gcc 2.96 optimization bug on RedHat7.0/alpha

Christopher C. Chimelis chris@debian.org
Tue Apr 17 17:05:00 GMT 2001


On Tue, 17 Apr 2001, Uncle George wrote:

> with -O2 i get the wrong answer
> 
>                 [gat@LX gat]$ gcc -g -O2 a.c -o a
>                 [gat@LX gat]$ ./a
>                 FlooR: x = 22.16, i0=1, i1=536871996
> 
> 
> with -O i get a passable answer
> 
>                 [gat@LX gat]$ gcc -g -O a.c -o a
>                 [gat@LX gat]$ ./a
>                 FlooR: x = 22.16, i0=1077291253, i1=-1030792151
> 
> 
> BTW what exactly is the page one uses to submit/fill in the bug report
> for gcc et al ?

This has not been fixed in upstream gcc CVS, FYI.  Using your source
and the gcc 3.0 snapshot from :

chris@spawn:
~/test > gcc-3.0 -g -O2 -o uncleg uncleg.c
chris@spawn:
~/test > ./uncleg 
FlooR: x = 22.16, i0=1, i1=536872108
chris@spawn:
~/test > gcc-3.0 -g -O -o uncleg uncleg.c
chris@spawn:
~/test > ./uncleg 
FlooR: x = 22.16, i0=1077291253, i1=-1030792151

My local copy of a patched 2.95.4pre returns the correct results, though,
so something broke between 2.95.x and 3.0pre.

Figured that you'd want to know :-)

C



More information about the Gcc-bugs mailing list