Re: Popcount optimization using AVX512

On Tue, Jul 30, 2024 at 04:32:07PM -0500, Nathan Bossart wrote:
> On Tue, Jul 30, 2024 at 02:07:01PM -0700, Andres Freund wrote:
>> Afaict we could just check for predefined preprocessor macros:
>>
>> echo|time gcc -c -mxsave -mavx512vpopcntdq -mavx512bw -xc -dM -E - -o -|grep -E '__XSAVE__|__AVX512BW__|__AVX512VPOPCNTDQ__'
>> #define __AVX512BW__ 1
>> #define __AVX512VPOPCNTDQ__ 1
>> #define __XSAVE__ 1
>> 0.00user 0.00system 0:00.00elapsed 100%CPU (0avgtext+0avgdata 13292maxresident)k
>>
>> echo|time gcc -c -march=nehalem -xc -dM -E - -o -|grep -E '__XSAVE__|__AVX512BW__|__AVX512VPOPCNTDQ__'
>> 0.00user 0.00system 0:00.00elapsed 100%CPU (0avgtext+0avgdata 10972maxresident)k
>
> Seems promising. I can't think of a reason that wouldn't work.
>
>> Now, a reasonable counter-argument would be that only some of these macros are
>> defined for msvc ([1]). However, as it turns out, the test is broken
>> today, as msvc doesn't error out when using an intrinsic that's not
>> "available" by the target architecture, it seems to assume that the caller did
>> a cpuid check ahead of time.

Hm. Upon further inspection, I see that MSVC appears to be missing
__XSAVE__ and __AVX512VPOPCNTDQ__, which is unfortunate. Still, I think
the worst case scenario is that the CPUID check fails and we don't use
AVX-512 instructions. AFAICT we aren't adding new function pointers in any
builds that don't already have them, just compiling some extra unused code.

--
nathan