Re: [Bug #14950] tbench regression with 2.6.33-rc1

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Monday 22 March 2010, Mike Galbraith wrote:
> On Sun, 2010-03-21 at 21:30 +0100, Rafael J. Wysocki wrote:
> > This message has been generated automatically as a part of a report
> > of regressions introduced between 2.6.32 and 2.6.33.
> > 
> > The following bug entry is on the current list of known regressions
> > introduced between 2.6.32 and 2.6.33.  Please verify if it still should
> > be listed and let the tracking team know (either way).
> > 
> > 
> > Bug-Entry	: http://bugzilla.kernel.org/show_bug.cgi?id=14950
> > Subject		: tbench regression with 2.6.33-rc1
> > Submitter	: Lin Ming <ming.m.lin@intel.com>
> > Date		: 2009-12-25 11:11 (87 days old)
> > References	: http://marc.info/?l=linux-kernel&m=126174044213172&w=4
> 
> This needs retest in tip.  I submitted some patchlets to shave a few
> cycles, with which my box (fwtw) shows zero regression 2.6.31 ->
> tip.today, whereas there was nothing but regression of up to ~6% in
> between, magnitude seemingly depending on phase-of-moon.  Actually,
> patched tip at submission time was a fraction above 31 throughput.
> Retesting today, it's a fraction of a percent below again (tbench is
> annoyingly jittery).
> 
> Hopefully, this tbench regression is on it's way to retirement.

Thanks for the update.

Rafael