- Sep 09, 2017
-
-
Davidlohr Bueso authored
We can work with a single rb_root_cached root to test both cached and non-cached rbtrees. In addition, also add a test to measure latencies between rb_first and its fast counterpart. Link: http://lkml.kernel.org/r/20170719014603.19029-7-dave@stgolabs.net Signed-off-by:
Davidlohr Bueso <dbueso@suse.de> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Signed-off-by:
Linus Torvalds <torvalds@linux-foundation.org>
-
Davidlohr Bueso authored
This adds a second test for regular rb-tree testing in that there is no need to repeat it for the augmented flavor. Link: http://lkml.kernel.org/r/20170719014603.19029-6-dave@stgolabs.net Signed-off-by:
-
Davidlohr Bueso authored
Allows for more flexible debugging. Link: http://lkml.kernel.org/r/20170719014603.19029-5-dave@stgolabs.net Signed-off-by:
-
Davidlohr Bueso authored
While overall the code is very nicely commented, it might not be immediately obvious from the diagrams what is going on. Add a very brief summary of each case. Opposite cases where the node is the left child are left untouched. Link: http://lkml.kernel.org/r/20170719014603.19029-4-dave@stgolabs.net Signed-off-by:
-
Davidlohr Bueso authored
The only times the nil-parent (root node) condition is true is when the node is the first in the tree, or after fixing rbtree rule #4 and the case 1 rebalancing made the node the root. Such conditions do not apply most of the time: (i) The common case in an rbtree is to have more than a single node, so this is only true for the first rb_insert(). (ii) While there is a chance only one first rotation is needed, cases where the node's uncle is black (cases 2,3) are more common as we can have the following scenarios during the rotation looping: case1 only, case1+1, case2+3, case1+2+3, case3 only, etc. This patch, therefore, adds an unlikely() optimization to this conditional. When profiling with CONFIG_PROFILE_ANNOTATED_BRANCHES, a kernel build shows that the incorrect rate is less than 15%, and for workloads that involve insert mostly trees overtime tend to have less than 2% incorrect rate. Link: http://lkml.kernel.org/r/20170719014603.19029-3-dave@stgolabs.net Signed-off-by: -
Davidlohr Bueso authored
Patch series "rbtree: Cache leftmost node internally", v4. A series to extending rbtrees to internally cache the leftmost node such that we can have fast overlap check optimization for all interval tree users[1]. The benefits of this series are that: (i) Unify users that do internal leftmost node caching. (ii) Optimize all interval tree users. (iii) Convert at least two new users (epoll and procfs) to the new interface. This patch (of 16): Red-black tree semantics imply that nodes with smaller or greater (or equal for duplicates) keys always be to the left and right, respectively. For the kernel this is extremely evident when considering our rb_first() semantics. Enabling lookups for the smallest node in the tree in O(1) can save a good chunk of cycles in not having to walk down the tree each time. To this end there are a few core users that explicitly do this, such as the scheduler and rtmutexes. There is also the desire for interval trees to have this optimization allowing faster overlap checking. This patch introduces a new 'struct rb_root_cached' which is just the root with a cached pointer to the leftmost node. The reason why the regular rb_root was not extended instead of adding a new structure was that this allows the user to have the choice between memory footprint and actual tree performance. The new wrappers on top of the regular rb_root calls are: - rb_first_cached(cached_root) -- which is a fast replacement for rb_first. - rb_insert_color_cached(node, cached_root, new) - rb_erase_cached(node, cached_root) In addition, augmented cached interfaces are also added for basic insertion and deletion operations; which becomes important for the interval tree changes. With the exception of the inserts, which adds a bool for updating the new leftmost, the interfaces are kept the same. To this end, porting rb users to the cached version becomes really trivial, and keeping current rbtree semantics for users that don't care about the optimization requires zero overhead. Link: http://lkml.kernel.org/r/20170719014603.19029-2-dave@stgolabs.net Signed-off-by:Jan Kara <jack@suse.cz> Acked-by:
Peter Zijlstra (Intel) <peterz@infradead.org> Signed-off-by:
-
Matthew Wilcox authored
[akpm@linux-foundation.org: minor tweaks] Link: http://lkml.kernel.org/r/20170720184539.31609-3-willy@infradead.org Signed-off-by:
Matthew Wilcox <mawilcox@microsoft.com> Cc: "H. Peter Anvin" <hpa@zytor.com> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: David Miller <davem@davemloft.net> Cc: Ingo Molnar <mingo@elte.hu> Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru> Cc: Matt Turner <mattst88@gmail.com> Cc: Michael Ellerman <mpe@ellerman.id.au> Cc: Minchan Kim <minchan@kernel.org> Cc: Ralf Baechle <ralf@linux-mips.org> Cc: Richard Henderson <rth@twiddle.net> Cc: Russell King <rmk+kernel@armlinux.org.uk> Cc: Sam Ravnborg <sam@ravnborg.org> Cc: Sergey Senozhatsky <sergey.senozhatsky@gmail.com> Cc: Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
-
Matthew Wilcox authored
Patch series "Multibyte memset variations", v4. A relatively common idiom we're missing is a function to fill an area of memory with a pattern which is larger than a single byte. I first noticed this with a zram patch which wanted to fill a page with an 'unsigned long' value. There turn out to be quite a few places in the kernel which can benefit from using an optimised function rather than a loop; sometimes text size, sometimes speed, and sometimes both. The optimised PowerPC version (not included here) improves performance by about 30% on POWER8 on just the raw memset_l(). Most of the extra lines of code come from the three testcases I added. This patch (of 8): memset16(), memset32() and memset64() are like memset(), but allow the caller to fill the destination with a value larger than a single byte. memset_l() and memset_p() allow the caller to use unsigned long and pointer values respectively. Link: http://lkml.kernel.org/r/20170720184539.31609-2-willy@infradead.org Signed-off-by:
-
- Aug 30, 2017
-
-
Alexander Kuleshov authored
Signed-off-by:
Alexander Kuleshov <kuleshovmail@gmail.com> Signed-off-by:
Jonathan Corbet <corbet@lwn.net>
-
Chris Mi authored
The following new APIs are added: int idr_alloc_ext(struct idr *idr, void *ptr, unsigned long *index, unsigned long start, unsigned long end, gfp_t gfp); void *idr_remove_ext(struct idr *idr, unsigned long id); void *idr_find_ext(const struct idr *idr, unsigned long id); void *idr_replace_ext(struct idr *idr, void *ptr, unsigned long id); void *idr_get_next_ext(struct idr *idr, unsigned long *nextid); Signed-off-by:Chris Mi <chrism@mellanox.com> Signed-off-by:
Jiri Pirko <jiri@mellanox.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
- Aug 29, 2017
-
-
Peter Zijlstra authored
Commit: e9149858 ("locking/lockdep/selftests: Add mixed read-write ABBA tests") adds an explicit FAILURE to the locking selftest but overlooked the fact that this kills lockdep. Fudge the test to avoid this. Signed-off-by:
Ingo Molnar <mingo@kernel.org>
-
- Aug 25, 2017
-
-
Denys Vlasenko authored
Signed-off-by:
Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: mingo@redhat.com Cc: Jim Kukunas <james.t.kukunas@linux.intel.com> Cc: Fenghua Yu <fenghua.yu@intel.com> Cc: Megha Dey <megha.dey@linux.intel.com> Cc: Gayatri Kammela <gayatri.kammela@intel.com> Cc: x86@kernel.org Cc: linux-kernel@vger.kernel.org Signed-off-by:
Shaohua Li <shli@fb.com>
-
Peter Zijlstra authored
Currently lockdep has limited support for recursive readers, add a few mixed read-write ABBA selftests to show the extend of these limitations. [ 0.000000] ---------------------------------------------------------------------------- [ 0.000000] | spin |wlock |rlock |mutex | wsem | rsem | [ 0.000000] -------------------------------------------------------------------------- [ 0.000000] mixed read-lock/lock-write ABBA: |FAILED| | ok | [ 0.000000] mixed read-lock/lock-read ABBA: | ok | | ok | [ 0.000000] mixed write-lock/lock-write ABBA: | ok | | ok | This clearly illustrates the case where lockdep fails to find a deadlock. Signed-off-by:
-
- Aug 22, 2017
-
-
Stephan Mueller authored
Using sg_miter_start and sg_miter_next, the buffer of an SG is kmap'ed to *buff. The current code calls sg_miter_stop (and thus kunmap) on the SG entry before the last access of *buff. The patch moves the sg_miter_stop call after the last access to *buff to ensure that the memory pointed to by *buff is still mapped. Fixes: 4816c940 ("lib/mpi: Fix SG miter leak") Cc: <stable@vger.kernel.org> Signed-off-by:
