GNU bug report logs - #55895
[PATCH] maint: Fix ptr_align signature to silence -Wmaybe-uninitialized

Previous Next

Full log

Message #8 received at 55895 <at> debbugs.gnu.org (full text, mbox):

From: Paul Eggert <eggert <at> cs.ucla.edu>
To: Anders Kaseorg <andersk <at> mit.edu>
Cc: 55895 <at> debbugs.gnu.org
Subject: Re: bug#55895: [PATCH] maint: Fix ptr_align signature to silence
 -Wmaybe-uninitialized
Date: Fri, 10 Jun 2022 20:23:19 -0700

On 6/10/22 15:24, Anders Kaseorg wrote:
> ptr_align is always called with a pointer to uninitialized memory, so
> it does not make sense for that pointer to be const.

I don't see why not. ptr_align does not modify the referenced storage so 
"void const *" is appropriate. If we changed the type to "void *" we'd 
unnecessarily limit ptr_align's applicability.


> This change avoids -Wmaybe-uninitialized warnings from GCC 11.

I can't reproduce the problem with the latest coreutils commit 
93e099e4c3b659b2e329f655fbdc73fdf594a66e on Savannah master on Ubuntu 
22.04 LTS x86-64, I don't get warnings with either gcc (Ubuntu 
11.2.0-19ubuntu1) 11.2.0 or with gcc-12 (Ubuntu 12-20220319-1ubuntu1) 
12.0.1 20220319 (experimental) [master r12-7719-g8ca61ad148f]. I also 
don't see a problem on Fedora 36 x86-64 with gcc (GCC) 12.1.1 20220507 
(Red Hat 12.1.1-1)
. In all cases I configured and built with "./configure 
--enable-gcc-warnings; make".

Perhaps the problem (whatever it was) has already been fixed in 
coreutils. If not, I guess that the issue has something to do with the 
particular platform and options you configured with. If it's an older 
compiler like GCC 11.3.0 I wouldn't worry too much about it, as older 
GCCs are notorious for false alarms and not worth the trouble of pacifying.

Most likely the warnings are false positives (though I haven't checked 
this).

Previous Next

GNU bug tracking system
Copyright (C) 1999 Darren O. Benham, 1997,2003 nCipher Corporation Ltd, 1994-97 Ian Jackson.