Re: pgsql: Suppress uninitialized-variable warning.

On Sun, Jan 23, 2011 at 19:36, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> Magnus Hagander <magnus(at)hagander(dot)net> writes:
>> On Sun, Jan 23, 2011 at 19:21, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>>> Magnus Hagander <magnus(at)hagander(dot)net> writes:
>>>> Which compiler did that come out of? I don't think I saw it on any of
>>>> the ones I tried..
>
>>> gcc 4.4.5, standard on Fedora 13.  I'm surprised that yours did not show
>>> it, because the usage pattern looks beyond gcc's ability to prove safe.
>
>> Interesting. I just make clean/make:ed it again to make sure, and it
>> doesn't. And I'm on:
>> gcc (Ubuntu/Linaro 4.4.4-14ubuntu5) 4.4.5
>
> [ scratches head... ]  Maybe you are using different switches, perhaps
> optimization higher than -O2?

Nope...

gcc -O2 -Wall -Wmissing-prototypes -Wpointer-arith
-Wdeclaration-after-statement -Wendif-labels -fno-strict-aliasing
-fwrapv -g -I../../../src/interfaces/libpq -I../../../src/include
-D_GNU_SOURCE -c -o pg_basebackup.o pg_basebackup.c -MMD -MP -MF
.deps/pg_basebackup.Po

gives no output at all.

> But anyway, I think the project policy for such things is that we want
> to suppress warnings when feasible on all common compilers, so the patch
> is appropriate no matter what the specific reason for the difference is.

Oh yes, I totally agree. I'm just trying to figure out why I didn't
see it myself.

--
 Magnus Hagander
 Me: http://www.hagander.net/
 Work: http://www.redpill-linpro.com/