On Wed, 3 Nov 2010, Thomas Renninger wrote: > On Wednesday 03 November 2010 00:47:41 Auke Kok wrote: >> On 11/02/10 07:40, Vitus Jensen wrote: >>> >>> Since some kernel versions I notice very high wakeup rates during idle ... >>> while the reasons for the wakeup stay at reasonable low numbers. In the >>> 'powertop -d' output below it's 27463 wakeups/sec but the highest number >>> for any reason is 48. How do I find out what causes the wakeups? > Sounds like C-states are not entered. > or does: > cat /proc/interrupts > top > show anything suspicious? > This guy has a similar problem, but it may be another root cause: > https://bugzilla.kernel.org/show_bug.cgi?id=12788 > > ~40,000 wakeups/sec after suspend/resume with Tickless kernel; idle=nomwait > or max_cstate=1 helps ...
Yes, this looks very similar. Except that it's always there and my system is only a uni-processor. But removing CONFIG_NOHZ fixes the issue as reported in that bug. ~150 wakeups, 12 W :-) > You may want to open a new bug on https://bugzilla.kernel.org and assign > it to the ACPI component. I will. And I will post a link to the other bug in id=12788. Thanks, Vitus -- Vitus Jensen, Hannover, Germany, Universe (current) pgp public key available from keyservers _______________________________________________ Power mailing list [email protected] http://www.bughost.org/mailman/listinfo/power
