https://bugzilla.novell.com/show_bug.cgi?id=325917#c16
Thomas Renninger changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
--- Comment #16 from Thomas Renninger 2007-09-20 10:57:57 MST ---
I just saw this is a:
Intel(R) Pentium(R) M processor 2.00GHz stepping 08
This one probably has a broken lapic timer in C2/C3 and broadcast timers should
be used here.
I sent a patch to not only hit AMDs but also Pentiums Ms and use broadcast
timers instead some time ago. Took me hours to find that out.
We should have an affected machine, it was a Sony Vaio.
There has been some discussion around this and it seems it fall out with the
clocksource patches. Not sure yet, but can you try:
Is it fast again with processor.max_cstate=1 parameter?
Maybe also nolapic_timer or nolapic helps?
The machine should also become faster again by keeping a key pressed.
This should already be able to verify at boot stage (after the processor module
got loaded at disk initialisation time it should get slow?), try to keep a key
pressed, if this helps you can use this as a verification for the other tests
above. Only a guess yet...
Hmm, IIRC 10.2 also had the problem and I added a patch to 10.2 after GM?
--
Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.