cpufreq Archive on lore.kernel.org
 help / color / mirror / Atom feed
* Temporary freezes in AMD Sempron 2800+
@ 2005-06-06 14:49 Ramkumar R
       [not found] ` <200506061659.19854.mpytasz@pai.net.pl>
  0 siblings, 1 reply; 7+ messages in thread
From: Ramkumar R @ 2005-06-06 14:49 UTC (permalink / raw)
  To: cpufreq

Hi!

I have a laptop with an AMD Sempron 2800+ (Acer Aspire 3002NLC). The
machine seems to get unresponsive every time the frequency changes
(the music stops for a second, say)...I tested this by using the
ondemand kernel governor. I use linux-2.6.12-rc4-ck1 with the inotify
patch. (With two more modules, fuse and cdfs, but I guess that
wouldn't matter much). Watching the current frequency file in /sys
shows that this sudden freeze occurs whenever the value changes.

Thanx,
Ramkumar.

-- 
The day Microsoft makes something that doesn't suck
is probably the day they start making vacuum cleaners.
                                               -Ernst Jan Plugge

^ permalink raw reply	[flat|nested] 7+ messages in thread
* RE: Temporary freezes in AMD Sempron 2800+
@ 2005-06-06 19:54 Langsdorf, Mark
  0 siblings, 0 replies; 7+ messages in thread
From: Langsdorf, Mark @ 2005-06-06 19:54 UTC (permalink / raw)
  To: Ramkumar R, Michal Pytasz, cpufreq

> yeah, one more question remains...does sempron have only 2 
> frequency levels...that's what dmesg and the sys file seem to 
> indicate....though i've read that amd processors have quite a 
> few speed levels (is that only for the athlon series ?)

The powernow driver reads the data provided from the BIOS.
If the BIOS provides an incorrect number of pstates, the
driver has no way of knowing that there are other states.

How many pstates you have is determined by the frequency of
the processor.  A Sempron 2800+ is a 1.6 GHz processor and
should only have 2 pstates - high and low.  It looks like
your system is set up correctly.

-Mark Langsdorf
AMD, Inc.

^ permalink raw reply	[flat|nested] 7+ messages in thread
* RE: Temporary freezes in AMD Sempron 2800+
@ 2005-06-06 20:45 Langsdorf, Mark
  0 siblings, 0 replies; 7+ messages in thread
From: Langsdorf, Mark @ 2005-06-06 20:45 UTC (permalink / raw)
  To: Ramkumar R, Michal Pytasz, cpufreq

> well...i guess i was a bit hasty...the p-states driver was 
> not the problem...rather, a more interesting pattern has come 
> up...it seems that the freeze occurs whenever i am on a X 
> display (i.e. it isn't even a problem when i have X running 
> and i am in one of the tty's)...in fact, to confirm my 
> doubts, I set up my .Xsession to simply loop, sleeping for a 
> second, and toggling between the two frequencies. It freezes 
> when i am on the X display, and the problem disappears the 
> moment i do a alt+f1...this is really strange...

Sounds like a potential motherboard issue.  Some low-end
motherboards don't have large enough PCI data buffers to
store all the way through a pstate transition.  There's 
not a lot I can do about it.

-Mark Langsdorf
AMD, Inc.

^ permalink raw reply	[flat|nested] 7+ messages in thread

end of thread, other threads:[~2005-06-06 20:45 UTC | newest]

Thread overview: 7+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-06-06 14:49 Temporary freezes in AMD Sempron 2800+ Ramkumar R
     [not found] ` <200506061659.19854.mpytasz@pai.net.pl>
     [not found]   ` <9a9bfc1705060609302b5d803f@mail.gmail.com>
2005-06-06 17:06     ` Michał Pytasz
     [not found]       ` <9a9bfc170506061021620e487a@mail.gmail.com>
     [not found]         ` <200506061956.16644.mpytasz@posejdon.wpk.p.lodz.pl>
2005-06-06 18:11           ` Ramkumar R
2005-06-06 19:23             ` Ramkumar R
2005-06-06 20:38               ` Ramkumar R
  -- strict thread matches above, loose matches on Subject: below --
2005-06-06 19:54 Langsdorf, Mark
2005-06-06 20:45 Langsdorf, Mark

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox