Re: [RFC -tip] perf, x86: Add PERF_COUNT_HW_NMI_WATCHDOG event v2

[Cyrill Gorcunov]

On Tue, Jun 21, 2011 at 08:48:20PM +0400, Cyrill Gorcunov wrote:
...
> 
>   The key here is that we use that named non-sleeping ticks (as oprofile
> did) for nmi-watchdog and it allows us to free "cpu-cycles" counter for
> user needs. Of course we pick up one counter for this but it doesn't intersect
> with "cpu-cycles" counter (because counters are grouped and can count only
> specified events in each group). Stepane, should I post more details?
> 
> 	Cyrill

  Stephane, I think I have a shorter explanation -- the counter is still borrowed
(there is no magic ;) but this counter is less loaded than anything else (in
terms of restrictions) so since CPU cycles is a way more popular now it's allowed
to count them simultaneously with nmi-watchdog, but again  -- this means
another set of events can't work and if (for some reason) user needs this counter
he has to disable nmi-watchdog.

	Cyrill
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/