Horton

Re: Popcount optimization using AVX512

Matthias van de Meent <boekewurm+postgres@gmail.com>

View thread
From: Matthias van de Meent <boekewurm+postgres@gmail.com>
To: "Amonson, Paul D" <paul.d.amonson@intel.com>
Cc: "pgsql-hackers@lists.postgresql.org" <pgsql-hackers@lists.postgresql.org>, "Shankaran, Akash" <akash.shankaran@intel.com>
Date: 2023-11-03T11:16:05Z
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. Fix __attribute__((target(...))) usage.

  2. Use __attribute__((target(...))) for AVX-512 support.

  3. Fix code for probing availability of AVX-512.

  4. Optimize visibilitymap_count() with AVX-512 instructions.

  5. Optimize pg_popcount() with AVX-512 instructions.

  6. Inline pg_popcount() for small buffers.

  7. Avoid function call overhead of pg_popcount() in syslogger.c.

  8. Refactor code for setting pg_popcount* function pointers.

  9. Inline pg_popcount{32,64} into pg_popcount().

  10. Remove MSVC scripts

  11. Use ARMv8 CRC instructions where available.

  12. Use Intel SSE 4.2 CRC instructions where available.

On Thu, 2 Nov 2023 at 15:22, Amonson, Paul D <paul.d.amonson@intel.com> wrote:
>
> This proposal showcases the speed-up provided to popcount feature when using AVX512 registers. The intent is to share the preliminary results with the community and get feedback for adding avx512 support for popcount.
>
> Revisiting the previous discussion/improvements around this feature, I have created a micro-benchmark based on the pg_popcount() in PostgreSQL's current implementations for x86_64 using the newer AVX512 intrinsics. Playing with this implementation has improved performance up to 46% on Intel's Sapphire Rapids platform on AWS. Such gains will benefit scenarios relying on popcount.

How does this compare to older CPUs, and more mixed workloads? IIRC,
the use of AVX512 (which I believe this instruction to be included in)
has significant implications for core clock frequency when those
instructions are being executed, reducing overall performance if
they're not a large part of the workload.

> My setup:
>
> Machine: AWS EC2 m7i - 16vcpu, 64gb RAM
> OS : Ubuntu 22.04
> GCC: 11.4 and 12.3 with flags "-mavx -mavx512vpopcntdq -mavx512vl -march=native -O2".
>
> 1. I copied the pg_popcount() implementation into a new C/C++ project using cmake/make.
>         a. Software only and
>         b. SSE 64 bit version
> 2. I created an implementation using the following AVX512 intrinsics:
>         a. _mm512_popcnt_epi64()
>         b. _mm512_reduce_add_epi64()
> 3. I tested random bit streams from 64 MiB to 1024 MiB in length (5 sizes; repeatable with RNG seed [std::mt19937_64])

Apart from the two type functions bytea_bit_count and bit_bit_count
(which are not accessed in postgres' own systems, but which could want
to cover bytestreams of >BLCKSZ) the only popcount usages I could find
were on objects that fit on a page, i.e. <8KiB in size. How does
performance compare for bitstreams of such sizes, especially after any
CPU clock implications are taken into account?

Kind regards,

Matthias van de Meent
Neon (https://neon.tech)