Stephan Mueller <smueller@chronox.de> Signed-off-by:
Herbert Xu <herbert@gondor.apana.org.au>
-
- Aug 18, 2017
-
-
Chris Wilson authored
This was the competing idea long ago, but it was only with the rewrite of the idr as an radixtree and using the radixtree directly ourselves, along with the realisation that we can store the vma directly in the radixtree and only need a list for the reverse mapping, that made the patch performant enough to displace using a hashtable. Though the vma ht is fast and doesn't require any extra allocation (as we can embed the node inside the vma), it does require a thread for resizing and serialization and will have the occasional slow lookup. That is hairy enough to investigate alternatives and favour them if equivalent in peak performance. One advantage of allocating an indirection entry is that we can support a single shared bo between many clients, something that was done on a first-come first-serve basis for shared GGTT vma previously. To offset the extra allocations, we create yet another kmem_cache for them. Signed-off-by:
Chris Wilson <chris@chris-wilson.co.uk> Reviewed-by:
Tvrtko Ursulin <tvrtko.ursulin@intel.com> Link: https://patchwork.freedesktop.org/patch/msgid/20170816085210.4199-5-chris@chris-wilson.co.uk
-
Thomas Gleixner authored
The hardlockup detector on x86 uses a performance counter based on unhalted CPU cycles and a periodic hrtimer. The hrtimer period is about 2/5 of the performance counter period, so the hrtimer should fire 2-3 times before the performance counter NMI fires. The NMI code checks whether the hrtimer fired since the last invocation. If not, it assumess a hard lockup. The calculation of those periods is based on the nominal CPU frequency. Turbo modes increase the CPU clock frequency and therefore shorten the period of the perf/NMI watchdog. With extreme Turbo-modes (3x nominal frequency) the perf/NMI period is shorter than the hrtimer period which leads to false positives. A simple fix would be to shorten the hrtimer period, but that comes with the side effect of more frequent hrtimer and softlockup thread wakeups, which is not desired. Implement a low pass filter, which checks the perf/NMI period against kernel time. If the perf/NMI fires before 4/5 of the watchdog period has elapsed then the event is ignored and postponed to the next perf/NMI. That solves the problem and avoids the overhead of shorter hrtimer periods and more frequent softlockup thread wakeups. Fixes: 58687acb ("lockup_detector: Combine nmi_watchdog and softlockup detector") Reported-and-tested-by:
Kan Liang <Kan.liang@intel.com> Signed-off-by:
Thomas Gleixner <tglx@linutronix.de> Cc: dzickus@redhat.com Cc: prarit@redhat.com Cc: ak@linux.intel.com Cc: babu.moger@oracle.com Cc: peterz@infradead.org Cc: eranian@google.com Cc: acme@redhat.com Cc: stable@vger.kernel.org Cc: atomlin@redhat.com Cc: akpm@linux-foundation.org Cc: torvalds@linux-foundation.org Link: http://lkml.kernel.org/r/alpine.DEB.2.20.1708150931310.1886@nanos
-
- Aug 17, 2017
-
-
Ingo Molnar authored
locking/lockdep: Make CONFIG_LOCKDEP_CROSSRELEASE and CONFIG_LOCKDEP_COMPLETIONS truly non-interactive The syntax to turn Kconfig options into non-interactive ones is to not offer interactive prompt help texts. Remove them. Cc: Boqun Feng <boqun.feng@gmail.com> Cc: Byungchul Park <byungchul.park@lge.com> Cc: Lai Jiangshan <jiangshanlai@gmail.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Tejun Heo <tj@kernel.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: linux-kernel@vger.kernel.org Signed-off-by: -
Byungchul Park authored
'complete' is an adjective and LOCKDEP_COMPLETE sounds like 'lockdep is complete', so pick a better name that uses a noun. Signed-off-by:
Byungchul Park <byungchul.park@lge.com> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: kernel-team@lge.com Link: http://lkml.kernel.org/r/1502960261-16206-3-git-send-email-byungchul.park@lge.com Signed-off-by:
-
Byungchul Park authored
Lockdep doesn't have to be made to work with crossrelease and just works with them. Reword the title so that what the option does is clear. Signed-off-by:
-
Byungchul Park authored
