Problems installing scipy 0.6 on FC5
Hi, I am running FC5 (I think, how to figure out) and I followed the installation instructions for FC5. The last step, however, fails with the following error messages:
python setup.py install ... ... g77:f77: scipy/fftpack/dfftpack/zfftb.f g77:f77: scipy/fftpack/dfftpack/dsinti.f g77:f77: scipy/fftpack/dfftpack/dfftb.f g77:f77: scipy/fftpack/dfftpack/zfftb1.f /tmp/ccuB5p4W.s: Assembler messages: /tmp/ccuB5p4W.s:589: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s:2948: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s: Assembler messages: /tmp/ccuB5p4W.s:589: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s:2948: Error: suffix or operands invalid for `movd' error: Command "/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer -malign-double -c -c scipy/fftpack/dfftpack/zfftb1.f -o build/temp.linux- i686-2.4/scipy/fftpack/dfftpack/zfftb1.o" failed with exit status 1
Any ideas how to solve this issue? Cheers Alex
Alexander Dietz wrote:
Hi,
I am running FC5 (I think, how to figure out) and I followed the installation instructions for FC5. The last step, however, fails with the following error messages:
python setup.py install ... ... g77:f77: scipy/fftpack/dfftpack/zfftb.f g77:f77: scipy/fftpack/dfftpack/dsinti.f g77:f77: scipy/fftpack/dfftpack/dfftb.f g77:f77: scipy/fftpack/dfftpack/zfftb1.f /tmp/ccuB5p4W.s: Assembler messages: /tmp/ccuB5p4W.s:589: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s:2948: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s: Assembler messages: /tmp/ccuB5p4W.s:589: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s:2948: Error: suffix or operands invalid for `movd' error: Command "/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer -malign-double -c -c scipy/fftpack/dfftpack/zfftb1.f -o build/temp.linux- i686-2.4/scipy/fftpack/dfftpack/zfftb1.o" failed with exit status 1
Any ideas how to solve this issue?
Cheers Alex
What's your machine ? More exactly, could you give us the result of the following command (to get the exact characteristics of your CPU): cat /proc/cpuinfo ? cheers, David
Hi, see the content of /proc/cpuinfo below. Cheers Alex processor : 0 vendor_id : GenuineIntel cpu family : 6 model : 14 model name : Genuine Intel(R) CPU T2300 @ 1.66GHz stepping : 8 cpu MHz : 1000.000 cache size : 2048 KB physical id : 0 siblings : 2 core id : 0 cpu cores : 2 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx constant_tsc pni monitor est tm2 xtpr bogomips : 3335.11 processor : 1 vendor_id : GenuineIntel cpu family : 6 model : 14 model name : Genuine Intel(R) CPU T2300 @ 1.66GHz stepping : 8 cpu MHz : 1000.000 cache size : 2048 KB physical id : 0 siblings : 2 core id : 1 cpu cores : 2 fdiv_bug : no hlt_bug : no f00f_bug : no coma_bug : no fpu : yes fpu_exception : yes cpuid level : 10 wp : yes flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx constant_tsc pni monitor est tm2 xtpr bogomips : 3329.30 On Nov 27, 2007 3:19 AM, David Cournapeau <david@ar.media.kyoto-u.ac.jp> wrote:
Alexander Dietz wrote:
Hi,
I am running FC5 (I think, how to figure out) and I followed the installation instructions for FC5. The last step, however, fails with the following error messages:
python setup.py install ... ... g77:f77: scipy/fftpack/dfftpack/zfftb.f g77:f77: scipy/fftpack/dfftpack/dsinti.f g77:f77: scipy/fftpack/dfftpack/dfftb.f g77:f77: scipy/fftpack/dfftpack/zfftb1.f /tmp/ccuB5p4W.s: Assembler messages: /tmp/ccuB5p4W.s:589: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s:2948: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s: Assembler messages: /tmp/ccuB5p4W.s:589: Error: suffix or operands invalid for `movd' /tmp/ccuB5p4W.s:2948: Error: suffix or operands invalid for `movd' error: Command "/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer -malign-double -c -c scipy/fftpack/dfftpack/zfftb1.f -o build/temp.linux- i686-2.4/scipy/fftpack/dfftpack/zfftb1.o" failed with exit status 1
Any ideas how to solve this issue?
Cheers Alex
What's your machine ? More exactly, could you give us the result of the following command (to get the exact characteristics of your CPU): cat /proc/cpuinfo ?
cheers,
David _______________________________________________ SciPy-user mailing list SciPy-user@scipy.org http://projects.scipy.org/mailman/listinfo/scipy-user
Alexander Dietz wrote:
Hi,
see the content of /proc/cpuinfo below.
