linux_dsm_epyc7002/kernel/locking
Waiman Long 4b486b535c locking/rwsem: Exit read lock slowpath if queue empty & no writer
It was discovered that a constant stream of readers with occassional
writers pounding on a rwsem may cause many of the readers to enter the
slowpath unnecessarily thus increasing latency and lowering performance.

In the current code, a reader entering the slowpath critical section
will unconditionally set the WAITING_BIAS, if not set yet, and clear
its active count even if no one is in the wait queue and no writer
is present. This causes some incoming readers to observe the presence
of waiters in the wait queue and hence have to go into the slowpath
themselves.

With sufficient numbers of readers and a relatively short lock hold time,
the WAITING_BIAS may be repeatedly turned on and off and a substantial
portion of the readers will go into the slowpath sustaining a rather
long queue in the wait queue spinlock and repeated WAITING_BIAS on/off
cycle until the logjam is broken opportunistically.

To avoid this situation from happening, an additional check is added to
detect the special case that the reader in the critical section is the
only one in the wait queue and no writer is present. When that happens,
it can just exit the slowpath and return immediately as its active count
has already been set in the lock.  Other incoming readers won't observe
the presence of waiters and so will not be forced into the slowpath.

The issue was found in a customer site where they had an application
that pounded on the pread64 syscalls heavily on an XFS filesystem. The
application was run in a recent 4-socket boxes with a lot of CPUs. They
saw significant spinlock contention in the rwsem_down_read_failed() call.
With this patch applied, the system CPU usage went down from 85% to 57%,
and the spinlock contention in the pread64 syscalls was gone.

Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Davidlohr Bueso <dbueso@suse.de>
Acked-by: Will Deacon <will.deacon@arm.com>
Cc: Joe Mario <jmario@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1532459425-19204-1-git-send-email-longman@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2018-09-10 10:16:39 +02:00
..
lockdep_internals.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
lockdep_proc.c proc: introduce proc_create_single{,_data} 2018-05-16 07:23:35 +02:00
lockdep_states.h locking/lockdep: Rework FS_RECLAIM annotation 2017-08-10 12:29:03 +02:00
lockdep.c tracing: Partial revert of "tracing: Centralize preemptirq tracepoints and unify their usage" 2018-08-10 15:11:25 -04:00
locktorture.c drm pull for 4.19-rc1 2018-08-15 17:39:07 -07:00
Makefile License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
mcs_spinlock.h locking/mcs: Use smp_cond_load_acquire() in MCS spin loop 2018-04-27 09:48:49 +02:00
mutex-debug.c locking/mutex: Rework mutex::owner 2016-10-25 11:31:50 +02:00
mutex-debug.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
mutex.c locking/mutex: Fix mutex debug call and ww_mutex documentation 2018-09-10 10:05:10 +02:00
mutex.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
osq_lock.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
percpu-rwsem.c locking/percpu-rwsem: Replace waitqueue with rcuwait 2017-01-14 11:14:35 +01:00
qrwlock.c locking/qrwlock: Prevent slowpath writers getting held up by fastpath 2017-10-25 10:57:25 +02:00
qspinlock_paravirt.h locking/qspinlock: Remove duplicate clear_pending() function from PV code 2018-04-27 12:55:22 +02:00
qspinlock_stat.h locking/qspinlock: Add stat tracking for pending vs. slowpath 2018-04-27 09:48:53 +02:00
qspinlock.c locking/qspinlock: Add stat tracking for pending vs. slowpath 2018-04-27 09:48:53 +02:00
rtmutex_common.h locking/rtmutex: Handle non enqueued waiters gracefully in remove_waiter() 2018-03-28 23:01:30 +02:00
rtmutex-debug.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
rtmutex-debug.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
rtmutex.c locking/rtmutex: Allow specifying a subclass for nested locking 2018-07-25 11:22:19 +02:00
rtmutex.h License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
rwsem-spinlock.c License cleanup: add SPDX GPL-2.0 license identifier to files with no license 2017-11-02 11:10:55 +01:00
rwsem-xadd.c locking/rwsem: Exit read lock slowpath if queue empty & no writer 2018-09-10 10:16:39 +02:00
rwsem.c locking/rwsem: Fix up_read_non_owner() warning with DEBUG_RWSEMS 2018-06-20 11:29:23 +02:00
rwsem.h locking/rwsem: Add a new RWSEM_ANONYMOUSLY_OWNED flag 2018-05-16 11:45:15 +02:00
semaphore.c sched/headers: Prepare for new header dependencies before moving code to <linux/sched/debug.h> 2017-03-02 08:42:34 +01:00
spinlock_debug.c locking/spinlock/debug: Remove spinlock lockup detection code 2017-02-10 09:09:49 +01:00
spinlock.c locking/core: Remove break_lock field when CONFIG_GENERIC_LOCKBREAK=y 2017-12-12 11:24:01 +01:00
test-ww_mutex.c locking: Implement an algorithm choice for Wound-Wait mutexes 2018-07-03 09:44:36 +02:00