summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* [SPARC64]: Fix pfn_pte() build failure.David S. Miller2006-08-221-1/+1
| | | | | | | | | | | | | The "%uhi" needs to be "%%uhi" because we want a real "%" character in the assembler here, instead of an assembler variable expansion. Aparently older GCCs were more liberal and interpreted this %-letter as a literal "%" for whatever reason. Based upon a build failure report from Meelis Roos. Signed-off-by: David S. Miller <davem@davemloft.net>
* elv_unregister: fix possible crash on module unloadOleg Nesterov2006-08-221-1/+2
| | | | | | | | | | An exiting task or process which didn't do I/O yet have no io context, elv_unregister() should check it is not NULL. Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Acked-by: Jens Axboe <axboe@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* Fix possible UDF deadlock and memory corruption (CVE-2006-4145)Jan Kara2006-08-222-26/+40
| | | | | | | | | | | | UDF code is not really ready to handle extents larger that 1GB. This is the easy way to forbid creating those. Also truncation code did not count with the case when there are no extents in the file and we are extending the file. Signed-off-by: Jan Kara <jack@suse.cz> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* Fix sctp privilege elevation (CVE-2006-3745)Sridhar Samudrala2006-08-225-53/+23
| | | | | | | | | | | | | | sctp_make_abort_user() now takes the msg_len along with the msg so that we don't have to recalculate the bytes in iovec. It also uses memcpy_fromiovec() so that we don't go beyond the length allocated. It is good to have this fix even if verify_iovec() is fixed to return error on overflow. Signed-off-by: Sridhar Samudrala <sri@us.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
* Merge gregkh@master.kernel.org:/home/rmk/linux-2.6-armGreg Kroah-Hartman2006-08-218-50/+184
|\
| * [ARM] 3746/2: Userspace helpers must be Thumb mode interworkableNicolas Pitre2006-08-181-9/+12
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch from Nicolas Pitre The userspace helpers in clean/arch/arm/kernel/entry-armv.S are called directly in/from userspace. They need to cope with being called from Thumb code. Patch below uses the bx interworking instruction when CONFIG_ARM_THUMB=y. Based on an earlier patch from Paul Brook <paul@codesourcery.com> Signed-off-by: Nicolas Pitre <nico@cam.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3757/1: Use PROCINFO_INITFUNC in head.SCatalin Marinas2006-08-181-1/+1
| | | | | | | | | | | | | | | | | | Patch from Catalin Marinas This is instead of a magic number. Signed-off-by: Catalin Marinas <catalin.marinas@arm.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3756/1: Assign value for HWCAP_IWMMXTPaul Gortmaker2006-08-181-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | Patch from Paul Gortmaker Some folks here at Wind River asked me if I'd push this out so that the value was generally agreed upon in advance by all folks interested in working with iWMMXt. Seems simple enough... Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3755/1: dmabounce: fix return value for find_safe_bufferKevin Hilman2006-08-181-3/+5
| | | | | | | | | | | | | | | | | | | | Patch from Kevin Hilman Previous locking changes to dmabounce incorrectly return non-NULL even when buffer not found. Fix it up. Signed-off-by: Kevin Hilman <khilman@mvista.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3754/1: S3C24XX: tidy arch/arm/mach-s3c2410/MakefileBen Dooks2006-08-181-17/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | Patch from Ben Dooks tidy up the makefile by using TABs to indent, and ensure that all items are indented the same. Move the DMA to its own section, ready for the next set of updates Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3753/1: S3C24XX: DMA fixesBen Dooks2006-08-182-19/+145
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Patch from Ben Dooks A number of small issues with the S3C24XX DMA have cropped up, which this patch fixes. These are: - check wether we can load another buff in start - update state handling in s3c2410_dma_lastxfer - only reload in irq if channel is not idle - more informative timeout errors (add source) - do not call request_irq() with irqs locked - added waitforstop function The patch also adds a S3C2410_DMAOP_STARTED for the occasions when the driver wants to ensure that the DMA system load state is resynced after loading. Signed-off-by: Ben Dooks <ben-linux@fluff.org> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
| * [ARM] 3752/1: fix versatile flash resource mapYoav Steinberg2006-08-131-1/+1
| | | | | | | | | | | | | | | | | | Patch from Yoav Steinberg Flash resource mapping for versatile machine included one extra byte for the end address. This results in failure to map other resources on physical address directly after the NOR flash. Signed-off-by: Yoav Steinberg <yoav@monfort.co.il> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
* | Merge branch 'for-linus' of git://brick.kernel.dk/data/git/linux-2.6-blockGreg Kroah-Hartman2006-08-214-25/+32
|\ \
