Messages in this thread | | | Date | Mon, 24 Oct 2016 14:12:02 +0200 | From | Peter Zijlstra <> | Subject | Re: hit a KASan bug related to Perf during stress test |
| |
On Mon, Oct 24, 2016 at 02:04:11PM +0200, Jiri Olsa wrote: > On Mon, Oct 24, 2016 at 01:29:45PM +0200, Peter Zijlstra wrote:
> > Hurm, then again, I imagine that after unhash_process the PID/TID could > > be instantly re-used and then we're still confused. > > sounds bad.. I haven't checked the related pid_alive code, > but shouldn't we already get the EXIT event in this case?
It has, perf_event_exit_task() happens before we unhash.
But a per-cpu event that has PID/TID reporting on will run into this.
We'll observe 'funny' values between the unhash and the next context switch.
| |