On Tue, Nov 19, 2013 at 08:13:35AM +0100, Mike Galbraith wrote: > Hi Peter, > > 3.12 escaped without this regression being fixed, nor does the fix have > a stable tag. Suspecting it may have just fallen through a crack...
Indeed, looks like a stable candidate, thanks for keeping track. > core2 boxen still have post mwait_idle() removal regressions (rainy day > job, started digging, rain went missing), but more modern westmere box > likes this fix quite a lot. What idle function does the core2 use? Something like: perf record -a sleep 1; perf report Should show you the idle function on top if ran on a mostly idle system. I would be expecting mwait_idle_with_hints() for core2. And I think I've covered all paths to it with the right magic, but how knows, these acpi people never reacted to any of these patches. > From: Peter Zijlstra <[email protected]> > Date: Wed, 11 Sep 2013 12:43:13 +0200 > Subject: sched, idle: Fix the idle polling state logic > > Commit ea8117478918a4734586d35ff530721b682425be upstream > > Mike reported that commit 7d1a9417 ("x86: Use generic idle loop") > regressed several workloads and caused excessive reschedule > interrupts. > > The patch in question failed to notice that the x86 code had an > inverted sense of the polling state versus the new generic code (x86: > default polling, generic: default !polling). > > Fix the two prominent x86 mwait based idle drivers and introduce a few > new generic polling helpers (fixing the wrong smp_mb__after_clear_bit > usage). > > (Mike: dropped need_resched() -> tif_need_resched() for @stable) > > Reported-by: Mike Galbraith <[email protected]> > Signed-off-by: Peter Zijlstra <[email protected]> > Cc: [email protected] > Cc: [email protected] > Cc: [email protected] # 3.10+ > Link: http://lkml.kernel.org/n/[email protected] > Signed-off-by: Ingo Molnar <[email protected]> -- To unsubscribe from this list: send the line "unsubscribe stable" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
