Re: Help understanding slow down

[Ingo Molnar]

* Andrew Morton <akpm@xxxxxxxx> wrote:

> Ingo Molnar <mingo@xxxxxxx> wrote:
> >
> >  with the patch below we will print a big fat warning. (I did not want to
> >  deny idle=poll altogether - future HT implementations might work fine
> >  with polling idle.)
> 
> idle=poll is handy when profiling the kernel with oprofile
> clock-unhalted events.  Because if you use the normal halt-based idle
> loop no profile "ticks" are accounted to idle time at all and the
> results are really hard to understand.

it makes it a bit more plausible, but kernel profiling based on ticks in
a HT environment is still quite unreliable, even with idle=poll. The HT
cores will yield to each other on various occasions - like spinlock
loops. This disproportionatly increases the hits of various looping
functions, creating false impressions of lock contention where there's
only little contention. Plus idle=poll is a constant ~20% performance
drain on the non-idle HT core, further distorting the profile. HT makes
profiling really hard, no matter what.

but ... we agree on the warning printk, right?

	Ingo
-
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/