apic: tracing IPI events
in [Kernel]
|
Prev: PATCH 3/6 - direct-io: do not merge logically non-contiguous requests
Next: Create generic alignment API (v8)
From: Andi Kleen on 6 Aug 2010 07:30 Jiri Olsa <jolsa(a)redhat.com> writes: > > please let me know what you think You forgot to state why you want it? In generall the only case that should happen anytime commonly is mask, the others only for obscure uncommon paths. However there is already accounting for it in /proc/interrupts that can be easily sampled. Does adding additional tracing really give any value? -Andi -- ak(a)linux.intel.com -- Speaking for myself only. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Jiri Olsa on 6 Aug 2010 07:50
On Fri, Aug 06, 2010 at 01:26:36PM +0200, Andi Kleen wrote: > Jiri Olsa <jolsa(a)redhat.com> writes: > > > > please let me know what you think > > You forgot to state why you want it? > > In generall the only case that should happen anytime commonly is mask, > the others only for obscure uncommon paths. > > However there is already accounting for it in /proc/interrupts > that can be easily sampled. Does adding additional tracing really give > any value? As I was asked to do that for other guys, I'm not completely sure. Anyway the reason I got was, that while debuging crash it seemed some IPIs might got lost. Also it'd be probably used as a tracepoint itself via additional code, than by ring buffer tracing. I'll check if I can get more justifing info.. jirka > > -Andi > -- > ak(a)linux.intel.com -- Speaking for myself only. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |