http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47817
--- Comment #5 from Andrew Pinski 2011-02-19
21:37:08 UTC ---
(In reply to comment #4)
> But given that reinterpret_cast does not, I would still suspect some tricky
> things happening inside gcc anyway..
No reinterpret_cast does not change the f
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47817
--- Comment #4 from Andrew Aksyonoff
2011-02-19 21:10:21 UTC ---
(In reply to comment #2)
> I think you have some violation of the C/C++ aliasing rules where you access
> an
> "unsigned int" via an "unsigned long long" which in turn causes undef
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47817
Andrew Pinski changed:
What|Removed |Added
Status|UNCONFIRMED |RESOLVED
Resolution|
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47817
--- Comment #2 from Andrew Pinski 2011-02-19
20:56:45 UTC ---
I think you have some violation of the C/C++ aliasing rules where you access an
"unsigned int" via an "unsigned long long" which in turn causes undefined
behavior.
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=47817
--- Comment #1 from Andrew Aksyonoff
2011-02-19 20:51:23 UTC ---
Created attachment 23407
--> http://gcc.gnu.org/bugzilla/attachment.cgi?id=23407
the preprocessed .ii file built with -O3 -save-temps
