Mike Galbraith 39c0cbe215 sched: Rate-limit nohz ... Entering nohz code on every micro-idle is costing ~10% throughput for netperf TCP_RR when scheduling cross-cpu. Rate limiting entry fixes this, but raises ticks a bit. On my Q6600, an idle box goes from ~85 interrupts/sec to 128. The higher the context switch rate, the more nohz entry costs. With this patch and some cycle recovery patches in my tree, max cross cpu context switch rate is improved by ~16%, a large portion of which of which is this ratelimiting. Signed-off-by: Mike Galbraith <efault@gmx.de> Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl> LKML-Reference: <1268301003.6785.28.camel@marge.simson.net> Signed-off-by: Ingo Molnar <mingo@elte.hu>		2010-03-11 18:32:49 +01:00
..
clockevents.c	clockevent: Don't remove broadcast device when cpu is dead	2010-01-18 14:44:50 +01:00
clocksource.c	Merge branch 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip	2010-03-01 08:48:25 -08:00
jiffies.c
Kconfig
Makefile
ntp.c	ntp: Cleanup xtime references in ntp.c	2010-01-29 10:15:19 +01:00
tick-broadcast.c	clockevents: Convert to raw_spinlock	2009-12-14 23:55:34 +01:00
tick-common.c	clockevents: Convert to raw_spinlock	2009-12-14 23:55:34 +01:00
tick-internal.h	clockevents: Make tick_device_lock static	2009-12-14 23:55:34 +01:00
tick-oneshot.c
tick-sched.c	sched: Rate-limit nohz	2010-03-11 18:32:49 +01:00
timecompare.c	timecompare: fix half-Y2K38 problem in timecompare_update while calculating offset	2009-12-16 07:19:57 -08:00
timeconv.c
timekeeping.c	Merge branch 'timers-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip	2010-03-01 08:48:25 -08:00
timer_list.c	cpumask: avoid dereferencing struct cpumask	2009-12-17 11:43:29 +10:30
timer_stats.c	hrtimers: Convert to raw_spinlocks	2009-12-14 23:55:34 +01:00