sched: Export ns irqtimes through /proc/stat
authorVenkatesh Pallipadi <venki@google.com>
Wed, 22 Dec 2010 01:09:03 +0000 (17:09 -0800)
committerIngo Molnar <mingo@elte.hu>
Wed, 26 Jan 2011 11:33:21 +0000 (12:33 +0100)
commitabb74cefa9c682fb38ba86c17ca3c86fed6cc464
tree6a2cbcbb8d509c607cd61051d18ebd6019855aaf
parent70a89a6620f658d47a1488515bada4b8ee6291d8
sched: Export ns irqtimes through /proc/stat

CONFIG_IRQ_TIME_ACCOUNTING adds ns granularity irq time on each CPU.
This info is already used in scheduler to do proper task chargeback
(earlier patches). This patch retro-fits this ns granularity
hardirq and softirq information to /proc/stat irq and softirq fields.

The update is still done on timer tick, where we look at accumulated
ns hardirq/softirq time and account the tick to user/system/irq/hardirq/guest
accordingly.

No new interface added.

Earlier versions looked at adding this as new fields in some /proc
files. This one seems to be the best in terms of impact to existing
apps, even though it has somewhat more kernel code than earlier versions.

Tested-by: Shaun Ruffell <sruffell@digium.com>
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1292980144-28796-5-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/sched.c