This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug fortran/52370] [4.7 Regression] Spurious "may be used uninitialized" warning for check of optional argument
- From: "Joost.VandeVondele at mat dot ethz.ch" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: Fri, 23 May 2014 19:22:36 +0000
- Subject: [Bug fortran/52370] [4.7 Regression] Spurious "may be used uninitialized" warning for check of optional argument
- Auto-submitted: auto-generated
- References: <bug-52370-4 at http dot gcc dot gnu dot org/bugzilla/>
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52370
Joost VandeVondele <Joost.VandeVondele at mat dot ethz.ch> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |Joost.VandeVondele at mat dot ethz
| |.ch
--- Comment #11 from Joost VandeVondele <Joost.VandeVondele at mat dot ethz.ch> ---
(In reply to Jakub Jelinek from comment #6)
> Until we get more fine-grained warning control, I believe the only way to
> fix this is make sure TREE_NO_WARNING is set on Fortran optional argument
> related variables (not just the data pointer (b.0 here), but I think the
> others too (bounds, size, ...).
This would be useful to have and would fix
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47386
*.dim[?].ubound
*.dim[?].lbound
*.dim[?].stride
*.offset
seem all compiler generated variables that trigger the warning 1000's of times
in a large Fortran package.