Messages in this thread | | | Date | Mon, 04 Apr 2011 11:20:53 +0200 | From | Alexander Holler <> | Subject | Re: Linux 2.6.33.8 |
| |
Hello,
Am 02.04.2011 13:27, schrieb Carsten Emde:
>> I'm using a >> model name : Intel(R) Core(TM) i7 CPU 860 @ 2.80GHz > This probably is not model 44; is it 26, 30 or 31?
Model 30, stepping 5. HT is enabled.
>> and can confirm some problem with power down, at least in 2.6.38.2. >> Sometimes I only see the powerdown message but the machine will stay on >> (and the fan starts getting louder). >> [..] >> Maybe it's a totally unrelated problem, but I thought it might be good >> to mention that I'm having a problem with powerdown in 2.6.38 too. > Yes, thank you, > > Would you mind to completely disable processor frequency scaling and > check whether the powerdown problem goes away on your system as well? I > do not know which distro you are using, but on some systems "chkconfig > cpuspeed off; reboot" would do the job. When the system then reboots, > "halt" should correctly power off the system.
I'm currently using F13, always with the latest stable vanilla kernel from kernel.org. No cpufreq-daemons or similiar are involved, just the governor "ondemand".
In regard to more detailed testing I can't say when I will find the time to do that. I'm using this machine regulary and the problem occurs only seldom (maybe 1 out of 5 shutdowns). So git bisect or similiar generic tests aren't very practical.
I will try to add some print statements, I assume native_machine_power_off() and stuff afterwards are the candidates where to do that.
Regards,
Alexander
| |