Re: [RFC PATCH 2/3] sched: add yield_to function

[Rik van Riel]

On 01/02/2011 06:43 AM, Hillf Danton wrote:

> The following work is based on what Rik posted, with a few changes.

I threw out my patch before Christmas, replacing it with
Mike Galbraith's work.

Today I finally discovered why I had not been getting
very useful results testing the PLE code on the test
system.

On this particular CPU model (Intel(R) Xeon(R) CPU X5670),
the value of ple_gap needs to be at least 48 !

A ple_gap value of 41 is simply too low to trigger reliably
with Avi's kvm-unit-tests PLE test and also only triggers
a few times a second at most (not too useful) with the other
tests I had been running...

--
All rights reversed
--
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/