Hello Andres,
>> There are some overflow checking with div and double to int cast, which were
>> added because of previous complaints, but which are not very useful to me.
>
> I think handling it inconsistently is the worst of all worlds.
Hmmm... I cannot say that inconsistency is a good thing, that would not
be consistent:-)
My 0.02€:
- I do not think that updating pgbench arithmetic for managing integer
overflows is worth Andres Freund time. My guess is that most
script would not trigger client-side overflows, so the change would
be a no-op in practice.
- I think that pgbench has more important defects/missing features, to
be fixed more urgently. Given the time patches spend in the cf queue,
obviously committers disagree on this one:-)
Now ISTM that it does not harm anything to add such a feature, so fine
with me. Maybe the global compiler option removal is worth the effort.
--
Fabien.