Ok, your CPU supports sse2 instruction set, so this is not the problem. Since you use an old FC version, it may be an old g77 bug. Could you give use the version of g77 (g77 -v). Also, you could try to see which flag make it fails, for example, does the following: /usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse -fomit-frame-pointer -malign-double -c scipy/fftpack/dfftpack/zfftb1.f (I suspect the -msse2 to be a problem with old g77; several problems related to -msse2 and movd were fixed in gcc 3.3). cheers, David
Hi, On Nov 27, 2007 9:06 AM, David Cournapeau <david@ar.media.kyoto-u.ac.jp> wrote:
Alexander Dietz wrote:
Hi,
see the content of /proc/cpuinfo below.
Ok, your CPU supports sse2 instruction set, so this is not the problem. Since you use an old FC version, it may be an old g77 bug. Could you give use the version of g77 (g77 -v).
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/3.2.3/specs Configured with: ../configure --prefix=/usr --mandir=/usr/share/man --infodir=/usr/share/info --enable-shared --enable-threads=posix --disable-checking --with-system-zlib --enable-__cxa_atexit --enable-languages=c,c++,f77 --disable-libgcj --host=i386-redhat-linux Thread model: posix gcc version 3.2.3 20030502 (Red Hat Linux 3.2.3-55.fc5)
Also, you could try to see which flag make it fails, for example, does the following:
/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse -fomit-frame-pointer -malign-double -c scipy/fftpack/dfftpack/zfftb1.f
That worked actually... Cheers Alex
(I suspect the -msse2 to be a problem with old g77; several problems related to -msse2 and movd were fixed in gcc 3.3).
cheers,
David _______________________________________________ SciPy-user mailing list SciPy-user@scipy.org http://projects.scipy.org/mailman/listinfo/scipy-user
Alexander Dietz wrote:
Also, you could try to see which flag make it fails, for example, does the following:
/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse -fomit-frame-pointer -malign-double -c scipy/fftpack/dfftpack/zfftb1.f
That worked actually...
Yes, so the problem is your gcc version. Maybe we could disable adding -msse2 altogether for gcc < 3.3, instead of enabling it for gcc > 3.2.2 (look at numpy/distutils/fcompiler/gnu.py, and look for -msse2; if Pearu or David are reading this, maybe they know a better solution ?). Note that FC5 is old. You should consider upgrading (FC 5 is not maintained anymore) if you can, cheers, David
David Cournapeau wrote:
Alexander Dietz wrote:
Also, you could try to see which flag make it fails, for example, does the following:
/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2 -funroll-loops -march=i686 -mmmx -msse -fomit-frame-pointer -malign-double -c scipy/fftpack/dfftpack/zfftb1.f
That worked actually...
Yes, so the problem is your gcc version. Maybe we could disable adding -msse2 altogether for gcc < 3.3, instead of enabling it for gcc > 3.2.2 (look at numpy/distutils/fcompiler/gnu.py, and look for -msse2; if Pearu or David are reading this, maybe they know a better solution ?).
Note that FC5 is old. You should consider upgrading (FC 5 is not maintained anymore) if you can, I forgot: another possibility is to use gfortran instead of g77, but you will have to recompile everything (you cannot really mix g77 and gfortran code together), including BLAS/LAPACK. For FC6, I remember that gfortran is the default ABI, so this is anyway the best choice, but I am not sure for FC5.
David
Hi, I finally found a work-around for my problems installing scipy for my Linux distribution: I needed just to comment the line in /numpy/distutils/fcompiler/gnu.py in which the term "msse2" is set as an compiler option. Then scipy installs without problems... Cheers Alex On Nov 27, 2007 9:46 AM, David Cournapeau <david@ar.media.kyoto-u.ac.jp> wrote:
David Cournapeau wrote:
Alexander Dietz wrote:
Also, you could try to see which flag make it fails, for example, does the following:
/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O2
-funroll-loops
-march=i686 -mmmx -msse -fomit-frame-pointer -malign-double -c scipy/fftpack/dfftpack/zfftb1.f
That worked actually...
Yes, so the problem is your gcc version. Maybe we could disable adding -msse2 altogether for gcc < 3.3, instead of enabling it for gcc > 3.2.2 (look at numpy/distutils/fcompiler/gnu.py, and look for -msse2; if Pearu or David are reading this, maybe they know a better solution ?).
Note that FC5 is old. You should consider upgrading (FC 5 is not maintained anymore) if you can, I forgot: another possibility is to use gfortran instead of g77, but you will have to recompile everything (you cannot really mix g77 and gfortran code together), including BLAS/LAPACK. For FC6, I remember that gfortran is the default ABI, so this is anyway the best choice, but I am not sure for FC5.
David _______________________________________________ SciPy-user mailing list SciPy-user@scipy.org http://projects.scipy.org/mailman/listinfo/scipy-user
participants (2)
-
Alexander Dietz -
David Cournapeau