Re: [PATCH] x86/mm: Don't try to change poison pages to uncacheable in a guest

[Luck, Tony]

But the guest isnât likely to do the right thing with a page fault. The guest just accessed a page that it knows is poisoned (VMM just told it once that it was poisoned). There is no reason that the VMM should let the guest actually touch the poison a second time. But if the guest does, then the guest should get the expected response.  I.e. another machine check.

Sent from my iPhone

> On May 16, 2020, at 08:03, Borislav Petkov <bp@xxxxxxxxx> wrote:
> 
> ïOn Sat, May 16, 2020 at 02:47:42PM +0000, Luck, Tony wrote:
>> There is only one actual machine check. But the VMM simulates a second
>> machine check to the guest when the guest tries to access the poisoned
>> page.
> 
> If the VMM unmaps the bad page, why doesn't the guest get a #PF instead
> injected by the VMM instead of latter injecting a second #MCE?
> 
> If the guest tries to access an unmapped page, it should get a #PF, I'd
> expect.
> 
> -- 
> Regards/Gruss,
>    Boris.
> 
> https://people.kernel.org/tglx/notes-about-netiquette