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]

gcc 2.95.3-test2 on m68k-openbsd


prompt> gcc -c  -DIN_GCC    --save-temps -g   -DHAVE_CONFIG_H    -I.
-I/usr/src/gnu/egcs/gcc -I/usr/src/gnu/egcs/gcc/config
-I/usr/src/gnu/egcs/gcc/../include /usr/src/gnu/egcs/gcc/expmed.c
/usr/src/gnu/egcs/gcc/expmed.c: In function `store_bit_field':
/usr/src/gnu/egcs/gcc/expmed.c:607: Internal compiler error in
`instantiate_virtual_regs_1', at function.c:3872
Please submit a full bug report.
See <URL:http://www.gnu.org/software/gcc/bugs.html> for instructions.
*** Error code 1

prompt> gcc -v
Reading specs from /usr/lib/gcc-lib/m68k-unknown-openbsd2.8/2.95.3/specs
gcc version 2.95.3 20010112 (prerelease)

<user excuses/rant>
For different reasons, I tried to have gcc not producing
m68060-code in libgcc2.a so that I wouldn't trap on long
divs/muls since compiling my apps with -m68020-60 makes gcc
jump to a stub routine that in the end contains a mulsl 
anyway.

Don't hit me on the head for that, I wont think bad of gcc
for my own ill-doings, _but_ I triggered a gcc bug when
-O2 is left out at that point, and I want to help improve
gcc.

So don't go ape on my futile attempts, just fix the error
and forget about my miscompiles. =)
</user excuses>

gcc -g <rest> triggers the bug, while adding -O1,-O2,-O9 and -Os
all make it compile.

Here's the .i-file: 


expmed.i.gz


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