|From:||Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>|
|To:||Noah Misch <noah(at)leadboat(dot)com>|
|Cc:||Aleksander Alekseev <a(dot)alekseev(at)postgrespro(dot)ru>, David Steele <david(at)pgmasters(dot)net>, PostgreSQL Developers <pgsql-hackers(at)postgresql(dot)org>|
|Subject:||Re: [PATCH] Suppress Clang 3.9 warnings|
|Views:||Raw Message | Whole Thread | Download mbox | Resend email|
Noah Misch <noah(at)leadboat(dot)com> writes:
> On Wed, Mar 15, 2017 at 10:57:15AM -0400, Tom Lane wrote:
>> Seems like the correct solution is to
>> absorb that fix, either by updating to a newer autoconf release or by
>> carrying our own version of AC_CHECK_DECLS until they come out with one.
> As you mention upthread, that Autoconf commit is still newer than every
> Autoconf release. (Last release was 58 months ago.) Altering configure.ac to
> work around the bug would be reasonable, but it feels heavy relative to the
> benefit of suppressing some warnings.
It does seem like rather a lot of work, but I think it's preferable to
hacking up the coding in port.h. Mainly because we could booby-trap the
substitute AC_CHECK_DECLS to make sure we revert it whenever autoconf 2.70
does materialize (a check on m4_PACKAGE_VERSION, like the one at
configure.in line 22, ought to do the trick); whereas I do not think
we'd remember to de-kluge port.h if we kluge around it there.
I'm fine with leaving it alone, too.
regards, tom lane
|Next Message||Dilip Kumar||2017-03-15 16:23:08||Re: Parallel Bitmap scans a bit broken|
|Previous Message||Stephen Frost||2017-03-15 16:21:00||Re: new set of psql patches for loading (saving) data from (to) text, binary files|