Messages in this thread |  | | From | Frederic Weisbecker <> | Subject | [PATCH 0/2] dynticks: A few updates | Date | Fri, 14 Feb 2014 13:12:38 +0100 |
| |
Hi,
A small pair of updates for the dynticks code. The patch from Viresh removes a few scheduler IPIs that I have seen on boot which are there to wake up CPUs when some deferrable timers are enqueued. Those were enqueued on all CPUs so there is a possible big round of IPI. And those can be avoided with the first patch. On my machine the main sources of these were MCE, vmstat/SLAB, cpufreq. They happen either on initcall or cpu hotplug.
But deferrable timers happen anytime, not just at boot. It depends on your config and your load. So the positive impact of the first patch should be broader.
git://git.kernel.org/pub/scm/linux/kernel/git/frederic/dynticks-testing.git timers/core
Thanks.
--- Paul Gortmaker (1): nohz: ensure users are aware boot CPU is not NO_HZ_FULL
Viresh Kumar (1): timer: Spare IPI when deferrable timer is queued on idle remote targets
kernel/time/Kconfig | 2 +- kernel/timer.c | 9 ++++++++- 2 files changed, 9 insertions(+), 2 deletions(-)
-- 1.8.3.1
|  |