Re: [2.6.12] x86-64 IO-APIC + timer doesn't work

From: Andi Kleen
Date: Sat Jun 18 2005 - 20:35:27 EST


> Despite the fact that this wasn't documented in the BIOS update, an update for
> my board (MS-7030 Neo Platinum by MSI) supposedly fixing "Fan Function"
> actually corrects the IO-APIC and NMI bugs. I now get the following in dmesg
> instead:
>
> Calibrating delay loop... 3973.12 BogoMIPS (lpj=1986560)
> Mount-cache hash table entries: 256
> CPU: L1 I Cache: 64K (64 bytes/line), D cache 64K (64 bytes/line)
> CPU: L2 Cache: 512K (64 bytes/line)
> CPU: AMD Athlon(tm) 64 Processor 3000+ stepping 00
> Using local APIC timer interrupts.
> Detected 12.561 MHz APIC timer.
> testing NMI watchdog ... OK.
>
> So I'm a happy man. Whether this is at all related to the problems I was
> having before, I don't really know. If the problem doesn't reoccur, I could
> very well have wasted your time.

Hmm - I suspect I know what's happening. The older BIOS probably had some long
running SMM code that breaks the BogoMips computation occasionally
and that breaks the timer check later which relies on usable BogoMips.

There was a patch for that from Venkatesh (because it showed on some
other machines too), but it didn't seem to have made it into 2.6.12

Venkatesh, can you push your calibrate_delay patch please? ?


-Andi
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/