS_PER_LONG==32 -#define MAX_LFS_FILESIZE (((u64)PAGE_CACHE_SIZE << (BITS_PER_LONG-1))-1) +#define MAX_LFS_FILESIZE (((loff_t)PAGE_CACHE_SIZE << (BITS_PER_LONG-1))-1) #elif BITS_PER_LONG==64 -#define MAX_LFS_FILESIZE 0x7fffffffffffffffUL +#define MAX_LFS_FILESIZE ((loff_t)0x7fffffffffffffff) #endif #define FL_POSIX 1 diff --git a/include/linux/ima.h b/include/linux/ima.h index 6ac8e50..2c7223d 100644 --- a/include/linux/ima.h +++ b/include/linux/ima.h @@ -39,5 +39,32 @@ static inline int ima_file_mmap(struct file *file, unsigned long prot) { return 0; } + #endif /* CONFIG_IMA_H */ + +#ifdef CONFIG_IMA_APPRAISE +extern void ima_inode_post_setattr(struct dentry *dentry); +extern int ima_inode_setxattr(struct dentry *dentry, const char *xattr_name, + const void *xattr_value, size_t xattr_value_len); +extern int ima_inode_removexattr(struct dentry *dentry, const char *xattr_name); +#else +static inline void ima_inode_post_setattr(struct dentry *dentry) +{ + return; +} + +static inline int ima_inode_setxattr(struct dentry *dentry, + const char *xattr_name, + const void *xattr_value, + size_t xattr_value_len) +{ + return 0; +} + +static inline int ima_inode_removexattr(struct dentry *dentry, + const char *xattr_name) +{ + return 0; +} +#endif /* CONFIG_IMA_APPRAISE_H */ #endif /* _LINUX_IMA_H */ diff --git a/include/linux/integrity.h b/include/linux/integrity.h index a0c4125..66c5fe9 100644 --- a/include/linux/integrity.h +++ b/include/linux/integrity.h @@ -22,13 +22,14 @@ enum integrity_status { /* List of EVM protected security xattrs */ #ifdef CONFIG_INTEGRITY -extern int integrity_inode_alloc(struct inode *inode); +extern struct integrity_iint_cache *integrity_inode_get(struct inode *inode); extern void integrity_inode_free(struct inode *inode); #else -static inline int integrity_inode_alloc(struct inode *inode) +static inline struct integrity_iint_cache * + integrity_inode_get(struct inode *inode) { - return 0; + return NULL; } static inline void integrity_inode_free(struct inode *inode) diff --git a/include/linux/libata.h b/include/linux/libata.h index 64f90e1..77eeeda 100644 --- a/include/linux/libata.h +++ b/include/linux/libata.h @@ -162,6 +162,7 @@ enum { ATA_DFLAG_DETACHED = (1 << 25), ATA_DFLAG_DA = (1 << 26), /* device supports Device Attention */ + ATA_DFLAG_DEVSLP = (1 << 27), /* device supports Device Sleep */ ATA_DEV_UNKNOWN = 0, /* unknown device */ ATA_DEV_ATA = 1, /* ATA device */ @@ -184,6 +185,7 @@ enum { ATA_LFLAG_DISABLED = (1 << 6), /* link is disabled */ ATA_LFLAG_SW_ACTIVITY = (1 << 7), /* keep activity stats */ ATA_LFLAG_NO_LPM = (1 << 8), /* disable LPM on this link */ + ATA_LFLAG_RST_ONCE = (1 << 9), /* limit recovery to one reset */ /* struct ata_port flags */ ATA_FLAG_SLAVE_POSS = (1 << 0), /* host supports slave dev */ @@ -649,6 +651,9 @@ struct ata_device { u32 gscr[SATA_PMP_GSCR_DWORDS]; /* PMP GSCR block */ }; + /* Identify Device Data Log (30h), SATA Settings (page 08h) */ + u8 sata_settings[ATA_SECT_SIZE]; + /* error history */ int spdn_cnt; /* ering is CLEAR_END, read comment above CLEAR_END */ @@ -986,8 +991,7 @@ extern int ata_host_activate(struct ata_host *host, int irq, irq_handler_t irq_handler, unsigned long irq_flags, struct scsi_host_template *sht); extern void ata_host_detach(struct ata_host *host); -extern void ata_host_init(struct ata_host *, struct device *, - unsigned long, struct ata_port_operations *); +extern void ata_host_init(struct ata_host *, struct device *, struct ata_port_operations *); extern int ata_scsi_detect(struct scsi_host_template *sht); extern int ata_scsi_ioctl(struct scsi_device *dev, int cmd, void __user *arg); extern int ata_scsi_queuecmd(struct Scsi_Host *h, struct scsi_cmnd *cmd); @@ -1012,6 +1016,17 @@ extern bool ata_link_offline(struct ata_link *link); #ifdef CONFIG_PM extern int ata_host_suspend(struct ata_host *host, pm_message_t mesg); extern void ata_host_resume(struct ata_host *host); +extern int ata_sas_port_async_suspend(struct ata_port *ap, int *async); +extern int ata_sas_port_async_resume(struct ata_port *ap, int *async); +#else +static inline int ata_sas_port_async_suspend(struct ata_port *ap, int *async) +{ + return 0; +} +static inline int ata_sas_port_async_resume(struct ata_port *ap, int *async) +{ + return 0; +} #endif extern int ata_ratelimit(void); extern void ata_msleep(struct ata_port *ap, unsigned int msecs); diff --git a/include/linux/mtd/partitions.h b/include/linux/mtd/partitions.h index 2475228..1f8d24b 100644 --- a/include/linux/mtd/partitions.h +++ b/include/linux/mtd/partitions.h @@ -79,9 +79,10 @@ struct mtd_part_parser { extern int register_mtd_parser(struct mtd_part_parser *parser); extern int deregister_mtd_parser(struct mtd_part_parser *parser); -int mtd_is_partition(struct mtd_info *mtd); +int mtd_is_partition(const struct mtd_info *mtd); int mtd_add_partition(struct mtd_info *master, char *name, long long offset, long long length); int mtd_del_partition(struct mtd_info *master, int partno); +uint64_t mtd_get_device_size(const struct mtd_info *mtd); #endif diff --git a/include/linux/net.h b/include/linux/net.h index 99276c3..6ab31ca 100644 --- a/include/linux/net.h +++ b/include/linux/net.h @@ -65,6 +65,7 @@ typedef enum { struct poll_table_struct; struct pipe_inode_info; struct inode; +struct file; struct net; #define SOCK_ASYNC_NOSPACE 0 @@ -246,7 +247,7 @@ extern int sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t len); extern int sock_recvmsg(struct socket *sock, struct msghdr *msg, size_t size, int flags); -extern int sock_map_fd(struct socket *sock, int flags); +extern struct file *sock_alloc_file(struct socket *sock, int flags, const char *dname); extern struct socket *sockfd_lookup(int fd, int *err); extern struct socket *sock_from_file(struct file *file, int *err); #define sockfd_put(sock) fput(sock->file) diff --git a/include/linux/opp.h b/include/linux/opp.h index 2a4e5fa..214e0ebc 100644 --- a/include/linux/opp.h +++ b/include/linux/opp.h @@ -48,6 +48,14 @@ int opp_disable(struct device *dev, unsigned long freq); struct srcu_notifier_head *opp_get_notifier(struct device *dev); +#ifdef CONFIG_OF +int of_init_opp_table(struct device *dev); +#else +static inline int of_init_opp_table(struct device *dev) +{ + return -EINVAL; +} +#endif /* CONFIG_OF */ #else static inline unsigned long opp_get_voltage(struct opp *opp) { diff --git a/include/linux/pci_ids.h b/include/linux/pci_ids.h index 8d3c427..33880f6 100644 --- a/include/linux/pci_ids.h +++ b/include/linux/pci_ids.h @@ -753,6 +753,7 @@ #define PCI_DEVICE_ID_HP_CISSD 0x3238 #define PCI_DEVICE_ID_HP_CISSE 0x323a #define PCI_DEVICE_ID_HP_CISSF 0x323b +#define PCI_DEVICE_ID_HP_CISSH 0x323c #define PCI_DEVICE_ID_HP_ZX2_IOC 0x4031 #define PCI_VENDOR_ID_PCTECH 0x1042 diff --git a/include/linux/pm.h b/include/linux/pm.h index 88f034a..007e687 100644 --- a/include/linux/pm.h +++ b/include/linux/pm.h @@ -510,12 +510,14 @@ struct dev_pm_info { bool is_prepared:1; /* Owned by the PM core */ bool is_suspended:1; /* Ditto */ bool ignore_children:1; + bool early_init:1; /* Owned by the PM core */ spinlock_t lock; #ifdef CONFIG_PM_SLEEP struct list_head entry; struct completion completion; struct wakeup_source *wakeup; bool wakeup_path:1; + bool syscore:1; #else unsigned int should_wakeup:1; #endif diff --git a/include/linux/pm_domain.h b/include/linux/pm_domain.h index a7d6172..7c1d252 100644 --- a/include/linux/pm_domain.h +++ b/include/linux/pm_domain.h @@ -114,7 +114,6 @@ struct generic_pm_domain_data { struct mutex lock; unsigned int refcount; bool need_restore; - bool always_on; }; #ifdef CONFIG_PM_GENERIC_DOMAINS @@ -139,36 +138,32 @@ extern int __pm_genpd_of_add_device(struct device_node *genpd_node, struct device *dev, struct gpd_timing_data *td); -static inline int pm_genpd_add_device(struct generic_pm_domain *genpd, - struct device *dev) -{ - return __pm_genpd_add_device(genpd, dev, NULL); -} - -static inline int pm_genpd_of_add_device(struct device_node *genpd_node, - struct device *dev) -{ - return __pm_genpd_of_add_device(genpd_node, dev, NULL); -} +extern int __pm_genpd_name_add_device(const char *domain_name, + struct device *dev, + struct gpd_timing_data *td); extern int pm_genpd_remove_device(struct generic_pm_domain *genpd, struct device *dev); -extern void pm_genpd_dev_always_on(struct device *dev, bool val); extern void pm_genpd_dev_need_restore(struct device *dev, bool val); extern int pm_genpd_add_subdomain(struct generic_pm_domain *genpd, struct generic_pm_domain *new_subdomain); +extern int pm_genpd_add_subdomain_names(const char *master_name, + const char *subdomain_name); extern int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd, struct generic_pm_domain *target); extern int pm_genpd_add_callbacks(struct device *dev, struct gpd_dev_ops *ops, struct gpd_timing_data *td); extern int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td); -extern int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state); -extern int genpd_detach_cpuidle(struct generic_pm_domain *genpd); +extern int pm_genpd_attach_cpuidle(struct generic_pm_domain *genpd, int state); +extern int pm_genpd_name_attach_cpuidle(const char *name, int state); +extern int pm_genpd_detach_cpuidle(struct generic_pm_domain *genpd); +extern int pm_genpd_name_detach_cpuidle(const char *name); extern void pm_genpd_init(struct generic_pm_domain *genpd, struct dev_power_governor *gov, bool is_off); extern int pm_genpd_poweron(struct generic_pm_domain *genpd); +extern int pm_genpd_name_poweron(const char *domain_name); extern bool default_stop_ok(struct device *dev); @@ -189,8 +184,15 @@ static inline int __pm_genpd_add_device(struct generic_pm_domain *genpd, { return -ENOSYS; } -static inline int pm_genpd_add_device(struct generic_pm_domain *genpd, - struct device *dev) +static inline int __pm_genpd_of_add_device(struct device_node *genpd_node, + struct device *dev, + struct gpd_timing_data *td) +{ + return -ENOSYS; +} +static inline int __pm_genpd_name_add_device(const char *domain_name, + struct device *dev, + struct gpd_timing_data *td) { return -ENOSYS; } @@ -199,13 +201,17 @@ static inline int pm_genpd_remove_device(struct generic_pm_domain *genpd, { return -ENOSYS; } -static inline void pm_genpd_dev_always_on(struct device *dev, bool val) {} static inline void pm_genpd_dev_need_restore(struct device *dev, bool val) {} static inline int pm_genpd_add_subdomain(struct generic_pm_domain *genpd, struct generic_pm_domain *new_sd) { return -ENOSYS; } +static inline int pm_genpd_add_subdomain_names(const char *master_name, + const char *subdomain_name) +{ + return -ENOSYS; +} static inline int pm_genpd_remove_subdomain(struct generic_pm_domain *genpd, struct generic_pm_domain *target) { @@ -221,11 +227,19 @@ static inline int __pm_genpd_remove_callbacks(struct device *dev, bool clear_td) { return -ENOSYS; } -static inline int genpd_attach_cpuidle(struct generic_pm_domain *genpd, int st) +static inline int pm_genpd_attach_cpuidle(struct generic_pm_domain *genpd, int st) { return -ENOSYS; } -static inline int genpd_detach_cpuidle(struct generic_pm_domain *genpd) +static inline int pm_genpd_name_attach_cpuidle(const char *name, int state) +{ + return -ENOSYS; +} +static inline int pm_genpd_detach_cpuidle(struct generic_pm_domain *genpd) +{ + return -ENOSYS; +} +static inline int pm_genpd_name_detach_cpuidle(const char *name) { return -ENOSYS; } @@ -237,6 +251,10 @@ static inline int pm_genpd_poweron(struct generic_pm_domain *genpd) { return -ENOSYS; } +static inline int pm_genpd_name_poweron(const char *domain_name) +{ + return -ENOSYS; +} static inline bool default_stop_ok(struct device *dev) { return false; @@ -245,6 +263,24 @@ static inline bool default_stop_ok(struct device *dev) #define pm_domain_always_on_gov NULL #endif +static inline int pm_genpd_add_device(struct generic_pm_domain *genpd, + struct device *dev) +{ + return __pm_genpd_add_device(genpd, dev, NULL); +} + +static inline int pm_genpd_of_add_device(struct device_node *genpd_node, + struct device *dev) +{ + return __pm_genpd_of_add_device(genpd_node, dev, NULL); +} + +static inline int pm_genpd_name_add_device(const char *domain_name, + struct device *dev) +{ + return __pm_genpd_name_add_device(domain_name, dev, NULL); +} + static inline int pm_genpd_remove_callbacks(struct device *dev) { return __pm_genpd_remove_callbacks(dev, true); @@ -258,4 +294,20 @@ static inline void genpd_queue_power_off_work(struct generic_pm_domain *gpd) {} static inline void pm_genpd_poweroff_unused(void) {} #endif +#ifdef CONFIG_PM_GENERIC_DOMAINS_SLEEP +extern void pm_genpd_syscore_switch(struct device *dev, bool suspend); +#else +static inline void pm_genpd_syscore_switch(struct device *dev, bool suspend) {} +#endif + +static inline void pm_genpd_syscore_poweroff(struct device *dev) +{ + pm_genpd_syscore_switch(dev, true); +} + +static inline void pm_genpd_syscore_poweron(struct device *dev) +{ + pm_genpd_syscore_switch(dev, false); +} + #endif /* _LINUX_PM_DOMAIN_H */ diff --git a/include/linux/ptrace.h b/include/linux/ptrace.h index 597e4fd..3db698a 100644 --- a/include/linux/ptrace.h +++ b/include/linux/ptrace.h @@ -130,8 +130,6 @@ extern void exit_ptrace(struct task_struct *tracer); #define PTRACE_MODE_READ 0x01 #define PTRACE_MODE_ATTACH 0x02 #define PTRACE_MODE_NOAUDIT 0x04 -/* Returns 0 on success, -errno on denial. */ -extern int __ptrace_may_access(struct task_struct *task, unsigned int mode); /* Returns true on success, false on denial. */ extern bool ptrace_may_access(struct task_struct *task, unsigned int mode); diff --git a/include/linux/sched.h b/include/linux/sched.h index 9c5612f..9d51e26 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -405,6 +405,7 @@ static inline void arch_pick_mmap_layout(struct mm_struct *mm) {} extern void set_dumpable(struct mm_struct *mm, int value); extern int get_dumpable(struct mm_struct *mm); +extern int __get_dumpable(unsigned long mm_flags); /* get/set_dumpable() values */ #define SUID_DUMPABLE_DISABLED 0 diff --git a/include/linux/security.h b/include/linux/security.h index 145acce..5b50c4e 100644 --- a/include/linux/security.h +++ b/include/linux/security.h @@ -3022,5 +3022,36 @@ static inline void free_secdata(void *secdata) { } #endif /* CONFIG_SECURITY */ +#ifdef CONFIG_SECURITY_YAMA +extern int yama_ptrace_access_check(struct task_struct *child, + unsigned int mode); +extern int yama_ptrace_traceme(struct task_struct *parent); +extern void yama_task_free(struct task_struct *task); +extern int yama_task_prctl(int option, unsigned long arg2, unsigned long arg3, + unsigned long arg4, unsigned long arg5); +#else +static inline int yama_ptrace_access_check(struct task_struct *child, + unsigned int mode) +{ + return 0; +} + +static inline int yama_ptrace_traceme(struct task_struct *parent) +{ + return 0; +} + +static inline void yama_task_free(struct task_struct *task) +{ +} + +static inline int yama_task_prctl(int option, unsigned long arg2, + unsigned long arg3, unsigned long arg4, + unsigned long arg5) +{ + return -ENOSYS; +} +#endif /* CONFIG_SECURITY_YAMA */ + #endif /* ! __LINUX_SECURITY_H */ diff --git a/include/linux/swiotlb.h b/include/linux/swiotlb.h index e872526..8d08b3e 100644 --- a/include/linux/swiotlb.h +++ b/include/linux/swiotlb.h @@ -25,6 +25,7 @@ extern int swiotlb_force; extern void swiotlb_init(int verbose); extern void swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, int verbose); extern unsigned long swiotlb_nr_tbl(void); +extern int swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs); /* * Enumeration for sync targets diff --git a/include/linux/tpm.h b/include/linux/tpm.h index fdc718a..fcb627f 100644 --- a/include/linux/tpm.h +++ b/include/linux/tpm.h @@ -32,6 +32,7 @@ extern int tpm_pcr_read(u32 chip_num, int pcr_idx, u8 *res_buf); extern int tpm_pcr_extend(u32 chip_num, int pcr_idx, const u8 *hash); extern int tpm_send(u32 chip_num, void *cmd, size_t buflen); +extern