Hello Michaël,
> \set aid 1 + 1
> pgbench -f addition.sql -t 50000000
>
> I have the following:
> HEAD: 3.5~3.7M TPS
> list method: 3.6~3.7M TPS
> array method: 3.4~3.5M TPS
> So all approaches have a comparable performance.
Yep, the execution trace is pretty similar in all cases, maybe with a
little more work for the array method, although I'm surprise that the
difference is discernable.
> Btw, patch 2 is returning a warning for me:
> It is trying to compare a 32b integer with an int64 value, evalFunc
> needed an int64.
Indeed. My gcc 4.8.4 with --Wall does not show the warning, too bad.
Attached is the fixed patch for the array method.
--
Fabien