Horton

Re: Change GUC hashtable to use simplehash?

John Naylor <johncnaylorls@gmail.com>

View thread
From: John Naylor <johncnaylorls@gmail.com>
To: Andres Freund <andres@anarazel.de>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, Jeff Davis <pgsql@j-davis.com>, Gurjeet Singh <gurjeet@singh.im>, pgsql-hackers@postgresql.org
Date: 2023-12-02T08:35:24Z
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. Silence warning in older versions of Valgrind

  2. Revert "Speed up tail processing when hashing aligned C strings, take two"

  3. Speed up tail processing when hashing aligned C strings, take two

  4. Teach fasthash_accum to use platform endianness for bytewise loads

  5. Add macro to disable address safety instrumentation

  6. Convert uses of hash_string_pointer to fasthash equivalent

  7. Speed up tail processing when hashing aligned C strings

  8. Add helper functions for dshash tables with string keys.

  9. Fix warnings in cpluspluscheck

  10. Further cosmetic review of hashfn_unstable.h

  11. Simplify initialization of incremental hash state

  12. Add optimized C string hashing

  13. Add inline incremental hash functions for in-memory use

  14. Make all Perl warnings fatal

On Wed, Nov 29, 2023 at 8:31 PM John Naylor <johncnaylorls@gmail.com> wrote:
>
> Attached is a rough start with Andres's earlier ideas, to get
> something concrete out there.

While looking at the assembly out of curiosity, I found a couple bugs
in the split API that I've fixed locally.

I think the path forward is:

- performance measurements with both byte-at-a-time and
word-at-a-time, once I make sure they're fixed
- based on the above decide which one is best for guc_name_hash
- clean up hash function implementation
- test with with a new guc_name_compare (using what we learned from my
guc_name_eq) and see how well we do with keeping dynahash vs.
simplehash

Separately, for string_hash:

- run SMHasher and see about reincorporating length in the
calculation. v5 should be a clear improvement in collision behavior
over the current guc_name_hash, but we need to make sure it's at least
as good as hash_bytes, and ideally not lose anything compared to
standard fast_hash.