This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug c++/13959] un-init local variable detection enhancement request
- From: "d dot binderman at virgin dot net" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 1 Feb 2004 20:04:46 -0000
- Subject: [Bug c++/13959] un-init local variable detection enhancement request
- References: <20040201141801.13959.d.binderman@virgin.net>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Additional Comments From d dot binderman at virgin dot net 2004-02-01 20:04 -------
>The reason for might is because there are time where might is true (for an
>example:
I agree that in your more complex case what the GNU compiler
does is reasonable. Saying "might" and requiring -O2 -Wall in that case is ok.
My enhancement request is about being more emphatic about a simpler situation.
>Turning on this warning at -O0 increases compile time because you need to know
>flow information.
I wouldn't regard increased compile time as a significant problem.
Especially in the case where the extra work finds certain run time
bugs at compile time.
As my original post should make clear, there isn't a lot of work in this
enhancement.
>There are large of cases in other software too (I do not persionally care
>about Redhat Fedora Core).
Fair enough.
>This is still a dup of 11182 no matter which way you look at it.
Only if I have expressed myself not clearly.
For which my apologies.
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=13959