Re: "APIC error on CPUx" - what does this mean?

Petr Vandrovec (VANDROVE@vc.cvut.cz)
Mon, 7 Jan 2002 14:16:28 +0100


On 8 Jan 02 at 2:08, Chris Wedgwood wrote:
> On Mon, Jan 07, 2002 at 01:29:42PM +0100, Petr Vandrovec wrote:
>
> They are spurious IRQ 7, just message is printed only once during
> kernel lifetime... I have about three spurious IRQ 7 per each 1000
> interrupts delivered to CPU. It is on A7V (Via KT133).
>
> Any idea _why_ these occur though? It seems some mainboards produce a
> plethora of these whilst others never produce these...

Nope. Probably when CPU is in local APIC mode, it acknowledges interrupts
to chipset with different timming, and from time to time CPU still
sees IRQ pending, so it asks for vector, but as chipset has no
interrupt pending, it answers with IRQ7. I did no analysis to find
whether IRQ7 happens directly when we send confirmation to 8259,
or whether it happens due to some noise on IRQ line.

AFAIK it happens only on VIA based boards, and only if (AMD) CPU is using
APIC.
Best regards,
Petr Vandrovec
vandrove@vc.cvut.cz

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