linux/ipc/sem.c, branch v4.11

sched/headers: Prepare for new header dependencies before moving code to

2017-03-02T07:42:26Z

We are going to split out of , which will have to be picked up from other headers and a couple of .c files. Create a trivial placeholder file that just maps to to make this patch obviously correct and bisectable. Include the new header in the files that are going to need it. Acked-by: Linus Torvalds Cc: Mike Galbraith Cc: Peter Zijlstra Cc: Thomas Gleixner Signed-off-by: Ingo Molnar

ipc/sem: add hysteresis

2017-02-28T02:43:46Z

sysv sem has two lock modes: One with per-semaphore locks, one lock mode with a single global lock for the whole array. When switching from the per-semaphore locks to the global lock, all per-semaphore locks must be scanned for ongoing operations. The patch adds a hysteresis for switching from the global lock to the per semaphore locks. This reduces how often the per-semaphore locks must be scanned. Compared to the initial patch, this is a simplified solution: Setting USE_GLOBAL_LOCK_HYSTERESIS to 1 restores the current behavior. In theory, a workload with exactly 10 simple sops and then one complex op now scales a bit worse, but this is pure theory: If there is concurrency, the it won't be exactly 10:1:10:1:10:1:... If there is no concurrency, then there is no need for scalability. Link: http://lkml.kernel.org/r/1476851896-3590-3-git-send-email-manfred@colorfullife.com Signed-off-by: Manfred Spraul Cc: Peter Zijlstra Cc: Davidlohr Bueso Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Cc: <1vier1@web.de> Cc: kernel test robot Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem.c: avoid using spin_unlock_wait()

2017-02-28T02:43:46Z

a) The ACQUIRE in spin_lock() applies to the read, not to the store, at least for powerpc. This forces to add a smp_mb() into the fast path. b) The memory barrier provided by spin_unlock_wait() is right now arch dependent. Therefore: Use spin_lock()/spin_unlock() instead of spin_unlock_wait(). Advantage: faster single op semop calls(), observed +8.9% on x86. (the other solution would be arch dependencies in ipc/sem). Disadvantage: slower complex op semop calls, if (and only if) there are no sleeping operations. The next patch adds hysteresis, this further reduces the probability that the slow path is used. Link: http://lkml.kernel.org/r/1476851896-3590-2-git-send-email-manfred@colorfullife.com Signed-off-by: Manfred Spraul Cc: Peter Zijlstra Cc: Davidlohr Bueso Cc: Thomas Gleixner Cc: Ingo Molnar Cc: H. Peter Anvin Cc: <1vier1@web.de> Cc: kernel test robot Cc: Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem.c: fix incorrect sem_lock pairing

2017-01-11T02:31:55Z

Based on the syzcaller test case from dvyukov: https://gist.githubusercontent.com/dvyukov/d0e5efefe4d7d6daed829f5c3ca26a40/raw/08d0a261fe3c987bed04fbf267e08ba04bd533ea/gistfile1.txt The slow (i.e.: failure to acquire) syscall exit from semtimedop() incorrectly assumed that the the same lock is acquired as it was at the initial syscall entry. This is wrong: - thread A: single semop semop(), sleeps - thread B: multi semop semop(), sleeps - thread A: woken up by signal/timeout With this sequence, the initial sem_lock() call locks the per-semaphore spinlock, and it is unlocked with sem_unlock(). The call at the syscall return locks the global spinlock. Because locknum is not updated, the following sem_unlock() call unlocks the per-semaphore spinlock, which is actually not locked. The fix is trivial: Use the return value from sem_lock. Fixes: 370b262c896e ("ipc/sem: avoid idr tree lookup for interrupted semop") Link: http://lkml.kernel.org/r/1482215645-22328-1-git-send-email-manfred@colorfullife.com Signed-off-by: Manfred Spraul Reported-by: Dmitry Vyukov Reported-by: Johanna Abrahamsson Tested-by: Johanna Abrahamsson Acked-by: Davidlohr Bueso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem: avoid idr tree lookup for interrupted semop

2016-12-15T00:04:08Z

We can avoid the idr tree lookup (albeit possibly avoiding idr_find_fast()) when being awoken in EINTR, as the semid will not change in this context while blocked. Use the sma pointer directly and take the sem_lock, then re-check for RMID races. We continue to re-check the queue.status with the lock held such that we can detect situations where we where are dealing with a spurious wakeup but another task that holds the sem_lock updated the queue.status while we were spinning for it. Once we take the lock it obviously won't change again. Being the only caller, get rid of sem_obtain_lock() altogether. Link: http://lkml.kernel.org/r/1478708774-28826-3-git-send-email-dave@stgolabs.net Signed-off-by: Davidlohr Bueso Cc: Manfred Spraul Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem: simplify wait-wake loop

2016-12-15T00:04:08Z

Instead of using the reverse goto, we can simplify the flow and make it more language natural by just doing do-while instead. One would hope this is the standard way (or obviously just with a while bucle) that we do wait/wakeup handling in the kernel. The exact same logic is kept, just more indented. [akpm@linux-foundation.org: coding-style fixes] Link: http://lkml.kernel.org/r/1478708774-28826-2-git-send-email-dave@stgolabs.net Signed-off-by: Davidlohr Bueso Cc: Manfred Spraul Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem: use proper list api for pending_list wakeups

