On Thu, Feb 21, 2008 at 1:42 PM, Ingo Molnar <mingo@elte.hu> wrote:
quoted text
>  I'd not exclude them fundamentally though, it's really the numbers that
>  matter. The code is certainly simple enough (albeit the .config and
>  sysctl controls are quite ugly and unacceptable - adaptive mutexes
>  should really be ... adaptive, with no magic constants in .configs or
>  else).
>
>  But ... i'm somewhat sceptic, after having played with spin-a-bit
>  mutexes before.

Yeah, it's yet another reason to get better instrumentation in -rt for
this purpose and why it's important to only log things in the slow
path so that the front end of the rtmutex doesn't see instrumentation
artifacts in the measurement. If I get a chance tonight, I'll extend
lockdep/lockstat for this purpose if peterz doesn't beat me to it. I
won't have access to the test harness to replicate this though.

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