dfftf1.f:0: error: CPU you selected does not support x86-64 instruction set

[Chris Edwards-9]

Hi.

  I'm trying to install scipy, and have run into what is probably a
simple problem.  I think I've successfully built numpy (as well as blas,
lapack, fftw, etc.), but  run into the following error after running
python setup.py build:

.
.
.
Fortran f77 compiler: /usr/bin/g77 -g -Wall -fno-second-underscore -fPIC
-O3 -funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer
creating build/temp.linux-x86_64-2.4
creating build/temp.linux-x86_64-2.4/scipy
creating build/temp.linux-x86_64-2.4/scipy/fftpack
creating build/temp.linux-x86_64-2.4/scipy/fftpack/dfftpack
compile options: '-c'
g77:f77: scipy/fftpack/dfftpack/dfftf1.f
scipy/fftpack/dfftpack/dfftf1.f:0: error: CPU you selected does not
support x86-64 instruction set
scipy/fftpack/dfftpack/dfftf1.f:0: error: CPU you selected does not
support x86-64 instruction set
scipy/fftpack/dfftpack/dfftf1.f:0: error: CPU you selected does not
support x86-64 instruction set
scipy/fftpack/dfftpack/dfftf1.f:0: error: CPU you selected does not
support x86-64 instruction set
error: Command "/usr/bin/g77 -g -Wall -fno-second-underscore -fPIC -O3
-funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer -c -c
scipy/fftpack/dfftpack/dfftf1.f -o
build/temp.linux-x86_64-2.4/scipy/fftpack/dfftpack/dfftf1.o" failed with
exit status 1

Looking through past posts to this group, this error seems similar to a
previous issue posted where the Core2 processor wasn't properly being
identified.  I'm running CentOS 5 on a  system with two Xeon 5400 series
chips.  Just in case this issue had been recently fixed, I downloaded
the latest scipy from the svn today, but had the same problem.

Just in case this is helpful, here is my /proc/cpuinfo:

processor    : 0
vendor_id    : GenuineIntel
cpu family    : 6
model        : 23
model name    : Intel(R) Xeon(R) CPU           E5440  @ 2.83GHz
stepping    : 6
cpu MHz        : 2833.504
cache size    : 6144 KB
physical id    : 0
siblings    : 4
core id        : 0
cpu cores    : 4
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 pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm syscall nx lm
constant_tsc pni monitor ds_cpl vmx est tm2 cx16 xtpr lahf_lm
bogomips    : 5669.15
clflush size    : 64
cache_alignment    : 64
address sizes    : 38 bits physical, 48 bits virtual
power management:

Then similar information is repeated 7 more times, with processor going
from 1 to 7.

Thanks for any help.

Chris Edwards
_______________________________________________
SciPy-user mailing list
[hidden email]
http://projects.scipy.org/mailman/listinfo/scipy-user

[Pearu Peterson-3]

Chris Edwards wrote:

> Just in case this is helpful, here is my /proc/cpuinfo:
...

Could you also send the output of
   $ python numpy/distutils/cpuinfo.py

Thanks,
Pearu

_______________________________________________
SciPy-user mailing list
[hidden email]
http://projects.scipy.org/mailman/listinfo/scipy-user

[Chris Edwards-9]

Hi.  Thanks!

$ python numpy/distutils/cpuinfo.py
CPU information: getNCPUs=8 has_mmx has_sse has_sse2 is_64bit is_Intel
is_XEON is_Xeon is_i686

Chris

Pearu Peterson wrote:

> Chris Edwards wrote:
>
>  
>> Just in case this is helpful, here is my /proc/cpuinfo:
>>    
> ...
>
> Could you also send the output of
>    $ python numpy/distutils/cpuinfo.py
>
> Thanks,
> Pearu
>
> _______________________________________________
> SciPy-user mailing list
> [hidden email]
> http://projects.scipy.org/mailman/listinfo/scipy-user
>  

--
----------------------------------------------------------------------------
Christopher A. Edwards                          [hidden email]
Ocean Sciences Department                       phone: [phone]
University of California                        fax: [phone]
Santa Cruz, CA  95064  
----------------------------------------------------------------------------

_______________________________________________
SciPy-user mailing list
[hidden email]
http://projects.scipy.org/mailman/listinfo/scipy-user

[Pearu Peterson-3]

On Mon, April 7, 2008 9:37 am, Chris Edwards wrote:

> Fortran f77 compiler: /usr/bin/g77 -g -Wall -fno-second-underscore -fPIC
> -O3 -funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer

Thanks for sending the cpuinfo.py output. Comparing it's output
and the numpy/distutils/fcompiler/gnu.py source, it seems that
you are using some older version of numpy. The -march value should
be 'nocona' for your case.

So, try upgrade numpy from svn (or just copy the gnu.py file from
svn to numpy installation directory).

HTH,
Pearu

_______________________________________________
SciPy-user mailing list
[hidden email]
http://projects.scipy.org/mailman/listinfo/scipy-user

[Chris Edwards-9]

Hi.  Thanks.  I had downloaded the latest build of scipy via svn, but
had not done so for numpy.  Thanks for what turned out to be a very
simple solution.

Chris

Pearu Peterson wrote:

> On Mon, April 7, 2008 9:37 am, Chris Edwards wrote:
>
>  
>> Fortran f77 compiler: /usr/bin/g77 -g -Wall -fno-second-underscore -fPIC
>> -O3 -funroll-loops -march=i686 -mmmx -msse2 -msse -fomit-frame-pointer
>>    
>
> Thanks for sending the cpuinfo.py output. Comparing it's output
> and the numpy/distutils/fcompiler/gnu.py source, it seems that
> you are using some older version of numpy. The -march value should
> be 'nocona' for your case.
>
> So, try upgrade numpy from svn (or just copy the gnu.py file from
> svn to numpy installation directory).
>
> HTH,
> Pearu
>
> _______________________________________________
> SciPy-user mailing list
> [hidden email]
> http://projects.scipy.org/mailman/listinfo/scipy-user
>  

--
----------------------------------------------------------------------------
Christopher A. Edwards                          [hidden email]
Ocean Sciences Department                       phone: [phone]
University of California                        fax: [phone]
Santa Cruz, CA  95064  
----------------------------------------------------------------------------

_______________________________________________
SciPy-user mailing list
[hidden email]
http://projects.scipy.org/mailman/listinfo/scipy-user