posix_cpu_timer: Reduce unnecessary sighand lock contention
authorJason Low <jason.low2@hp.com>
Wed, 14 Oct 2015 19:07:56 +0000 (12:07 -0700)
committerThomas Gleixner <tglx@linutronix.de>
Thu, 15 Oct 2015 09:23:41 +0000 (11:23 +0200)
commitc8d75aa47dd585c9538a8205e9bb9847e12cfb84
tree2d02f5e5041dcec03648a69730420ae9fbafbfcb
parentd5c373eb5610686162ff50429f63f4c00c554799
posix_cpu_timer: Reduce unnecessary sighand lock contention

It was found while running a database workload on large systems that
significant time was spent trying to acquire the sighand lock.

The issue was that whenever an itimer expired, many threads ended up
simultaneously trying to send the signal. Most of the time, nothing
happened after acquiring the sighand lock because another thread
had just already sent the signal and updated the "next expire" time.
The fastpath_timer_check() didn't help much since the "next expire"
time was updated after the threads exit fastpath_timer_check().

This patch addresses this by having the thread_group_cputimer structure
maintain a boolean to signify when a thread in the group is already
checking for process wide timers, and adds extra logic in the fastpath
to check the boolean.

Signed-off-by: Jason Low <jason.low2@hp.com>
Reviewed-by: Oleg Nesterov <oleg@redhat.com>
Reviewed-by: George Spelvin <linux@horizon.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: hideaki.kimura@hpe.com
Cc: terry.rudd@hpe.com
Cc: scott.norton@hpe.com
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1444849677-29330-5-git-send-email-jason.low2@hp.com
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
include/linux/init_task.h
include/linux/sched.h
kernel/time/posix-cpu-timers.c