* Thomas Gleixner <tglx@xxxxxxxxxxxxx> [2009-03-17 14:07:44]: > On Mon, 16 Mar 2009, Arun R Bharadwaj wrote: > > * Arun R Bharadwaj <arun@xxxxxxxxxxxxxxxxxx> [2009-03-16 16:40:45]: > > > > The following pinned hrtimers have been identified and marked: > > 1)sched_rt_period_timer > > 2)tick_sched_timer > > 3)stack_trace_timer_fn > > How did you verify that these are the only timers which need to be > pinned ? > Until 2.6.28, there was a callback mode called HRTIMER_CB_IRQSAFE_PERCPU which did not allow timers to be migrated, if set. These were the only 3 hrtimers which were using that callback mode. --arun > Can we be sure that there is no code which relies on the current > behaviour to keep functionality tied together ? Networking and block > layer comes to mind. > > Thanks, > > tglx _______________________________________________ linux-pm mailing list linux-pm@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/linux-pm