int tpm_get_random(u32 chip_num, u8 *data, size_t max); #else static inline int tpm_pcr_read(u32 chip_num, int pcr_idx, u8 *res_buf) { return -ENODEV; @@ -42,5 +43,8 @@ static inline int tpm_pcr_extend(u32 chip_num, int pcr_idx, const u8 *hash) { static inline int tpm_send(u32 chip_num, void *cmd, size_t buflen) { return -ENODEV; } +static inline int tpm_get_random(u32 chip_num, u8 *data, size_t max) { + return -ENODEV; +} #endif #endif diff --git a/include/linux/xattr.h b/include/linux/xattr.h index 2ace7a6..cc13e11 100644 --- a/include/linux/xattr.h +++ b/include/linux/xattr.h @@ -33,6 +33,9 @@ #define XATTR_EVM_SUFFIX "evm" #define XATTR_NAME_EVM XATTR_SECURITY_PREFIX XATTR_EVM_SUFFIX +#define XATTR_IMA_SUFFIX "ima" +#define XATTR_NAME_IMA XATTR_SECURITY_PREFIX XATTR_IMA_SUFFIX + #define XATTR_SELINUX_SUFFIX "selinux" #define XATTR_NAME_SELINUX XATTR_SECURITY_PREFIX XATTR_SELINUX_SUFFIX diff --git a/include/mtd/ubi-user.h b/include/mtd/ubi-user.h index 8787349..53cae1e 100644 --- a/include/mtd/ubi-user.h +++ b/include/mtd/ubi-user.h @@ -222,6 +222,7 @@ enum { * @ubi_num: UBI device number to create * @mtd_num: MTD device number to attach * @vid_hdr_offset: VID header offset (use defaults if %0) + * @max_beb_per1024: maximum expected number of bad PEB per 1024 PEBs * @padding: reserved for future, not used, has to be zeroed * * This data structure is used to specify MTD device UBI has to attach and the @@ -245,12 +246,25 @@ enum { * be 2KiB-64 bytes = 1984. Note, that this position is not even 512-bytes * aligned, which is OK, as UBI is clever enough to realize this is 4th * sub-page of the first page and add needed padding. + * + * The @max_beb_per1024 is the maximum amount of bad PEBs UBI expects on the + * UBI device per 1024 eraseblocks. This value is often given in an other form + * in the NAND datasheet (min NVB i.e. minimal number of valid blocks). The + * maximum expected bad eraseblocks per 1024 is then: + * 1024 * (1 - MinNVB / MaxNVB) + * Which gives 20 for most NAND devices. This limit is used in order to derive + * amount of eraseblock UBI reserves for handling new bad blocks. If the device + * has more bad eraseblocks than this limit, UBI does not reserve any physical + * eraseblocks for new bad eraseblocks, but attempts to use available + * eraseblocks (if any). The accepted range is 0-768. If 0 is given, the + * default kernel value of %CONFIG_MTD_UBI_BEB_LIMIT will be used. */ struct ubi_attach_req { __s32 ubi_num; __s32 mtd_num; __s32 vid_hdr_offset; - __s8 padding[12]; + __s16 max_beb_per1024; + __s8 padding[10]; }; /** diff --git a/include/scsi/libsas.h b/include/scsi/libsas.h index ae33706..ef937b5 100644 --- a/include/scsi/libsas.h +++ b/include/scsi/libsas.h @@ -79,7 +79,8 @@ enum phy_event { PHYE_OOB_DONE = 1, PHYE_OOB_ERROR = 2, PHYE_SPINUP_HOLD = 3, /* hot plug SATA, no COMWAKE sent */ - PHY_NUM_EVENTS = 4, + PHYE_RESUME_TIMEOUT = 4, + PHY_NUM_EVENTS = 5, }; enum discover_event { @@ -87,8 +88,10 @@ enum discover_event { DISCE_REVALIDATE_DOMAIN = 1, DISCE_PORT_GONE = 2, DISCE_PROBE = 3, - DISCE_DESTRUCT = 4, - DISC_NUM_EVENTS = 5, + DISCE_SUSPEND = 4, + DISCE_RESUME = 5, + DISCE_DESTRUCT = 6, + DISC_NUM_EVENTS = 7, }; /* ---------- Expander Devices ---------- */ @@ -128,7 +131,7 @@ struct ex_phy { u8 attached_sas_addr[SAS_ADDR_SIZE]; u8 attached_phy_id; - u8 phy_change_count; + int phy_change_count; enum routing_attribute routing_attr; u8 virtual:1; @@ -141,7 +144,7 @@ struct ex_phy { struct expander_device { struct list_head children; - u16 ex_change_count; + int ex_change_count; u16 max_route_indexes; u8 num_phys; @@ -169,6 +172,7 @@ struct sata_device { enum ata_command_set command_set; struct smp_resp rps_resp; /* report_phy_sata_resp */ u8 port_no; /* port number, if this is a PM (Port) */ + int pm_result; struct ata_port *ap; struct ata_host ata_host; @@ -182,6 +186,7 @@ struct ssp_device { enum { SAS_DEV_GONE, + SAS_DEV_FOUND, /* device notified to lldd */ SAS_DEV_DESTROY, SAS_DEV_EH_PENDING, SAS_DEV_LU_RESET, @@ -273,6 +278,7 @@ struct asd_sas_port { enum sas_linkrate linkrate; struct sas_work work; + int suspended; /* public: */ int id; @@ -321,6 +327,7 @@ struct asd_sas_phy { unsigned long phy_events_pending; int error; + int suspended; struct sas_phy *phy; @@ -687,6 +694,9 @@ struct sas_domain_function_template { extern int sas_register_ha(struct sas_ha_struct *); extern int sas_unregister_ha(struct sas_ha_struct *); +extern void sas_prep_resume_ha(struct sas_ha_struct *sas_ha); +extern void sas_resume_ha(struct sas_ha_struct *sas_ha); +extern void sas_suspend_ha(struct sas_ha_struct *sas_ha); int sas_set_phy_speed(struct sas_phy *phy, struct sas_phy_linkrates *rates); diff --git a/include/scsi/sas_ata.h b/include/scsi/sas_ata.h index 2dfbdaa..ff71a56 100644 --- a/include/scsi/sas_ata.h +++ b/include/scsi/sas_ata.h @@ -45,6 +45,8 @@ void sas_ata_eh(struct Scsi_Host *shost, struct list_head *work_q, void sas_ata_schedule_reset(struct domain_device *dev); void sas_ata_wait_eh(struct domain_device *dev); void sas_probe_sata(struct asd_sas_port *port); +void sas_suspend_sata(struct asd_sas_port *port); +void sas_resume_sata(struct asd_sas_port *port); void sas_ata_end_eh(struct ata_port *ap); #else @@ -82,6 +84,14 @@ static inline void sas_probe_sata(struct asd_sas_port *port) { } +static inline void sas_suspend_sata(struct asd_sas_port *port) +{ +} + +static inline void sas_resume_sata(struct asd_sas_port *port) +{ +} + static inline int sas_get_ata_info(struct domain_device *dev, struct ex_phy *phy) { return 0; diff --git a/include/scsi/scsi_bsg_fc.h b/include/scsi/scsi_bsg_fc.h index 91a4e4f..3031b90 100644 --- a/include/scsi/scsi_bsg_fc.h +++ b/include/scsi/scsi_bsg_fc.h @@ -26,8 +26,6 @@ * This file intended to be included by both kernel and user space */ -#include <scsi/scsi.h> - /* * FC Transport SGIO v4 BSG Message Support */ diff --git a/include/scsi/scsi_device.h b/include/scsi/scsi_device.h index 9895f69..88fae8d 100644 --- a/include/scsi/scsi_device.h +++ b/include/scsi/scsi_device.h @@ -156,6 +156,7 @@ struct scsi_device { unsigned is_visible:1; /* is the device visible in sysfs */ unsigned can_power_off:1; /* Device supports runtime power off */ unsigned wce_default_on:1; /* Cache is ON by default */ + unsigned no_dif:1; /* T10 PI (DIF) should be disabled */ DECLARE_BITMAP(supported_events, SDEV_EVT_MAXBITS); /* supported events */ struct list_head event_list; /* asserted events */ @@ -476,6 +477,9 @@ static inline int scsi_device_enclosure(struct scsi_device *sdev) static inline int scsi_device_protection(struct scsi_device *sdev) { + if (sdev->no_dif) + return 0; + return sdev->scsi_level > SCSI_2 && sdev->inquiry[5] & (1<<0); } diff --git a/include/scsi/scsi_devinfo.h b/include/scsi/scsi_devinfo.h index b4ddd3b..cc1f3e7 100644 --- a/include/scsi/scsi_devinfo.h +++ b/include/scsi/scsi_devinfo.h @@ -30,4 +30,5 @@ #define BLIST_RETRY_HWERROR 0x400000 /* retry HARDWARE_ERROR */ #define BLIST_MAX_512 0x800000 /* maximum 512 sector cdb length */ #define BLIST_ATTACH_PQ3 0x1000000 /* Scan: Attach to PQ3 devices */ +#define BLIST_NO_DIF 0x2000000 /* Disable T10 PI (DIF) */ #endif diff --git a/include/scsi/scsi_host.h b/include/scsi/scsi_host.h index 5f7d5b3..4908480 100644 --- a/include/scsi/scsi_host.h +++ b/include/scsi/scsi_host.h @@ -873,6 +873,9 @@ static inline unsigned int scsi_host_dif_capable(struct Scsi_Host *shost, unsign SHOST_DIF_TYPE2_PROTECTION, SHOST_DIF_TYPE3_PROTECTION }; + if (target_type > SHOST_DIF_TYPE3_PROTECTION) + return 0; + return shost->prot_capabilities & cap[target_type] ? target_type : 0; } @@ -884,6 +887,9 @@ static inline unsigned int scsi_host_dix_capable(struct Scsi_Host *shost, unsign SHOST_DIX_TYPE2_PROTECTION, SHOST_DIX_TYPE3_PROTECTION }; + if (target_type > SHOST_DIX_TYPE3_PROTECTION) + return 0; + return shost->prot_capabilities & cap[target_type]; #endif return 0; diff --git a/include/xen/grant_table.h b/include/xen/grant_table.h index f19fff8..aecee9d 100644 --- a/include/xen/grant_table.h +++ b/include/xen/grant_table.h @@ -190,4 +190,16 @@ int gnttab_unmap_refs(struct gnttab_unmap_grant_ref *unmap_ops, struct gnttab_map_grant_ref *kunmap_ops, struct page **pages, unsigned int count); +/* Perform a batch of grant map/copy operations. Retry every batch slot + * for which the hypervisor returns GNTST_eagain. This is typically due + * to paged out target frames. + * + * Will retry for 1, 2, ... 255 ms, i.e. 256 times during 32 seconds. + * + * Return value in each iand every status field of the batch guaranteed + * to not be GNTST_eagain. + */ +void gnttab_batch_map(struct gnttab_map_grant_ref *batch, unsigned count); +void gnttab_batch_copy(struct gnttab_copy *batch, unsigned count); + #endif /* __ASM_GNTTAB_H__ */ diff --git a/include/xen/interface/grant_table.h b/include/xen/interface/grant_table.h index a17d844..f9f8b97 100644 --- a/include/xen/interface/grant_table.h +++ b/include/xen/interface/grant_table.h @@ -338,7 +338,7 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_dump_table); #define GNTTABOP_transfer 4 struct gnttab_transfer { /* IN parameters. */ - unsigned long mfn; + xen_pfn_t mfn; domid_t domid; grant_ref_t ref; /* OUT parameters. */ @@ -375,7 +375,7 @@ struct gnttab_copy { struct { union { grant_ref_t ref; - unsigned long gmfn; + xen_pfn_t gmfn; } u; domid_t domid; uint16_t offset; @@ -519,7 +519,9 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_get_version); #define GNTST_no_device_space (-7) /* Out of space in I/O MMU. */ #define GNTST_permission_denied (-8) /* Not enough privilege for operation. */ #define GNTST_bad_page (-9) /* Specified page was invalid for op. */ -#define GNTST_bad_copy_arg (-10) /* copy arguments cross page boundary */ +#define GNTST_bad_copy_arg (-10) /* copy arguments cross page boundary. */ +#define GNTST_address_too_big (-11) /* transfer page address too large. */ +#define GNTST_eagain (-12) /* Operation not done; try again. */ #define GNTTABOP_error_msgs { \ "okay", \ @@ -532,7 +534,9 @@ DEFINE_GUEST_HANDLE_STRUCT(gnttab_get_version); "no spare translation slot in the I/O MMU", \ "permission denied", \ "bad page", \ - "copy arguments cross page boundary" \ + "copy arguments cross page boundary", \ + "page address size too large", \ + "operation not done; try again" \ } #endif /* __XEN_PUBLIC_GRANT_TABLE_H__ */ diff --git a/include/xen/interface/memory.h b/include/xen/interface/memory.h index eac3ce1..d8e33a9 100644 --- a/include/xen/interface/memory.h +++ b/include/xen/interface/memory.h @@ -31,7 +31,7 @@ struct xen_memory_reservation { * OUT: GMFN bases of extents that were allocated * (NB. This command also updates the mach_to_phys translation table) */ - GUEST_HANDLE(ulong) extent_start; + GUEST_HANDLE(xen_pfn_t) extent_start; /* Number of extents, and size/alignment of each (2^extent_order pages). */ unsigned long nr_extents; @@ -130,7 +130,7 @@ struct xen_machphys_mfn_list { * any large discontiguities in the machine address space, 2MB gaps in * the machphys table will be represented by an MFN base of zero. */ - GUEST_HANDLE(ulong) extent_start; + GUEST_HANDLE(xen_pfn_t) extent_start; /* * Number of extents written to the above array. This will be smaller @@ -163,6 +163,9 @@ struct xen_add_to_physmap { /* Which domain to change the mapping for. */ domid_t domid; + /* Number of pages to go through for gmfn_range */ + uint16_t size; + /* Source mapping space. */ #define XENMAPSPACE_shared_info 0 /* shared info page */ #define XENMAPSPACE_grant_table 1 /* grant table page */ @@ -172,7 +175,7 @@ struct xen_add_to_physmap { unsigned long idx; /* GPFN where the source mapping page should appear. */ - unsigned long gpfn; + xen_pfn_t gpfn; }; DEFINE_GUEST_HANDLE_STRUCT(xen_add_to_physmap); diff --git a/include/xen/interface/platform.h b/include/xen/interface/platform.h index 61fa661..54ad6f9 100644 --- a/include/xen/interface/platform.h +++ b/include/xen/interface/platform.h @@ -54,7 +54,7 @@ DEFINE_GUEST_HANDLE_STRUCT(xenpf_settime_t); #define XENPF_add_memtype 31 struct xenpf_add_memtype { /* IN variables. */ - unsigned long mfn; + xen_pfn_t mfn; uint64_t nr_mfns; uint32_t type; /* OUT variables. */ @@ -84,7 +84,7 @@ struct xenpf_read_memtype { /* IN variables. */ uint32_t reg; /* OUT variables. */ - unsigned long mfn; + xen_pfn_t mfn; uint64_t nr_mfns; uint32_t type; }; @@ -112,6 +112,7 @@ DEFINE_GUEST_HANDLE_STRUCT(xenpf_platform_quirk_t); #define XEN_FW_DISK_INFO 1 /* from int 13 AH=08/41/48 */ #define XEN_FW_DISK_MBR_SIGNATURE 2 /* from MBR offset 0x1b8 */ #define XEN_FW_VBEDDC_INFO 3 /* from int 10 AX=4f15 */ +#define XEN_FW_KBD_SHIFT_FLAGS 5 /* Int16, Fn02: Get keyboard shift flags. */ struct xenpf_firmware_info { /* IN variables. */ uint32_t type; @@ -142,6 +143,8 @@ struct xenpf_firmware_info { /* must refer to 128-byte buffer */ GUEST_HANDLE(uchar) edid; } vbeddc_info; /* XEN_FW_VBEDDC_INFO */ + + uint8_t kbd_shift_flags; /* XEN_FW_KBD_SHIFT_FLAGS */ } u; }; DEFINE_GUEST_HANDLE_STRUCT(xenpf_firmware_info_t); diff --git a/include/xen/interface/version.h b/include/xen/interface/version.h index e8b6519..dd58cf5 100644 --- a/include/xen/interface/version.h +++ b/include/xen/interface/version.h @@ -60,4 +60,7 @@ struct xen_feature_info { /* arg == NULL; returns host memory page size. */ #define XENVER_pagesize 7 +/* arg == xen_domain_handle_t. */ +#define XENVER_guest_handle 8 + #endif /* __XEN_PUBLIC_VERSION_H__ */ diff --git a/include/xen/interface/xen.h b/include/xen/interface/xen.h index 0801468..886a5d8 100644 --- a/include/xen/interface/xen.h +++ b/include/xen/interface/xen.h @@ -10,7 +10,6 @@ #define __XEN_PUBLIC_XEN_H__ #include <asm/xen/interface.h> -#include <asm/pvclock-abi.h> /* * XEN "SYSTEM CALLS" (a.k.a. HYPERCALLS). @@ -190,7 +189,7 @@ struct mmuext_op { unsigned int cmd; union { /* [UN]PIN_TABLE, NEW_BASEPTR, NEW_USER_BASEPTR */ - unsigned long mfn; + xen_pfn_t mfn; /* INVLPG_LOCAL, INVLPG_ALL, SET_LDT */ unsigned long linear_addr; } arg1; @@ -430,11 +429,11 @@ struct start_info { unsigned long nr_pages; /* Total pages allocated to this domain. */ unsigned long shared_info; /* MACHINE address of shared info struct. */ uint32_t flags; /* SIF_xxx flags. */ - unsigned long store_mfn; /* MACHINE page number of shared page. */ + xen_pfn_t store_mfn; /* MACHINE page number of shared page. */ uint32_t store_evtchn; /* Event channel for store communication. */ union { struct { - unsigned long mfn; /* MACHINE page number of console page. */ + xen_pfn_t mfn; /* MACHINE page number of console page. */ uint32_t evtchn; /* Event channel for console page. */ } domU; struct { @@ -455,6 +454,7 @@ struct dom0_vga_console_info { uint8_t video_type; #define XEN_VGATYPE_TEXT_MODE_3 0x03 #define XEN_VGATYPE_VESA_LFB 0x23 +#define XEN_VGATYPE_EFI_LFB 0x70 union { struct { diff --git a/include/xen/privcmd.h b/include/xen/privcmd.h index 17857fb..a853168 100644 --- a/include/xen/privcmd.h +++ b/include/xen/privcmd.h @@ -35,8 +35,7 @@ #include <linux/types.h> #include <linux/compiler.h> - -typedef unsigned long xen_pfn_t; +#include <xen/interface/xen.h> struct privcmd_hypercall { __u64 op; @@ -59,13 +58,33 @@ struct privcmd_mmapbatch { int num; /* number of pages to populate */ domid_t dom; /* target domain */ __u64 addr; /* virtual address */ - xen_pfn_t __user *arr; /* array of mfns - top nibble set on err */ + xen_pfn_t __user *arr; /* array of mfns - or'd with + PRIVCMD_MMAPBATCH_*_ERROR on err */ +}; + +#define PRIVCMD_MMAPBATCH_MFN_ERROR 0xf0000000U +#define PRIVCMD_MMAPBATCH_PAGED_ERROR 0x80000000U + +struct privcmd_mmapbatch_v2 { + unsigned int num; /* number of pages to populate */ + domid_t dom; /* target domain */ + __u64 addr; /* virtual address */ + const xen_pfn_t __user *arr; /* array of mfns */ + int __user *err; /* array of error codes */ }; /* * @cmd: IOCTL_PRIVCMD_HYPERCALL * @arg: &privcmd_hypercall_t * Return: Value returned from execution of the specified hypercall. + * + * @cmd: IOCTL_PRIVCMD_MMAPBATCH_V2 + * @arg: &struct privcmd_mmapbatch_v2 + * Return: 0 on success (i.e., arg->err contains valid error codes for + * each frame). On an error other than a failed frame remap, -1 is + * returned and errno is set to EINVAL, EFAULT etc. As an exception, + * if the operation was otherwise successful but any frame failed with + * -ENOENT, then -1 is returned and errno is set to ENOENT. */ #define IOCTL_PRIVCMD_HYPERCALL \ _IOC(_IOC_NONE, 'P', 0, sizeof(struct privcmd_hypercall)) @@ -73,5 +92,7 @@ struct privcmd_mmapbatch { _IOC(_IOC_NONE, 'P', 2, sizeof(struct privcmd_mmap)) #define IOCTL_PRIVCMD_MMAPBATCH \ _IOC(_IOC_NONE, 'P', 3, sizeof(struct privcmd_mmapbatch)) +#define IOCTL_PRIVCMD_MMAPBATCH_V2 \ + _IOC(_IOC_NONE, 'P', 4, sizeof(struct privcmd_mmapbatch_v2)) #endif /* __LINUX_PUBLIC_PRIVCMD_H__ */ diff --git a/include/xen/swiotlb-xen.h b/include/xen/swiotlb-xen.h index 4f4d449..de8bcc6 100644 --- a/include/xen/swiotlb-xen.h +++ b/include/xen/swiotlb-xen.h @@ -3,7 +3,7 @@ #include <linux/swiotlb.h> -extern void xen_swiotlb_init(int verbose); +extern int xen_swiotlb_init(int verbose, bool early); extern void *xen_swiotlb_alloc_coherent(struct device *hwdev, size_t size, @@ -23,15 +23,6 @@ extern dma_addr_t xen_swiotlb_map_page(struct device *dev, struct page *page, extern void xen_swiotlb_unmap_page(struct device *hwdev, dma_addr_t dev_addr, size_t size, enum dma_data_direction dir, struct dma_attrs *attrs); -/* -extern int -xen_swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nents, - enum dma_data_direction dir); - -extern void -xen_swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents, - enum dma_data_direction dir); -*/ extern int xen_swiotlb_map_sg_attrs(struct device *hwdev, struct scatterlist *sgl, int nelems, enum dma_data_direction dir, diff --git a/ipc/mqueue.c b/ipc/mqueue.c index 9a08acc..6d255e5 100644 --- a/ipc/mqueue.c +++ b/ipc/mqueue.c @@ -944,7 +944,7 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr, size_t, msg_len, unsigned int, msg_prio, const struct timespec __user *, u_abs_timeout) { - struct file *filp; + struct fd f; struct inode *inode; struct ext_wait_queue wait; struct ext_wait_queue *receiver; @@ -967,21 +967,21 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr, audit_mq_sendrecv(mqdes, msg_len, msg_prio, timeout ? &ts : NULL); - filp = fget(mqdes); - if (unlikely(!filp)) { + f = fdget(mqdes); + if (unlikely(!f.file)) { ret = -EBADF; goto out; } - inode = filp->f_path.dentry->d_inode; - if (unlikely(filp->f_op != &mqueue_file_operations)) { + inode = f.file->f_path.dentry->d_inode; + if (unlikely(f.file->f_op != &mqueue_file_operations)) { ret = -EBADF; goto out_fput; } info = MQUEUE_I(inode); - audit_inode(NULL, filp->f_path.dentry); + audit_inode(NULL, f.file->f_path.dentry); - if (unlikely(!(filp->f_mode & FMODE_WRITE))) { + if (unlikely(!(f.file->f_mode & FMODE_WRITE))) { ret = -EBADF; goto out_fput; } @@ -1023,7 +1023,7 @@ SYSCALL_DEFINE5(mq_timedsend, mqd_t, mqdes, const char __user *, u_msg_ptr, } if (info->attr.mq_curmsgs == info->attr.mq_maxmsg) { - if (filp->f_flags & O_NONBLOCK) { + if (f.file->f_flags & O_NONBLOCK) { ret = -EAGAIN; } else { wait.task = current; @@ -1056,7 +1056,7 @@ out_free: if (ret) free_msg(msg_ptr); out_fput: - fput(filp); + fdput(f); out: return ret; } @@ -1067,7 +1067,7 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, { ssize_t ret; struct msg_msg *msg_ptr; - struct file *filp; + struct fd f; struct inode *inode; struct mqueue_inode_info *info; struct ext_wait_queue wait; @@ -1084,21 +1084,21 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, audit_mq_sendrecv(mqdes, msg_len, 0, timeout ? &ts : NULL); - filp = fget(mqdes); - if (unlikely(!filp)) { + f = fdget(mqdes); + if (unlikely(!f.file)) { ret = -EBADF; goto out; } - inode = filp->f_path.dentry->d_inode; - if (unlikely(filp->f_op != &mqueue_file_operations)) { + inode = f.file->f_path.dentry->d_inode; + if (unlikely(f.file->f_op != &mqueue_file_operations)) { ret = -EBADF; goto out_fput; } info = MQUEUE_I(inode); - audit_inode(NULL, filp->f_path.dentry); + audit_inode(NULL, f.file->f_path.dentry); - if (unlikely(!(filp->f_mode & FMODE_READ))) { + if (unlikely(!(f.file->f_mode & FMODE_READ))) { ret = -EBADF; goto out_fput; } @@ -1130,7 +1130,7 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, } if (info->attr.mq_curmsgs == 0) { - if (filp->f_flags & O_NONBLOCK) { + if (f.file->f_flags & O_NONBLOCK) { spin_unlock(&info->lock); ret = -EAGAIN; } else { @@ -1160,7 +1160,7 @@ SYSCALL_DEFINE5(mq_timedreceive, mqd_t, mqdes, char __user *, u_msg_ptr, free_msg(msg_ptr); } out_fput: - fput(filp); + fdput(f); out: return ret; } @@ -1174,7 +1174,7 @@ SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes, const struct sigevent __user *, u_notification) { int ret; - struct file *filp; + struct fd f; struct sock *sock; struct inode *inode; struct sigevent notification; @@ -1220,13 +1220,13 @@ SYSCALL_DEFINE2(mq_notify, mqd_t, mqdes, skb_put(nc, NOTIFY_COOKIE_LEN); /* and attach it to the socket */ retry: - filp = fget(notification.sigev_signo); - if (!filp) { + f = fdget(notification.sigev_signo); + if (!f.file) { ret = -EBADF; goto out; } - sock = netlink_getsockbyfilp(filp); - fput(filp); + sock = netlink_getsockbyfilp(f.file); + fdput(f); if (IS_ERR(sock)) { ret = PTR_ERR(sock); sock = NULL; @@ -1245,14 +1245,14 @@ retry: } } - filp = fget(mqdes); - if (!filp) { + f = fdget(mqdes); + if (!f.file) { ret = -EBADF; goto out; } - inode = filp->f_path.dentry->d_inode; - if (unlikely(filp->f_op != &mqueue_file_operations)) { + inode = f.file->f_path.dentry->d_inode; + if (unlikely(f.file->f_op != &mqueue_file_operations)) { ret = -EBADF; goto out_fput; } @@ -1292,7 +1292,7 @@ retry: } spin_unlock(&info->lock); out_fput: - fput(filp); + fdput(f); out: if (sock) { netlink_detachskb(sock, nc); @@ -1308,7 +1308,7 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes, { int ret; struct mq_attr mqstat, omqstat; - struct file *filp; + struct fd f; struct inode *inode; struct mqueue_inode_info *info; @@ -1319,14 +1319,14 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes, return -EINVAL; } - filp = fget(mqdes); - if (!filp) { + f = fdget(mqdes); + if (!f.file) { ret = -EBADF; goto out; } - inode = filp->f_path.dentry->d_inode; - if (unlikely(filp->f_op != &mqueue_file_operations)) { + inode = f.file->f_path.dentry->d_inode; + if (unlikely(f.file->f_op != &mqueue_file_operations)) { ret = -EBADF; goto out_fput; } @@ -1335,15 +1335,15 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes, spin_lock(&info->lock); omqstat = info->attr; - omqstat.mq_flags = filp->f_flags & O_NONBLOCK; + omqstat.mq_flags = f.file->f_flags & O_NONBLOCK; if (u_mqstat) { audit_mq_getsetattr(mqdes, &mqstat); - spin_lock(&filp->f_lock); + spin_lock(&f.file->f_lock); if (mqstat.mq_flags & O_NONBLOCK) - filp->f_flags |= O_NONBLOCK; + f.file->f_flags |= O_NONBLOCK; else - filp->f_flags &= ~O_NONBLOCK; - spin_unlock(&filp->f_lock); + f.file->f_flags &= ~O_NONBLOCK; + spin_unlock(&f.file->f_lock); inode->i_atime = inode->i_ctime = CURRENT_TIME; } @@ -1356,7 +1356,7 @@ SYSCALL_DEFINE3(mq_getsetattr, mqd_t, mqdes, ret = -EFAULT; out_fput: - fput(filp); + fdput(f); out: return ret; } diff --git a/kernel/auditsc.c b/kernel/auditsc.c index ff4798f..29e090c 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -1146,13 +1146,44 @@ error_path: EXPORT_SYMBOL(audit_log_task_context); -static void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk) +void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk) { + const struct cred *cred; char name[sizeof(tsk->comm)]; struct mm_struct *mm = tsk->mm; struct vm_area_struct *vma; + char *tty; + + if (!ab) + return; /* tsk == current */ + cred = current_cred(); + + spin_lock_irq(&tsk->sighand->siglock); + if (tsk->signal && tsk->signal->tty && tsk->signal->tty->name) + tty = tsk->signal->tty->name; + else + tty = "(none)"; + spin_unlock_irq(&tsk->sighand->siglock); + + + audit_log_format(ab, + " ppid=%ld pid=%d auid=%u uid=%u gid=%u" + " euid=%u suid=%u fsuid=%u" + " egid=%u sgid=%u fsgid=%u ses=%u tty=%s", + sys_getppid(), + tsk->pid, + from_kuid(&init_user_ns, tsk->loginuid), + from_kuid(&init_user_ns, cred->uid), + from_kgid(&init_user_ns, cred->gid), + from_kuid(&init_user_ns, cred->euid), + from_kuid(&init_user_ns, cred->suid), + from_kuid(&init_user_ns, cred->fsuid), + from_kgid(&init_user_ns, cred->egid), + from_kgid(&init_user_ns, cred->sgid), + from_kgid(&init_user_ns, cred->fsgid), + tsk->sessionid, tty); get_task_comm(name, tsk); audit_log_format(ab, " comm="); @@ -1175,6 +1206,8 @@ static void audit_log_task_info(struct audit_buffer *ab, struct task_struct *tsk audit_log_task_context(ab); } +EXPORT_SYMBOL(audit_log_task_info); + static int audit_log_pid_context(struct audit_context *context, pid_t pid, kuid_t auid, kuid_t uid, unsigned int sessionid, u32 sid, char *comm) @@ -1580,26 +1613,12 @@ static void audit_log_name(struct audit_context *context, struct audit_names *n, static void audit_log_exit(struct audit_context *context, struct task_struct *tsk) { - const struct cred *cred; int i, call_panic = 0; struct audit_buffer *ab; struct audit_aux_data *aux; - const char *tty; struct audit_names *n; /* tsk == current */ - context->pid = tsk->pid; - if (!context->ppid) - context->ppid = sys_getppid(); - cred = current_cred(); - context->uid = cred->uid; - context->gid = cred->gid; - context->euid = cred->euid; - context->suid = cred->suid; - context->fsuid = cred->fsuid; - context->egid = cred->egid; - context->sgid = cred->sgid; - context->fsgid = cred->fsgid; context->personality = tsk->personality; ab = audit_log_start(context, GFP_KERNEL, AUDIT_SYSCALL); @@ -1614,37 +1633,13 @@ static void audit_log_exit(struct audit_context *context, struct task_struct *ts (context->return_valid==AUDITSC_SUCCESS)?"yes":"no", context->return_code); - spin_lock_irq(&tsk->sighand->siglock); - if (tsk->signal && tsk->signal->tty && tsk->signal->tty->name) - tty = tsk->signal->tty->name; - else - tty = "(none)"; - spin_unlock_irq(&tsk->sighand->siglock); - audit_log_format(ab, - " a0=%lx a1=%lx a2=%lx a3=%lx items=%d" - " ppid=%d pid=%d auid=%u uid=%u gid=%u" - " euid=%u suid=%u fsuid=%u" - " egid=%u sgid=%u fsgid=%u tty=%s ses=%u", - context->argv[0], - context->argv[1], - context->argv[2], - context->argv[3], - context->name_count, - context->ppid, - context->pid, - from_kuid(&init_user_ns, tsk->loginuid), - from_kuid(&init_user_ns, context->uid), - from_kgid(&init_user_ns, context->gid), - from_kuid(&init_user_ns, context->euid), - from_kuid(&init_user_ns, context->suid), - from_kuid(&init_user_ns, context->fsuid), - from_kgid(&init_user_ns, context->egid), - from_kgid(&init_user_ns, context->sgid), - from_kgid(&init_user_ns, context->fsgid), - tty, - tsk->sessionid); - + " a0=%lx a1=%lx a2=%lx a3=%lx items=%d", + context->argv[0], + context->argv[1], + context->argv[2], + context->argv[3], + context->name_count); audit_log_task_info(ab, tsk); audit_log_key(ab, context->filterkey); diff --git a/kernel/events/core.c b/kernel/events/core.c index deec4e5..f16f3c5 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -468,14 +468,13 @@ static inline int perf_cgroup_connect(int fd, struct perf_event *event, { struct perf_cgroup *cgrp; struct cgroup_subsys_state *css; - struct file *file; - int ret = 0, fput_needed; + struct fd f = fdget(fd); + int ret = 0; - file = fget_light(fd, &fput_needed); - if (!file) + if (!f.file) return -EBADF; - css = cgroup_css_from_dir(file, perf_subsys_id); + css = cgroup_css_from_dir(f.file, perf_subsys_id); if (IS_ERR(css)) { ret = PTR_ERR(css); goto out; @@ -501,7 +500,7 @@ static inline int perf_cgroup_connect(int fd, struct perf_event *event, ret = -EINVAL; } out: - fput_light(file, fput_needed); + fdput(f); return ret; } @@ -3234,21 +3233,18 @@ unlock: static const struct file_operations perf_fops; -static struct file *perf_fget_light(int fd, int *fput_needed) +static inline int perf_fget_light(int fd, struct fd *p) { - struct file *file; - - file = fget_light(fd, fput_needed); - if (!file) - return ERR_PTR(-EBADF); + struct fd f = fdget(fd); + if (!f.file) + return -EBADF; - if (file->f_op != &perf_fops) { - fput_light(file, *fput_needed); - *fput_needed = 0; - return ERR_PTR(-EBADF); + if (f.file->f_op != &perf_fops) { + fdput(f); + return -EBADF; } - - return file; + *p = f; + return 0; } static int perf_event_set_output(struct perf_event *event, @@ -3280,22 +3276,19 @@ static long perf_ioctl(struct file *file, unsigned int cmd, unsigned long arg) case PERF_EVENT_IOC_SET_OUTPUT: { - struct file *output_file = NULL; - struct perf_event *output_event = NULL; - int fput_needed = 0; int ret; - if (arg != -1) { - output_file = perf_fget_light(arg, &fput_needed); - if (IS_ERR(output_file)) - return PTR_ERR(output_file); - output_event = output_file->private_data; + struct perf_event *output_event; + struct fd output; + ret = perf_fget_light(arg, &output); + if (ret) + return ret; + output_event = output.file->private_data; + ret = perf_event_set_output(event, output_event); + fdput(output); + } else { + ret = perf_event_set_output(event, NULL); } - - ret = perf_event_set_output(event, output_event); - if (output_event) - fput_light(output_file, fput_needed); - return ret; } @@ -6443,12 +6436,11 @@ SYSCALL_DEFINE5(perf_event_open, struct perf_event_attr attr; struct perf_event_context *ctx; struct file *event_file = NULL; - struct file *group_file = NULL; + struct fd group = {NULL, 0}; struct task_struct *task = NULL; struct pmu *pmu; int event_fd; int move_group = 0; - int fput_needed = 0; int err; /* for future expandability... */ @@ -6478,17 +6470,15 @@ SYSCALL_DEFINE5(perf_event_open, if ((flags & PERF_FLAG_PID_CGROUP) && (pid == -1 || cpu == -1)) return -EINVAL; - event_fd = get_unused_fd_flags(O_RDWR); + event_fd = get_unused_fd(); if (event_fd < 0) return event_fd; if (group_fd != -1) { - group_file = perf_fget_light(group_fd, &fput_needed); - if (IS_ERR(group_file)) { - err = PTR_ERR(group_file); + err = perf_fget_light(group_fd, &group); + if (err) goto err_fd; - } - group_leader = group_file->private_data; + group_leader = group.file->private_data; if (flags & PERF_FLAG_FD_OUTPUT) output_event = group_leader; if (flags & PERF_FLAG_FD_NO_GROUP) @@ -6664,7 +6654,7 @@ SYSCALL_DEFINE5(perf_event_open, * of the group leader will find the pointer to itself in * perf_group_detach(). */ - fput_light(group_file, fput_needed); + fdput(group); fd_install(event_fd, event_file); return event_fd; @@ -6678,7 +6668,7 @@ err_task: if (task) put_task_struct(task); err_group_fd: - fput_light(group_file, fput_needed); + fdput(group); err_fd: put_unused_fd(event_fd); return err; diff --git a/kernel/exit.c b/kernel/exit.c index 42f2595..346616c00 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -457,108 +457,13 @@ void daemonize(const char *name, ...) /* Become as one with the init task */ daemonize_fs_struct(); - exit_files(current); - current->files = init_task.files; - atomic_inc(¤t->files->count); + daemonize_descriptors(); reparent_to_kthreadd(); } EXPORT_SYMBOL(daemonize); -static void close_files(struct files_struct * files) -{ - int i, j; - struct fdtable *fdt; - - j = 0; - - /* - * It is safe to dereference the fd table without RCU or - * ->file_lock because this is the last reference to the - * files structure. But use RCU to shut RCU-lockdep up. - */ - rcu_read_lock(); - fdt = files_fdtable(files); - rcu_read_unlock(); - for (;;) { - unsigned long set; - i = j * BITS_PER_LONG; - if (i >= fdt->max_fds) - break; - set = fdt->open_fds[j++]; - while (set) { - if (set & 1) { - struct file * file = xchg(&fdt->fd[i], NULL); - if (file) { - filp_close(file, files); - cond_resched(); - } - } - i++; - set >>= 1; - } - } -} - -struct files_struct *get_files_struct(struct task_struct *task) -{ - struct files_struct *files; - - task_lock(task); - files = task->files; - if (files) - atomic_inc(&files->count); - task_unlock(task); - - return files; -} - -void put_files_struct(struct files_struct *files) -{ - struct fdtable *fdt; - - if (atomic_dec_and_test(&files->count)) { - close_files(files); - /* - * Free the fd and fdset arrays if we expanded them. - * If the fdtable was embedded, pass files for freeing - * at the end of the RCU grace period. Otherwise, - * you can free files immediately. - */ - rcu_read_lock(); - fdt = files_fdtable(files); - if (fdt != &files->fdtab) - kmem_cache_free(files_cachep, files); - free_fdtable(fdt); - rcu_read_unlock(); - } -} - -void reset_files_struct(struct files_struct *files) -{ - struct task_struct *tsk = current; - struct files_struct *old; - - old = tsk->files; - task_lock(tsk); - tsk->files = files; - task_unlock(tsk); - put_files_struct(old); -} - -void exit_files(struct task_struct *tsk) -{ - struct files_struct * files = tsk->files; - - if (files) { - task_lock(tsk); - tsk->files = NULL; - task_unlock(tsk); - put_files_struct(files); - } -} - #ifdef CONFIG_MM_OWNER /* * A task is exiting. If it owned this mm, find a new owner for the mm. diff --git a/kernel/power/Kconfig b/kernel/power/Kconfig index a70518c..5dfdc9e 100644 --- a/kernel/power/Kconfig +++ b/kernel/power/Kconfig @@ -263,6 +263,10 @@ config PM_GENERIC_DOMAINS bool depends on PM +config PM_GENERIC_DOMAINS_SLEEP + def_bool y + depends on PM_SLEEP && PM_GENERIC_DOMAINS + config PM_GENERIC_DOMAINS_RUNTIME def_bool y depends on PM_RUNTIME && PM_GENERIC_DOMAINS diff --git a/kernel/power/poweroff.c b/kernel/power/poweroff.c index d523593..68197a4 100644 --- a/kernel/power/poweroff.c +++ b/kernel/power/poweroff.c @@ -37,7 +37,7 @@ static struct sysrq_key_op sysrq_poweroff_op = { .enable_mask = SYSRQ_ENABLE_BOOT, }; -static int pm_sysrq_init(void) +static int __init pm_sysrq_init(void) { register_sysrq_key('o', &sysrq_poweroff_op); return 0; diff --git a/kernel/power/process.c b/kernel/power/process.c index 19db29f..87da817 100644 --- a/kernel/power/process.c +++ b/kernel/power/process.c @@ -79,7 +79,7 @@ static int try_to_freeze_tasks(bool user_only) /* * We need to retry, but first give the freezing tasks some - * time to enter the regrigerator. + * time to enter the refrigerator. */ msleep(10); } diff --git a/kernel/power/qos.c b/kernel/power/qos.c index 6a031e6..846bd42 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -139,6 +139,7 @@ static inline int pm_qos_get_value(struct pm_qos_constraints *c) default: /* runtime check for not using enum */ BUG(); + return PM_QOS_DEFAULT_VALUE; } } diff --git a/kernel/ptrace.c b/kernel/ptrace.c index a232bb5..1f5e55d 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -180,7 +180,8 @@ static int ptrace_has_cap(struct user_namespace *ns, unsigned int mode) return has_ns_capability(current, ns, CAP_SYS_PTRACE); } -int __ptrace_may_access(struct task_struct *task, unsigned int mode) +/* Returns 0 on success, -errno on denial. */ +static int __ptrace_may_access(struct task_struct *task, unsigned int mode) { const struct cred *cred = current_cred(), *tcred; diff --git a/kernel/sys.c b/kernel/sys.c index 241507f..f949228 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1788,15 +1788,15 @@ SYSCALL_DEFINE1(umask, int, mask) #ifdef CONFIG_CHECKPOINT_RESTORE static int prctl_set_mm_exe_file(struct mm_struct *mm, unsigned int fd) { - struct file *exe_file; + struct fd exe; struct dentry *dentry; int err; - exe_file = fget(fd); - if (!exe_file) + exe = fdget(fd); + if (!exe.file) return -EBADF; - dentry = exe_file->f_path.dentry; + dentry = exe.file->f_path.dentry; /* * Because the original mm->exe_file points to executable file, make @@ -1805,7 +1805,7 @@ static int prctl_set_mm_exe_file(struct mm_struct *mm, unsigned int fd) */ err = -EACCES; if (!S_ISREG(dentry->d_inode->i_mode) || - exe_file->f_path.mnt->mnt_flags & MNT_NOEXEC) + exe.file->f_path.mnt->mnt_flags & MNT_NOEXEC) goto exit; err = inode_permission(dentry->d_inode, MAY_EXEC); @@ -1839,12 +1839,12 @@ static int prctl_set_mm_exe_file(struct mm_struct *mm, unsigned int fd) goto exit_unlock; err = 0; - set_mm_exe_file(mm, exe_file); + set_mm_exe_file(mm, exe.file); /* this grabs a reference to exe.file */ exit_unlock: up_write(&mm->mmap_sem); exit: - fput(exe_file); + fdput(exe); return err; } diff --git a/kernel/taskstats.c b/kernel/taskstats.c index 5eab1f3..610f083 100644 --- a/kernel/taskstats.c +++ b/kernel/taskstats.c @@ -424,16 +424,15 @@ static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) struct nlattr *na; size_t size; u32 fd; - struct file *file; - int fput_needed; + struct fd f; na = info->attrs[CGROUPSTATS_CMD_ATTR_FD]; if (!na) return -EINVAL; fd = nla_get_u32(info->attrs[CGROUPSTATS_CMD_ATTR_FD]); - file = fget_light(fd, &fput_needed); - if (!file) + f = fdget(fd); + if (!f.file) return 0; size = nla_total_size(sizeof(struct cgroupstats)); @@ -453,7 +452,7 @@ static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) stats = nla_data(na); memset(stats, 0, sizeof(*stats)); - rc = cgroupstats_build(stats, file->f_dentry); + rc = cgroupstats_build(stats, f.file->f_dentry); if (rc < 0) { nlmsg_free(rep_skb); goto err; @@ -462,7 +461,7 @@ static int cgroupstats_user_cmd(struct sk_buff *skb, struct genl_info *info) rc = send_reply(rep_skb, info); err: - fput_light(file, fput_needed); + fdput(f); return rc; } diff --git a/kernel/time/clockevents.c b/kernel/time/clockevents.c index 7e1ce01..30b6de0 100644 --- a/kernel/time/clockevents.c +++ b/kernel/time/clockevents.c @@ -397,6 +397,30 @@ void clockevents_exchange_device(struct clock_event_device *old, local_irq_restore(flags); } +/** + * clockevents_suspend - suspend clock devices + */ +void clockevents_suspend(void) +{ + struct clock_event_device *dev; + + list_for_each_entry_reverse(dev, &clockevent_devices, list) + if (dev->suspend) + dev->suspend(dev); +} + +/** + * clockevents_resume - resume clock devices + */ +void clockevents_resume(void) +{ + struct clock_event_device *dev; + + list_for_each_entry(dev, &clockevent_devices, list) + if (dev->resume) + dev->resume(dev); +} + #ifdef CONFIG_GENERIC_CLOCKEVENTS /** * clockevents_notify - notification about relevant events diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c index d3b91e7..5ce06a3 100644 --- a/kernel/time/timekeeping.c +++ b/kernel/time/timekeeping.c @@ -776,6 +776,7 @@ static void timekeeping_resume(void) read_persistent_clock(&ts); + clockevents_resume(); clocksource_resume(); write_seqlock_irqsave(&tk->lock, flags); @@ -835,6 +836,7 @@ static int timekeeping_suspend(void) clockevents_notify(CLOCK_EVT_NOTIFY_SUSPEND, NULL); clocksource_suspend(); + clockevents_suspend(); return 0; } diff --git a/lib/swiotlb.c b/lib/swiotlb.c index 45bc1f8..f114bf6 100644 --- a/lib/swiotlb.c +++ b/lib/swiotlb.c @@ -170,7 +170,7 @@ void __init swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, int verbose) * Statically reserve bounce buffer space and initialize bounce buffer data * structures for the software IO TLB used to implement the DMA API. */ -void __init +static void __init swiotlb_init_with_default_size(size_t default_size, int verbose) { unsigned long bytes; @@ -206,8 +206,9 @@ swiotlb_init(int verbose) int swiotlb_late_init_with_default_size(size_t default_size) { - unsigned long i, bytes, req_nslabs = io_tlb_nslabs; + unsigned long bytes, req_nslabs = io_tlb_nslabs; unsigned int order; + int rc = 0; if (!io_tlb_nslabs) { io_tlb_nslabs = (default_size >> IO_TLB_SHIFT); @@ -229,16 +230,32 @@ swiotlb_late_init_with_default_size(size_t default_size) order--; } - if (!io_tlb_start) - goto cleanup1; - + if (!io_tlb_start) { + io_tlb_nslabs = req_nslabs; + return -ENOMEM; + } if (order != get_order(bytes)) { printk(KERN_WARNING "Warning: only able to allocate %ld MB " "for software IO TLB\n", (PAGE_SIZE << order) >> 20); io_tlb_nslabs = SLABS_PER_PAGE << order; - bytes = io_tlb_nslabs << IO_TLB_SHIFT; } + rc = swiotlb_late_init_with_tbl(io_tlb_start, io_tlb_nslabs); + if (rc) + free_pages((unsigned long)io_tlb_start, order); + return rc; +} + +int +swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs) +{ + unsigned long i, bytes; + + bytes = nslabs << IO_TLB_SHIFT; + + io_tlb_nslabs = nslabs; + io_tlb_start = tlb; io_tlb_end = io_tlb_start + bytes; + memset(io_tlb_start, 0, bytes); /* @@ -288,10 +305,8 @@ cleanup3: io_tlb_list = NULL; cleanup2: io_tlb_end = NULL; - free_pages((unsigned long)io_tlb_start, order); io_tlb_start = NULL; -cleanup1: - io_tlb_nslabs = req_nslabs; + io_tlb_nslabs = 0; return -ENOMEM; } diff --git a/mm/fadvise.c b/mm/fadvise.c index 9b75a04..a47f0f5 100644 --- a/mm/fadvise.c +++ b/mm/fadvise.c @@ -26,7 +26,7 @@ */ SYSCALL_DEFINE(fadvise64_64)(int fd, loff_t offset, loff_t len, int advice) { - struct file *file = fget(fd); + struct fd f = fdget(fd); struct address_space *mapping; struct backing_dev_info *bdi; loff_t endbyte; /* inclusive */ @@ -35,15 +35,15 @@ SYSCALL_DEFINE(fadvise64_64)(int fd, loff_t offset, loff_t len, int advice) unsigned long nrpages; int ret = 0; - if (!file) + if (!f.file) return -EBADF; - if (S_ISFIFO(file->f_path.dentry->d_inode->i_mode)) { + if (S_ISFIFO(f.file->f_path.dentry->d_inode->i_mode)) { ret = -ESPIPE; goto out; } - mapping = file->f_mapping; + mapping = f.file->f_mapping; if (!mapping || len < 0) { ret = -EINVAL; goto out; @@ -76,21 +76,21 @@ SYSCALL_DEFINE(fadvise64_64)(int fd, loff_t offset, loff_t len, int advice) switch (advice) { case POSIX_FADV_NORMAL: - file->f_ra.ra_pages = bdi->ra_pages; - spin_lock(&file->f_lock); - file->f_mode &= ~FMODE_RANDOM; - spin_unlock(&file->f_lock); + f.file->f_ra.ra_pages = bdi->ra_pages; + spin_lock(&f.file->f_lock); + f.file->f_mode &= ~FMODE_RANDOM; + spin_unlock(&f.file->f_lock); break; case POSIX_FADV_RANDOM: - spin_lock(&file->f_lock); - file->f_mode |= FMODE_RANDOM; - spin_unlock(&file->f_lock); + spin_lock(&f.file->f_lock); + f.file->f_mode |= FMODE_RANDOM; + spin_unlock(&f.file->f_lock); break; case POSIX_FADV_SEQUENTIAL: - file->f_ra.ra_pages = bdi->ra_pages * 2; - spin_lock(&file->f_lock); - file->f_mode &= ~FMODE_RANDOM; - spin_unlock(&file->f_lock); + f.file->f_ra.ra_pages = bdi->ra_pages * 2; + spin_lock(&f.file->f_lock); + f.file->f_mode &= ~FMODE_RANDOM; + spin_unlock(&f.file->f_lock); break; case POSIX_FADV_WILLNEED: /* First and last PARTIAL page! */ @@ -106,7 +106,7 @@ SYSCALL_DEFINE(fadvise64_64)(int fd, loff_t offset, loff_t len, int advice) * Ignore return value because fadvise() shall return * success even if filesystem can't retrieve a hint, */ - force_page_cache_readahead(mapping, file, start_index, + force_page_cache_readahead(mapping, f.file, start_index, nrpages); break; case POSIX_FADV_NOREUSE: @@ -128,7 +128,7 @@ SYSCALL_DEFINE(fadvise64_64)(int fd, loff_t offset, loff_t len, int advice) ret = -EINVAL; } out: - fput(file); + fdput(f); return ret; } #ifdef CONFIG_HAVE_SYSCALL_WRAPPERS diff --git a/mm/fremap.c b/mm/fremap.c index 9ed4fd4..048659c 100644 --- a/mm/fremap.c +++ b/mm/fremap.c @@ -195,10 +195,9 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size, */ if (mapping_cap_account_dirty(mapping)) { unsigned long addr; - struct file *file = vma->vm_file; + struct file *file = get_file(vma->vm_file); flags &= MAP_NONBLOCK; - get_file(file); addr = mmap_region(file, start, size, flags, vma->vm_flags, pgoff); fput(file); diff --git a/mm/frontswap.c b/mm/frontswap.c index 6b3e71a..2890e67d 100644 --- a/mm/frontswap.c +++ b/mm/frontswap.c @@ -44,6 +44,13 @@ EXPORT_SYMBOL(frontswap_enabled); */ static bool frontswap_writethrough_enabled __read_mostly; +/* + * If enabled, the underlying tmem implementation is capable of doing + * exclusive gets, so frontswap_load, on a successful tmem_get must + * mark the page as no longer in frontswap AND mark it dirty. + */ +static bool frontswap_tmem_exclusive_gets_enabled __read_mostly; + #ifdef CONFIG_DEBUG_FS /* * Counters available via /sys/kernel/debug/frontswap (if debugfs is @@ -97,6 +104,15 @@ void frontswap_writethrough(bool enable) EXPORT_SYMBOL(frontswap_writethrough); /* + * Enable/disable frontswap exclusive gets (see above). + */ +void frontswap_tmem_exclusive_gets(bool enable) +{ + frontswap_tmem_exclusive_gets_enabled = enable; +} +EXPORT_SYMBOL(frontswap_tmem_exclusive_gets); + +/* * Called when a swap device is swapon'd. */ void __frontswap_init(unsigned type) @@ -174,8 +190,13 @@ int __frontswap_load(struct page *page) BUG_ON(sis == NULL); if (frontswap_test(sis, offset)) ret = frontswap_ops.load(type, offset, page); - if (ret == 0) + if (ret == 0) { inc_frontswap_loads(); + if (frontswap_tmem_exclusive_gets_enabled) { + SetPageDirty(page); + frontswap_clear(sis, offset); + } + } return ret; } EXPORT_SYMBOL(__frontswap_load); @@ -263,6 +284,11 @@ static int __frontswap_unuse_pages(unsigned long total, unsigned long *unused, return ret; } +/* + * Used to check if it's necessory and feasible to unuse pages. + * Return 1 when nothing to do, 0 when need to shink pages, + * error code when there is an error. + */ static int __frontswap_shrink(unsigned long target_pages, unsigned long *pages_to_unuse, int *type) @@ -275,7 +301,7 @@ static int __frontswap_shrink(unsigned long target_pages, if (total_pages <= target_pages) { /* Nothing to do */ *pages_to_unuse = 0; - return 0; + return 1; } total_pages_to_unuse = total_pages - target_pages; return __frontswap_unuse_pages(total_pages_to_unuse, pages_to_unuse, type); @@ -292,7 +318,7 @@ static int __frontswap_shrink(unsigned long target_pages, void frontswap_shrink(unsigned long target_pages) { unsigned long pages_to_unuse = 0; - int type, ret; + int uninitialized_var(type), ret; /* * we don't want to hold swap_lock while doing a very @@ -302,7 +328,7 @@ void frontswap_shrink(unsigned long target_pages) spin_lock(&swap_lock); ret = __frontswap_shrink(target_pages, &pages_to_unuse, &type); spin_unlock(&swap_lock); - if (ret == 0 && pages_to_unuse) + if (ret == 0) try_to_unuse(type, true, pages_to_unuse); return; } @@ -1301,8 +1301,7 @@ munmap_back: goto free_vma; correct_wcount = 1; } - vma->vm_file = file; - get_file(file); + vma->vm_file = get_file(file); error = file->f_op->mmap(file, vma); if (error) goto unmap_and_free_vma; @@ -1282,10 +1282,8 @@ unsigned long do_mmap_pgoff(struct file *file, vma->vm_pgoff = pgoff; if (file) { - region->vm_file = file; - get_file(file); - vma->vm_file = file; - get_file(file); + region->vm_file = get_file(file); + vma->vm_file = get_file(file); if (vm_flags & VM_EXECUTABLE) { added_exe_file_vma(current->mm); vma->vm_mm = current->mm; diff --git a/mm/readahead.c b/mm/readahead.c index ea8f8fa..7963f23 100644 --- a/mm/readahead.c +++ b/mm/readahead.c @@ -579,19 +579,19 @@ do_readahead(struct address_space *mapping, struct file *filp, SYSCALL_DEFINE(readahead)(int fd, loff_t offset, size_t count) { ssize_t ret; - struct file *file; + struct fd f; ret = -EBADF; - file = fget(fd); - if (file) { - if (file->f_mode & FMODE_READ) { - struct address_space *mapping = file->f_mapping; + f = fdget(fd); + if (f.file) { + if (f.file->f_mode & FMODE_READ) { + struct address_space *mapping = f.file->f_mapping; pgoff_t start = offset >> PAGE_CACHE_SHIFT; pgoff_t end = (offset + count - 1) >> PAGE_CACHE_SHIFT; unsigned long len = end - start + 1; - ret = do_readahead(mapping, file, start, len); + ret = do_readahead(mapping, f.file, start, len); } - fput(file); + fdput(f); } return ret; } diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c index 505f0ce3..15656b8 100644 --- a/net/9p/trans_fd.c +++ b/net/9p/trans_fd.c @@ -793,30 +793,28 @@ static int p9_fd_open(struct p9_client *client, int rfd, int wfd) static int p9_socket_open(struct p9_client *client, struct socket *csocket) { struct p9_trans_fd *p; - int ret, fd; + struct file *file; + int ret; p = kmalloc(sizeof(struct p9_trans_fd), GFP_KERNEL); if (!p) return -ENOMEM; csocket->sk->sk_allocation = GFP_NOIO; - fd = sock_map_fd(csocket, 0); - if (fd < 0) { + file = sock_alloc_file(csocket, 