Re: add AVX2 support to simd.h

Nathan Bossart <nathandbossart@gmail.com>

From: Nathan Bossart <nathandbossart@gmail.com>

To: John Naylor <johncnaylorls@gmail.com>

Cc: Ants Aasma <ants@cybertec.at>, pgsql-hackers@postgresql.org

Date: 2024-03-21T18:38:23Z

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 →

Improve style of pg_lfind32().
- 7188a7806d20 17.0 landed
Fix compiler warning for pg_lfind32().
- 1f42337be535 17.0 landed
Micro-optimize pg_lfind32().
- 7644a7340c8a 17.0 landed
Introduce helper SIMD functions for small byte arrays
- 9f225e992bed 17.0 cited
Optimize xid/subxid searches in XidInMVCCSnapshot().
- 37a6e5df3713 16.0 cited

Attachments

2reg.jpg (image/jpeg)
2reg_extended.jpg (image/jpeg)

On Thu, Mar 21, 2024 at 12:09:44PM -0500, Nathan Bossart wrote:
> On Thu, Mar 21, 2024 at 11:30:30AM +0700, John Naylor wrote:
>> Further, now that the algorithm is more SIMD-appropriate, I wonder
>> what doing 4 registers at a time is actually buying us for either SSE2
>> or AVX2. It might just be a matter of scale, but that would be good to
>> understand.
> 
> I'll follow up with these numbers shortly.

It looks like the 4-register code still outperforms the 2-register code,
except for a handful of cases where there aren't many elements.

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