Messages in this thread | | | Date | Fri, 20 Nov 2009 15:19:59 +0100 | From | Heiko Carstens <> | Subject | Re: [tip:timers/urgent] hrtimer: Fix /proc/timer_list regression |
| |
On Fri, Nov 20, 2009 at 10:30:39AM +0000, tip-bot for Feng Tang wrote: > Commit-ID: 8629ea2eaba8ca0de2e38ce1b4a825e16255976e > Gitweb: http://git.kernel.org/tip/8629ea2eaba8ca0de2e38ce1b4a825e16255976e > Author: Feng Tang <feng.tang@intel.com> > AuthorDate: Thu, 3 Sep 2009 16:32:53 +0800 > Committer: Thomas Gleixner <tglx@linutronix.de> > CommitDate: Fri, 20 Nov 2009 11:25:48 +0100 > > hrtimer: Fix /proc/timer_list regression > > commit 507e1231 (timer stats: Optimize by adding quick check to avoid > function calls) introduced a regression in /proc/timer_list. > > /proc/timer_list shows now > #0: <c27d46b0>, tick_sched_timer, S:01, <(null)>, /-1 > instead of > #0: <c27d46b0>, tick_sched_timer, S:01, hrtimer_start, swapper/0
For a 'real' fix I see three possible ways if timer_stats are disabled:
Print something like:
#0: <c27d46b0>, tick_sched_timer, S:01, unknown, unknown/-1
That way the format would be still the same and it should be quite obvious that timer_stats are disabled and should be enabled.
Alternatively leave start_site, taskname and pid away if the correspending timer has a NULL start_site. That would be the same like for the !TIMER_STATS case:
#0: <c27d46b0>, tick_sched_timer, S:01
or (worst choice) revert the whole optimization and live with extra cpu cost.
| |