Horton

Re: Improve CRC32C performance on SSE4.2

Nathan Bossart <nathandbossart@gmail.com>

View thread
From: Nathan Bossart <nathandbossart@gmail.com>
To: "Devulapalli, Raghuveer" <raghuveer.devulapalli@intel.com>
Cc: John Naylor <johncnaylorls@gmail.com>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>, "Shankaran, Akash" <akash.shankaran@intel.com>
Date: 2025-02-12T22:19:37Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Include _mm512_zextsi128_si512() in AVX-512 configure probes.

  2. Properly fix AVX-512 CRC calculation bug

  3. Workaround code generation bug in clang

  4. Compute CRC32C using AVX-512 instructions where available

  5. Inline CRC computation for small fixed-length input on x86

  6. Be more paranoid in configure's checks for CRC and POPCNT intrinsics.

On Wed, Feb 12, 2025 at 10:12:20PM +0000, Devulapalli, Raghuveer wrote:
>> Well, I suspect the AVX-512 version will pretty much always need the runtime
>> check given that its not available on a lot of newer hardware and requires a
>> bunch of extra runtime checks (see pg_popcount_avx512.c).  But it might be
>> worth doing for PCLMUL.  Otherwise, I think we'd have to leave out the PCLMUL
>> optimizations if built with -msse4.2 -mpclmul because we don't want to regress
>> existing -msse4.2 users with a runtime check.
> 
> Sounds good to me. Although, users building with just -msse4.2 will now encounter an 
> an additional pclmul runtime check. That would be a regression unless they update to 
> building with both -msse4.2 and -mpclmul. 

My thinking was that building with just -msse4.2 would cause the existing
SSE 4.2 implementation to be used (without the function pointer).  That's
admittedly a bit goofy because they'd miss out on the PCLMUL optimization,
but things at least don't get any worse for them.

-- 
nathan