- 13 Jan, 2012 16 commits
-
-
Jason Baron authored
The current epoll code can be tickled to run basically indefinitely in both loop detection path check (on ep_insert()), and in the wakeup paths. The programs that tickle this behavior set up deeply linked networks of epoll file descriptors that cause the epoll algorithms to traverse them indefinitely. A couple of these sample programs have been previously posted in this thread: https://lkml.org/lkml/2011/2/25/297. To fix the loop detection path check algorithms, I simply keep track of the epoll nodes that have been already visited. Thus, the loop detection becomes proportional to the number of epoll file descriptor and links. This dramatically decreases the run-time of the loop check algorithm. In one diabolical case I tried it reduced the run-time from 15 mintues (all in kernel time) to .3 seconds. Fixing the wakeup paths could be done at wakeup time in a similar manner by keeping track of nodes that have already been visited, but the complexity is harder, since there can be multiple wakeups on different cpus...Thus, I've opted to limit the number of possible wakeup paths when the paths are created. This is accomplished, by noting that the end file descriptor points that are found during the loop detection pass (from the newly added link), are actually the sources for wakeup events. I keep a list of these file descriptors and limit the number and length of these paths that emanate from these 'source file descriptors'. In the current implemetation I allow 1000 paths of length 1, 500 of length 2, 100 of length 3, 50 of length 4 and 10 of length 5. Note that it is sufficient to check the 'source file descriptors' reachable from the newly added link, since no other 'source file descriptors' will have newly added links. This allows us to check only the wakeup paths that may have gotten too long, and not re-check all possible wakeup paths on the system. In terms of the path limit selection, I think its first worth noting that the most common case for epoll, is probably the model where you have 1 epoll file descriptor that is monitoring n number of 'source file descriptors'. In this case, each 'source file descriptor' has a 1 path of length 1. Thus, I believe that the limits I'm proposing are quite reasonable and in fact may be too generous. Thus, I'm hoping that the proposed limits will not prevent any workloads that currently work to fail. In terms of locking, I have extended the use of the 'epmutex' to all epoll_ctl add and remove operations. Currently its only used in a subset of the add paths. I need to hold the epmutex, so that we can correctly traverse a coherent graph, to check the number of paths. I believe that this additional locking is probably ok, since its in the setup/teardown paths, and doesn't affect the running paths, but it certainly is going to add some extra overhead. Also, worth noting is that the epmuex was recently added to the ep_ctl add operations in the initial path loop detection code using the argument that it was not on a critical path. Another thing to note here, is the length of epoll chains that is allowed. Currently, eventpoll.c defines: /* Maximum number of nesting allowed inside epoll sets */ #define EP_MAX_NESTS 4 This basically means that I am limited to a graph depth of 5 (EP_MAX_NESTS + 1). However, this limit is currently only enforced during the loop check detection code, and only when the epoll file descriptors are added in a certain order. Thus, this limit is currently easily bypassed. The newly added check for wakeup paths, stricly limits the wakeup paths to a length of 5, regardless of the order in which ep's are linked together. Thus, a side-effect of the new code is a more consistent enforcement of the graph depth. Thus far, I've tested this, using the sample programs previously mentioned, which now either return quickly or return -EINVAL. I've also testing using the piptest.c epoll tester, which showed no difference in performance. I've also created a number of different epoll networks and tested that they behave as expectded. I believe this solves the original diabolical test cases, while still preserving the sane epoll nesting. Signed-off-by:
Jason Baron <jbaron@redhat.com> Cc: Nelson Elhage <nelhage@ksplice.com> Cc: Davide Libenzi <davidel@xmailserver.org> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Sasha Levin authored
When a user with the CAP_SYS_RESOURCE cap tries to F_SETPIPE_SZ a pipe with size bigger than kmalloc() can alloc it spits out an ugly warning: ------------[ cut here ]------------ WARNING: at mm/page_alloc.c:2095 __alloc_pages_nodemask+0x5d3/0x7a0() Pid: 733, comm: a.out Not tainted 3.2.0-rc1+ #4 Call Trace: warn_slowpath_common+0x75/0xb0 warn_slowpath_null+0x15/0x20 __alloc_pages_nodemask+0x5d3/0x7a0 __get_free_pages+0x12/0x50 __kmalloc+0x12b/0x150 pipe_set_size+0x75/0x120 pipe_fcntl+0xf8/0x140 do_fcntl+0x2d4/0x410 sys_fcntl+0x66/0xa0 system_call_fastpath+0x16/0x1b ---[ end trace 432f702e6db7b5ee ]--- Instead, make kcalloc() handle the overflow case and fail quietly. [akpm@linux-foundation.org: switch to sizeof(*bufs) for 80-column niceness] Signed-off-by:Sasha Levin <levinsasha928@gmail.com> Cc: Alexander Viro <viro@zeniv.linux.org.uk> Acked-by:
Pekka Enberg <penberg@kernel.org> Signed-off-by:
-
Stanislaw Gruszka authored
Acked-by:
David Rientjes <rientjes@google.com> Cc: Pekka Enberg <penberg@cs.helsinki.fi> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Signed-off-by:
Stanislaw Gruszka <sgruszka@redhat.com> Signed-off-by:
-
Mathias Krause authored
The address limit is already set in flush_old_exec() so those calls to set_fs(USER_DS) are redundant. Signed-off-by:
Mathias Krause <minipli@googlemail.com> Cc: Kyle McMartin <kyle@mcmartin.ca> Cc: Helge Deller <deller@gmx.de> Signed-off-by:
-
Mathias Krause authored
The address limit is already set in flush_old_exec() so this set_fs(USER_DS) is redundant. Signed-off-by:
-
Andrew Morton authored
Fix the int/bool confusion in there. drivers/video/nvidia/nvidia.c:1602: warning: return from incompatible pointer type Cc: Florian Tobias Schandinat <FlorianSchandinat@gmx.de> Signed-off-by:
-
Heiko Carstens authored
Move CMPXCHG_DOUBLE and rename it to HAVE_CMPXCHG_DOUBLE so architectures can simply select the option if it is supported. Signed-off-by:
Heiko Carstens <heiko.carstens@de.ibm.com> Acked-by:
Christoph Lameter <cl@linux.com> Cc: Pekka Enberg <penberg@kernel.org> Cc: Ingo Molnar <mingo@elte.hu> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: "H. Peter Anvin" <hpa@zytor.com> Signed-off-by:
-
Heiko Carstens authored
Move CMPXCHG_LOCAL and rename it to HAVE_CMPXCHG_LOCAL so architectures can simply select the option if it is supported. Signed-off-by:
-
Heiko Carstens authored
While implementing cmpxchg_double() on s390 I realized that we don't set CONFIG_CMPXCHG_LOCAL despite the fact that we have support for it. However setting that option will increase the size of struct page by eight bytes on 64 bit, which we certainly do not want. Also, it doesn't make sense that a present cpu feature should increase the size of struct page. Besides that it looks like the dependency to CMPXCHG_LOCAL is wrong and that it should depend on CMPXCHG_DOUBLE instead. This patch: If an architecture supports CMPXCHG_LOCAL this shouldn't result automatically in larger struct pages if the SLUB allocator is used. Instead introduce a new config option "HAVE_ALIGNED_STRUCT_PAGE" which can be selected if a double word aligned struct page is required. Also update x86 Kconfig so that it should work as before. Signed-off-by:
-
Joe Perches authored
The uses have been renamed so delete the unused macro. Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Joe Perches authored
Use the more commonly used __noreturn instead of ATTRIB_NORETURN. [akpm@linux-foundation.org: coding-style fixes] Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Acked-by:
Ralf Baechle <ralf@linux-mips.org> Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> Cc: Paul Mackerras <paulus@samba.org> Cc: Martin Schwidefsky <schwidefsky@de.ibm.com> Cc: Heiko Carstens <heiko.carstens@de.ibm.com> Cc: Chris Metcalf <cmetcalf@tilera.com> Signed-off-by:
-
Joe Perches authored
It's a very old and now unused prototype marking so just delete it. Neaten panic pointer argument style to keep checkpatch quiet. Signed-off-by:
-
Joe Perches authored
The only use in kernel.h is gone so remove the macro. Signed-off-by:
-
Joe Perches authored
Use __printf macro. Convert NORET_AND to ATTRIB_NORET. Use the normal kernel style for pointer arguments. Signed-off-by:
-
Stephen Boyd authored
Enabling DEBUG_STRICT_USER_COPY_CHECKS causes the following warning: In file included from arch/x86/include/asm/uaccess.h:573, from kernel/kprobes.c:55: In function 'copy_from_user', inlined from 'write_enabled_file_bool' at kernel/kprobes.c:2191: arch/x86/include/asm/uaccess_64.h:65: warning: call to 'copy_from_user_overflow' declared with attribute warning: copy_from_user() buffer size is not provably correct presumably due to buf_size being signed causing GCC to fail to see that buf_size can't become negative. Signed-off-by:Stephen Boyd <sboyd@codeaurora.org> Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com> Cc: Anil S Keshavamurthy <anil.s.keshavamurthy@intel.com> Cc: David S. Miller <davem@davemloft.net> Acked-by:
Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com> Signed-off-by:
-
Xiaotian Feng authored
get_proc_task() can fail to search the task and return NULL, put_task_struct() will then bomb the kernel with following oops: BUG: unable to handle kernel NULL pointer dereference at 0000000000000010 IP: [<ffffffff81217d34>] proc_pid_permission+0x64/0xe0 PGD 112075067 PUD 112814067 PMD 0 Oops: 0002 [#1] PREEMPT SMP This is a regression introduced by commit 0499680a ("procfs: add hidepid= and gid= mount options"). The kernel should return -ESRCH if get_proc_task() failed. Signed-off-by:
Xiaotian Feng <dannyfeng@tencent.com> Cc: Al Viro <viro@zeniv.linux.org.uk> Cc: Vasiliy Kulikov <segoon@openwall.com> Cc: Stephen Wilson <wilsons@start.ca> Acked-by:
-
- 12 Jan, 2012 24 commits
-
-
Anton Vorontsov authored
This very noisy sparse warning appears on almost every file in the kernel: CHECK init/main.c arch/x86/include/asm/thread_info.h:43:55: error: dubious one-bit signed bitfield arch/x86/include/asm/thread_info.h:44:46: error: dubious one-bit signed bitfield This patch changes sig_on_uaccess_error and uaccess_err flags to unsigned type and thus fixes the warning. Signed-off-by:
Anton Vorontsov <cbouatmailru@gmail.com> Acked-by:
Andy Lutomirski <luto@mit.edu> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/soundLinus Torvalds authored
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tiwai/sound: (526 commits) ASoC: twl6040 - Add method to query optimum PDM_DL1 gain ALSA: hda - Fix the lost power-setup of seconary pins after PM resume ALSA: usb-audio: add Yamaha MOX6/MOX8 support ALSA: virtuoso: add S/PDIF input support for all Xonars ALSA: ice1724 - Support for ooAoo SQ210a ALSA: ice1724 - Allow card info based on model only ALSA: ice1724 - Create capture pcm only for ADC-enabled configurations ALSA: hdspm - Provide unique driver id based on card serial ASoC: Dynamically allocate the rtd device for a non-empty release() ASoC: Fix recursive dependency due to select ATMEL_SSC in SND_ATMEL_SOC_SSC ALSA: hda - Fix the detection of "Loopback Mixing" control for VIA codecs ALSA: hda - Return the error from get_wcaps_type() for invalid NIDs ALSA: hda - Use auto-parser for HP laptops with cx20459 codec ALSA: asihpi - Fix potential Oops in snd_asihpi_cmode_info() ALSA: hdsp - Fix potential Oops in snd_hdsp_info_pref_sync_ref() ALSA: hda/cirrus - support for iMac12,2 model ASoC: cx20442: add bias control over a platform provided regulator ALSA: usb-audio - Avoid flood of frame-active debug messages ALSA: snd-usb-us122l: Delete calls to preempt_disable mfd: Put WM8994 into cache only mode when suspending ... Fix up trivial conflicts in: - arch/arm/mach-s3c64xx/mach-crag6410.c: renamed speyside_wm8962 to tobermory, added littlemill right next to it - drivers/base/regmap/{regcache.c,regmap.c}: duplicate diff that had already come in with other changes in the regmap tree
-
Bjorn Helgaas authored
We only need amd_bus.o for AMD systems with PCI. arch/x86/pci/Makefile already depends on CONFIG_PCI=y, so this patch just adds the dependency on CONFIG_AMD_NB. Cc: Yinghai Lu <yinghai@kernel.org> Cc: stable@kernel.org # 2.6.34+ (needs adjustment for k8 -> amd rename) Signed-off-by:
