gcc 2.95.2, bug in gcse?

[Franz Sirl]

Am Sun, 12 Mar 2000 schrieb Martin v. Loewis:
>> egcs-20000306 on i686-redhat-linux gives an a.out that doesn't
>> segfault, so it appears that the bug was fixed in the mainline.
>
>Nevertheless, I think this bug is important enough to be fixed in the
>release branch as well; today's 2.95.3 still generates wrong code.

The current gcc-2_95-branch works OK on powerpc-linux-gnu. Are you sure you
didn't get confused by the strange testcase? It aborts for correct behaviour
and segfaults for the bug.

Franz.