0, NULL); + if (IS_ERR(file)) { pr_err("%s (%d): failed to map fd\n", __func__, task_pid_nr(current)); sock_release(csocket); kfree(p); - return fd; + return PTR_ERR(file); } - get_file(csocket->file); - get_file(csocket->file); - p->wr = p->rd = csocket->file; + get_file(file); + p->wr = p->rd = file; client->trans = p; client->status = Connected; - sys_close(fd); /* still racy */ - p->rd->f_flags |= O_NONBLOCK; p->conn = p9_conn_create(client); diff --git a/net/compat.c b/net/compat.c index 74ed1d7..79ae884 100644 --- a/net/compat.c +++ b/net/compat.c @@ -301,8 +301,7 @@ void scm_detach_fds_compat(struct msghdr *kmsg, struct scm_cookie *scm) break; } /* Bump the usage count and install the file. */ - get_file(fp[i]); - fd_install(new_fd, fp[i]); + fd_install(new_fd, get_file(fp[i])); } if (i > 0) { diff --git a/net/core/netprio_cgroup.c b/net/core/netprio_cgroup.c index 4a83fb3..79285a3 100644 --- a/net/core/netprio_cgroup.c +++ b/net/core/netprio_cgroup.c @@ -239,38 +239,24 @@ out_free_devname: return ret; } +static int update_netprio(const void *v, struct file *file, unsigned n) +{ + int err; + struct socket *sock = sock_from_file(file, &err); + if (sock) + sock->sk->sk_cgrp_prioidx = (u32)(unsigned long)v; + return 0; +} + void net_prio_attach(struct cgroup *cgrp, struct cgroup_taskset *tset) { struct task_struct *p; + void *v; cgroup_taskset_for_each(p, cgrp, tset) { - unsigned int fd; - struct fdtable *fdt; - struct files_struct *files; - task_lock(p); - files = p->files; - if (!files) { - task_unlock(p); - continue; - } - - spin_lock(&files->file_lock); - fdt = files_fdtable(files); - for (fd = 0; fd < fdt->max_fds; fd++) { - struct file *file; - struct socket *sock; - int err; - - file = fcheck_files(files, fd); - if (!file) - continue; - - sock = sock_from_file(file, &err); - if (sock) - sock_update_netprioidx(sock->sk, p); - } - spin_unlock(&files->file_lock); + v = (void *)(unsigned long)task_netprioidx(p); + iterate_fd(p->files, 0, update_netprio, v); task_unlock(p); } } diff --git a/net/core/scm.c b/net/core/scm.c index 9c1c63d..ab57084 100644 --- a/net/core/scm.c +++ b/net/core/scm.c @@ -301,11 +301,10 @@ void scm_detach_fds(struct msghdr *msg, struct scm_cookie *scm) break; } /* Bump the usage count and install the file. */ - get_file(fp[i]); sock = sock_from_file(fp[i], &err); if (sock) sock_update_netprioidx(sock->sk, current); - fd_install(new_fd, fp[i]); + fd_install(new_fd, get_file(fp[i])); } if (i > 0) diff --git a/net/sctp/socket.c b/net/sctp/socket.c index d37d24f..59d16ea 100644 --- a/net/sctp/socket.c +++ b/net/sctp/socket.c @@ -70,6 +70,7 @@ #include <linux/init.h> #include <linux/crypto.h> #include <linux/slab.h> +#include <linux/file.h> #include <net/ip.h> #include <net/icmp.h> @@ -4292,6 +4293,7 @@ static int sctp_getsockopt_peeloff(struct sock *sk, int len, char __user *optval { sctp_peeloff_arg_t peeloff; struct socket *newsock; + struct file *newfile; int retval = 0; if (len < sizeof(sctp_peeloff_arg_t)) @@ -4305,22 +4307,35 @@ static int sctp_getsockopt_peeloff(struct sock *sk, int len, char __user *optval goto out; /* Map the socket to an unused fd that can be returned to the user. */ - retval = sock_map_fd(newsock, 0); + retval = get_unused_fd(); if (retval < 0) { sock_release(newsock); goto out; } + newfile = sock_alloc_file(newsock, 0, NULL); + if (unlikely(IS_ERR(newfile))) { + put_unused_fd(retval); + sock_release(newsock); + return PTR_ERR(newfile); + } + SCTP_DEBUG_PRINTK("%s: sk: %p newsk: %p sd: %d\n", __func__, sk, newsock->sk, retval); /* Return the fd mapped to the new socket. */ + if (put_user(len, optlen)) { + fput(newfile); + put_unused_fd(retval); + return -EFAULT; + } peeloff.sd = retval; - if (put_user(len, optlen)) + if (copy_to_user(optval, &peeloff, len)) { + fput(newfile); + put_unused_fd(retval); return -EFAULT; - if (copy_to_user(optval, &peeloff, len)) - retval = -EFAULT; - + } + fd_install(retval, newfile); out: return retval; } diff --git a/net/socket.c b/net/socket.c index 80dc7e8..d92c490 100644 --- a/net/socket.c +++ b/net/socket.c @@ -347,17 +347,11 @@ static struct file_system_type sock_fs_type = { * but we take care of internal coherence yet. */ -static int sock_alloc_file(struct socket *sock, struct file **f, int flags, - const char *dname) +struct file *sock_alloc_file(struct socket *sock, int flags, const char *dname) { struct qstr name = { .name = "" }; struct path path; struct file *file; - int fd; - - fd = get_unused_fd_flags(flags); - if (unlikely(fd < 0)) - return fd; if (dname) { name.name = dname; @@ -367,10 +361,8 @@ static int sock_alloc_file(struct socket *sock, struct file **f, int flags, name.len = strlen(name.name); } path.dentry = d_alloc_pseudo(sock_mnt->mnt_sb, &name); - if (unlikely(!path.dentry)) { - put_unused_fd(fd); - return -ENOMEM; - } + if (unlikely(!path.dentry)) + return ERR_PTR(-ENOMEM); path.mnt = mntget(sock_mnt); d_instantiate(path.dentry, SOCK_INODE(sock)); @@ -382,30 +374,33 @@ static int sock_alloc_file(struct socket *sock, struct file **f, int flags, /* drop dentry, keep inode */ ihold(path.dentry->d_inode); path_put(&path); - put_unused_fd(fd); - return -ENFILE; + return ERR_PTR(-ENFILE); } sock->file = file; file->f_flags = O_RDWR | (flags & O_NONBLOCK); file->f_pos = 0; file->private_data = sock; - - *f = file; - return fd; + return file; } +EXPORT_SYMBOL(sock_alloc_file); -int sock_map_fd(struct socket *sock, int flags) +static int sock_map_fd(struct socket *sock, int flags) { struct file *newfile; - int fd = sock_alloc_file(sock, &newfile, flags, NULL); + int fd = get_unused_fd_flags(flags); + if (unlikely(fd < 0)) + return fd; - if (likely(fd >= 0)) + newfile = sock_alloc_file(sock, flags, NULL); + if (likely(!IS_ERR(newfile))) { fd_install(fd, newfile); + return fd; + } - return fd; + put_unused_fd(fd); + return PTR_ERR(newfile); } -EXPORT_SYMBOL(sock_map_fd); struct socket *sock_from_file(struct file *file, int *err) { @@ -1466,17 +1461,32 @@ SYSCALL_DEFINE4(socketpair, int, family, int, type, int, protocol, if (err < 0) goto out_release_both; - fd1 = sock_alloc_file(sock1, &newfile1, flags, NULL); + fd1 = get_unused_fd_flags(flags); if (unlikely(fd1 < 0)) { err = fd1; goto out_release_both; } - - fd2 = sock_alloc_file(sock2, &newfile2, flags, NULL); + fd2 = get_unused_fd_flags(flags); if (unlikely(fd2 < 0)) { err = fd2; + put_unused_fd(fd1); + goto out_release_both; + } + + newfile1 = sock_alloc_file(sock1, flags, NULL); + if (unlikely(IS_ERR(newfile1))) { + err = PTR_ERR(newfile1); + put_unused_fd(fd1); + put_unused_fd(fd2); + goto out_release_both; + } + + newfile2 = sock_alloc_file(sock2, flags, NULL); + if (IS_ERR(newfile2)) { + err = PTR_ERR(newfile2); fput(newfile1); put_unused_fd(fd1); + put_unused_fd(fd2); sock_release(sock2); goto out; } @@ -1608,13 +1618,19 @@ SYSCALL_DEFINE4(accept4, int, fd, struct sockaddr __user *, upeer_sockaddr, */ __module_get(newsock->ops->owner); - newfd = sock_alloc_file(newsock, &newfile, flags, - sock->sk->sk_prot_creator->name); + newfd = get_unused_fd_flags(flags); if (unlikely(newfd < 0)) { err = newfd; sock_release(newsock); goto out_put; } + newfile = sock_alloc_file(newsock, flags, sock->sk->sk_prot_creator->name); + if (unlikely(IS_ERR(newfile))) { + err = PTR_ERR(newfile); + put_unused_fd(newfd); + sock_release(newsock); + goto out_put; + } err = security_socket_accept(sock, newsock); if (err) diff --git a/samples/seccomp/Makefile b/samples/seccomp/Makefile index 16aa2d4..bbbd276 100644 --- a/samples/seccomp/Makefile +++ b/samples/seccomp/Makefile @@ -18,14 +18,22 @@ HOSTCFLAGS_bpf-direct.o += -idirafter $(objtree)/include bpf-direct-objs := bpf-direct.o # Try to match the kernel target. -ifeq ($(CONFIG_64BIT),) -HOSTCFLAGS_bpf-direct.o += -m32 -HOSTCFLAGS_dropper.o += -m32 -HOSTCFLAGS_bpf-helper.o += -m32 -HOSTCFLAGS_bpf-fancy.o += -m32 -HOSTLOADLIBES_bpf-direct += -m32 -HOSTLOADLIBES_bpf-fancy += -m32 -HOSTLOADLIBES_dropper += -m32 +ifndef CONFIG_64BIT + +# s390 has -m31 flag to build 31 bit binaries +ifndef CONFIG_S390 +MFLAG = -m32 +else +MFLAG = -m31 +endif + +HOSTCFLAGS_bpf-direct.o += $(MFLAG) +HOSTCFLAGS_dropper.o += $(MFLAG) +HOSTCFLAGS_bpf-helper.o += $(MFLAG) +HOSTCFLAGS_bpf-fancy.o += $(MFLAG) +HOSTLOADLIBES_bpf-direct += $(MFLAG) +HOSTLOADLIBES_bpf-fancy += $(MFLAG) +HOSTLOADLIBES_dropper += $(MFLAG) endif # Tell kbuild to always build the programs diff --git a/samples/seccomp/bpf-helper.h b/samples/seccomp/bpf-helper.h index 643279d..38ee70f 100644 --- a/samples/seccomp/bpf-helper.h +++ b/samples/seccomp/bpf-helper.h @@ -59,6 +59,16 @@ void seccomp_bpf_print(struct sock_filter *filter, size_t count); #define FIND_LABEL(labels, label) seccomp_bpf_label((labels), #label) #define EXPAND(...) __VA_ARGS__ + +/* Ensure that we load the logically correct offset. */ +#if __BYTE_ORDER == __LITTLE_ENDIAN +#define LO_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) +#elif __BYTE_ORDER == __BIG_ENDIAN +#define LO_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) + sizeof(__u32) +#else +#error "Unknown endianness" +#endif + /* Map all width-sensitive operations */ #if __BITS_PER_LONG == 32 @@ -70,21 +80,16 @@ void seccomp_bpf_print(struct sock_filter *filter, size_t count); #define JLE(x, jt) JLE32(x, EXPAND(jt)) #define JA(x, jt) JA32(x, EXPAND(jt)) #define ARG(i) ARG_32(i) -#define LO_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) #elif __BITS_PER_LONG == 64 /* Ensure that we load the logically correct offset. */ #if __BYTE_ORDER == __LITTLE_ENDIAN #define ENDIAN(_lo, _hi) _lo, _hi -#define LO_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) #define HI_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) + sizeof(__u32) #elif __BYTE_ORDER == __BIG_ENDIAN #define ENDIAN(_lo, _hi) _hi, _lo -#define LO_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) + sizeof(__u32) #define HI_ARG(idx) offsetof(struct seccomp_data, args[(idx)]) -#else -#error "Unknown endianness" #endif union arg64 { diff --git a/security/integrity/evm/evm_main.c b/security/integrity/evm/evm_main.c index 8901501..eb54845 100644 --- a/security/integrity/evm/evm_main.c +++ b/security/integrity/evm/evm_main.c @@ -34,6 +34,9 @@ char *evm_config_xattrnames[] = { #ifdef CONFIG_SECURITY_SMACK XATTR_NAME_SMACK, #endif +#ifdef CONFIG_IMA_APPRAISE + XATTR_NAME_IMA, +#endif XATTR_NAME_CAPS, NULL }; diff --git a/security/integrity/iint.c b/security/integrity/iint.c index 399641c..d82a5a1 100644 --- a/security/integrity/iint.c +++ b/security/integrity/iint.c @@ -22,7 +22,7 @@ #include "integrity.h" static struct rb_root integrity_iint_tree = RB_ROOT; -static DEFINE_SPINLOCK(integrity_iint_lock); +static DEFINE_RWLOCK(integrity_iint_lock); static struct kmem_cache *iint_cache __read_mostly; int iint_initialized; @@ -35,8 +35,6 @@ static struct integrity_iint_cache *__integrity_iint_find(struct inode *inode) struct integrity_iint_cache *iint; struct rb_node *n = integrity_iint_tree.rb_node; - assert_spin_locked(&integrity_iint_lock); - while (n) { iint = rb_entry(n, struct integrity_iint_cache, rb_node); @@ -63,9 +61,9 @@ struct integrity_iint_cache *integrity_iint_find(struct inode *inode) if (!IS_IMA(inode)) return NULL; - spin_lock(&integrity_iint_lock); + read_lock(&integrity_iint_lock); iint = __integrity_iint_find(inode); - spin_unlock(&integrity_iint_lock); + read_unlock(&integrity_iint_lock); return iint; } @@ -74,59 +72,53 @@ static void iint_free(struct integrity_iint_cache *iint) { iint->version = 0; iint->flags = 0UL; + iint->ima_status = INTEGRITY_UNKNOWN; iint->evm_status = INTEGRITY_UNKNOWN; kmem_cache_free(iint_cache, iint); } /** - * integrity_inode_alloc - allocate an iint associated with an inode + * integrity_inode_get - find or allocate an iint associated with an inode * @inode: pointer to the inode + * @return: allocated iint + * + * Caller must lock i_mutex */ -int integrity_inode_alloc(struct inode *inode) +struct integrity_iint_cache *integrity_inode_get(struct inode *inode) { struct rb_node **p; - struct rb_node *new_node, *parent = NULL; - struct integrity_iint_cache *new_iint, *test_iint; - int rc; + struct rb_node *node, *parent = NULL; + struct integrity_iint_cache *iint, *test_iint; - new_iint = kmem_cache_alloc(iint_cache, GFP_NOFS); - if (!new_iint) - return -ENOMEM; + iint = integrity_iint_find(inode); + if (iint) + return iint; - new_iint->inode = inode; - new_node = &new_iint->rb_node; + iint = kmem_cache_alloc(iint_cache, GFP_NOFS); + if (!iint) + return NULL; - mutex_lock(&inode->i_mutex); /* i_flags */ - spin_lock(&integrity_iint_lock); + write_lock(&integrity_iint_lock); p = &integrity_iint_tree.rb_node; while (*p) { parent = *p; test_iint = rb_entry(parent, struct integrity_iint_cache, rb_node); - rc = -EEXIST; if (inode < test_iint->inode) p = &(*p)->rb_left; - else if (inode > test_iint->inode) - p = &(*p)->rb_right; else - goto out_err; + p = &(*p)->rb_right; } + iint->inode = inode; + node = &iint->rb_node; inode->i_flags |= S_IMA; - rb_link_node(new_node, parent, p); - rb_insert_color(new_node, &integrity_iint_tree); + rb_link_node(node, parent, p); + rb_insert_color(node, &integrity_iint_tree); - spin_unlock(&integrity_iint_lock); - mutex_unlock(&inode->i_mutex); /* i_flags */ - - return 0; -out_err: - spin_unlock(&integrity_iint_lock); - mutex_unlock(&inode->i_mutex); /* i_flags */ - iint_free(new_iint); - - return rc; + write_unlock(&integrity_iint_lock); + return iint; } /** @@ -142,10 +134,10 @@ void integrity_inode_free(struct inode *inode) if (!IS_IMA(inode)) return; - spin_lock(&integrity_iint_lock); + write_lock(&integrity_iint_lock); iint = __integrity_iint_find(inode); rb_erase(&iint->rb_node, &integrity_iint_tree); - spin_unlock(&integrity_iint_lock); + write_unlock(&integrity_iint_lock); iint_free(iint); } @@ -157,7 +149,7 @@ static void init_once(void *foo) memset(iint, 0, sizeof *iint); iint->version = 0; iint->flags = 0UL; - mutex_init(&iint->mutex); + iint->ima_status = INTEGRITY_UNKNOWN; iint->evm_status = INTEGRITY_UNKNOWN; } diff --git a/security/integrity/ima/Kconfig b/security/integrity/ima/Kconfig index b9c1219..d232c73 100644 --- a/security/integrity/ima/Kconfig +++ b/security/integrity/ima/Kconfig @@ -11,6 +11,7 @@ config IMA select CRYPTO_SHA1 select TCG_TPM if HAS_IOMEM && !UML select TCG_TIS if TCG_TPM && X86 + select TCG_IBMVTPM if TCG_TPM && PPC64 help The Trusted Computing Group(TCG) runtime Integrity Measurement Architecture(IMA) maintains a list of hash @@ -55,3 +56,18 @@ config IMA_LSM_RULES default y help Disabling this option will disregard LSM based policy rules. + +config IMA_APPRAISE + bool "Appraise integrity measurements" + depends on IMA + default n + help + This option enables local measurement integrity appraisal. + It requires the system to be labeled with a security extended + attribute containing the file hash measurement. To protect + the security extended attributes from offline attack, enable + and configure EVM. + + For more information on integrity appraisal refer to: + <http://linux-ima.sourceforge.net> + If unsure, say N. diff --git a/security/integrity/ima/Makefile b/security/integrity/ima/Makefile index 5f740f6..3f2ca6b 100644 --- a/security/integrity/ima/Makefile +++ b/security/integrity/ima/Makefile @@ -8,3 +8,4 @@ obj-$(CONFIG_IMA) += ima.o ima-y := ima_fs.o ima_queue.o ima_init.o ima_main.o ima_crypto.o ima_api.o \ ima_policy.o ima-$(CONFIG_IMA_AUDIT) += ima_audit.o +ima-$(CONFIG_IMA_APPRAISE) += ima_appraise.o diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h index e7c99fd..8180add 100644 --- a/security/integrity/ima/ima.h +++ b/security/integrity/ima/ima.h @@ -40,6 +40,7 @@ enum tpm_pcrs { TPM_PCR0 = 0, TPM_PCR8 = 8 }; extern int ima_initialized; extern int ima_used_chip; extern char *ima_hash; +extern int ima_appraise; /* IMA inode template definition */ struct ima_template_data { @@ -107,11 +108,14 @@ static inline unsigned long ima_hash_key(u8 *digest) } /* LIM API function definitions */ +int ima_get_action(struct inode *inode, int mask, int function); int ima_must_measure(struct inode *inode, int mask, int function); int ima_collect_measurement(struct integrity_iint_cache *iint, struct file *file); void ima_store_measurement(struct integrity_iint_cache *iint, struct file *file, const unsigned char *filename); +void ima_audit_measurement(struct integrity_iint_cache *iint, + const unsigned char *filename); int ima_store_template(struct ima_template_entry *entry, int violation, struct inode *inode); void ima_template_show(struct seq_file *m, void *e, enum ima_show_type show); @@ -123,14 +127,45 @@ struct integrity_iint_cache *integrity_iint_insert(struct inode *inode); struct integrity_iint_cache *integrity_iint_find(struct inode *inode); /* IMA policy related functions */ -enum ima_hooks { FILE_CHECK = 1, FILE_MMAP, BPRM_CHECK }; +enum ima_hooks { FILE_CHECK = 1, FILE_MMAP, BPRM_CHECK, POST_SETATTR }; -int ima_match_policy(struct inode *inode, enum ima_hooks func, int mask); +int ima_match_policy(struct inode *inode, enum ima_hooks func, int mask, + int flags); void ima_init_policy(void); void ima_update_policy(void); ssize_t ima_parse_add_rule(char *); void ima_delete_rules(void); +/* Appraise integrity measurements */ +#define IMA_APPRAISE_ENFORCE 0x01 +#define IMA_APPRAISE_FIX 0x02 + +#ifdef CONFIG_IMA_APPRAISE +int ima_appraise_measurement(struct integrity_iint_cache *iint, + struct file *file, const unsigned char *filename); +int ima_must_appraise(struct inode *inode, enum ima_hooks func, int mask); +void ima_update_xattr(struct integrity_iint_cache *iint, struct file *file); + +#else +static inline int ima_appraise_measurement(struct integrity_iint_cache *iint, + struct file *file, + const unsigned char *filename) +{ + return INTEGRITY_UNKNOWN; +} + +static inline int ima_must_appraise(struct inode *inode, + enum ima_hooks func, int mask) +{ + return 0; +} + +static inline void ima_update_xattr(struct integrity_iint_cache *iint, + struct file *file) +{ +} +#endif + /* LSM based policy rules require audit */ #ifdef CONFIG_IMA_LSM_RULES diff --git a/security/integrity/ima/ima_api.c b/security/integrity/ima/ima_api.c index 032ff03..b356884 100644 --- a/security/integrity/ima/ima_api.c +++ b/security/integrity/ima/ima_api.c @@ -9,13 +9,17 @@ * License. * * File: ima_api.c - * Implements must_measure, collect_measurement, store_measurement, - * and store_template. + * Implements must_appraise_or_measure, collect_measurement, + * appraise_measurement, store_measurement and store_template. */ #include <linux/module.h> #include <linux/slab.h> - +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/xattr.h> +#include <linux/evm.h> #include "ima.h" + static const char *IMA_TEMPLATE_NAME = "ima"; /* @@ -93,7 +97,7 @@ err_out: } /** - * ima_must_measure - measure decision based on policy. + * ima_get_action - appraise & measure decision based on policy. * @inode: pointer to inode to measure * @mask: contains the permission mask (MAY_READ, MAY_WRITE, MAY_EXECUTE) * @function: calling function (FILE_CHECK, BPRM_CHECK, FILE_MMAP) @@ -105,15 +109,22 @@ err_out: * mask: contains the permission mask * fsmagic: hex value * - * Return 0 to measure. For matching a DONT_MEASURE policy, no policy, - * or other error, return an error code. -*/ -int ima_must_measure(struct inode *inode, int mask, int function) + * Returns IMA_MEASURE, IMA_APPRAISE mask. + * + */ +int ima_get_action(struct inode *inode, int mask, int function) { - int must_measure; + int flags = IMA_MEASURE | IMA_AUDIT | IMA_APPRAISE; + + if (!ima_appraise) + flags &= ~IMA_APPRAISE; - must_measure = ima_match_policy(inode, function, mask); - return must_measure ? 0 : -EACCES; + return ima_match_policy(inode, function, mask, flags); +} + +int ima_must_measure(struct inode *inode, int mask, int function) +{ + return ima_match_policy(inode, function, mask, IMA_MEASURE); } /* @@ -129,16 +140,24 @@ int ima_must_measure(struct inode *inode, int mask, int function) int ima_collect_measurement(struct integrity_iint_cache *iint, struct file *file) { - int result = -EEXIST; + struct inode *inode = file->f_dentry->d_inode; + const char *filename = file->f_dentry->d_name.name; + int result = 0; - if (!(iint->flags & IMA_MEASURED)) { + if (!(iint->flags & IMA_COLLECTED)) { u64 i_version = file->f_dentry->d_inode->i_version; - memset(iint->digest, 0, IMA_DIGEST_SIZE); - result = ima_calc_hash(file, iint->digest); - if (!result) + iint->ima_xattr.type = IMA_XATTR_DIGEST; + result = ima_calc_hash(file, iint->ima_xattr.digest); + if (!result) { iint->version = i_version; + iint->flags |= IMA_COLLECTED; + } } + if (result) + integrity_audit_msg(AUDIT_INTEGRITY_DATA, inode, + filename, "collect_data", "failed", + result, 0); return result; } @@ -167,6 +186,9 @@ void ima_store_measurement(struct integrity_iint_cache *iint, struct ima_template_entry *entry; int violation = 0; + if (iint->flags & IMA_MEASURED) + return; + entry = kmalloc(sizeof(*entry), GFP_KERNEL); if (!entry) { integrity_audit_msg(AUDIT_INTEGRITY_PCR, inode, filename, @@ -174,7 +196,7 @@ void ima_store_measurement(struct integrity_iint_cache *iint, return; } memset(&entry->template, 0, sizeof(entry->template)); - memcpy(entry->template.digest, iint->digest, IMA_DIGEST_SIZE); + memcpy(entry->template.digest, iint->ima_xattr.digest, IMA_DIGEST_SIZE); strcpy(entry->template.file_name, (strlen(filename) > IMA_EVENT_NAME_LEN_MAX) ? file->f_dentry->d_name.name : filename); @@ -185,3 +207,33 @@ void ima_store_measurement(struct integrity_iint_cache *iint, if (result < 0) kfree(entry); } + +void ima_audit_measurement(struct integrity_iint_cache *iint, + const unsigned char *filename) +{ + struct audit_buffer *ab; + char hash[(IMA_DIGEST_SIZE * 2) + 1]; + int i; + + if (iint->flags & IMA_AUDITED) + return; + + for (i = 0; i < IMA_DIGEST_SIZE; i++) + hex_byte_pack(hash + (i * 2), iint->ima_xattr.digest[i]); + hash[i * 2] = '\0'; + + ab = audit_log_start(current->audit_context, GFP_KERNEL, + AUDIT_INTEGRITY_RULE); + if (!ab) + return; + + audit_log_format(ab, "file="); + audit_log_untrustedstring(ab, filename); + audit_log_format(ab, " hash="); + audit_log_untrustedstring(ab, hash); + + audit_log_task_info(ab, current); + audit_log_end(ab); + + iint->flags |= IMA_AUDITED; +} diff --git a/security/integrity/ima/ima_appraise.c b/security/integrity/ima/ima_appraise.c new file mode 100644 index 0000000..0aa43bd --- /dev/null +++ b/security/integrity/ima/ima_appraise.c @@ -0,0 +1,263 @@ +/* + * Copyright (C) 2011 IBM Corporation + * + * Author: + * Mimi Zohar <zohar@us.ibm.com> + * + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License as published by + * the Free Software Foundation, version 2 of the License. + */ +#include <linux/module.h> +#include <linux/file.h> +#include <linux/fs.h> +#include <linux/xattr.h> +#include <linux/magic.h> +#include <linux/ima.h> +#include <linux/evm.h> + +#include "ima.h" + +static int __init default_appraise_setup(char *str) +{ + if (strncmp(str, "off", 3) == 0) + ima_appraise = 0; + else if (strncmp(str, "fix", 3) == 0) + ima_appraise = IMA_APPRAISE_FIX; + return 1; +} + +__setup("ima_appraise=", default_appraise_setup); + +/* + * ima_must_appraise - set appraise flag + * + * Return 1 to appraise + */ +int ima_must_appraise(struct inode *inode, enum ima_hooks func, int mask) +{ + if (!ima_appraise) + return 0; + + return ima_match_policy(inode, func, mask, IMA_APPRAISE); +} + +static void ima_fix_xattr(struct dentry *dentry, + struct integrity_iint_cache *iint) +{ + iint->ima_xattr.type = IMA_XATTR_DIGEST; + __vfs_setxattr_noperm(dentry, XATTR_NAME_IMA, (u8 *)&iint->ima_xattr, + sizeof iint->ima_xattr, 0); +} + +/* + * ima_appraise_measurement - appraise file measurement + * + * Call evm_verifyxattr() to verify the integrity of 'security.ima'. + * Assuming success, compare the xattr hash with the collected measurement. + * + * Return 0 on success, error code otherwise + */ +int ima_appraise_measurement(struct integrity_iint_cache *iint, + struct file *file, const unsigned char *filename) +{ + struct dentry *dentry = file->f_dentry; + struct inode *inode = dentry->d_inode; + struct evm_ima_xattr_data *xattr_value = NULL; + enum integrity_status status = INTEGRITY_UNKNOWN; + const char *op = "appraise_data"; + char *cause = "unknown"; + int rc; + + if (!ima_appraise) + return 0; + if (!inode->i_op->getxattr) + return INTEGRITY_UNKNOWN; + + if (iint->flags & IMA_APPRAISED) + return iint->ima_status; + + rc = vfs_getxattr_alloc(dentry, XATTR_NAME_IMA, (char **)&xattr_value, + 0, GFP_NOFS); + if (rc <= 0) { + if (rc && rc != -ENODATA) + goto out; + + cause = "missing-hash"; + status = + (inode->i_size == 0) ? INTEGRITY_PASS : INTEGRITY_NOLABEL; + goto out; + } + + status = evm_verifyxattr(dentry, XATTR_NAME_IMA, xattr_value, rc, iint); + if ((status != INTEGRITY_PASS) && (status != INTEGRITY_UNKNOWN)) { + if ((status == INTEGRITY_NOLABEL) + || (status == INTEGRITY_NOXATTRS)) + cause = "missing-HMAC"; + else if (status == INTEGRITY_FAIL) + cause = "invalid-HMAC"; + goto out; + } + + switch (xattr_value->type) { + case IMA_XATTR_DIGEST: + rc = memcmp(xattr_value->digest, iint->ima_xattr.digest, + IMA_DIGEST_SIZE); + if (rc) { + cause = "invalid-hash"; + status = INTEGRITY_FAIL; + print_hex_dump_bytes("security.ima: ", DUMP_PREFIX_NONE, + xattr_value, sizeof(*xattr_value)); + print_hex_dump_bytes("collected: ", DUMP_PREFIX_NONE, + (u8 *)&iint->ima_xattr, + sizeof iint->ima_xattr); + break; + } + status = INTEGRITY_PASS; + break; + case EVM_IMA_XATTR_DIGSIG: + iint->flags |= IMA_DIGSIG; + rc = integrity_digsig_verify(INTEGRITY_KEYRING_IMA, + xattr_value->digest, rc - 1, + iint->ima_xattr.digest, + IMA_DIGEST_SIZE); + if (rc == -EOPNOTSUPP) { + status = INTEGRITY_UNKNOWN; + } else if (rc) { + cause = "invalid-signature"; + status = INTEGRITY_FAIL; + } else { + status = INTEGRITY_PASS; + } + break; + default: + status = INTEGRITY_UNKNOWN; + cause = "unknown-ima-data"; + break; + } + +out: + if (status != INTEGRITY_PASS) { + if ((ima_appraise & IMA_APPRAISE_FIX) && + (!xattr_value || + xattr_value->type != EVM_IMA_XATTR_DIGSIG)) { + ima_fix_xattr(dentry, iint); + status = INTEGRITY_PASS; + } + integrity_audit_msg(AUDIT_INTEGRITY_DATA, inode, filename, + op, cause, rc, 0); + } else { + iint->flags |= IMA_APPRAISED; + } + iint->ima_status = status; + kfree(xattr_value); + return status; +} + +/* + * ima_update_xattr - update 'security.ima' hash value + */ +void ima_update_xattr(struct integrity_iint_cache *iint, struct file *file) +{ + struct dentry *dentry = file->f_dentry; + int rc = 0; + + /* do not collect and update hash for digital signatures */ + if (iint->flags & IMA_DIGSIG) + return; + + rc = ima_collect_measurement(iint, file); + if (rc < 0) + return; + + ima_fix_xattr(dentry, iint); +} + +/** + * ima_inode_post_setattr - reflect file metadata changes + * @dentry: pointer to the affected dentry + * + * Changes to a dentry's metadata might result in needing to appraise. + * + * This function is called from notify_change(), which expects the caller + * to lock the inode's i_mutex. + */ +void ima_inode_post_setattr(struct dentry *dentry) +{ + struct inode *inode = dentry->d_inode; + struct integrity_iint_cache *iint; + int must_appraise, rc; + + if (!ima_initialized || !ima_appraise || !S_ISREG(inode->i_mode) + || !inode->i_op->removexattr) + return; + + must_appraise = ima_must_appraise(inode, MAY_ACCESS, POST_SETATTR); + iint = integrity_iint_find(inode); + if (iint) { + if (must_appraise) + iint->flags |= IMA_APPRAISE; + else + iint->flags &= ~(IMA_APPRAISE | IMA_APPRAISED); + } + if (!must_appraise) + rc = inode->i_op->removexattr(dentry, XATTR_NAME_IMA); + return; +} + +/* + * ima_protect_xattr - protect 'security.ima' + * + * Ensure that not just anyone can modify or remove 'security.ima'. + */ +static int ima_protect_xattr(struct dentry *dentry, const char *xattr_name, + const void *xattr_value, size_t xattr_value_len) +{ + if (strcmp(xattr_name, XATTR_NAME_IMA) == 0) { + if (!capable(CAP_SYS_ADMIN)) + return -EPERM; + return 1; + } + return 0; +} + +static void ima_reset_appraise_flags(struct inode *inode) +{ + struct integrity_iint_cache *iint; + + if (!ima_initialized || !ima_appraise || !S_ISREG(inode->i_mode)) + return; + + iint = integrity_iint_find(inode); + if (!iint) + return; + + iint->flags &= ~IMA_DONE_MASK; + return; +} + +int ima_inode_setxattr(struct dentry *dentry, const char *xattr_name, + const void *xattr_value, size_t xattr_value_len) +{ + int result; + + result = ima_protect_xattr(dentry, xattr_name, xattr_value, + xattr_value_len); + if (result == 1) { + ima_reset_appraise_flags(dentry->d_inode); + result = 0; + } + return result; +} + +int ima_inode_removexattr(struct dentry *dentry, const char *xattr_name) +{ + int result; + + result = ima_protect_xattr(dentry, xattr_name, NULL, 0); + if (result == 1) { + ima_reset_appraise_flags(dentry->d_inode); + result = 0; + } + return result; +} diff --git a/security/integrity/ima/ima_crypto.c b/security/integrity/ima/ima_crypto.c index 9b3ade7..b21ee5b 100644 --- a/security/integrity/ima/ima_crypto.c +++ b/security/integrity/ima/ima_crypto.c @@ -48,7 +48,7 @@ int ima_calc_hash(struct file *file, char *digest) struct scatterlist sg[1]; loff_t i_size, offset = 0; char *rbuf; - int rc; + int rc, read = 0; rc = init_desc(&desc); if (rc != 0) @@ -59,6 +59,10 @@ int ima_calc_hash(struct file *file, char *digest) rc = -ENOMEM; goto out; } + if (!(file->f_mode & FMODE_READ)) { + file->f_mode |= FMODE_READ; + read = 1; + } i_size = i_size_read(file->f_dentry->d_inode); while (offset < i_size) { int rbuf_len; @@ -80,6 +84,8 @@ int ima_calc_hash(struct file *file, char *digest) kfree(rbuf); if (!rc) rc = crypto_hash_final(&desc, digest); + if (read) + file->f_mode &= ~FMODE_READ; out: crypto_free_hash(desc.tfm); return rc; diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c index be82949..73c9a26 100644 --- a/security/integrity/ima/ima_main.c +++ b/security/integrity/ima/ima_main.c @@ -22,12 +22,19 @@ #include <linux/mount.h> #include <linux/mman.h> #include <linux/slab.h> +#include <linux/xattr.h> #include <linux/ima.h> #include "ima.h" int ima_initialized; +#ifdef CONFIG_IMA_APPRAISE +int ima_appraise = IMA_APPRAISE_ENFORCE; +#else +int ima_appraise; +#endif + char *ima_hash = "sha1"; static int __init hash_setup(char *str) { @@ -52,7 +59,7 @@ static void ima_rdwr_violation_check(struct file *file) struct dentry *dentry = file->f_path.dentry; struct inode *inode = dentry->d_inode; fmode_t mode = file->f_mode; - int rc; + int must_measure; bool send_tomtou = false, send_writers = false; unsigned char *pathname = NULL, *pathbuf = NULL; @@ -67,8 +74,8 @@ static void ima_rdwr_violation_check(struct file *file) goto out; } - rc = ima_must_measure(inode, MAY_READ, FILE_CHECK); - if (rc < 0) + must_measure = ima_must_measure(inode, MAY_READ, FILE_CHECK); + if (!must_measure) goto out; if (atomic_read(&inode->i_writecount) > 0) @@ -100,17 +107,21 @@ out: } static void ima_check_last_writer(struct integrity_iint_cache *iint, - struct inode *inode, - struct file *file) + struct inode *inode, struct file *file) { fmode_t mode = file->f_mode; - mutex_lock(&iint->mutex); - if (mode & FMODE_WRITE && - atomic_read(&inode->i_writecount) == 1 && - iint->version != inode->i_version) - iint->flags &= ~IMA_MEASURED; - mutex_unlock(&iint->mutex); + if (!(mode & FMODE_WRITE)) + return; + + mutex_lock(&inode->i_mutex); + if (atomic_read(&inode->i_writecount) == 1 && + iint->version != inode->i_version) { + iint->flags &= ~IMA_DONE_MASK; + if (iint->flags & IMA_APPRAISE) + ima_update_xattr(iint, file); + } + mutex_unlock(&inode->i_mutex); } /** @@ -140,28 +151,37 @@ static int process_measurement(struct file *file, const unsigned char *filename, struct inode *inode = file->f_dentry->d_inode; struct integrity_iint_cache *iint; unsigned char *pathname = NULL, *pathbuf = NULL; - int rc = 0; + int rc = -ENOMEM, action, must_appraise; if (!ima_initialized || !S_ISREG(inode->i_mode)) return 0; - rc = ima_must_measure(inode, mask, function); - if (rc != 0) - return rc; -retry: - iint = integrity_iint_find(inode); - if (!iint) { - rc = integrity_inode_alloc(inode); - if (!rc || rc == -EEXIST) - goto retry; - return rc; - } + /* Determine if in appraise/audit/measurement policy, + * returns IMA_MEASURE, IMA_APPRAISE, IMA_AUDIT bitmask. */ + action = ima_get_action(inode, mask, function); + if (!action) + return 0; - mutex_lock(&iint->mutex); + must_appraise = action & IMA_APPRAISE; - rc = iint->flags & IMA_MEASURED ? 1 : 0; - if (rc != 0) + mutex_lock(&inode->i_mutex); + + iint = integrity_inode_get(inode); + if (!iint) + goto out; + + /* Determine if already appraised/measured based on bitmask + * (IMA_MEASURE, IMA_MEASURED, IMA_APPRAISE, IMA_APPRAISED, + * IMA_AUDIT, IMA_AUDITED) */ + iint->flags |= action; + action &= ~((iint->flags & IMA_DONE_MASK) >> 1); + + /* Nothing to do, just return existing appraised status */ + if (!action) { + if (iint->flags & IMA_APPRAISED) + rc = iint->ima_status; goto out; + } rc = ima_collect_measurement(iint, file); if (rc != 0) @@ -177,11 +197,18 @@ retry: pathname = NULL; } } - ima_store_measurement(iint, file, !pathname ? filename : pathname); + if (action & IMA_MEASURE) + ima_store_measurement(iint, file, + !pathname ? filename : pathname); + if (action & IMA_APPRAISE) + rc = ima_appraise_measurement(iint, file, + !pathname ? filename : pathname); + if (action & IMA_AUDIT) + ima_audit_measurement(iint, !pathname ? filename : pathname); kfree(pathbuf); out: - mutex_unlock(&iint->mutex); - return rc; + mutex_unlock(&inode->i_mutex); + return (rc && must_appraise) ? -EACCES : 0; } /** @@ -197,14 +224,14 @@ out: */ int ima_file_mmap(struct file *file, unsigned long prot) { - int rc; + int rc = 0; if (!file) return 0; if (prot & PROT_EXEC) rc = process_measurement(file, file->f_dentry->d_name.name, MAY_EXEC, FILE_MMAP); - return 0; + return (ima_appraise & IMA_APPRAISE_ENFORCE) ? rc : 0; } /** @@ -228,7 +255,7 @@ int ima_bprm_check(struct linux_binprm *bprm) (strcmp(bprm->filename, bprm->interp) == 0) ? bprm->filename : bprm->interp, MAY_EXEC, BPRM_CHECK); - return 0; + return (ima_appraise & IMA_APPRAISE_ENFORCE) ? rc : 0; } /** @@ -249,7 +276,7 @@ int ima_file_check(struct file *file, int mask) rc = process_measurement(file, file->f_dentry->d_name.name, mask & (MAY_READ | MAY_WRITE | MAY_EXEC), FILE_CHECK); - return 0; + return (ima_appraise & IMA_APPRAISE_ENFORCE) ? rc : 0; } EXPORT_SYMBOL_GPL(ima_file_check); diff --git a/security/integrity/ima/ima_policy.c b/security/integrity/ima/ima_policy.c index c84df05..c7dacd2 100644 --- a/security/integrity/ima/ima_policy.c +++ b/security/integrity/ima/ima_policy.c @@ -24,22 +24,29 @@ #define IMA_MASK 0x0002 #define IMA_FSMAGIC 0x0004 #define IMA_UID 0x0008 +#define IMA_FOWNER 0x0010 -enum ima_action { UNKNOWN = -1, DONT_MEASURE = 0, MEASURE }; +#define UNKNOWN 0 +#define MEASURE 0x0001 /* same as IMA_MEASURE */ +#define DONT_MEASURE 0x0002 +#define APPRAISE 0x0004 /* same as IMA_APPRAISE */ +#define DONT_APPRAISE 0x0008 +#define AUDIT 0x0040 #define MAX_LSM_RULES 6 enum lsm_rule_types { LSM_OBJ_USER, LSM_OBJ_ROLE, LSM_OBJ_TYPE, LSM_SUBJ_USER, LSM_SUBJ_ROLE, LSM_SUBJ_TYPE }; -struct ima_measure_rule_entry { +struct ima_rule_entry { struct list_head list; - enum ima_action action; + int action; unsigned int flags; enum ima_hooks func; int mask; unsigned long fsmagic; kuid_t uid; + kuid_t fowner; struct { void *rule; /* LSM file metadata specific */ int type; /* audit type */ @@ -48,7 +55,7 @@ struct ima_measure_rule_entry { /* * Without LSM specific knowledge, the default policy can only be - * written in terms of .action, .func, .mask, .fsmagic, and .uid + * written in terms of .action, .func, .mask, .fsmagic, .uid, and .fowner */ /* @@ -57,7 +64,7 @@ struct ima_measure_rule_entry { * normal users can easily run the machine out of memory simply building * and running executables. */ -static struct ima_measure_rule_entry default_rules[] = { +static struct ima_rule_entry default_rules[] = { {.action = DONT_MEASURE,.fsmagic = PROC_SUPER_MAGIC,.flags = IMA_FSMAGIC}, {.action = DONT_MEASURE,.fsmagic = SYSFS_MAGIC,.flags = IMA_FSMAGIC}, {.action = DONT_MEASURE,.fsmagic = DEBUGFS_MAGIC,.flags = IMA_FSMAGIC}, @@ -75,19 +82,41 @@ static struct ima_measure_rule_entry default_rules[] = { .flags = IMA_FUNC | IMA_MASK | IMA_UID}, }; -static LIST_HEAD(measure_default_rules); -static LIST_HEAD(measure_policy_rules); -static struct list_head *ima_measure; +static struct ima_rule_entry default_appraise_rules[] = { + {.action = DONT_APPRAISE,.fsmagic = PROC_SUPER_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = SYSFS_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = DEBUGFS_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = TMPFS_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = RAMFS_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = DEVPTS_SUPER_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = BINFMTFS_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = SECURITYFS_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = SELINUX_MAGIC,.flags = IMA_FSMAGIC}, + {.action = DONT_APPRAISE,.fsmagic = CGROUP_SUPER_MAGIC,.flags = IMA_FSMAGIC}, + {.action = APPRAISE,.fowner = GLOBAL_ROOT_UID,.flags = IMA_FOWNER}, +}; + +static LIST_HEAD(ima_default_rules); +static LIST_HEAD(ima_policy_rules); +static struct list_head *ima_rules; -static DEFINE_MUTEX(ima_measure_mutex); +static DEFINE_MUTEX(ima_rules_mutex); static bool ima_use_tcb __initdata; -static int __init default_policy_setup(char *str) +static int __init default_measure_policy_setup(char *str) { ima_use_tcb = 1; return 1; } -__setup("ima_tcb", default_policy_setup); +__setup("ima_tcb", default_measure_policy_setup); + +static bool ima_use_appraise_tcb __initdata; +static int __init default_appraise_policy_setup(char *str) +{ + ima_use_appraise_tcb = 1; + return 1; +} +__setup("ima_appraise_tcb", default_appraise_policy_setup); /** * ima_match_rules - determine whether an inode matches the measure rule. @@ -98,7 +127,7 @@ __setup("ima_tcb", default_policy_setup); * * Returns true on rule match, false on failure. */ -static bool ima_match_rules(struct ima_measure_rule_entry *rule, +static bool ima_match_rules(struct ima_rule_entry *rule, struct inode *inode, enum ima_hooks func, int mask) { struct task_struct *tsk = current; @@ -114,6 +143,8 @@ static bool ima_match_rules(struct ima_measure_rule_entry *rule, return false; if ((rule->flags & IMA_UID) && !uid_eq(rule->uid, cred->uid)) return false; + if ((rule->flags & IMA_FOWNER) && !uid_eq(rule->fowner, inode->i_uid)) + return false; for (i = 0; i < MAX_LSM_RULES; i++) { int rc = 0; u32 osid, sid; @@ -163,39 +194,61 @@ static bool ima_match_rules(struct ima_measure_rule_entry *rule, * as elements in the list are never deleted, nor does the list * change.) */ -int ima_match_policy(struct inode *inode, enum ima_hooks func, int mask) +int ima_match_policy(struct inode *inode, enum ima_hooks func, int mask, + int flags) { - struct ima_measure_rule_entry *entry; + struct ima_rule_entry *entry; + int action = 0, actmask = flags | (flags << 1); + + list_for_each_entry(entry, ima_rules, list) { + + if (!(entry->action & actmask)) + continue; + + if (!ima_match_rules(entry, inode, func, mask)) + continue; - list_for_each_entry(entry, ima_measure, list) { - bool rc; + action |= entry->action & IMA_DO_MASK; + if (entry->action & IMA_DO_MASK) + actmask &= ~(entry->action | entry->action << 1); + else + actmask &= ~(entry->action | entry->action >> 1); - rc = ima_match_rules(entry, inode, func, mask); - if (rc) - return entry->action; + if (!actmask) + break; } - return 0; + + return action; } /** * ima_init_policy - initialize the default measure rules. * - * ima_measure points to either the measure_default_rules or the - * the new measure_policy_rules. + * ima_rules points to either the ima_default_rules or the + * the new ima_policy_rules. */ void __init ima_init_policy(void) { - int i, entries; + int i, measure_entries, appraise_entries; /* if !ima_use_tcb set entries = 0 so we load NO default rules */ - if (ima_use_tcb) - entries = ARRAY_SIZE(default_rules); - else - entries = 0; - - for (i = 0; i < entries; i++) - list_add_tail(&default_rules[i].list, &measure_default_rules); - ima_measure = &measure_default_rules; + measure_entries = ima_use_tcb ? ARRAY_SIZE(default_rules) : 0; + appraise_entries = ima_use_appraise_tcb ? + ARRAY_SIZE(default_appraise_rules) : 0; + + for (i = 0; i < measure_entries + appraise_entries; i++) { + if (i < measure_entries) + list_add_tail(&default_rules[i].list, + &ima_default_rules); + else { + int j = i - measure_entries; + + list_add_tail(&default_appraise_rules[j].list, + &ima_default_rules); + } + } + + ima_rules = &ima_default_rules; } /** @@ -212,8 +265,8 @@ void ima_update_policy(void) int result = 1; int audit_info = 0; - if (ima_measure == &measure_default_rules) { - ima_measure = &measure_policy_rules; + if (ima_rules == &ima_default_rules) { + ima_rules = &ima_policy_rules; cause = "complete"; result = 0; } @@ -224,14 +277,19 @@ void ima_update_policy(void) enum { Opt_err = -1, Opt_measure = 1, Opt_dont_measure, + Opt_appraise, Opt_dont_appraise, + Opt_audit, Opt_obj_user, Opt_obj_role, Opt_obj_type, Opt_subj_user, Opt_subj_role, Opt_subj_type, - Opt_func, Opt_mask, Opt_fsmagic, Opt_uid + Opt_func, Opt_mask, Opt_fsmagic, Opt_uid, Opt_fowner }; static match_table_t policy_tokens = { {Opt_measure, "measure"}, {Opt_dont_measure, "dont_measure"}, + {Opt_appraise, "appraise"}, + {Opt_dont_appraise, "dont_appraise"}, + {Opt_audit, "audit"}, {Opt_obj_user, "obj_user=%s"}, {Opt_obj_role, "obj_role=%s"}, {Opt_obj_type, "obj_type=%s"}, @@ -242,10 +300,11 @@ static match_table_t policy_tokens = { {Opt_mask, "mask=%s"}, {Opt_fsmagic, "fsmagic=%s"}, {Opt_uid, "uid=%s"}, + {Opt_fowner, "fowner=%s"}, {Opt_err, NULL} }; -static int ima_lsm_rule_init(struct ima_measure_rule_entry *entry, +static int ima_lsm_rule_init(struct ima_rule_entry *entry, char *args, int lsm_rule, int audit_type) { int result; @@ -269,7 +328,7 @@ static void ima_log_string(struct audit_buffer *ab, char *key, char *value) audit_log_format(ab, " "); } -static int ima_parse_rule(char *rule, struct ima_measure_rule_entry *entry) +static int ima_parse_rule(char *rule, struct ima_rule_entry *entry) { struct audit_buffer *ab; char *p; @@ -278,6 +337,7 @@ static int ima_parse_rule(char *rule, struct ima_measure_rule_entry *entry) ab = audit_log_start(NULL, GFP_KERNEL, AUDIT_INTEGRITY_RULE); entry->uid = INVALID_UID; + entry->fowner = INVALID_UID; entry->action = UNKNOWN; while ((p = strsep(&rule, " \t")) != NULL) { substring_t args[MAX_OPT_ARGS]; @@ -306,11 +366,35 @@ static int ima_parse_rule(char *rule, struct ima_measure_rule_entry *entry) entry->action = DONT_MEASURE; break; + case Opt_appraise: + ima_log_string(ab, "action", "appraise"); + + if (entry->action != UNKNOWN) + result = -EINVAL; + + entry->action = APPRAISE; + break; + case Opt_dont_appraise: + ima_log_string(ab, "action", "dont_appraise"); + + if (entry->action != UNKNOWN) + result = -EINVAL; + + entry->action = DONT_APPRAISE; + break; + case Opt_audit: + ima_log_string(ab, "action", "audit"); + + if (entry->action != UNKNOWN) + result = -EINVAL; + + entry->action = AUDIT; + break; case Opt_func: ima_log_string(ab, "func", args[0].from); if (entry->func) - result = -EINVAL; + result = -EINVAL; if (strcmp(args[0].from, "FILE_CHECK") == 0) entry->func = FILE_CHECK; @@ -375,6 +459,23 @@ static int ima_parse_rule(char *rule, struct ima_measure_rule_entry *entry) entry->flags |= IMA_UID; } break; + case Opt_fowner: + ima_log_string(ab, "fowner", args[0].from); + + if (uid_valid(entry->fowner)) { + result = -EINVAL; + break; + } + + result = strict_strtoul(args[0].from, 10, &lnum); + if (!result) { + entry->fowner = make_kuid(current_user_ns(), (uid_t)lnum); + if (!uid_valid(entry->fowner) || (((uid_t)lnum) != lnum)) + result = -EINVAL; + else + entry->flags |= IMA_FOWNER; + } + break; case Opt_obj_user: ima_log_string(ab, "obj_user", args[0].from); result = ima_lsm_rule_init(entry, args[0].from, @@ -426,7 +527,7 @@ static int ima_parse_rule(char *rule, struct ima_measure_rule_entry *entry) } /** - * ima_parse_add_rule - add a rule to measure_policy_rules + * ima_parse_add_rule - add a rule to ima_policy_rules * @rule - ima measurement policy rule * * Uses a mutex to protect the policy list from multiple concurrent writers. @@ -436,12 +537,12 @@ ssize_t ima_parse_add_rule(char *rule) { const char *op = "update_policy"; char *p; - struct ima_measure_rule_entry *entry; + struct ima_rule_entry *entry; ssize_t result, len; int audit_info = 0; /* Prevent installed policy from changing */ - if (ima_measure != &measure_default_rules) { + if (ima_rules != &ima_default_rules) { integrity_audit_msg(AUDIT_INTEGRITY_STATUS, NULL, NULL, op, "already exists", -EACCES, audit_info); @@ -474,9 +575,9 @@ ssize_t ima_parse_add_rule(char *rule) return result; } - mutex_lock(&ima_measure_mutex); - list_add_tail(&entry->list, &measure_policy_rules); - mutex_unlock(&ima_measure_mutex); + mutex_lock(&ima_rules_mutex); + list_add_tail(&entry->list, &ima_policy_rules); + mutex_unlock(&ima_rules_mutex); return len; } @@ -484,12 +585,12 @@ ssize_t ima_parse_add_rule(char *rule) /* ima_delete_rules called to cleanup invalid policy */ void ima_delete_rules(void) { - struct ima_measure_rule_entry *entry, *tmp; + struct ima_rule_entry *entry, *tmp; - mutex_lock(&ima_measure_mutex); - list_for_each_entry_safe(entry, tmp, &measure_policy_rules, list) { + mutex_lock(&ima_rules_mutex); + list_for_each_entry_safe(entry, tmp, &ima_policy_rules, list) { list_del(&entry->list); kfree(entry); } - mutex_unlock(&ima_measure_mutex); + mutex_unlock(&ima_rules_mutex); } diff --git a/security/integrity/integrity.h b/security/integrity/integrity.h index 7a25ece..e9db763 100644 --- a/security/integrity/integrity.h +++ b/security/integrity/integrity.h @@ -15,8 +15,22 @@ #include <linux/integrity.h> #include <crypto/sha.h> +/* iint action cache flags */ +#define IMA_MEASURE 0x0001 +#define IMA_MEASURED 0x0002 +#define IMA_APPRAISE 0x0004 +#define IMA_APPRAISED 0x0008 +/*#define IMA_COLLECT 0x0010 do not use this flag */ +#define IMA_COLLECTED 0x0020 +#define IMA_AUDIT 0x0040 +#define IMA_AUDITED 0x0080 + /* iint cache flags */ -#define IMA_MEASURED 0x01 +#define IMA_DIGSIG 0x0100 + +#define IMA_DO_MASK (IMA_MEASURE | IMA_APPRAISE | IMA_AUDIT) +#define IMA_DONE_MASK (IMA_MEASURED | IMA_APPRAISED | IMA_AUDITED \ + | IMA_COLLECTED) enum evm_ima_xattr_type { IMA_XATTR_DIGEST = 0x01, @@ -34,9 +48,9 @@ struct integrity_iint_cache { struct rb_node rb_node; /* rooted in integrity_iint_tree */ struct inode *inode; /* back pointer to inode in question */ u64 version; /* track inode changes */ - unsigned char flags; - u8 digest[SHA1_DIGEST_SIZE]; - struct mutex mutex; /* protects: version, flags, digest */ + unsigned short flags; + struct evm_ima_xattr_data ima_xattr; + enum integrity_status ima_status; enum integrity_status evm_status; }; diff --git a/security/keys/trusted.c b/security/keys/trusted.c index 2d5d041..3f163d0 100644 --- a/security/keys/trusted.c +++ b/security/keys/trusted.c @@ -369,38 +369,6 @@ static int trusted_tpm_send(const u32 chip_num, unsigned char *cmd, } /* - * get a random value from TPM - */ -static int tpm_get_random(struct tpm_buf *tb, unsigned char *buf, uint32_t len) -{ - int ret; - - INIT_BUF(tb); - store16(tb, TPM_TAG_RQU_COMMAND); - store32(tb, TPM_GETRANDOM_SIZE); - store32(tb, TPM_ORD_GETRANDOM); - store32(tb, len); - ret = trusted_tpm_send(TPM_ANY_NUM, tb->data, sizeof tb->data); - if (!ret) - memcpy(buf, tb->data + TPM_GETRANDOM_SIZE, len); - return ret; -} - -static int my_get_random(unsigned char *buf, int