This is the mail archive of the gcc-bugs@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug middle-end/17657] [4.0 Regression] ICE in expand_case


------- Additional Comments From stevenb at suse dot de  2004-10-06 13:24 -------
Subject: Re:  [4.0 Regression] ICE in expand_case

> void
> bar (void)
> {
>   signed char tmp = foo (0);
>   int t1 = tmp;
>   switch (t1)

The problem is not the cast, but the fact that here in the source
t1 is an int so the case 255 can match, but somewhere along the
way the switch condition is turned into a signed int.
I claim that is an invalid transformation.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17657


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]