[PATCH 3/5] x86: support AVX10.1/512

[Jan Beulich]

On 30.08.2023 17:28, H.J. Lu wrote:
> On Wed, Aug 30, 2023 at 12:48 AM Jan Beulich <jbeulich@suse.com> wrote:
>>
>> On 30.08.2023 03:10, Jiang, Haochen wrote:
>>>> I don't think we need to make AVX10 to imply AES and VPCLMULQDQ in
>>>> assembler.  The AES and PCLMULQDQ changes aren't needed.
>>>>
>>>
>>> AVX10 should not imply them. However, I suppose we are talking about
>>> the relation between AES/VAES and PCLMULQDQ/VPCLMULQDQ.
>>
>> Correct. Some change needs to be done; the question is whether as
>> posted, or whether with a prereq patch adding those two connections
>> (with, as said before, me intending to go the latter route unless I
>> hear otherwise).
> 
> I agree.  VAES should imply AES and VPCLMULQDQ should imply
> PCLMULQDQ.

Actually there's something similar with AVX10 itself: AVX512F includes
equivalents right away of what comes under separate extensions for AVX:
F16C and FMA. AVX10, otoh, is presently specified to only guarantee
AVX and AVX2. Does that mean that VEX-encoded vfm{add,sub}* and ps<->ph
conversion insns aren't guaranteed to also be available? Doesn't seem
logical to me, so I'm inclined to make FMA and F16C prereqs of AVX10.1
as well (or alternatively of AVX512F, but I think this would have
undesirable effects). AVX2 isn't an explicit prereq only because it
already is one of AVX512F.

Jan