len) -{ - struct tpm_buf *tb; - int ret; - - tb = kmalloc(sizeof *tb, GFP_KERNEL); - if (!tb) - return -ENOMEM; - ret = tpm_get_random(tb, buf, len); - - kfree(tb); - return ret; -} - -/* * Lock a trusted key, by extending a selected PCR. * * Prevents a trusted key that is sealed to PCRs from being accessed. @@ -413,8 +381,8 @@ static int pcrlock(const int pcrnum) if (!capable(CAP_SYS_ADMIN)) return -EPERM; - ret = my_get_random(hash, SHA1_DIGEST_SIZE); - if (ret < 0) + ret = tpm_get_random(TPM_ANY_NUM, hash, SHA1_DIGEST_SIZE); + if (ret != SHA1_DIGEST_SIZE) return ret; return tpm_pcr_extend(TPM_ANY_NUM, pcrnum, hash) ? -EINVAL : 0; } @@ -429,8 +397,8 @@ static int osap(struct tpm_buf *tb, struct osapsess *s, unsigned char ononce[TPM_NONCE_SIZE]; int ret; - ret = tpm_get_random(tb, ononce, TPM_NONCE_SIZE); - if (ret < 0) + ret = tpm_get_random(TPM_ANY_NUM, ononce, TPM_NONCE_SIZE); + if (ret != TPM_NONCE_SIZE) return ret; INIT_BUF(tb); @@ -524,8 +492,8 @@ static int tpm_seal(struct tpm_buf *tb, uint16_t keytype, if (ret < 0) goto out; - ret = tpm_get_random(tb, td->nonceodd, TPM_NONCE_SIZE); - if (ret < 0) + ret = tpm_get_random(TPM_ANY_NUM, td->nonceodd, TPM_NONCE_SIZE); + if (ret != TPM_NONCE_SIZE) goto out; ordinal = htonl(TPM_ORD_SEAL); datsize = htonl(datalen); @@ -634,8 +602,8 @@ static int tpm_unseal(struct tpm_buf *tb, ordinal = htonl(TPM_ORD_UNSEAL); keyhndl = htonl(SRKHANDLE); - ret = tpm_get_random(tb, nonceodd, TPM_NONCE_SIZE); - if (ret < 0) { + ret = tpm_get_random(TPM_ANY_NUM, nonceodd, TPM_NONCE_SIZE); + if (ret != TPM_NONCE_SIZE) { pr_info("trusted_key: tpm_get_random failed (%d)\n", ret); return ret; } @@ -935,6 +903,7 @@ static int trusted_instantiate(struct key *key, const void *data, char *datablob; int ret = 0; int key_cmd; + size_t key_len; if (datalen <= 0 || datalen > 32767 || !data) return -EINVAL; @@ -974,8 +943,9 @@ static int trusted_instantiate(struct key *key, const void *data, pr_info("trusted_key: key_unseal failed (%d)\n", ret); break; case Opt_new: - ret = my_get_random(payload->key, payload->key_len); - if (ret < 0) { + key_len = payload->key_len; + ret = tpm_get_random(TPM_ANY_NUM, payload->key, key_len); + if (ret != key_len) { pr_info("trusted_key: key_create failed (%d)\n", ret); goto out; } diff --git a/security/security.c b/security/security.c index f9a2f2e..3724029 100644 --- a/security/security.c +++ b/security/security.c @@ -136,11 +136,23 @@ int __init register_security(struct security_operations *ops) int security_ptrace_access_check(struct task_struct *child, unsigned int mode) { +#ifdef CONFIG_SECURITY_YAMA_STACKED + int rc; + rc = yama_ptrace_access_check(child, mode); + if (rc) + return rc; +#endif return security_ops->ptrace_access_check(child, mode); } int security_ptrace_traceme(struct task_struct *parent) { +#ifdef CONFIG_SECURITY_YAMA_STACKED + int rc; + rc = yama_ptrace_traceme(parent); + if (rc) + return rc; +#endif return security_ops->ptrace_traceme(parent); } @@ -561,6 +573,9 @@ int security_inode_setxattr(struct dentry *dentry, const char *name, ret = security_ops->inode_setxattr(dentry, name, value, size, flags); if (ret) return ret; + ret = ima_inode_setxattr(dentry, name, value, size); + if (ret) + return ret; return evm_inode_setxattr(dentry, name, value, size); } @@ -596,6 +611,9 @@ int security_inode_removexattr(struct dentry *dentry, const char *name) ret = security_ops->inode_removexattr(dentry, name); if (ret) return ret; + ret = ima_inode_removexattr(dentry, name); + if (ret) + return ret; return evm_inode_removexattr(dentry, name); } @@ -761,6 +779,9 @@ int security_task_create(unsigned long clone_flags) void security_task_free(struct task_struct *task) { +#ifdef CONFIG_SECURITY_YAMA_STACKED + yama_task_free(task); +#endif security_ops->task_free(task); } @@ -876,6 +897,12 @@ int security_task_wait(struct task_struct *p) int security_task_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5) { +#ifdef CONFIG_SECURITY_YAMA_STACKED + int rc; + rc = yama_task_prctl(option, arg2, arg3, arg4, arg5); + if (rc != -ENOSYS) + return rc; +#endif return security_ops->task_prctl(option, arg2, arg3, arg4, arg5); } diff --git a/security/selinux/hooks.c b/security/selinux/hooks.c index 6c77f63..651d845 100644 --- a/security/selinux/hooks.c +++ b/security/selinux/hooks.c @@ -2088,15 +2088,19 @@ static int selinux_bprm_secureexec(struct linux_binprm *bprm) return (atsecure || cap_bprm_secureexec(bprm)); } +static int match_file(const void *p, struct file *file, unsigned fd) +{ + return file_has_perm(p, file, file_to_av(file)) ? fd + 1 : 0; +} + /* Derived from fs/exec.c:flush_old_files. */ static inline void flush_unauthorized_files(const struct cred *cred, struct files_struct *files) { struct file *file, *devnull = NULL; struct tty_struct *tty; - struct fdtable *fdt; - long j = -1; int drop_tty = 0; + unsigned n; tty = get_current_tty(); if (tty) { @@ -2123,58 +2127,23 @@ static inline void flush_unauthorized_files(const struct cred *cred, no_tty(); /* Revalidate access to inherited open files. */ - spin_lock(&files->file_lock); - for (;;) { - unsigned long set, i; - int fd; - - j++; - i = j * BITS_PER_LONG; - fdt = files_fdtable(files); - if (i >= fdt->max_fds) - break; - set = fdt->open_fds[j]; - if (!set) - continue; - spin_unlock(&files->file_lock); - for ( ; set ; i++, set >>= 1) { - if (set & 1) { - file = fget(i); - if (!file) - continue; - if (file_has_perm(cred, - file, - file_to_av(file))) { - sys_close(i); - fd = get_unused_fd(); - if (fd != i) { - if (fd >= 0) - put_unused_fd(fd); - fput(file); - continue; - } - if (devnull) { - get_file(devnull); - } else { - devnull = dentry_open( - &selinux_null, - O_RDWR, cred); - if (IS_ERR(devnull)) { - devnull = NULL; - put_unused_fd(fd); - fput(file); - continue; - } - } - fd_install(fd, devnull); - } - fput(file); - } - } - spin_lock(&files->file_lock); + n = iterate_fd(files, 0, match_file, cred); + if (!n) /* none found? */ + return; + devnull = dentry_open(&selinux_null, O_RDWR, cred); + if (!IS_ERR(devnull)) { + /* replace all the matching ones with this */ + do { + replace_fd(n - 1, get_file(devnull), 0); + } while ((n = iterate_fd(files, n, match_file, cred)) != 0); + fput(devnull); + } else { + /* just close all the matching ones */ + do { + replace_fd(n - 1, NULL, 0); + } while ((n = iterate_fd(files, n, match_file, cred)) != 0); } - spin_unlock(&files->file_lock); } /* diff --git a/security/smack/smack_lsm.c b/security/smack/smack_lsm.c index 8221514..2874c73 100644 --- a/security/smack/smack_lsm.c +++ b/security/smack/smack_lsm.c @@ -1691,40 +1691,19 @@ static int smack_task_kill(struct task_struct *p, struct siginfo *info, * smack_task_wait - Smack access check for waiting * @p: task to wait for * - * Returns 0 if current can wait for p, error code otherwise + * Returns 0 */ static int smack_task_wait(struct task_struct *p) { - struct smk_audit_info ad; - char *sp = smk_of_current(); - char *tsp = smk_of_forked(task_security(p)); - int rc; - - /* we don't log here, we can be overriden */ - rc = smk_access(tsp, sp, MAY_WRITE, NULL); - if (rc == 0) - goto out_log; - /* - * Allow the operation to succeed if either task - * has privilege to perform operations that might - * account for the smack labels having gotten to - * be different in the first place. - * - * This breaks the strict subject/object access - * control ideal, taking the object's privilege - * state into account in the decision as well as - * the smack value. + * Allow the operation to succeed. + * Zombies are bad. + * In userless environments (e.g. phones) programs + * get marked with SMACK64EXEC and even if the parent + * and child shouldn't be talking the parent still + * may expect to know when the child exits. */ - if (smack_privileged(CAP_MAC_OVERRIDE) || - has_capability(p, CAP_MAC_OVERRIDE)) - rc = 0; - /* we log only if we didn't get overriden */ - out_log: - smk_ad_init(&ad, __func__, LSM_AUDIT_DATA_TASK); - smk_ad_setfield_u_tsk(&ad, p); - smack_log(tsp, sp, MAY_WRITE, rc, &ad); - return rc; + return 0; } /** @@ -2705,9 +2684,7 @@ static int smack_getprocattr(struct task_struct *p, char *name, char **value) static int smack_setprocattr(struct task_struct *p, char *name, void *value, size_t size) { - int rc; struct task_smack *tsp; - struct task_smack *oldtsp; struct cred *new; char *newsmack; @@ -2737,21 +2714,13 @@ static int smack_setprocattr(struct task_struct *p, char *name, if (newsmack == smack_known_web.smk_known) return -EPERM; - oldtsp = p->cred->security; new = prepare_creds(); if (new == NULL) return -ENOMEM; - tsp = new_task_smack(newsmack, oldtsp->smk_forked, GFP_KERNEL); - if (tsp == NULL) { - kfree(new); - return -ENOMEM; - } - rc = smk_copy_rules(&tsp->smk_rules, &oldtsp->smk_rules, GFP_KERNEL); - if (rc != 0) - return rc; + tsp = new->security; + tsp->smk_task = newsmack; - new->security = tsp; commit_creds(new); return size; } diff --git a/security/smack/smackfs.c b/security/smack/smackfs.c index b1b768e..99929a5 100644 --- a/security/smack/smackfs.c +++ b/security/smack/smackfs.c @@ -49,6 +49,7 @@ enum smk_inos { SMK_LOAD_SELF2 = 15, /* load task specific rules with long labels */ SMK_ACCESS2 = 16, /* make an access check with long labels */ SMK_CIPSO2 = 17, /* load long label -> CIPSO mapping */ + SMK_REVOKE_SUBJ = 18, /* set rules with subject label to '-' */ }; /* @@ -1992,6 +1993,77 @@ static const struct file_operations smk_access2_ops = { }; /** + * smk_write_revoke_subj - write() for /smack/revoke-subject + * @file: file pointer + * @buf: data from user space + * @count: bytes sent + * @ppos: where to start - must be 0 + */ +static ssize_t smk_write_revoke_subj(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + char *data = NULL; + const char *cp = NULL; + struct smack_known *skp; + struct smack_rule *sp; + struct list_head *rule_list; + struct mutex *rule_lock; + int rc = count; + + if (*ppos != 0) + return -EINVAL; + + if (!smack_privileged(CAP_MAC_ADMIN)) + return -EPERM; + + if (count == 0 || count > SMK_LONGLABEL) + return -EINVAL; + + data = kzalloc(count, GFP_KERNEL); + if (data == NULL) + return -ENOMEM; + + if (copy_from_user(data, buf, count) != 0) { + rc = -EFAULT; + goto free_out; + } + + cp = smk_parse_smack(data, count); + if (cp == NULL) { + rc = -EINVAL; + goto free_out; + } + + skp = smk_find_entry(cp); + if (skp == NULL) { + rc = -EINVAL; + goto free_out; + } + + rule_list = &skp->smk_rules; + rule_lock = &skp->smk_rules_lock; + + mutex_lock(rule_lock); + + list_for_each_entry_rcu(sp, rule_list, list) + sp->smk_access = 0; + + mutex_unlock(rule_lock); + +free_out: + kfree(data); + kfree(cp); + return rc; +} + +static const struct file_operations smk_revoke_subj_ops = { + .write = smk_write_revoke_subj, + .read = simple_transaction_read, + .release = simple_transaction_release, + .llseek = generic_file_llseek, +}; + +/** * smk_fill_super - fill the /smackfs superblock * @sb: the empty superblock * @data: unused @@ -2037,6 +2109,9 @@ static int smk_fill_super(struct super_block *sb, void *data, int silent) "access2", &smk_access2_ops, S_IRUGO|S_IWUGO}, [SMK_CIPSO2] = { "cipso2", &smk_cipso2_ops, S_IRUGO|S_IWUSR}, + [SMK_REVOKE_SUBJ] = { + "revoke-subject", &smk_revoke_subj_ops, + S_IRUGO|S_IWUSR}, /* last one */ {""} }; diff --git a/security/yama/Kconfig b/security/yama/Kconfig index 51d6709..20ef514 100644 --- a/security/yama/Kconfig +++ b/security/yama/Kconfig @@ -11,3 +11,11 @@ config SECURITY_YAMA Further information can be found in Documentation/security/Yama.txt. If you are unsure how to answer this question, answer N. + +config SECURITY_YAMA_STACKED + bool "Yama stacked with other LSMs" + depends on SECURITY_YAMA + default n + help + When Yama is built into the kernel, force it to stack with the + selected primary LSM. diff --git a/security/yama/yama_lsm.c b/security/yama/yama_lsm.c index 0cc99a3..b4c2984 100644 --- a/security/yama/yama_lsm.c +++ b/security/yama/yama_lsm.c @@ -100,7 +100,7 @@ static void yama_ptracer_del(struct task_struct *tracer, * yama_task_free - check for task_pid to remove from exception list * @task: task being removed */ -static void yama_task_free(struct task_struct *task) +void yama_task_free(struct task_struct *task) { yama_ptracer_del(task, task); } @@ -116,7 +116,7 @@ static void yama_task_free(struct task_struct *task) * Return 0 on success, -ve on error. -ENOSYS is returned when Yama * does not handle the given option. */ -static int yama_task_prctl(int option, unsigned long arg2, unsigned long arg3, +int yama_task_prctl(int option, unsigned long arg2, unsigned long arg3, unsigned long arg4, unsigned long arg5) { int rc; @@ -143,7 +143,7 @@ static int yama_task_prctl(int option, unsigned long arg2, unsigned long arg3, if (arg2 == 0) { yama_ptracer_del(NULL, myself); rc = 0; - } else if (arg2 == PR_SET_PTRACER_ANY) { + } else if (arg2 == PR_SET_PTRACER_ANY || (int)arg2 == -1) { rc = yama_ptracer_add(NULL, myself); } else { struct task_struct *tracer; @@ -243,7 +243,7 @@ static int ptracer_exception_found(struct task_struct *tracer, * * Returns 0 if following the ptrace is allowed, -ve on error. */ -static int yama_ptrace_access_check(struct task_struct *child, +int yama_ptrace_access_check(struct task_struct *child, unsigned int mode) { int rc; @@ -293,7 +293,7 @@ static int yama_ptrace_access_check(struct task_struct *child, * * Returns 0 if following the ptrace is allowed, -ve on error. */ -static int yama_ptrace_traceme(struct task_struct *parent) +int yama_ptrace_traceme(struct task_struct *parent) { int rc; @@ -324,6 +324,7 @@ static int yama_ptrace_traceme(struct task_struct *parent) return rc; } +#ifndef CONFIG_SECURITY_YAMA_STACKED static struct security_operations yama_ops = { .name = "yama", @@ -332,6 +333,7 @@ static struct security_operations yama_ops = { .task_prctl = yama_task_prctl, .task_free = yama_task_free, }; +#endif #ifdef CONFIG_SYSCTL static int yama_dointvec_minmax(struct ctl_table *table, int write, @@ -378,13 +380,17 @@ static struct ctl_table yama_sysctl_table[] = { static __init int yama_init(void) { +#ifndef CONFIG_SECURITY_YAMA_STACKED if (!security_module_enable(&yama_ops)) return 0; +#endif printk(KERN_INFO "Yama: becoming mindful.\n"); +#ifndef CONFIG_SECURITY_YAMA_STACKED if (register_security(&yama_ops)) panic("Yama: kernel registration failed.\n"); +#endif #ifdef CONFIG_SYSCTL if (!register_sysctl_paths(yama_sysctl_path, yama_sysctl_table)) diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c index 53b5ada..20554ef 100644 --- a/sound/core/pcm_native.c +++ b/sound/core/pcm_native.c @@ -1563,25 +1563,25 @@ static int snd_pcm_drop(struct snd_pcm_substream *substream) /* WARNING: Don't forget to fput back the file */ -static struct file *snd_pcm_file_fd(int fd) +static struct file *snd_pcm_file_fd(int fd, int *fput_needed) { struct file *file; struct inode *inode; unsigned int minor; - file = fget(fd); + file = fget_light(fd, fput_needed); if (!file) return NULL; inode = file->f_path.dentry->d_inode; if (!S_ISCHR(inode->i_mode) || imajor(inode) != snd_major) { - fput(file); + fput_light(file, *fput_needed); return NULL; } minor = iminor(inode); if (!snd_lookup_minor_data(minor, SNDRV_DEVICE_TYPE_PCM_PLAYBACK) && !snd_lookup_minor_data(minor, SNDRV_DEVICE_TYPE_PCM_CAPTURE)) { - fput(file); + fput_light(file, *fput_needed); return NULL; } return file; @@ -1597,8 +1597,9 @@ static int snd_pcm_link(struct snd_pcm_substream *substream, int fd) struct snd_pcm_file *pcm_file; struct snd_pcm_substream *substream1; struct snd_pcm_group *group; + int fput_needed; - file = snd_pcm_file_fd(fd); + file = snd_pcm_file_fd(fd, &fput_needed); if (!file) return -EBADFD; pcm_file = file->private_data; @@ -1633,7 +1634,7 @@ static int snd_pcm_link(struct snd_pcm_substream *substream, int fd) write_unlock_irq(&snd_pcm_link_rwlock); up_write(&snd_pcm_link_rwsem); _nolock: - fput(file); + fput_light(file, fput_needed); if (res < 0) kfree(group); return res; |
