Re: [PATCH] x86: Properly encode vmovd with 64-bit memeory

["H.J. Lu" <hjl.tools@gmail.com>]

On Mon, Jan 8, 2018 at 3:54 AM, Jan Beulich <JBeulich@suse.com> wrote:
>>>> On 08.01.18 at 12:48, <hjl.tools@gmail.com> wrote:
>> On Mon, Jan 8, 2018 at 3:35 AM, Jan Beulich <JBeulich@suse.com> wrote:
>>>>>> On 08.01.18 at 12:22, <hjl.tools@gmail.com> wrote:
>>>> On Mon, Jan 8, 2018 at 3:14 AM, H.J. Lu <hjl.tools@gmail.com> wrote:
>>>>> On Mon, Jan 8, 2018 at 12:48 AM, Jan Beulich <JBeulich@suse.com> wrote:
>>>>>>>>> On 08.01.18 at 02:18, <hongjiu.lu@intel.com> wrote:
>>>>>>> For historical reason, we allow movd/vmovd with 64-bit register and
>>>>>>> memeory operands.  But for vmovd, we failed to handle 64-bit memeory
>>>>>>> operand.  This has been gone unnoticed since AT&T syntax always treats
>>>>>>> memory operand as 32-bit memory.  This patch properly encodes vmovd
>>>>>>> with 64-bit memeory operands.
>>>>>>
>>>>>> Interesting coincidence - just over the weekend I've run into this
>>>>>> issue too. My intended solution is quite different, though: Since
>>>>>> VMOVD (other than MOVD) doesn't have a 64-bit operand variant
>>>>>> in either Intel's SDM nor AMD's PM, I'd rather remove memory
>>>>>> operand support from it:
>>>>>> - generate code was wrong so far, so anyone having used it would
>>>>>>   have run buggy code anyway,
>>>>>> - old gcc only ever uses the 64-bit variants with register operands.
>>>>>
>>>>> Works for me.  Can you submit a patch?
>>>
>>> Hopefully later this week.
>>>
>>>> If we do that, should we also remove MOVD with 64-bit memory?
>>>
>>> We can't, as even up-to-date AMD PM still specifies this name
>>> instead of MOVQ.
>>
>> I consider it is a defect in AMD manual.
>>
>>>> Otherwise, -msse2avx won't work on MOVD with 64-bit memory.
>>>
>>> Hmm, good point - perhaps the SSE2AVX pattern then needs the
>>> change that you've been doing, while the plain one could have its
>>> memory alternative removed?
>>
>> If we can't remove MOVD with 64-bit memory, I will go with my patch.
>
> Fine with me for the AVX variant, but please consider dropping
> the AVX512 one.

Did you mean dropping AVX512 vmovd with 64-bit memory operand?

-- 
H.J.