Messages in this thread | | | Date | Sat, 7 Dec 2013 11:01:30 -0500 | Subject | Re: 50 Watt idle power regression bisected to Linux-3.10 | From | Len Brown <> |
| |
On Sat, Dec 7, 2013 at 3:39 AM, Mike Galbraith <bitbucket@online.de> wrote: > On Sat, 2013-12-07 at 03:00 -0500, Len Brown wrote: > >> No, Linux-3.13-rc3 does not fix this issue, even though it contains >> the following patch, claiming to address an issue with the commit above: >> >> commit ea8117478918a4734586d35ff530721b682425be >> Author: Peter Zijlstra <peterz@infradead.org> >> Date: Wed Sep 11 12:43:13 2013 +0200 >> >> sched, idle: Fix the idle polling state logic >> >> Mike reported that commit 7d1a9417 ("x86: Use generic idle loop") >> regressed several workloads and caused excessive reschedule >> interrupts. > > It fixes that, except for my Q6600 box. Too bad mwait_idle() went away, > beloved old box doesn't play hints game, so it continues to flog itself.
Hi Mike, What C-states are available on your Q6600 box? Is it running HALT only? (if so, I would expect C1 MWAIT be about the same, if it still existed)
But I would expect the box to have deeper C-states than just C1, so...
Can you show me the output of dmesg | grep idle grep . /sys/devices/system/cpu/cpu0/cpuidle/*/*
Please elaborate on exactly what symptom you see. Do you have a power meter?
thanks, Len Brown, Intel Open Source Technology Center
| |