Re: [PATCH] x86: Execute WP test after interrupts are enabled

[Jeremy Fitzhardinge]

On 08/25/09 12:38, Pekka Enberg wrote:
> On Tue, 2009-08-25 at 12:34 -0700, Jeremy Fitzhardinge wrote:
>   
>> I don't think this is the real fix, and it seems a bit ugly.  I'm OK
>> with it as a workaround, but I think it will end up getting reverted if
>> applied.
>>     
> Yeah, I'm fine with that. It's just that we're really late in the
> release cycle now so I'd like to keep the fix as simple as possible.

Yeah, that's fine, but I think something like this would be simpler. 
Arnd, could you test it please?

From: Jeremy Fitzhardinge <jeremy.fitzhardinge@xxxxxxxxxx>
Date: Tue, 25 Aug 2009 12:47:17 -0700
Subject: [PATCH] xen: suppress WP test

Xen always runs on CPUs which properly support WP enforcement in
privileged mode, so there's no need to test for it.

This also works around a crash reported by Arnd Hannemann, though
I think its just a band-aid for that case.

Reported-by: Arnd Hannemann <hannemann@xxxxxxxxxxxxxxxxxxx>
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@xxxxxxxxxx>
Cc: Pekka Enberg <penberg@xxxxxxxxxxxxxx>

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index 16d0d70..e5b903b 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1069,6 +1069,7 @@ asmlinkage void __init xen_start_kernel(void)
 	/* set up basic CPUID stuff */
 	cpu_detect(&new_cpu_data);
 	new_cpu_data.hard_math = 1;
+	new_cpu_data.wp_works_ok = 1;
 	new_cpu_data.x86_capability[0] = cpuid_edx(1);
 #endif



>  Do you think the real bug lies in Xen or...?
>   

I'm unsure.  If handling a trap ends up enabling interrupts for some
reason, then things would break horribly all over the place, so it would
have to be something a bit more subtle.  I'm looking into it.

    J
--
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/