Messages in this thread | | | Date | Wed, 15 Apr 2015 16:20:02 +0200 | From | Peter Zijlstra <> | Subject | Re: [PATCH 3/3] perf: Fix mux_interval hrtimer wreckage |
| |
On Wed, Apr 15, 2015 at 07:48:02AM -0600, David Ahern wrote: > Hi Peter: > > On 4/15/15 3:41 AM, Peter Zijlstra wrote: > >Thomas stumbled over the hrtimer_forward_now() in > >perf_event_mux_interval_ms_store() and noticed its broken-ness. > > > >You cannot just change the expiry time of an active timer, it will > >destroy the red-black tree order and cause havoc. > > > >Change it to (re)start the timer instead, (re)starting a timer will > >dequeue and enqueue a timer and therefore preserve rb-tree order. > > Should this be fixed in stable trees as well?
Hmm, I was going to write how this all relies on the hrtimer rework, but I think we might just be fine because all this is strictly cpu local and serialized through irq-disable.
But yes, its can't be worse than it is now.
| |