Re: [PATCH 00/20] x86: drop bogus IgnoreSize attributes

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

On Wed, Sep 5, 2018 at 11:38 PM, Jan Beulich <JBeulich@suse.com> wrote:
> Their extremely widespread use today is, afaict, a result of blindly
> having retained/copied them from the original MMX insn templates,
> where they are indeed necessary (without a distinction between
> qword and mmword operands). This series gets rid of most of them;
> a few instances where the general patterns I've looked for don't
> apply may be left - I expect to clean these up over time, as they
> turn up.
>
> Redundant *word operand attributes also get squashed in a number
> of cases, when lines are being touched anyway.
>
> The splitting into multiple patches is to limit individual patch size
> only. Also a few of the patches will only apply cleanly on top of
> relevant parts of the other series sent yesterday.
>
> 01: drop bogus IgnoreSize from SSE insns
> 02: drop bogus IgnoreSize from SSE2 insns
> 03: drop bogus IgnoreSize from SSE3 insns
> 04: drop bogus IgnoreSize from SSSE3 insns
> 05: drop bogus IgnoreSize from SSE4.1 insns
> 06: drop bogus IgnoreSize from SSE4.2 insns
> 07: drop bogus IgnoreSize from AES/VAES insns
> 08: drop bogus IgnoreSize from PCLMUL/VPCLMUL insns
> 09: drop bogus IgnoreSize from GFNI insns
> 10: drop bogus IgnoreSize from AVX insns
> 11: drop bogus IgnoreSize from AVX2 insns
> 12: drop bogus IgnoreSize from XOP and SSE4a insns
> 13: drop bogus IgnoreSize from SHA insns
> 14: drop bogus IgnoreSize from AVX512F insns
> 15: drop bogus IgnoreSize from AVX512ER insns
> 16: drop bogus IgnoreSize from AVX512VL insns
> 17: drop bogus IgnoreSize from AVX512BW insns
> 18: drop bogus IgnoreSize from AVX512DQ insns
> 19: drop bogus IgnoreSize from AVX512_4* insns
> 20: drop bogus IgnoreSize from a few further insns
>

OK.

Thanks.

-- 
H.J.