| * | [PATCH] cfq_cic_link: fix usage of wrong cfq_io_contextOleg Nesterov2006-08-211-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | Obviously, cfq_cic_link() shouldn't free a just allocated cfq_io_context? The dead key is from __cic, so drop that. Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Jens Axboe <axboe@suse.de>
| * | [PATCH] uninline ioprio_best()Oleg Nesterov2006-08-212-22/+24
| | | | | | | | | | | | | | | | | | | | | Saves 376 bytes (5 callers) for me. Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Jens Axboe <axboe@suse.de>
| * | [PATCH] Fix current_io_context() vs set_task_ioprio() raceOleg Nesterov2006-08-212-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I know nothing about io scheduler, but I suspect set_task_ioprio() is not safe. current_io_context() initializes "struct io_context", then sets ->io_context. set_task_ioprio() running on another cpu may see the changes out of order, so ->set_ioprio(ioc) may use io_context which was not initialized properly. Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Jens Axboe <axboe@suse.de>
| * | [PATCH] sys_ioprio_set: minor do_each_thread+break fixOleg Nesterov2006-08-211-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | From include/linux/sched.h: * Careful: do_each_thread/while_each_thread is a double loop so * 'break' will not work as expected - use goto instead. */ Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru> Signed-off-by: Jens Axboe <axboe@suse.de>
| | |
| \ \
| \ \
| \ \
| \ \
| \ \
| \ \
| \ \
*-------. \ \ Merge trivial low-risk suspend hotkey bugzilla-5918 into releaseLen Brown2006-08-2016-179/+229
|\ \ \ \ \ \ \ | |_|_|_|_|/ / |/| | | | | |
| | | | | * | PCI: quirk to disable e100 interrupt if RESET failed toBjorn Helgaas2006-07-261-0/+57
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Without this quirk, e100 can be pulling on a shared interrupt line when another device (eg. USB) loads, causing the interrupt to scream and get disabled. http://bugzilla.kernel.org/show_bug.cgi?id=5918 Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com> Signed-off-by: Len Brown <len.brown@intel.com>
| | | | * | | ACPI: hotkey.c fixes, fix for potential crash of hotkey.cHandle X2006-08-161-162/+119
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | While going through the code, I found out some memory leaks and potential crashes in drivers/acpi/hotkey.c Please find the patch to fix them. This patch does the following, 1. Fixes memory leaks in error paths of hotkey_write_config 2. Fixes freeing unallocated pointers in the error paths of hotkey_write_config 3. Uses a loop instead of linear searching for parsing the userspace input in get_params 4. Uses array of char * instead of passing 4 pointer parameters explicitly into the init_{poll_}hotkey_* static functions Signed-off-by: Andrew Morton <akpm@osdl.org> Acked-by: Luming Yu <luming.yu@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| | | * | | | ACPI: Handle BIOS that resumes from S3 to suspend routine rather than resume ↵William Morrrow2006-08-161-1/+4
| | | |/ / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | vector A BIOS has been found that resumes from S3 to the routine that invoked suspend, ignoring the resume vector. This appears to the OS as a failed S3 attempt. This same system suspend/resume's properly with Windows. It is possible to invoke the protected mode register restore routine (which would normally restore the sysenter registers) when the BIOS returns from S3. This has no effect on a correctly running system and repairs the damage from the deviant BIOS. Signed-off-by: William Morrow <william.morrow@amd.com> Signed-off-by: Jordan Crouse <jordan.crouse@amd.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: relax BAD_MADT_ENTRY check to allow LSAPIC variable length string UIDsStarikovskiy, Alexey Y2006-08-182-2/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ACPI 3.0 appended a variable length UID string to the LAPIC structure as part of support for > 256 processors. So the BAD_MADT_ENTRY() sanity check can no longer compare for equality with a fixed structure length. Signed-off-by: Alexey Y Starikovskiy <alexey.y.starikovskiy@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: avoid irqrouter_resume might_sleep oops on resume from S4Len Brown2006-08-162-1/+11
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | __might_sleep+0x8e/0x93 acpi_os_wait_semaphore+0x50/0xa3 acpi_ut_acquire_mutex+0x28/0x6a acpi_ns_get_node+0x46/0x88 acpi_ns_evaluate+0x2d/0xfc acpi_rs_set_srs_method_data+0xc5/0xe1 acpi_set_current_resources+0x31/0x3f acpi_pci_link_set+0xfc/0x1a5 irqrouter_resume+0x48/0x5f and __might_sleep+0x8e/0x93 kmem_cache_alloc+0x2a/0x8f acpi_evaluate_integer+0x32/0x96 acpi_bus_get_status+0x30/0x84 acpi_pci_link_set+0x12a/0x1a5 irqrouter_resume+0x48/0x5f http://bugzilla.kernel.org/show_bug.cgi?id=6810 Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPIPHP: allow acpiphp to build without ACPI_DOCKKristen Carlson Accardi2006-08-161-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Change the build options for acpiphp so that it may build without being dependent on the ACPI_DOCK option, but yet does not allow the option of acpiphp being built-in when dock is built as a module. This does not change the previous patch for ACPI_IBM_DOCK Kconfig. For the following matrix of config options, I built an i386 kernel. Dock acpiphp should it build? confirmed