Hi
gcc-3.2.3 was released today. Created a diff using 3.2.2, attached.
Tried to compile using latest gpc alpha. Got as far as dwarf2out.c then got an internal compiler error message, dwarf2out_finish, at dwarf2out.c:12278
Russ
Russell Whitaker wrote:
gcc-3.2.3 was released today. Created a diff using 3.2.2, attached.
Tried to compile using latest gpc alpha. Got as far as dwarf2out.c then got an internal compiler error message, dwarf2out_finish, at dwarf2out.c:12278
I've decided not to include multiple patches which differ only by the offsets. Patch can easily ignore such differences -- compared to my effort of keeping several almost identical files up to date (now that I seem to find a new backend bug every week).
So I'll just create a link to gcc-3.2.2.diff and give a note in configure that gcc-3.2.3 is not yet working well (as you found out, I think Waldek is working on it) ...
Frank
Russell Whitaker wrote:
gcc-3.2.3 was released today. Created a diff using 3.2.2, attached.
Tried to compile using latest gpc alpha. Got as far as dwarf2out.c then got an internal compiler error message, dwarf2out_finish, at dwarf2out.c:12278
With the .diff, gcc-3.2.3 builds on Mac OS X, where stabs is the default debugging format. The testsuite reveals known back-end bugs (announced to be fixed in gcc-3.3).
Frank Heckenbach wrote:
After some other changes I made, I found that this was rather easy to implement, and the next GPC version will support it with `--longjmp-all-nonlocal-labels'.
Even if it's fixed in gcc-3.3 on Mac OS, this might be useful. Who knows on which other platforms the bug may still exist or what other new bugs gcc-3.3 will bring ...
Thanks for the new option.
Regards,
Adriaan van Os
-
Adriaan van Os -
Frank Heckenbach -
Russell Whitaker