Bjorn Helgaas <bhelgaas@google.com> Signed-off-by:
-
Takashi Iwai authored
-
Takashi Iwai authored
-
Takashi Iwai authored
Merge branch 'for-3.3' of git://git.kernel.org/pub/scm/linux/kernel/git/broonie/sound into topic/asoc
-
git://github.com/pmundt/linux-shLinus Torvalds authored
SH/R-Mobile updates for 3.3 merge window. * tag 'rmobile-for-linus' of git://github.com/pmundt/linux-sh: (32 commits) arm: mach-shmobile: add a resource name for shdma ARM: mach-shmobile: r8a7779 SMP support V3 ARM: mach-shmobile: Add kota2 defconfig. ARM: mach-shmobile: Add marzen defconfig. ARM: mach-shmobile: r8a7779 power domain support V2 ARM: mach-shmobile: Fix up marzen build for recent GIC changes. ARM: mach-shmobile: r8a7779 PFC function support ARM: mach-shmobile: Flush caches in platform_cpu_die() ARM: mach-shmobile: Allow SoC specific CPU kill code ARM: mach-shmobile: Fix headsmp.S code to use CPUINIT ARM: mach-shmobile: clock-r8a7779: clkz/clkzs support ARM: mach-shmobile: clock-r8a7779: add DIV4 clock support ARM: mach-shmobile: Marzen LAN89218 support ARM: mach-shmobile: Marzen SCIF2/SCIF4 support ARM: mach-shmobile: r8a7779 PFC GPIO-only support V2 ARM: mach-shmobile: r8a7779 and Marzen base support V2 sh: pfc: Unlock register support sh: pfc: Variable bitfield width config register support sh: pfc: Add config_reg_helper() function sh: pfc: Convert index to field and value pair ...
-
git://github.com/pmundt/linux-shLinus Torvalds authored
SuperH updates for 3.3 merge window. * tag 'sh-for-linus' of git://github.com/pmundt/linux-sh: (38 commits) sh: magicpanelr2: Update for parse_mtd_partitions() fallout. sh: mach-rsk: Update for parse_mtd_partitions() fallout. sh: sh2a: Improve cache flush/invalidate functions sh: also without PM_RUNTIME pm_runtime.o must be built sh: add a resource name for shdma sh: Remove redundant try_to_freeze() invocations. sh: Ensure IRQs are enabled across do_notify_resume(). sh: Fix up store queue code for subsys_interface changes. sh: clkfwk: sh_clk_init_parent() should be called after clk_register() sh: add platform_device for renesas_usbhs in board-sh7757lcr sh: modify clock-sh7757 for renesas_usbhs sh: pfc: ioremap() support sh: use ioread32/iowrite32 and mapped_reg for div6 sh: use ioread32/iowrite32 and mapped_reg for div4 sh: use ioread32/iowrite32 and mapped_reg for mstp32 sh: extend clock struct with mapped_reg member sh: clkfwk: clock-sh73a0: all div6_clks use SH_CLK_DIV6_EXT() sh: clkfwk: clock-sh7724: all div6_clks use SH_CLK_DIV6_EXT() sh: clock-sh7723: add CLKDEV_ICK_ID for cleanup serial: sh-sci: Handle GPIO function requests. ...
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
* 'sched-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: sched: Fix lockup by limiting load-balance retries on lock-break sched: Fix CONFIG_CGROUP_SCHED dependency sched: Remove empty #ifdefs
-
Paul Mundt authored
Follows the RSK+ change for the same rationale. Signed-off-by:Paul Mundt <lethal@linux-sh.org>
-
Paul Mundt authored
The RSK+ setup code was doing some pretty dubious things with parse_mtd_partitions() in order to populate the physmap-flash map platform data. The physmap-flash driver contains all of the functionality that we require already, so simply drop the special casing and pad out the platform data accordingly. Signed-off-by: -
Paul Mundt authored
-
Phil Edworthy authored
The cache functions lock out interrupts for long periods; this patch reduces the impact when operating on large address ranges. In such cases it will: - Invalidate the entire cache rather than individual addresses. - Do nothing when flushing the operand cache in write-through mode. - When flushing the operand cache in write-back mdoe, index the search for matching addresses on the cache entires instead of the addresses to flush Note: sh2a__flush_purge_region was only invalidating the operand cache, this adds flush. Signed-off-by:
Phil Edworthy <phil.edworthy@renesas.com> Signed-off-by:
-
Paul Mundt authored
-
Paul Mundt authored
Conflicts: arch/sh/kernel/cpu/sh4a/clock-sh7723.c Signed-off-by: -
Guennadi Liakhovetski authored
When CONFIG_PM_RUNTIME is off, drivers/sh/pm_runtime.o still has to be built on sh platforms, because then it provides means to statically switch on device PM clocks. Signed-off-by:
