Re: [Xen-devel] [PATCH 4/9] x86/HVM: move NOFLUSH handling out of hvm_set_cr3()

[Jan Beulich <jbeulich@xxxxxxxx>]

On 12.09.2019 17:42, Roger Pau Monné  wrote:
> On Thu, Sep 12, 2019 at 04:47:17PM +0200, Jan Beulich wrote:
>> On 12.09.2019 16:44, Roger Pau Monné  wrote:
>>> On a different question, AFAICT hvm_set_cr3 should never be called
>>> with X86_CR3_NOFLUSH set? If so, do you think it would make sense to
>>> add an assert to that regard?
>>
>> I've been debating this with myself, and decided against for now.
>> We don't know what meaning the bit may gain eventually in the
>> actual register.
> 
> I'm slightly lost here, the noflush bit is actually defined in the
> Intel SDM for cr3, and hence won't gain any other meaning?

The noflush bit is a operation one, i.e. taking effect on the
MOV-to-CR3, without getting written to the underlying register.
Therefore there may well appear a meaning for the actual
register bit, but I agree it doesn't seem very likely for such
an overload to get put in place.

Jan

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel