Re: Popcount optimization using AVX512

Nathan Bossart <nathandbossart@gmail.com>

From: Nathan Bossart <nathandbossart@gmail.com>

To: "Amonson, Paul D" <paul.d.amonson@intel.com>

Cc: Tom Lane <tgl@sss.pgh.pa.us>, David Rowley <dgrowleyml@gmail.com>, Andres Freund <andres@anarazel.de>, Alvaro Herrera <alvherre@alvh.no-ip.org>, "Shankaran, Akash" <akash.shankaran@intel.com>, Noah Misch <noah@leadboat.com>, Matthias van de Meent <boekewurm+postgres@gmail.com>, "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>

Date: 2024-03-30T03:22:09Z

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 →

Fix __attribute__((target(...))) usage.
- 41b98ddb77bf 18.0 landed
Use __attribute__((target(...))) for AVX-512 support.
- f78667bd910e 18.0 landed
Fix code for probing availability of AVX-512.
- 598e0114a3b1 17.0 landed
Optimize visibilitymap_count() with AVX-512 instructions.
- 41c51f0c68b2 17.0 landed
Optimize pg_popcount() with AVX-512 instructions.
- 792752af4eb5 17.0 landed
Inline pg_popcount() for small buffers.
- deb1486c7d36 17.0 landed
Avoid function call overhead of pg_popcount() in syslogger.c.
- 4133c1f45c54 17.0 landed
Refactor code for setting pg_popcount* function pointers.
- 6687430c98f3 17.0 landed
Inline pg_popcount{32,64} into pg_popcount().
- cc4826dd5e52 17.0 landed
Remove MSVC scripts
- 1301c80b2167 17.0 cited
Use ARMv8 CRC instructions where available.
- f044d71e331d 11.0 cited
Use Intel SSE 4.2 CRC instructions where available.
- 3dc2d62d0486 9.5.0 cited

Attachments

v17-0001-AVX512-popcount-support.patch (text/x-diff)

Here's a v17 of the patch.  This one has configure checks for everything
(i.e., CPUID, XGETBV, and the AVX512 intrinsics) as well as the relevant
runtime checks (i.e., we call CPUID to check for XGETBV and AVX512 POPCNT
availability, and we call XGETBV to ensure the ZMM registers are enabled).
I restricted the AVX512 configure checks to x86_64 since we know we won't
have TRY_POPCNT_FAST on 32-bit, and we rely on pg_popcount_fast() as our
fallback implementation in the AVX512 version.  Finally, I removed the
inline assembly in favor of using the _xgetbv() intrinsic on all systems.
It looks like that's available on gcc, clang, and msvc, although it
sometimes requires -mxsave, so that's applied to
pg_popcount_avx512_choose.o as needed.  I doubt this will lead to SIGILLs,
but it's admittedly a little shaky.

-- 
Nathan Bossart
Amazon Web Services: https://aws.amazon.com