I wrote:
> The test is attached and the test function is part of the patch. It's
> based on the test used in the commit above. The test searches for a
> string that's at the end of a ~1 million byte string. This is on gcc
> 11 with 2-3 runs to ensure repeatability, but I didn't bother with
> statistics because the differences are pretty big:
>
> patch | no match | ascii | mulitbyte
> -----------------------------------------+----------+-------+-----------
> PG11 | 1120 | 1100 | 900
> master | 381 | 2350 | 1900
> DFA | 386 | 1640 | 1640
> branchless utf mblen | 387 | 4100 | 2600
> inline pg_utf_mblen() | 380 | 1080 | 920
> inline pg_utf_mblen() + ascii fast path | 382 | 470 | 918
I failed to mention that the above numbers are milliseconds, so
smaller is better.
--
John Naylor
EDB: http://www.enterprisedb.com