y y y y y n y y y m y y m y no - acpiphp should acpiphp was convert to m converted to m m n y y m m y y n y y y n n y y n m y y Signed-off-by: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: memory hotplug: remove useless message at boot timeYasunori Goto2006-08-161-6/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This is to remove noisy useless message at boot. The message is a ton of "ACPI Exception (acpi_memory-0492): AE_ERROR, handle is no memory device" In my emulation, number of memory devices are not so many (only 6), but, this messages are displayed 114 times. It is showed by acpi_memory_register_notify_handler() which is called by acpi_walk_namespace(). acpi_walk_namespace() parses all of ACPI's namespace and execute acpi_memory_register_notify_handler(). So, it is called for all of the device which is defined in namespace. If the parsing device is not memory, acpi_memhotplug ignores it due to "no match" and will parse next device. This is normal route, not an exception. Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: verbose on kset/kobject_register errorsRandy Dunlap2006-08-151-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: add message if firmware_register() init failsRandy Dunlap2006-08-151-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: skip smart battery init when acpi=offLen Brown2006-08-151-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: fix boot with acpi=offPavel Machek2006-08-152-0/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix acpi_ac/battery boot with acpi=off Signed-off-by: Pavel Machek <pavel@suse.cz> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Len Brown <len.brown@intel.com>
| | * | | | ACPI: fix kfree in i2c_ec error pathJean Delvare2006-08-111-1/+1
| | |/ / / | | | | | | | | | | | | | | | | | | | | Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | | ACPI: add Dock Station driver to MAINTAINERS fileKristen Carlson Accardi2006-07-261-0/+6
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Kristen Carlson Accardi <kristen.c.accardi@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | | ACPI: restore some dmesg to DEBUG-only, ala 2.6.17Len Brown2006-07-251-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The ACPI_EXCEPTION() patch enabled a bunch of messages to print even in the non-DEBUG kernel. Need to change a couple back, and note that ACPI_EXCEPTION takes no \n, but ACPI_DEBUG_PRINT does. No context for object [%p]\n Device `[%s]' is not power manageable\n Signed-off-by: Len Brown <len.brown@intel.com>
| * | | | ACPI: scan: handle kset/kobject errorsRandy Dunlap2006-07-121-2/+10
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Check and handle kset_register() and kobject_register() init errors. Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Len Brown <len.brown@intel.com>
| * | | | ACPI: handle firmware_register init errorsRandy Dunlap2006-07-121-1/+5
| | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Randy Dunlap <rdunlap@xenotime.net> Signed-off-by: Len Brown <len.brown@intel.com>
* | | | | Merge gregkh@master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6Greg Kroah-Hartman2006-08-1819-120/+175
|\ \ \ \ \
| * | | | | [BRIDGE]: Disable SG/GSO if TX checksum is offHerbert Xu2006-08-172-1/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | When the bridge recomputes features, it does not maintain the constraint that SG/GSO must be off if TX checksum is off. This patch adds that constraint. On a completely unrelated note, I've also added TSO6 and TSO_ECN feature bits if GSO is enabled on the underlying device through the new NETIF_F_GSO_SOFTWARE macro. Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [NETFILTER]: ip_tables: fix table locking in ipt_do_tablePatrick McHardy2006-08-171-1/+2
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | table->private might change because of ruleset changes, don't use it without holding the lock. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [NETFILTER]: ctnetlink: fix deadlock in table dumpingPatrick McHardy2006-08-172-20/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | ip_conntrack_put must not be called while holding ip_conntrack_lock since destroy_conntrack takes it again. Signed-off-by: Patrick McHardy <kaber@trash.net> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [IPV4]: severe locking bug in fib_semantics.cAlexey Kuznetsov2006-08-171-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Found in 2.4 by Yixin Pan <yxpan@hotmail.com>. > When I read fib_semantics.c of Linux-2.4.32, write_lock(&fib_info_lock) = > is used in fib_release_info() instead of write_lock_bh(&fib_info_lock). = > Is the following case possible: a BH interrupts fib_release_info() while = > holding the write lock, and calls ip_check_fib_default() which calls = > read_lock(&fib_info_lock), and spin forever. Signed-off-by: Alexey Kuznetsov <kuznet@ms2.inr.ac.ru> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [MCAST]: Fix filter leak on device removal.David L Stevens2006-08-172-17/+25
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This fixes source filter leakage when a device is removed and a process leaves the group thereafter. This also includes corresponding fixes for IPv6 multicast source filters on device removal. Signed-off-by: David L Stevens <dlstevens@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [NET]: Disallow whitespace in network device names.David S. Miller2006-08-171-5/+14
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It causes way too much trouble and confusion in userspace. Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [PPP]: handle kmalloc failures and convert to using kzallocPanagiotis Issaris2006-08-171-11/+19
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The PPP code contains two kmalloc()s followed by memset()s without handling a possible memory allocation failure. (Suggested by Joe Perches). And furthermore, conversions from kmalloc+memset to kzalloc. [akpm@osdl.org: fix error-path leak] [akpm@osdl.org: cleanups] [paulus@samba.org: don't add useless printk and cardmap_destroy calls] Signed-off-by: Panagiotis Issaris <takis@issaris.org> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Paul Mackerras <paulus@samba.org> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [PKT_SCHED] cls_u32: Fix typo.Ralf Hildebrandt2006-08-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Signed-off-by: Ralf Hildebrandt <Ralf.Hildebrandt@charite.de> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [ATM]: Compile error on ARMKevin Hilman2006-08-171-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | atm_proc_exit() is declared as __exit, and thus in .exit.text. On some architectures (ARM) .exit.text is discarded at compile time, and since atm_proc_exit() is called by some other __init functions, it results in a link error. Signed-off-by: Kevin Hilman <khilman@mvista.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [BNX2]: Convert to netdev_alloc_skb()Michael Chan2006-08-171-8/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Convert dev_alloc_skb() to netdev_alloc_skb() and increase default rx ring size to 255. The old ring size of 100 was too small. Update version to 1.4.44. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [BNX2]: Fix tx race condition.Michael Chan2006-08-172-23/+24
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Fix a subtle race condition between bnx2_start_xmit() and bnx2_tx_int() similar to the one in tg3 discovered by Herbert Xu: CPU0 CPU1 bnx2_start_xmit() if (tx_ring_full) { tx_lock bnx2_tx() if (!netif_queue_stopped) netif_stop_queue() if (!tx_ring_full) update_tx_ring netif_wake_queue() tx_unlock } Even though tx_ring is updated before the if statement in bnx2_tx_int() in program order, it can be re-ordered by the CPU as shown above. This scenario can cause the tx queue to be stopped forever if bnx2_tx_int() has just freed up the entire tx_ring. The possibility of this happening should be very rare though. The following changes are made, very much identical to the tg3 fix: 1. Add memory barrier to fix the above race condition. 2. Eliminate the private tx_lock altogether and rely solely on netif_tx_lock. This eliminates one spinlock in bnx2_start_xmit() when the ring is full. 3. Because of 2, use netif_tx_lock in bnx2_tx_int() before calling netif_wake_queue(). 4. Add memory barrier to bnx2_tx_avail(). 5. Add bp->tx_wake_thresh which is set to half the tx ring size. 6. Check for the full wake queue condition before getting netif_tx_lock in tg3_tx(). This reduces the number of unnecessary spinlocks when the tx ring is full in a steady-state condition. Signed-off-by: Michael Chan <mchan@broadcom.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [NET]: Terminology in ip-sysctl.txtJan "Yenya" Kasprzak2006-08-171-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | this minor patch fixes the description of net.ipv4.tcp_mem sysctl in ip-sysctl.txt - the headline names the values "min, pressure, max", while the description uses the "low, pressure, high" values. Both tcp_rmem and tcp_wmem descriptions use the "min, pressure, max" values, so I have changed the tcp_mem to match this and not vice versa. Signed-off-by: Jan "Yenya" Kasprzak <kas@fi.muni.cz> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [IPV4]: Possible leak of multicast source filter sctructureMichal Ruzicka2006-08-171-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There is a leak of a socket's multicast source filter list structure on closing a socket with a multicast source filter set on an interface that does not exist any more. Signed-off-by: Michal Ruzicka <michal.ruzicka@comstar.cz> Acked-by: David L Stevens <dlstevens@us.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [IPV6] lockdep: annotate __icmpv6_socketIngo Molnar2006-08-171-0/+13
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Split off __icmpv6_socket's sk->sk_dst_lock class, because it gets used from softirqs, which is safe for __icmpv6_sockets (because they never get directly used via userspace syscalls), but unsafe for normal sockets. Has no effect on non-lockdep kernels. Signed-off-by: Ingo Molnar <mingo@elte.hu> Acked-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
| * | | | | [NETFILTER]: xt_physdev build fixAndrew Morton2006-08-171-0/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It needs netfilter_bridge.h for brnf_deferred_hooks Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: David S. Miller <davem@davemloft.net>
OpenPOWER on IntegriCloud