linux/kernel/time
Jason Low 1018016c70 sched, timer: Replace spinlocks with atomics in thread_group_cputimer(), to improve scalability
While running a database workload, we found a scalability issue with itimers.

Much of the problem was caused by the thread_group_cputimer spinlock.
Each time we account for group system/user time, we need to obtain a
thread_group_cputimer's spinlock to update the timers. On larger systems
(such as a 16 socket machine), this caused more than 30% of total time
spent trying to obtain this kernel lock to update these group timer stats.

This patch converts the timers to 64-bit atomic variables and use
atomic add to update them without a lock. With this patch, the percent
of total time spent updating thread group cputimer timers was reduced
from 30% down to less than 1%.

Note: On 32-bit systems using the generic 64-bit atomics, this causes
sample_group_cputimer() to take locks 3 times instead of just 1 time.
However, we tested this patch on a 32-bit system ARM system using the
generic atomics and did not find the overhead to be much of an issue.
An explanation for why this isn't an issue is that 32-bit systems usually
have small numbers of CPUs, and cacheline contention from extra spinlocks
called periodically is not really apparent on smaller systems.

Signed-off-by: Jason Low <jason.low2@hp.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Aswin Chandramouleeswaran <aswin@hp.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: Mike Galbraith <umgwanakikbuti@gmail.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Preeti U Murthy <preeti@linux.vnet.ibm.com>
Cc: Scott J Norton <scott.norton@hp.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Waiman Long <Waiman.Long@hp.com>
Link: http://lkml.kernel.org/r/1430251224-5764-4-git-send-email-jason.low2@hp.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
2015-05-08 12:15:31 +02:00
..
alarmtimer.c all arches, signal: move restart_block to struct task_struct 2015-02-12 18:54:12 -08:00
clockevents.c Char/Misc driver patches for 4.1-rc1 2015-04-21 09:42:58 -07:00
clocksource.c clocksource: Improve comment explaining clocks_calc_max_nsecs()'s 50% safety margin 2015-04-03 08:18:35 +02:00
hrtimer.c clockevents: Cleanup dead cpu explicitely 2015-04-03 08:44:37 +02:00
itimer.c time/timers: Move all time(r) related files into kernel/time 2014-06-23 11:22:35 +02:00
jiffies.c tick: Move clocksource related stuff to timekeeping.h 2015-04-01 14:22:58 +02:00
Kconfig clockevents: Remove CONFIG_GENERIC_CLOCKEVENTS_BUILD 2015-04-01 14:22:57 +02:00
Makefile clockevents: Remove CONFIG_GENERIC_CLOCKEVENTS_BUILD 2015-04-01 14:22:57 +02:00
ntp_internal.h timekeeping: Convert timekeeping core to use timespec64s 2014-07-23 10:17:54 -07:00
ntp.c time: Add y2038 safe update_persistent_clock64() 2015-04-03 08:18:20 +02:00
posix-clock.c kernel: Fix files explicitly needing EXPORT_SYMBOL infrastructure 2011-10-31 19:30:05 -04:00
posix-cpu-timers.c sched, timer: Replace spinlocks with atomics in thread_group_cputimer(), to improve scalability 2015-05-08 12:15:31 +02:00
posix-timers.c posix-timers: Fix stack info leak in timer_create() 2014-10-25 10:43:15 +02:00
sched_clock.c timers, sched/clock: Clean up the code a bit 2015-03-27 08:34:01 +01:00
test_udelay.c time: Rename udelay_test.c to test_udelay.c 2014-11-21 11:59:55 -08:00
tick-broadcast-hrtimer.c timers/tick/broadcast-hrtimer: Fix suspicious RCU usage in idle loop 2015-03-23 10:50:05 +01:00
tick-broadcast.c clockevents: Cleanup dead cpu explicitely 2015-04-03 08:44:37 +02:00
tick-common.c timers/PM: Drop unnecessary braces from tick_freeze() 2015-04-03 15:15:52 +02:00
tick-internal.h clockevents: Cleanup dead cpu explicitely 2015-04-03 08:44:37 +02:00
tick-oneshot.c clockevents: Manage device's state separately for the core 2015-03-27 10:26:19 +01:00
tick-sched.c tick: Move core only declarations and functions to core 2015-04-01 14:22:58 +02:00
tick-sched.h arm/bL_switcher: Kill tick suspend hackery 2015-04-01 14:23:00 +02:00
time.c Merge branch 'timers-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip 2015-01-25 17:47:34 -08:00
timeconst.bc time/timers: Move all time(r) related files into kernel/time 2014-06-23 11:22:35 +02:00
timeconv.c
timecounter.c timecounter: keep track of accumulated fractional nanoseconds 2014-12-30 18:29:27 -05:00
timekeeping_debug.c timekeeping: Convert timekeeping core to use timespec64s 2014-07-23 10:17:54 -07:00
timekeeping_internal.h clocksource: Move cycle_last validation to core code 2014-07-23 15:01:51 -07:00
timekeeping.c timekeeping: Get rid of stale comment 2015-04-03 08:44:37 +02:00
timekeeping.h tick: Move clocksource related stuff to timekeeping.h 2015-04-01 14:22:58 +02:00
timer_list.c tick: Move core only declarations and functions to core 2015-04-01 14:22:58 +02:00
timer_stats.c timer stats: Add a 'Collection: active/inactive' line to timer usage statistics 2013-10-10 09:59:25 +02:00
timer.c timer: Further simplify the SMP and HOTPLUG logic 2015-04-02 17:46:21 +02:00