]> pilppa.com Git - linux-2.6-omap-h63xx.git/commit
timers: more consistently use clock vs timer
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Thu, 12 Feb 2009 14:00:52 +0000 (15:00 +0100)
committerIngo Molnar <mingo@elte.hu>
Fri, 13 Feb 2009 12:04:05 +0000 (13:04 +0100)
commit3997ad317fdf9ecdb5702e2b4fd1f8229814ff8c
treebe0b1802f65e85157d97acac38f1bd310ba6d626
parent37bed90094fdb1eea6e4afec6a200d4e60143e55
timers: more consistently use clock vs timer

While reviewing the manpages, I noticed I'd missed some clock vs timer sites.

Make sure that all timer functions call cpu_timer_sample_group() and not
cpu_clock_sample_group(). This ensures that we enable the process wide timer
in time, and therefore pay the O(n) thread group cost from the syscall.

Not doing it here, will result in the first jiffy tick after setting the timer
doing this, resulting in a very expensive tick (but only once) and a delay in
actually starting the timer.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
kernel/posix-cpu-timers.c