Guennadi Liakhovetski <g.liakhovetski@gmx.de> Signed-off-by:
-
Shimoda, Yoshihiro authored
Signed-off-by:
Yoshihiro Shimoda <yoshihiro.shimoda.uh@renesas.com> Signed-off-by:
-
Paul Mundt authored
-
Shimoda, Yoshihiro authored
Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
* 'x86-platform-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/intel config: Fix the APB_TIMER selection x86/mrst: Add additional debug prints for pb_keys x86/intel config: Revamp configuration to allow for Moorestown and Medfield x86/intel/scu/ipc: Match the changes in the x86 configuration x86/apb: Fix configuration constraints x86: Fix INTEL_MID silly x86/Kconfig: Cyclone-timer depends on x86-summit x86: Reduce clock calibration time during slave cpu startup x86/config: Revamp configuration for MID devices x86/sfi: Kill the IRQ as id hack
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
* 'x86-debug-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86, reboot: Fix typo in nmi reboot path x86, NMI: Add to_cpumask() to silence compile warning x86, NMI: NMI selftest depends on the local apic x86: Add stack top margin for stack overflow checking x86, NMI: NMI-selftest should handle the UP case properly x86: Fix the 32-bit stackoverflow-debug build x86, NMI: Add knob to disable using NMI IPIs to stop cpus x86, NMI: Add NMI IPI selftest x86, reboot: Use NMI instead of REBOOT_VECTOR to stop cpus x86: Clean up the range of stack overflow checking x86: Panic on detection of stack overflow x86: Check stack overflow in detail
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
* 'x86-efi-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86, efi: Break up large initrd reads x86, efi: EFI boot stub support efi: Add EFI file I/O data types efi.h: Add boottime->locate_handle search types efi.h: Add graphics protocol guids efi.h: Add allocation types for boottime->allocate_pages() efi.h: Add efi_image_loaded_t efi.h: Add struct definition for boot time services x86: Don't use magic strings for EFI loader signature x86: Add missing bzImage fields to struct setup_header
-
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
* 'x86-mm-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86/numa: Add constraints check for nid parameters mm, x86: Remove debug_pagealloc_enabled x86/mm: Initialize high mem before free_all_bootmem() arch/x86/kernel/e820.c: quiet sparse noise about plain integer as NULL pointer arch/x86/kernel/e820.c: Eliminate bubble sort from sanitize_e820_map() x86: Fix mmap random address range x86, mm: Unify zone_sizes_init() x86, mm: Prepare zone_sizes_init() for unification x86, mm: Use max_low_pfn for ZONE_NORMAL on 64-bit x86, mm: Wrap ZONE_DMA32 with CONFIG_ZONE_DMA32 x86, mm: Use max_pfn instead of highend_pfn x86, mm: Move zone init from paging_init() on 64-bit x86, mm: Use MAX_DMA_PFN for ZONE_DMA on 32-bit
-
git://git.kernel.org/pub/scm/linux/kernel/git/davej/cpufreqLinus Torvalds authored
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/davej/cpufreq: (23 commits) [CPUFREQ] EXYNOS: Removed useless headers and codes [CPUFREQ] EXYNOS: Make EXYNOS common cpufreq driver [CPUFREQ] powernow-k8: Update copyright, maintainer and documentation information [CPUFREQ] powernow-k8: Fix indexing issue [CPUFREQ] powernow-k8: Avoid Pstate MSR accesses on systems supporting CPB [CPUFREQ] update lpj only if frequency has changed [CPUFREQ] cpufreq:userspace: fix cpu_cur_freq updation [CPUFREQ] Remove wall variable from cpufreq_gov_dbs_init() [CPUFREQ] EXYNOS4210: cpufreq code is changed for stable working [CPUFREQ] EXYNOS4210: Update frequency table for cpu divider [CPUFREQ] EXYNOS4210: Remove code about bus on cpufreq [CPUFREQ] s3c64xx: Use pr_fmt() for consistent log messages cpufreq: OMAP: fixup for omap_device changes, include <linux/module.h> cpufreq: OMAP: fix freq_table leak cpufreq: OMAP: put clk if cpu_init failed cpufreq: OMAP: only supports OPP library cpufreq: OMAP: dont support !freq_table cpufreq: OMAP: deny initialization if no mpudev cpufreq: OMAP: move clk name decision to init cpufreq: OMAP: notify even with bad boot frequency ...
-
