Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Peter Zijlstra
On Fri, May 31, 2013 at 05:55:05PM +0200, Stephane Eranian wrote: > On Fri, May 31, 2013 at 5:48 PM, Peter Zijlstra wrote: > > On Fri, May 31, 2013 at 04:21:04PM +0200, Stephane Eranian wrote: > >> Hi, > >> > >> Any comment on this patch? > >> It does really help with stack pressure and will help

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Peter Zijlstra
On Thu, May 23, 2013 at 11:07:03AM -0700, Andrew Hunter wrote: > schedule_events caches event constraints on the stack during > scheduling. Given the number of possible events, this is 512 bytes of > stack; since it can be invoked under schedule() under god-knows-what, > this is causing stack

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Stephane Eranian
On Fri, May 31, 2013 at 5:48 PM, Peter Zijlstra wrote: > On Fri, May 31, 2013 at 04:21:04PM +0200, Stephane Eranian wrote: >> Hi, >> >> Any comment on this patch? >> It does really help with stack pressure and will help fix a PEBS-LL issue. > > Yeah, its on my todo list, but I've been spending

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Peter Zijlstra
On Fri, May 31, 2013 at 04:21:04PM +0200, Stephane Eranian wrote: > Hi, > > Any comment on this patch? > It does really help with stack pressure and will help fix a PEBS-LL issue. Yeah, its on my todo list, but I've been spending all of my time trying to fix all the bugs Vince has been throwing

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Stephane Eranian
Hi, Any comment on this patch? It does really help with stack pressure and will help fix a PEBS-LL issue. Thanks. On Thu, May 23, 2013 at 8:07 PM, Andrew Hunter wrote: > schedule_events caches event constraints on the stack during > scheduling. Given the number of possible events, this is 512

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Stephane Eranian
Hi, Any comment on this patch? It does really help with stack pressure and will help fix a PEBS-LL issue. Thanks. On Thu, May 23, 2013 at 8:07 PM, Andrew Hunter a...@google.com wrote: schedule_events caches event constraints on the stack during scheduling. Given the number of possible events,

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Peter Zijlstra
On Fri, May 31, 2013 at 04:21:04PM +0200, Stephane Eranian wrote: Hi, Any comment on this patch? It does really help with stack pressure and will help fix a PEBS-LL issue. Yeah, its on my todo list, but I've been spending all of my time trying to fix all the bugs Vince has been throwing my

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Stephane Eranian
On Fri, May 31, 2013 at 5:48 PM, Peter Zijlstra pet...@infradead.org wrote: On Fri, May 31, 2013 at 04:21:04PM +0200, Stephane Eranian wrote: Hi, Any comment on this patch? It does really help with stack pressure and will help fix a PEBS-LL issue. Yeah, its on my todo list, but I've been

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Peter Zijlstra
On Thu, May 23, 2013 at 11:07:03AM -0700, Andrew Hunter wrote: schedule_events caches event constraints on the stack during scheduling. Given the number of possible events, this is 512 bytes of stack; since it can be invoked under schedule() under god-knows-what, this is causing stack

Re: [PATCH] perf: reduce stack usage of schedule_events

2013-05-31 Thread Peter Zijlstra
On Fri, May 31, 2013 at 05:55:05PM +0200, Stephane Eranian wrote: On Fri, May 31, 2013 at 5:48 PM, Peter Zijlstra pet...@infradead.org wrote: On Fri, May 31, 2013 at 04:21:04PM +0200, Stephane Eranian wrote: Hi, Any comment on this patch? It does really help with stack pressure and will

[PATCH] perf: reduce stack usage of schedule_events

2013-05-23 Thread Andrew Hunter
schedule_events caches event constraints on the stack during scheduling. Given the number of possible events, this is 512 bytes of stack; since it can be invoked under schedule() under god-knows-what, this is causing stack blowouts. Trade some space usage for stack safety: add a place to cache

[PATCH] perf: reduce stack usage of schedule_events

2013-05-23 Thread Andrew Hunter
schedule_events caches event constraints on the stack during scheduling. Given the number of possible events, this is 512 bytes of stack; since it can be invoked under schedule() under god-knows-what, this is causing stack blowouts. Trade some space usage for stack safety: add a place to cache