This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: c/8339: insn does not satisfy its constraints
- From: Jason Borden <jborden at sorenson dot com>
- To: double at inebraska dot com, gcc-gnats at gcc dot gnu dot org, gcc-prs at gcc dot gnu dot org, gcc-bugs at gcc dot gnu dot org, nobody at gcc dot gnu dot org
- Date: Fri, 25 Oct 2002 14:04:57 -0600
- Subject: Re: c/8339: insn does not satisfy its constraints
- Organization: Sorenson Media
- Reply-to: Jason Borden <jborden at sorenson dot com>
http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=8339
I just thought that I'd add to this bug, since I've had the same issue. The
problem seems to be caused by the -mfpmath=sse flag in combination with the
pentium4 (as far as I can tell). I've run into this bug only twice. Once when
compiling the gsmatrix.c file previously mentioned and the other when
compiling postgresql-7.2.3, the src/backend/utils/adt/float.c file. I've
compiled these same files on a pentium3 (using -march=pentium3 and no -sse2)
and on an athlon-xp (using -march=athlon-xp and no -sse2, but also having
-m3dnow) and the bug does not occur in these situations. I've reproduced this
bug on pentium4 and pentium4-m (mobile) processors. Hope this information
will be of help.
Jason