Re: [PATCH 1/2] x86_64: Fold pda into per cpu area v3

From: Mike Travis
Date: Wed Feb 20 2008 - 13:57:45 EST


Ingo Molnar wrote:
> * Mike Travis <travis@xxxxxxx> wrote:
>
>> * Declare the pda as a per cpu variable. This will move the pda area
>> to an address accessible by the x86_64 per cpu macros.
>> Subtraction of __per_cpu_start will make the offset based from the
>> beginning of the per cpu area. Since %gs is pointing to the pda,
>> it will then also point to the per cpu variables and can be
>> accessed thusly:
>>
>> %gs:[&per_cpu_xxxx - __per_cpu_start]
>
> randconfig QA on x86.git found a crash on x86.git#testing with
> nmi_watchdog=2 (config attached) - and i bisected it down to this patch.
>
> config and crashlog attached. You can pick up x86.git#testing via:
>
> http://people.redhat.com/mingo/x86.git/README
>
> (since i had to hand-merge the patch when integrating it, i've attached
> the merged version below.)
>
> Ingo
>

I must need some different test machines as my AMD box does not fail with
either yours or Thomas's configs, and the Intel box complains about the
PCI-e e1000 driver and dies. I'll see about configuring a new box.

Did you try Eric's patch to see if that fixed the failure?

Thanks,
Mike

--
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/