This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: invalid address bug in egcs optimizer (snapshot)
- To: law at cygnus dot com
- Subject: Re: invalid address bug in egcs optimizer (snapshot)
- From: Michal Jaegermann <michal at ellpspace dot math dot ualberta dot ca>
- Date: Tue, 28 Jul 1998 18:48:04 -0600 (MDT)
- Cc: bcurrie at tssc dot co dot nz, egcs-bugs at cygnus dot com
>
> > Sounds like the problem I had over the weekend building dosemu (I had
> > problems with inlining a not quite trivial function whos result was
> > beeing asigned to a pointer).
> BTW, the file you need to update to get the fix is regmove.c.
>
It does not look like that this is the same trouble I reported
last week (failed compilation on Alpha of slhc.o from Linux kernel
sources). I just tried egcs-19980727 snapshot, where regmove.c is
modified, and the reported problem persists.
make[2]: Entering directory `/usr/src/expm/linux-2.0.35.patched/drivers/net'
gcc -D__KERNEL__ -I/usr/src/expm/linux-2.0.35.patched/include -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -fno-strength-reduce -pipe -mno-fp-regs -Wa,-m21164a -DBWX_USABLE -DMODULE -c -o slhc.o slhc.c
cc1: memory violation at pc=12010dcac rp=12010ddc4 (bad address = 11f7fffe0)
{standard input}: Assembler messages:
{standard input}:0: Warning: end of file not at end of a line; newline inserted
gcc: Internal compiler error: program cc1 got fatal signal 11
make[2]: *** [slhc.o] Error 1
I did not try yet if the other reported by me problem (series of kernel
"oops" on shutdown for a "snapshot-compiled" kernel) remains or not.
Michal
michal@ellpspace.math.ualberta.ca
michal@harddata.com