Re: APIC error interrupt routine.

From: Chris Chiappa (chris@chiappa.net)
Date: Sun Sep 10 2000 - 12:18:05 EST


On Sat, Aug 19, 2000 at 12:34:17AM +0200, Ingo Molnar wrote:
> > You're one of the lucky ones then. I have a non-overclocked BP6, CPUs
> > at default voltage and cooling fans up the wazoo (I even have a 486
> > fan with extra grease strapped to the BX) and if I run CPU intensive
> > apps, I still generate a bunch of APIC errors. [...]
> could you try to lower the CPU voltage by 0.05V or 0.10V (below the
> default), does it make a difference? If not, please try to increase CPU
> volate by the same amount - any difference? What CPU speed, system bus
> speed and PCI/AGP bus speed do you have - you can see it in the Soft-CPU
> menu. The ABP6 system here has two 466MHz Celerons, 66MHz(1/2) system
> bus/PCI speed, x4.5 multiplier, AGPCLK 2/3 and 2.05v core voltage. And for
> the duration of the test, could you keep the cover(s) open, so that air
> circulates better.
Finally got all this done (been a bit busy moving etc). I was able to
generate APIC errors while running the Seti@Home client at 1.90, 1.95, 2.00,
2.05 and 2.10 Volts. Cover was off for all tests. It seemed like APIC
errors were less common at the higher voltages, but I didn't bother to make
any records to compare etc.

Setup is:
400MHz CPUs, 66MHz bus rate, Default AGPCLK (I assume 2/3), QQ Beta BIOS
Power supply is a PC Power & Cooling turbo cool 300, I have another fan in the
front of the box and a 486 fan over the BX chipset.
Other peripherals:
1 Toshiba SCSI DVD drive
3 x IBM UW SCSI hard drives
Matrox G400 AGP
Diamond Dual Fireport 40 (Symbios 876 chip)
Creative Ensoniq Soundscape
3C905A Ethernet
ISA modem

It's not really a big deal to me since I'm getting rid of the BP6 in 2 or 3
weeks, but it's definitely my last ABIT product.

-- 

..ooOO chris@chiappa.net | My opinions are my own OOoo.. ..ooOO cchiappa@us.oracle.com | and certainly not those OOoo.. ..ooOO http://www.snurgle.org/~griffon/ | of my employer OOoo.. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Fri Sep 15 2000 - 21:00:13 EST