On Thu, Nov 11, 2021 at 03:51:35PM -0500, Tom Lane wrote:
> That'd be considerably messier wouldn't it?
Yes, that would be a bit messier. For example, we could do that with
something saved in Makefile.global.in by ./configure that plperl feeds
on to add this extra CFLAGS in its own local Makefile
On Thu, Nov 11, 2021 at 01:18:32PM -0500, Tom Lane wrote:
> I see you tried to do that by adding -Werror to CPPFLAGS. That doesn't
> work because it breaks configure's tests. (I wonder if meson is any
> smarter than autoconf about that.) The way I do it on my own
> animals is like
Nya. (Japani
Andrew Dunstan writes:
> Wouldn't it be better in any case just to add the clang fix for building
> plperl rather than globally?
That'd be considerably messier wouldn't it? (And I wonder if the
meson system will be able to do it at all.) If you feel motivated
to make that happen, go for it, but
On 11/11/21 13:18, Tom Lane wrote:
> Michael Paquier writes:
>> Backpatched this one as of 9ff47ea. That should allow the addition of
>> -Werror on dangomushi.
> I see you tried to do that by adding -Werror to CPPFLAGS. That doesn't
> work because it breaks configure's tests. (I wonder if mes
Michael Paquier writes:
> Backpatched this one as of 9ff47ea. That should allow the addition of
> -Werror on dangomushi.
I see you tried to do that by adding -Werror to CPPFLAGS. That doesn't
work because it breaks configure's tests. (I wonder if meson is any
smarter than autoconf about that.)
Michael Paquier writes:
> Backpatched this one as of 9ff47ea. That should allow the addition of
> -Werror on dangomushi.
Cool. I have also enabled -Werror on florican.
regards, tom lane
On Wed, Nov 10, 2021 at 07:35:21AM +0900, Michael Paquier wrote:
> Tom has suggested that we could add -Wno-compound-token-split-by-macro
> to take care of the issue on our side, and attached is a patch to do
> so.
>
> Any objections? I'd like to get this back-patched.
Backpatched this one as of
