Alvaro Herrera <alvherre@2ndquadrant.com> writes:
> Tom Lane wrote:
>> My own compilers don't generate errors either, only warnings. Maybe
>> we could catch this sort of thing mechanically with a critter configured
>> with -Werror as well as --disable-integer-datetimes, but I'm a tad
>> hesitant to have a buildfarm machine configured with -Werror. We'd
>> be raising the stakes quite a bit on keeping the code warning-free
>> for a specific compiler that might prove to be picky in silly ways.
>>
>> But if it were a pretty mainstream compiler that hadn't shown much
>> evidence of wanting to throw useless warnings, maybe that's fine.
> I think we're at a point where one gcc plus one clang animals set up
> with -Werror would be a good thing to have, and keep green. In
> practice, this list fulfills the role of such a compiler, only with much
> higher communication overhead.
Yeah. I have longfin which is running Apple's clang, and is on a machine
that doesn't have much to do otherwise. I propose to turn on -Werror in
its configuration, and to configure a second critter on the same hardware
that runs with -Werror as well as --disable-integer-datetimes. Somebody
else should do similarly with a reasonably modern/stable gcc release.
In the longer term, we might want to encourage more buildfarm owners to
use -Werror, but only with mainstream compilers that are generally pretty
sane about warnings. Some of the older critters generate
hundreds/thousands of useless warnings; I'm not interested in trying to
suppress all of those.
regards, tom lane
