This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: check-g++ fails
- To: GCC list <gcc at gcc dot gnu dot org>
- Subject: Re: check-g++ fails
- From: Michael Ritzert <gcc at realtech dot de>
- Date: Fri, 27 Apr 2001 13:24:45 +0200
Well, sometimes you just have bad luck...
both errors are due to something else:
- on the gcc 3.0 branch I get
Bootstrap comparison failure!
cp/class.o differs
make[1]: *** [compare] Error 1
- on the HEAD (other machine), the
Forbidden identifiers: i386 linux unix
stage1/xgcc -Bstage1/ -B/opt/gcc/i686-pc-linux-gnu/bin/ -c -DIN_GCC -DHAVE_CONFIG_H -g -O2 -W -Wall -Wwrite-strings -Wstrict-prototypes -Wmissing-prototypes -Wtraditional -pedantic -Wno-long-long -I. -I.. -I../../../gcc/gcc/fixinc -I../../../gcc/gcc/fixinc/.. -I../../../gcc/gcc/fixinc/../config -I../../../gcc/gcc/fixinc/../../include ../../../gcc/gcc/fixinc/fixincl.c
make[3]: stage1/xgcc: Command not found
error came back.
I didn't notice before because the last thing I saw above the check-g++ error
was the successful check-gcc run...
I can't check if the same things happened yesterday because I accidentally
dropped that mail...
At the moment I try to compile the two source trees on the respective other
machine with fresh CVS sources. I'll provide that results ASAP if something
changes.
Ok. The first one just finished - the
make[3]: stage1/xgcc: Command not found
make[3]: *** [fixincl.o] Error 127
error is also back on the 3.0 branch.
The second bootstrap will take longer because this is a slower machine and
the error happens later in the bootstrap.
Regards,
Michael