Crossrelease support added the CONFIG_LOCKDEP_CROSSRELEASE and CONFIG_LOCKDEP_COMPLETE options. It makes little sense to enable them when PROVE_LOCKING is disabled. Make them non-interative options and part of PROVE_LOCKING to simplify the UI. Signed-off-by:
-
Stefan Agner authored
Use just @ to denote comments which works with gcc and clang. Otherwise clang reports an escape sequence error: error: invalid % escape in inline assembly string Use %0-%3 as operand references, this avoids: error: invalid operand in inline asm: 'umull ${1:r}, ${0:r}, ${2:r}, ${3:r}' Also remove superfluous casts on output operands to avoid warnings such as: warning: invalid use of a cast in an inline asm context requiring an l-value Signed-off-by:Stefan Agner <stefan@agner.ch> Acked-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
- Aug 14, 2017
-
-
Waiman Long authored
The allocated debug objects are either on the free list or in the hashed bucket lists. So they won't get lost. However if both debug objects and kmemleak are enabled and kmemleak scanning is done while some of the debug objects are transitioning from one list to the others, false negative reporting of memory leaks may happen for those objects. For example, [38687.275678] kmemleak: 12 new suspected memory leaks (see /sys/kernel/debug/kmemleak) unreferenced object 0xffff92e98aabeb68 (size 40): comm "ksmtuned", pid 4344, jiffies 4298403600 (age 906.430s) hex dump (first 32 bytes): 00 00 00 00 00 00 00 00 d0 bc db 92 e9 92 ff ff ................ 01 00 00 00 00 00 00 00 38 36 8a 61 e9 92 ff ff ........86.a.... backtrace: [<ffffffff8fa5378a>] kmemleak_alloc+0x4a/0xa0 [<ffffffff8f47c019>] kmem_cache_alloc+0xe9/0x320 [<ffffffff8f62ed96>] __debug_object_init+0x3e6/0x400 [<ffffffff8f62ef01>] debug_object_activate+0x131/0x210 [<ffffffff8f330d9f>] __call_rcu+0x3f/0x400 [<ffffffff8f33117d>] call_rcu_sched+0x1d/0x20 [<ffffffff8f4a183c>] put_object+0x2c/0x40 [<ffffffff8f4a188c>] __delete_object+0x3c/0x50 [<ffffffff8f4a18bd>] delete_object_full+0x1d/0x20 [<ffffffff8fa535c2>] kmemleak_free+0x32/0x80 [<ffffffff8f47af07>] kmem_cache_free+0x77/0x350 [<ffffffff8f453912>] unlink_anon_vmas+0x82/0x1e0 [<ffffffff8f440341>] free_pgtables+0xa1/0x110 [<ffffffff8f44af91>] exit_mmap+0xc1/0x170 [<ffffffff8f29db60>] mmput+0x80/0x150 [<ffffffff8f2a7609>] do_exit+0x2a9/0xd20 The references in the debug objects may also hide a real memory leak. As there is no point in having kmemleak to track debug object allocations, kmemleak checking is now disabled for debug objects. Signed-off-by:Waiman Long <longman@redhat.com> Signed-off-by:
-
- Aug 10, 2017
-
-
Akinobu Mita authored
Commit 1203c8e6 ("fault-inject: simplify access check for fail-nth") unintentionally broke a conditional statement in should_fail(). Any faults are not injected in the task context by the change when the systematic fault injection is not used. This change restores to the previous correct behaviour. Link: http://lkml.kernel.org/r/1501633700-3488-1-git-send-email-akinobu.mita@gmail.com Fixes: 1203c8e6 ("fault-inject: simplify access check for fail-nth") Signed-off-by:
Akinobu Mita <akinobu.mita@gmail.com> Reported-by:
Lu Fengqi <lufq.fnst@cn.fujitsu.com> Tested-by:
-
Dan Carpenter authored
The break was in the wrong place so file system tests don't work as intended, leaking memory at each test switch. [mcgrof@kernel.org: massaged commit subject, noted memory leak issue without the fix] Link: http://lkml.kernel.org/r/20170802211450.27928-6-mcgrof@kernel.org Fixes: 39258f448d71 ("kmod: add test driver to stress test the module loader") Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
Luis R. Rodriguez <mcgrof@kernel.org> Reported-by:
David Binderman <dcb314@hotmail.com> Cc: Colin Ian King <colin.king@canonical.com> Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com> Cc: Eric W. Biederman <ebiederm@xmission.com> Cc: Jessica Yu <jeyu@redhat.com> Cc: Josh Poimboeuf <jpoimboe@redhat.com> Cc: Kees Cook <keescook@chromium.org> Cc: Michal Marek <mmarek@suse.com> Cc: Miroslav Benes <mbenes@suse.cz> Cc: Petr Mladek <pmladek@suse.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Shuah Khan <shuah@kernel.org> Signed-off-by:
-
Dan Carpenter authored
We accidentally just drop the lock twice instead of taking it and then releasing it. This isn't a big issue unless you are adding more than one device to test on, and the kmod.sh doesn't do that yet, however this obviously is the correct thing to do. [mcgrof@kernel.org: massaged subject, explain what happens] Link: http://lkml.kernel.org/r/20170802211450.27928-5-mcgrof@kernel.org Fixes: 39258f448d71 ("kmod: add test driver to stress test the module loader") Signed-off-by:
-
Luis R. Rodriguez authored
Parsing with kstrtol() enables values to be negative, and we failed to check for negative values when parsing with test_dev_config_update_uint_sync() or test_dev_config_update_uint_range(). test_dev_config_update_uint_range() has a minimum check though so an issue is not present there. test_dev_config_update_uint_sync() is only used for the number of threads to use (config_num_threads_store()), and indeed this would fail with an attempt for a large allocation. Although the issue is only present in practice with the first fix both by using kstrtoul() instead of kstrtol(). Link: http://lkml.kernel.org/r/20170802211450.27928-4-mcgrof@kernel.org Fixes: 39258f448d71 ("kmod: add test driver to stress test the module loader") Signed-off-by:
-
Colin Ian King authored
Trivial fix to spelling mistake in snprintf text [mcgrof@kernel.org: massaged commit message] Link: http://lkml.kernel.org/r/20170802211450.27928-3-mcgrof@kernel.org Fixes: 39258f448d71 ("kmod: add test driver to stress test the module loader") Signed-off-by:
Colin Ian King <colin.king@canonical.com> Signed-off-by:
-
Luis R. Rodriguez authored
The firmware API has a feature to enable batching requests for the same fil e under one worker, so only one lookup is done. This only triggers if we so happen to schedule two lookups for same file around the same time, or if release_firmware() has not been called for a successful firmware call. This can happen for instance if you happen to have multiple devices and one device driver for certain drivers where the stars line up scheduling wise. This adds a new sync and async test trigger. Instead of adding a new trigger for each new test type we make the tests a bit configurable so that we could configure the tests in userspace and just kick a test through a few basic triggers. With this, for instance the two types of sync requests: o request_firmware() and o request_firmware_direct() can be modified with a knob. Likewise the two type of async requests: o request_firmware_nowait(uevent=true) and o request_firmware_nowait(uevent=false) can be configured with another knob. The call request_firmware_into_buf() has no users... yet. The old tests are left in place as-is given they serve a few other purposes which we are currently not interested in also testing yet. This will change later as we will be able to just consolidate all tests under a few basic triggers with just one general configuration setup. We perform two types of tests, one for where the file is present and one for where the file is not present. All test tests go tested and they now pass for the following 3 kernel builds possible for the firmware API: 0. Most distro setup: CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n CONFIG_FW_LOADER_USER_HELPER=y 1. Android: CONFIG_FW_LOADER_USER_HELPER_FALLBACK=y CONFIG_FW_LOADER_USER_HELPER=y 2. Rare build: CONFIG_FW_LOADER_USER_HELPER_FALLBACK=n CONFIG_FW_LOADER_USER_HELPER=n Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Byungchul Park authored
Although wait_for_completion() and its family can cause deadlock, the lock correctness validator could not be applied to them until now, because things like complete() are usually called in a different context from the waiting context, which violates lockdep's assumption. Thanks to CONFIG_LOCKDEP_CROSSRELEASE, we can now apply the lockdep detector to those completion operations. Applied it. Signed-off-by:
-
Byungchul Park authored
Lockdep is a runtime locking correctness validator that detects and reports a deadlock or its possibility by checking dependencies between locks. It's useful since it does not report just an actual deadlock but also the possibility of a deadlock that has not actually happened yet. That enables problems to be fixed before they affect real systems. However, this facility is only applicable to typical locks, such as spinlocks and mutexes, which are normally released within the context in which they were acquired. However, synchronization primitives like page locks or completions, which are allowed to be released in any context, also create dependencies and can cause a deadlock. So lockdep should track these locks to do a better job. The 'crossrelease' implementation makes these primitives also be tracked. Signed-off-by:
-
- Aug 09, 2017
-
-
Daniel Borkmann authored
Currently, eBPF only understands BPF_JGT (>), BPF_JGE (>=), BPF_JSGT (s>), BPF_JSGE (s>=) instructions, this means that particularly *JLT/*JLE counterparts involving immediates need to be rewritten from e.g. X < [IMM] by swapping arguments into [IMM] > X, meaning the immediate first is required to be loaded into a register Y := [IMM], such that then we can compare with Y > X. Note that the destination operand is always required to be a register. This has the downside of having unnecessarily increased register pressure, meaning complex program would need to spill other registers temporarily to stack in order to obtain an unused register for the [IMM]. Loading to registers will thus also affect state pruning since we need to account for that register use and potentially those registers that had to be spilled/filled again. As a consequence slightly more stack space might have been used due to spilling, and BPF programs are a bit longer due to extra code involving the register load and potentially required spill/fills. Thus, add BPF_JLT (<), BPF_JLE (<=), BPF_JSLT (s<), BPF_JSLE (s<=) counterparts to the eBPF instruction set. Modifying LLVM to remove the NegateCC() workaround in a PoC patch at [1] and allowing it to also emit the new instructions resulted in cilium's BPF programs that are injected into the fast-path to have a reduced program length in the range of 2-3% (e.g. accumulated main and tail call sections from one of the object file reduced from 4864 to 4729 insns), reduced complexity in the range of 10-30% (e.g. accumulated sections reduced in one of the cases from 116432 to 88428 insns), and reduced stack usage in the range of 1-5% (e.g. accumulated sections from one of the object files reduced from 824 to 784b). The modification for LLVM will be incorporated in a backwards compatible way. Plan is for LLVM to have i) a target specific option to offer a possibility to explicitly enable the extension by the user (as we have with -m target specific extensions today for various CPU insns), and ii) have the kernel checked for presence of the extensions and enable them transparently when the user is selecting more aggressive options such as -march=native in a bpf target context. (Other frontends generating BPF byte code, e.g. ply can probe the kernel directly for its code generation.) [1] https://github.com/borkmann/llvm/tree/bpf-insns Signed-off-by:
Daniel Borkmann <daniel@iogearbox.net> Acked-by:
Alexei Starovoitov <ast@kernel.org> Signed-off-by:
-
Ard Biesheuvel authored
Provide a NEON accelerated implementation of the recovery algorithm, which supersedes the default byte-by-byte one. Signed-off-by:
Ard Biesheuvel <ard.biesheuvel@linaro.org> Signed-off-by:
Catalin Marinas <catalin.marinas@arm.com>
-
Ard Biesheuvel authored
The P/Q left side optimization in the delta syndrome simply involves repeatedly multiplying a value by polynomial 'x' in GF(2^8). Given that 'x * x * x * x' equals 'x^4' even in the polynomial world, we can accelerate this substantially by performing up to 4 such operations at once, using the NEON instructions for polynomial multiplication. Results on a Cortex-A57 running in 64-bit mode: Before: ------- raid6: neonx1 xor() 1680 MB/s raid6: neonx2 xor() 2286 MB/s raid6: neonx4 xor() 3162 MB/s raid6: neonx8 xor() 3389 MB/s After: ------ raid6: neonx1 xor() 2281 MB/s raid6: neonx2 xor() 3362 MB/s raid6: neonx4 xor() 3787 MB/s raid6: neonx8 xor() 4239 MB/s While we're at it, simplify MASK() by using a signed shift rather than a vector compare involving a temp register. Signed-off-by:
-
- Jul 31, 2017
-
-
Phil Sutter authored
This is similar to strdup() for netlink string attributes. Signed-off-by:
Phil Sutter <phil@nwl.cc> Signed-off-by:
Pablo Neira Ayuso <pablo@netfilter.org>
-
Jamal Hadi Salim authored
