octave-maintainers
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: CVS Build problem (what.m missing?)


From: Fredrik Lingvall
Subject: Re: CVS Build problem (what.m missing?)
Date: Thu, 01 Nov 2007 13:43:04 +0100
User-agent: Thunderbird 2.0.0.6 (X11/20070807)

John W. Eaton wrote:
>
> Have you tried editing the configure script to change the lines
>
>   LIBS="$XLIBS"
>   case "`./conftest$ac_exeext`" in
>     *succeeded*)
>
> to
>
>   LIBS="$XLIBS"
>   ./conftest$ac_exeext
>   case "`./conftest$ac_exeext`" in
>     *succeeded*)
>
> ?  If you do this, what happens when the program is run outside of the
> "``" quoting in the case statement?
>   

The output (with your suggestion above) is:

failed
 (  0.00000000000000     ,  0.00000000000000     )
./configure: line 14137: 25877 Segmentation fault

I added an echo "..." of the actual compile command in the configure
script, which then was:

gfortran -o conftest -O3 -pipe -mtune=pentium-m -march=pentium-m
-mfpmath=sse -msse2 -fomit-frame-pointer -funroll-loops
-fprefetch-loop-arrays -pthread -mieee-fp  conftest.f
/usr/local/lib/libLAPACK.so /usr/local/lib/libBLAS.so 
-L/usr/lib/gcc/i686-pc-linux-gnu/4.1.2
-L/usr/lib/gcc/i686-pc-linux-gnu/4.1.2/../../../../i686-pc-linux-gnu/lib
-L/usr/lib/gcc/i686-pc-linux-gnu/4.1.2/../../.. -lz -lgfortranbegin
-lgfortran -lm -lpthread -lz -lm

Compiling foo.f with the same flags also segfaults. The problem turned
out to be the -fomit-frame-pointer flag. Removing that one from FFLAGS
solved the problem. I don't know if this is a compiler bug or if it's a
problem with my particular Gentoo setup. I will re-emerge the gcc tools
(rebuild gcc) to see if that makes any difference.

/Fredrik


reply via email to

[Prev in Thread] Current Thread [Next in Thread]