2016-12-15T00:04:08Z

... saves some LoC and looks cleaner than re-implementing the calls. Link: http://lkml.kernel.org/r/1474225896-10066-6-git-send-email-dave@stgolabs.net Signed-off-by: Davidlohr Bueso Acked-by: Manfred Spraul Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem: explicitly inline check_restart

2016-12-15T00:04:08Z

The compiler already does this, but make it explicit. This helper is really small and also used in update_queue's main loop, which is O(N^2) scanning. Inline and avoid the function overhead. Link: http://lkml.kernel.org/r/1474225896-10066-5-git-send-email-dave@stgolabs.net Signed-off-by: Davidlohr Bueso Cc: Manfred Spraul Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem: optimize perform_atomic_semop()

2016-12-15T00:04:08Z

This is the main workhorse that deals with semop user calls such that the waitforzero or semval update operations, on the set, can complete on not as the sma currently stands. Currently, the set is iterated twice (setting semval, then backwards for the sempid value). Slowpaths, and particularly SEM_UNDO calls, must undo any altered sem when it is detected that the caller must block or has errored-out. With larger sets, there can occur situations where this involves a lot of cycles and can obviously be a suboptimal use of cached resources in shared memory. Ie, discarding CPU caches that are also calling semop and have the sembuf cached (and can complete), while the current lock holder doing the semop will block, error, or does a waitforzero operation. This patch proposes still iterating the set twice, but the first scan is read-only, and we perform the actual updates afterward, once we know that the call will succeed. In order to not suffer from the overhead of dealing with sops that act on the same sem_num, such (rare) cases use perform_atomic_semop_slow(), which is exactly what we have now. Duplicates are detected before grabbing sem_lock, and uses simple a 32/64-bit hash array variable to based on the sem_num we are working on. In addition add some comments to when we expect to the caller to block. [akpm@linux-foundation.org: coding-style fixes] [colin.king@canonical.com: ensure we left shift a ULL rather than a 32 bit integer] Link: http://lkml.kernel.org/r/20161028181129.7311-1-colin.king@canonical.com Link: http://lkml.kernel.org/r/20160921194603.GB21438@linux-80c1.suse Signed-off-by: Davidlohr Bueso Cc: Manfred Spraul Signed-off-by: Colin Ian King Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds

ipc/sem: rework task wakeups

2016-12-15T00:04:08Z

Our sysv sems have been using the notion of lockless wakeups for a while, ever since commit 0a2b9d4c7967 ("ipc/sem.c: move wake_up_process out of the spinlock section"), in order to reduce the sem_lock hold times. This in-house pending queue can be replaced by wake_q (just like all the rest of ipc now), in that it provides the following advantages: o Simplifies and gets rid of unnecessary code. o We get rid of the IN_WAKEUP complexities. Given that wake_q_add() grabs reference to the task, if awoken due to an unrelated event, between the wake_q_add() and wake_up_q() window, we cannot race with sys_exit and the imminent call to wake_up_process(). o By not spinning IN_WAKEUP, we no longer need to disable preemption. In consequence, the wakeup paths (after schedule(), that is) must acknowledge an external signal/event, as well spurious wakeup occurring during the pending wakeup window. Obviously no changes in semantics that could be visible to the user. The fastpath is _only_ for when we know for sure that we were awoken due to a the waker's successful semop call (queue.status is not -EINTR). On a 48-core Haswell, running the ipcscale 'waitforzero' test, the following is seen with increasing thread counts: v4.8-rc5 v4.8-rc5 semopv2 Hmean sembench-sem-2 574733.00 ( 0.00%) 578322.00 ( 0.62%) Hmean sembench-sem-8 811708.00 ( 0.00%) 824689.00 ( 1.59%) Hmean sembench-sem-12 842448.00 ( 0.00%) 845409.00 ( 0.35%) Hmean sembench-sem-21 933003.00 ( 0.00%) 977748.00 ( 4.80%) Hmean sembench-sem-48 935910.00 ( 0.00%) 1004759.00 ( 7.36%) Hmean sembench-sem-79 937186.00 ( 0.00%) 983976.00 ( 4.99%) Hmean sembench-sem-234 974256.00 ( 0.00%) 1060294.00 ( 8.83%) Hmean sembench-sem-265 975468.00 ( 0.00%) 1016243.00 ( 4.18%) Hmean sembench-sem-296 991280.00 ( 0.00%) 1042659.00 ( 5.18%) Hmean sembench-sem-327 975415.00 ( 0.00%) 1029977.00 ( 5.59%) Hmean sembench-sem-358 1014286.00 ( 0.00%) 1049624.00 ( 3.48%) Hmean sembench-sem-389 972939.00 ( 0.00%) 1043127.00 ( 7.21%) Hmean sembench-sem-420 981909.00 ( 0.00%) 1056747.00 ( 7.62%) Hmean sembench-sem-451 990139.00 ( 0.00%) 1051609.00 ( 6.21%) Hmean sembench-sem-482 965735.00 ( 0.00%) 1040313.00 ( 7.72%) [akpm@linux-foundation.org: coding-style fixes] [sfr@canb.auug.org.au: merge fix for WAKE_Q to DEFINE_WAKE_Q rename] Link: http://lkml.kernel.org/r/20161122210410.5eca9fc2@canb.auug.org.au Link: http://lkml.kernel.org/r/1474225896-10066-3-git-send-email-dave@stgolabs.net Signed-off-by: Davidlohr Bueso Acked-by: Manfred Spraul Signed-off-by: Stephen Rothwell Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds