Re: [PATCH 1/2 v2] x86, mce, severity: extend the the mce_severity

[Borislav Petkov]

On Thu, Nov 06, 2014 at 09:24:01PM +0000, Luck, Tony wrote:
> Architecturally it is true that MCIP will be set when machine check is
> signaled. But, sometimes there are bugs. BIOS has a hook to get an SMI
> to see the event before the OS sees the machine check - which gives
> lots of scope for things to not happen by the book. If MCIP isn't set
> correctly, I'd like to get on and panic quickly - because all sorts
> of bad things will happen if a nested machine check happens and isn't
> caught because MCIP wasn't set in the first machine check.

I won't even ask about your actual experience with stuff not setting
MCIP :-)

Anyway, noted. So for the mce_severity rewrite we should make sure to
note from which context we're being called as the more reliable method,
instead of looking at MCGSTATUS.MCIP.

Thanks.

-- 
Regards/Gruss,
    Boris.

Sent from a fat crate under my desk. Formatting is fine.
--
--
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/