[Bug other/27417] wrong code or aliasing violation with missed diagnostic?

2006-05-05 Thread pluto at agmk dot net
--- Comment #4 from pluto at agmk dot net 2006-05-05 15:10 --- (In reply to comment #3) You cannot. Though you can try extending c-common.c:strict_aliasing_warning and cp/typeck.c:build_reinterpret_cast_1 to warn in this case. sorry, but i don't know compiler internals to do this.

[Bug other/27417] wrong code or aliasing violation with missed diagnostic?

2006-05-04 Thread rguenth at gcc dot gnu dot org
--- Comment #1 from rguenth at gcc dot gnu dot org 2006-05-04 11:58 --- This missed diagnostic is known, as enabling a warning here would cause too much false positives. But yes, you are violating strict-aliasing rules here. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27417

[Bug other/27417] wrong code or aliasing violation with missed diagnostic?

2006-05-04 Thread pluto at agmk dot net
--- Comment #2 from pluto at agmk dot net 2006-05-04 12:08 --- (In reply to comment #1) This missed diagnostic is known, as enabling a warning here would cause too much false positives. but what about -Wstrict-aliasing=2? it doesn't report anything, so how can i check possible

[Bug other/27417] wrong code or aliasing violation with missed diagnostic?

2006-05-04 Thread rguenth at gcc dot gnu dot org
--- Comment #3 from rguenth at gcc dot gnu dot org 2006-05-04 14:00 --- You cannot. Though you can try extending c-common.c:strict_aliasing_warning and cp/typeck.c:build_reinterpret_cast_1 to warn in this case. -- http://gcc.gnu.org/bugzilla/show_bug.cgi?id=27417