Generic bitflags attribute content sent to the kernel by user. With this netlink attr type the user can either set or unset a flag in the kernel. The value is a bitmap that defines the bit values being set The selector is a bitmask that defines which value bit is to be considered. A check is made to ensure the rules that a kernel subsystem always conforms to bitflags the kernel already knows about. i.e if the user tries to set a bit flag that is not understood then the _it will be rejected_. In the most basic form, the user specifies the attribute policy as: [ATTR_GOO] = { .type = NLA_BITFIELD32, .validation_data = &myvalidflags }, where myvalidflags is the bit mask of the flags the kernel understands. If the user _does not_ provide myvalidflags then the attribute will also be rejected. Examples: value = 0x0, and selector = 0x1 implies we are selecting bit 1 and we want to set its value to 0. value = 0x2, and selector = 0x2 implies we are selecting bit 2 and we want to set its value to 1. Suggested-by:Jamal Hadi Salim <jhs@mojatatu.com> Signed-off-by:
-
- Jul 26, 2017
-
-
Jeff Layton authored
Nothing calls this wrapper anymore, so just remove it and rename the old function to get rid of the double underscore prefix. Signed-off-by:Jeff Layton <jlayton@redhat.com>
-
Josh Poimboeuf authored
A couple of Kconfig changes which make it much easier to switch to the new CONFIG_ORC_UNWINDER: 1) Remove x86 dependencies on CONFIG_FRAME_POINTER for lockdep, latencytop, and fault injection. x86 has a 'guess' unwinder which just scans the stack for kernel text addresses. It's not 100% accurate but in many cases it's good enough. This allows those users who don't want the text overhead of the frame pointer or ORC unwinders to still use these features. More importantly, this also makes it much more straightforward to disable frame pointers. 2) Make CONFIG_ORC_UNWINDER depend on !CONFIG_FRAME_POINTER. While it would be possible to have both enabled, it doesn't really make sense to do so. So enforce a sane configuration to prevent the user from making a dumb mistake. With these changes, when you disable CONFIG_FRAME_POINTER, "make oldconfig" will ask if you want to enable CONFIG_ORC_UNWINDER. Signed-off-by:
Josh Poimboeuf <jpoimboe@redhat.com> Cc: Andy Lutomirski <luto@kernel.org> Cc: Borislav Petkov <bp@alien8.de> Cc: Brian Gerst <brgerst@gmail.com> Cc: Denys Vlasenko <dvlasenk@redhat.com> Cc: H. Peter Anvin <hpa@zytor.com> Cc: Jiri Slaby <jslaby@suse.cz> Cc: Linus Torvalds <torvalds@linux-foundation.org> Cc: Mike Galbraith <efault@gmx.de> Cc: Peter Zijlstra <peterz@infradead.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: live-patching@vger.kernel.org Link: http://lkml.kernel.org/r/9985fb91ce5005fe33ea5cc2a20f14bd33c61d03.1500938583.git.jpoimboe@redhat.com Signed-off-by:
-
Josh Poimboeuf authored
Add the new ORC unwinder which is enabled by CONFIG_ORC_UNWINDER=y. It plugs into the existing x86 unwinder framework. It relies on objtool to generate the needed .orc_unwind and .orc_unwind_ip sections. For more details on why ORC is used instead of DWARF, see Documentation/x86/orc-unwinder.txt - but the short version is that it's a simplified, fundamentally more robust debugninfo data structure, which also allows up to two orders of magnitude faster lookups than the DWARF unwinder - which matters to profiling workloads like perf. Thanks to Andy Lutomirski for the performance improvement ideas: splitting the ORC unwind table into two parallel arrays and creating a fast lookup table to search a subset of the unwind table. Signed-off-by:
-
- Jul 25, 2017
-
-
Phil Sutter authored
I forgot one spot when introducing struct test_obj_val. Fixes: e859afe1 ("lib: test_rhashtable: fix for large entry counts") Reported by: kernel test robot <fengguang.wu@intel.com> Signed-off-by:
-
- Jul 24, 2017
-
-
Phil Sutter authored
During concurrent access testing, threadfunc() concatenated thread ID and object index to create a unique key like so: | tdata->objs[i].value = (tdata->id << 16) | i; This breaks if a user passes an entries parameter of 64k or higher, since 'i' might use more than 16 bits then. Effectively, this will lead to duplicate keys in the table. Fix the problem by introducing a struct holding object and thread ID and using that as key instead of a single integer type field. Fixes: f4a3e90b ("rhashtable-test: extend to test concurrency") Reported by: Manuel Messner <mm@skelett.io> Signed-off-by:
-
