Fuqian Huang <huangfq.daxian@xxxxxxxxx> writes:
> Vitaly Kuznetsov <vkuznets@xxxxxxxxxx> 於 2019年9月12日週四 下午4:51寫道：
>>
>> Fuqian Huang <huangfq.daxian@xxxxxxxxx> writes:
>>
>> > Emulation of VMPTRST can incorrectly inject a page fault
>> > when passed an operand that points to an MMIO address.
>> > The page fault will use uninitialized kernel stack memory
>> > as the CR2 and error code.
>> >
>> > The right behavior would be to abort the VM with a KVM_EXIT_INTERNAL_ERROR
>> > exit to userspace;
>>
>> Hm, why so? KVM_EXIT_INTERNAL_ERROR is basically an error in KVM, this
>> is not a proper reaction to a userspace-induced condition (or ever).
>>
>> I also looked at VMPTRST's description in Intel's manual and I can't
>> find and explicit limitation like "this must be normal memory". We're
>> just supposed to inject #PF "If a page fault occurs in accessing the
>> memory destination operand."
>>
>> In case it seems to be too cumbersome to handle VMPTRST to MMIO and we
>> think that nobody should be doing that I'd rather prefer injecting #GP.
>>
>> Please tell me what I'm missing :-)
>
> I found it during the code review, and it looks like the problem the
> commit 353c0956a618 ("KVM: x86: work around leak of uninitialized
> stack contents (CVE-2019-7222)")
> mentions. So I fixed it in a similar way.
>
Oh, yes, I'm not against the fix at all, I was just wondering about why
you think we need to kill the guest in this case.
--
Vitaly
