Re: [Xen-devel] [PATCH V7 1/3] x86/xsaves: fix overwriting between non-lazy/lazy xsaves

["Jan Beulich" <JBeulich@xxxxxxxx>]

>>> On 29.04.16 at 03:36, <shuai.ruan@xxxxxxxxxxxxxxx> wrote:
> On Mon, Apr 25, 2016 at 12:51:44AM -0600, Jan Beulich wrote:
>> >>> On 31.03.16 at 10:57, <shuai.ruan@xxxxxxxxxxxxxxx> wrote:
>> > +#define XRSTOR(pfx) \
>> > +        if ( v->arch.xcr0_accum & XSTATE_XSAVES_ONLY ) \
>> > +        { \
>> > +            if ( unlikely(!(ptr->xsave_hdr.xcomp_bv & \
>> > +                            XSTATE_COMPACTION_ENABLED)) ) \
>> > +                ptr->xsave_hdr.xcomp_bv |= ptr->xsave_hdr.xstate_bv | \
>> > +                                           XSTATE_COMPACTION_ENABLED; \
>> 
>> From v5 to v6 this changed from just = to |=, without any
>> explanation, and without me really noticing - why? Weren't
>> the other changes done specifically to guarantee xcomp_bv
>> to be zero up to this point? In which case I'd prefer to make
>> this obvious/explicit, by using = and perhaps an ASSERT()
>> here. (I have a patch ready, but I'd like to understand if
>> there was a reason for this change that I don't see.)
> 
> Using "=" is better. xcomp_bv can be guarantee to be zero to this
> point.

Thanks. I already have a patch, which I'll submit after 4.7 got
branched off.

Jan


_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel