summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Do not sent 120 TEST UNIT READY requests on generic NOT READY statuses.mav2013-04-111-5/+2
| | | | | | | | | | Some failing disks tend to return vendor-specific ASC/ASCQ codes with NOT READY sense key. It caused extremely long recovery attempts, repeating these 120 TURs (it takes at least 1 minute) for every I/O request. Instead of that use default error handling, doing just few retries. Reviewed by: ken, gibbs MFC after: 1 month
* Make the code to check if VMX is enabled more readable by using macrosneel2013-04-112-1/+7
| | | | | | instead of magic numbers. Discussed with: Chris Torek
* Update ciss(4) man page. We absolutely support performant mode of operationsbruno2013-04-101-10/+7
| | | | | | | | | | on this controller. Indicate that the 5300 is the *only* controller that will only work in simple mode. Bus rescans should not be needed now and I consider it a bug if disks do not appear or dissapear when created or destroyed. Obtained from: Yahoo! Inc
* While investigating a p/r I noted that the camcontrol devlist output forsbruno2013-04-101-3/+8
| | | | | | | | | | | | | volumes behind a ciss(4) controller were being reported with malformeed names and identifiers. Repair that reporting by using the CAM values for the three SCSI indents reported via camcontrol devlist PR: kern/171650 Reviewed by: scottl Obtained from: Yahoo! Inc. MFC after: 2 weeks
* options DPT_HANDLE_TIMEOUTS hasn't worked since dpt(4) was converted to CAMsbruno2013-04-104-180/+1
| | | | | | | | | | | | | | | somewhere around svn r39402 to r39234. I don't know of anyone who really wants to test these changes, but they only remove the deprecated code in question. This shreds the driver down a bit and *removes* options from the kernel configs. These don't appear to be referenced in the man page, so no need to check it there. PR: kern/44587 Obtained from: Yahoo! Inc. MFC after: 2 weeks
* Add a callback to the ada(4) driver so that it knows when GEOM has releasedken2013-04-101-0/+26
| | | | | | | | | | | | | | | | | | | | | | | references to it. This is the functional equivalent to change r237518, which added this functionality to the cd(4) and da(4) drivers. This fix prevents a panic caused by GEOM calling adaopen() while the device is going away. We now keep the device around until GEOM has finished cleaning up its state. ata_da.c: In adaregister(), add a d_gone callback to the GEOM disk structure registered for the ada driver. Increment the peripheral reference count for GEOM. Add a new callback, adadiskgonecb(), that GEOM calls when it is done with its resources. This callback releases the reference acquired in adaregister(). Submitted by: Po-Li Soong Sponsored by: Spectra Logic MFC After: 5 days
* Create controller-level DMA tag, handling range of supported addresses.mav2013-04-102-8/+28
| | | | | | | That simplifies logic for channels and gives the bus information about what device actually allocated the tag. Submitted by: jhb@
* Use kvm_counter_u64_fetch() to fix obtaining ipstat and tcpstat fromglebius2013-04-103-29/+69
| | | | | | kernel core files. Sponsored by: Nginx, Inc.
* o Provide function kvm_read_zpcpu() to access UMA_ZONE_PCPU allocations.glebius2013-04-104-7/+89
| | | | | | o Provide function kvm_counter_u64_fetch() to fetch valut of a counter(9). Sponsored by: Nginx, Inc.
* Need to call init_mem() to really initialize the MMIO range lookups.neel2013-04-101-0/+1
| | | | | | | | This was working by accident because: - the RB_HEADs were being initialized to zero as part of BSS - the pthread_rwlock functions were implicitly initializing the lock object Obtained from: NetApp
* Remove obsolete comment about lack of locking for MMIO range lookup.neel2013-04-101-6/+0
| | | | Pointed out by: Tycho Nightingale (tycho.nightingale@plurisbusnetworks.com)
* Simplify allocate_legacy code, txr pointer was breaking LEGACY compile,jfv2013-04-101-2/+1
| | | | thanks to Nick Rogers for pointing this out.
* Add ID for ASMedia ASM1042 USB 3.0 controller.mav2013-04-101-0/+3
| | | | MFC after: 1 week
* Since UMA_ZONE_PCPU zones put a constraint on sizeof(struct pcpu), declaredglebius2013-04-101-7/+1
| | | | | | as CTASSERT in MI pcpu.h, stop including all possible mutually exclusive PCPU_MD_FIELDS fields into LINT kernels, due to brekaing aforementioned CTASSERT.
