Re: [PATCH v2 02/11] perf/x86: Improve HT workaround GP counter constraint

[Andi Kleen]

On Wed, May 27, 2015 at 09:48:09AM +0200, Peter Zijlstra wrote:
> On Tue, May 26, 2015 at 04:33:42PM -0700, Andi Kleen wrote:
> > Peter Zijlstra <peterz@xxxxxxxxxxxxx> writes:
> > > +		 */
> > > +		if (is_ht_workaround_enabled() && !cpuc->is_fake &&
> > 
> > Could this function also check if at least one leaking event is
> > scheduled somewhere? (e.g. from a global count)
> 
> You truncated one line too many:
> 
> +               if (is_ht_workaround_enabled() && !cpuc->is_fake &&
> +                   READ_ONCE(cpuc->excl_cntrs->exclusive_present))
> +                       gpmax /= 2;
> 
> Guess what that READ_ONCE() does? It checks if there's one such leaky
> event on the current core.

Great. Thanks.

-Andi


-- 
ak@xxxxxxxxxxxxxxx -- Speaking for myself only.
--
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/