* Fix a time calculation error in ctlstat_standard().ken2013-04-101-2/+3
| | | | | | | | | | | ctlstat.c: When converting a timeval to a floating point number in ctlstat_standard(), cast the nanoseconds calculation to a long double, so we don't lose precision. Without the cast, we wind up with a time in whole seconds only. Sponsored by: Spectra Logic MFC after: 3 days
* Remove the number of addresses restriction from sctp_connectx().tuexen2013-04-101-202/+94
| | | | | | | Remove unused code. While there, do some cleanup of the code. MFC after: 1 week
* Make ctladm(8) try to load ctl.ko kernel module if needed.trasz2013-04-101-0/+10
| | | | Reviewed by: ken
* Fix build.glebius2013-04-104-4/+4
|
* Cast to (void *)(uintptr_t) on copyout and copyin of zfs_iocparm_t.zfs_cmdmm2013-04-101-2/+2
| | | | MFC after: 9 days
* Unsynchronized TSCs on the host require special handling in bhyve:neel2013-04-104-16/+109
| | | | | | | | | | | | - use clock_gettime(2) as the time base for the emulated ACPI timer instead of directly using rdtsc(). - don't advertise the invariant TSC capability to the guest to discourage it from using the TSC as its time base. Discussed with: jhb@ (about making 'smp_tsc' a global) Reported by: Dan Mack on freebsd-virtualization@ Obtained from: NetApp
* swapcontext wrapper can not be implemented in C, the stack pointer saved indavidxu2013-04-101-10/+1
| | | | | | the context becomes invalid when the function returns, same as setjmp, it must be implemented in assemble language, see discussions in PR misc/177624.
* Change name of variable from 'rwlock' to more descriptive 'mmio_rwlock'neel2013-04-101-11/+11
| | | | | Requested by: grehan Obtained from: NetApp
* Improve PCI BAR emulation:neel2013-04-107-60/+320
| | | | | | | | - Respect the MEMEN and PORTEN bits in the command register - Allow the guest to reprogram the address decoded by the BAR Submitted by: Gopakumar T Obtained from: NetApp
* Fix the build.jimharris2013-04-101-1/+1
|
* ZFS expects a copyout of zfs_cmd_t on an ioctl error. Our sys_ioctl()mm2013-04-094-27/+88
| | | | | | | | | | | | | | | doesn't copyout in this case. To solve this issue a new struct zfs_iocparm_t is introduced consisting of: - zfs_ioctl_version (future backwards compatibility purposes) - user space pointer to zfs_cmd_t (copyin and copyout) - size of zfs_cmd_t (verification purposes) The copyin and copyout of zfs_cmd_t is now done the illumos (vendor) way what makes porting of new changes easier and ensures correct behavior if returning an error. MFC after: 10 days
* Change certain heavily used network related mutexes and rwlocks toandre2013-04-098-11/+11
| | | | | | | | | | reside on their own cache line to prevent false sharing with other nearby structures, especially for those in the .bss segment. NB: Those mutexes and rwlocks with variables next to them that get changed on every invocation do not benefit from their own cache line. Actually it may be net negative because two cache misses would be incurred in those cases.
* Fix a race condition on tcp listen socket teardown with pendingandre2013-04-091-0/+9
| | | | | | | | | | | | | | connections in the accept queue and contiguous new incoming SYNs. Compared to the original submitters patch I've moved the test next to the SYN handling to have it together in a logical unit and reworded the comment explaining the issue. Submitted by: Matt Miller <matt@matthewjmiller.net> Submitted by: Juan Mojica <jmojica@gmail.com> Reviewed by: Matt Miller (changes) Tested by: pho MFC after: 1 week
* Ensure make -j N universe works correctly, by checking for an up-to-datedim2013-04-091-2/+2
| | | | | | | | | | | | | | | | make before starting the universe targets themselves. Otherwise, all of the targets would attempt to build make simultaneously, overwriting each other's copies of the make object files and executable. This could lead to strange errors, for example when partially-written make executables are invoked. Also amend r216620, to make the rest of universe wait properly until the upgrade_checks target is finished, by adding universe_${target}_prologue to the .ORDER target. Otherwise, make will be too smart for its own good, and start building the universe targets simultaneously with the prologues anyway. MFC after: 1 week
* Proxy allocation requests for the PCI ROM BAR from child devices similarjhb2013-04-091-25/+35
| | | | | | | to how the VGA bus driver currently proxies allocation requests for other PCI BARs. MFC after: 1 week
* Pass the segmented address of the counter, based on %fs, i.e. offsetkib2013-04-091-1/+3
| | | | from the pcpu[0] to the counter base, instead of the linear address.
* Convert UMA code to C99 uintXX_t types.glebius2013-04-094-97/+97
|
* Swap us_freecount and us_flags, achieving same structure sizeglebius2013-04-091-2/+2
| | | | | | as before previous commit. Submitted by: alc
* Add static/const keywords to the arrays.ed2013-04-093-8/+17
| | | | | | | | | This theoretically allows a compiler to optimize (parts of) the array away if unused. While there, make the array size implicit and use a _Static_assert() to ensure that the definition matches up with the number of elements in the list.
* Since now we support 256 items per slab, we need more bitsglebius2013-04-091-1/+1
| | | | | | | | | | for us_freecount. This grows uma_slab_head on 32-bit arches, but growth isn't significant. Taking kmem zones as example, only the 32 byte zone is affected, ipers is reduced from 113 to 112. In collaboration with: kib
* Last remaining prototype issue that still hasn't been fixed upstream.des2013-04-092-3/+1
|
* Revert local changes and pull in r3828 from upstream.des2013-04-098-269/+12
|
* Fix KASSERTs: maximum number of items per slab is 256.glebius2013-04-091-3/+3
|
* Fix build for AIM 64bit.kib2013-04-091-0/+2
|
* Fix the assertions for the state of the object under the map entrykib2013-04-091-6/+16
| | | | | | | | | | | with the MAP_ENTRY_VN_WRITECNT flag: - Move the assertion that verifies the state of the v_writecount and vnp.writecount, under the block where the object is locked. - Check that the object type is OBJT_VNODE before asserting. Reported by: avg Reviewed by: alc MFC after: 1 week
* Fix VIMAGE build.glebius2013-04-091-2/+2
|
* Use IP6STAT_INC/IP6STAT_DEC macros to update ip6 stats.ae2013-04-0915-101/+102
| | | | MFC after: 1 week
* Import a new version of NetBSD's mtree.ed2013-04-098-15/+186
|\ | | | | | | | | | | | | | | This version of mtree implements a new flag (-O) that can be used to restrict the tool to certain pathnames. Also, it fixes a compiler warning generated by -Wmissing-variable-declarations. Acked by: brooks
| * Vendor import of NetBSD's mtree at 2013-04-08.ed2013-04-087-14/+185
| |
* | FireWire: Don't allow a tlabel to reference an xfer after free.will2013-04-081-3/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | sys/dev/firewire/firewire.c: - fw_xfer_unload(): Since we are about to free this xfer, call fw_tl_free() to remove the xfer from its tlabel's list, if it has a tlabel. - In every occasion when a xfer is removed from a tlabel's list, reset xfer->tl to -1 while holding fc->tlabel_lock, so that the xfer isn't mis-identified as belonging to a tlabel. This doesn't fix all the use-after-free problems for M_FWMEM, but is an incremental towards that goal. Reviewed by: kan, sbruno Sponsored by: Spectra Logic
* | Fix this to compile when ATH_DEBUG_ALQ is defined but ATH_DEBUG isn't.adrian2013-04-081-1/+1
| |
* | Add FreeBSD 8.4.pluknet2013-04-081-0/+1
| | | | | | | | MFC after: 3 days
* | The per-page act_count can be made very-easily protected by theattilio2013-04-082-5/+5
| | | | | | | | | | | | | | | | | | per-page lock rather than vm_object lock, without any further overhead. Make the formal switch. Sponsored by: EMC / Isilon storage division Reviewed by: alc Tested by: pho
* | Switch some "low-hanging fruit" to acquire read lock on vmobjectsattilio2013-04-083-19/+19
| | | | | | | | | | | | | | | | rather than write locks. Sponsored by: EMC / Isilon storage division Reviewed by: alc Tested by: pho
* | Merge from projects/counters: TCP/IP stats.glebius2013-04-085-289/+550
| | | | | | | | | | | | | | | | | | Convert 'struct ipstat' and 'struct tcpstat' to counter(9). This speeds up IP forwarding at extreme packet rates, and makes accounting more precise. Sponsored by: Nginx, Inc.
* | Merge from projects/counters: counter(9).glebius2013-04-0814-0/+819
| | | | | | | | | | | | | | | | | | | | | | | | | | Introduce counter(9) API, that implements fast and raceless counters, provided (but not limited to) for gathering of statistical data. See http://lists.freebsd.org/pipermail/freebsd-arch/2013-April/014204.html for more details. In collaboration with: kib Reviewed by: luigi Tested by: ae, ray Sponsored by: Nginx, Inc.
OpenPOWER on IntegriCloud