> Next step is to try icc instead of gcc, and if that works, try
> the latest MKL (10.0.3.020).
OK, either I've got a corrupted copy of MKL 10.0.3.020, or it has
a problem. Building with icc & MKL 10.0.1.014 works.
Erik, are you reading this? If so, roll back to MKL 10.0.014 and it
should work,
n some way, because
> something is likely to influence how MKL is initialized. But what
> exactly, I have no idea.
AFAIK, there is one MKL version for each release. The only
difference is the licensing.
Next step is to try icc instead of gcc, and if that works, try
the latest MKL (10.0.3.020).
Thanks,
-rex
--
I almost stole another tagline. I'm so ashamed.
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
when MKL
> > is used and you don't?
> I don't know. I would ask Intel about this error if the above does not
> work, maybe you did not install it correctly, or there was a bug in your
> version (my version is a bit more recent, I downloaded it a few days
> ago).
In your list post you show mkl/10.0.1.014/. I'm using 10.0.3.020,
but I've tried an older 10.0.X version with no better luck. BTW,
my build runs, i.e., I can run some programs that use NumPy.
I have posted to the Intel MKL forum. No responses yet.
-rex
I am not goin' to buy my kids an encyclopedia. Let them walk to school the way
I did.
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
David, how do these environment variables compare with yours? Are you sure
MKL is being used?
Adjusted for your local path, what does the ldd command below show?
ldd /usr/local/lib/python2.5/site-packages/numpy/linalg/lapack_lite.so
linux-gate.so.1 => (0xe000)
libmkl_lapack.so => /opt
David wrote:
> I have not tried with icc, but the following works for me with the
> last mkl (I have only tried numpy).
>
> [mkl]
> library_dirs = /home/david/intel/mkl/10.0.1.014/lib/32
> lapack_libs = mkl_lapack
> mkl_libs = mkl, guide
>
> (of course, adapt the library_dirs accordingly). All test
Intel examples, I think
(but could be wrong) that vml doesn't need to be explicitly
linked to get its functionality, but mkl_lapack does. This requires
changing:
class mkl_info(system_info):
section = 'mkl'
dir_env_var = 'MKL'
_lib_mkl = ['mkl','vml
> I am trying to build numpy with intel icc and mkl. I don't understand
> a lot of what I am doing.
Me, too. I have built it with icc & MKL several times in the past,
but cannot build the numpy svn with MKL now. I can build it with
icc and no MKL, and it passes all the tests with no errors.
I
that it would help. It did not.
I suggest that efforts to produce a product that ordinary 150
IQ mortals can install rather than add new features would grow the
user population faster. For example, distutils has been broken for
months for Intel MKL. I've posted 3x about it now.
-rex
--
This is the 3rd time I have reported this problem and a fix.
-rex
- Forwarded message from rex <[EMAIL PROTECTED]> -
Date: Fri, 9 Nov 2007 11:16:17 -0800
From: rex <[EMAIL PROTECTED]>
To: Discussion of Numerical Python
Subject: NumPy 1.04, MKL 10.0, & Intel 10.1 icc &a
uential.alibmkl_vml_p4m.s olibmkl_blacs.a
libmkl_gf.alibmkl_lapack.a libmkl_sequential.so
libmkl_vml_p4p.so libmkl_blacs_intelmpi20.a libmkl_gf.so
libmkl_lapack.so libmkl.so libmkl_vml_p4.so
libmkl_blacs_intelmpi.alibmkl_gnu_thread.a libmkl_p3.so
libmkl_solver.alibmkl_blacs_openmpi.a libmkl_gnu_thread.so
libmkl_p4m.so libmkl_solver_sequential.a libmkl_cdft.a
libmkl_ia32.a libmkl_p4p.so libmkl_vml_def.so
Thanks to all who have helped me with earlier versions.
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
y
>>> print numpy.__version__
1.0.1
>>> print scipy.__version__
0.5.2
>>> numpy.test(level=1,verbosity=2)
[...]
Ran 526 tests in 0.432s
OK
>>> scipy.test(level=1,verbosity=2)
[...]
Ran 1596 tests in 2.942s
OK
There doesn'
Charles R Harris <[EMAIL PROTECTED]> [2007-06-24 06:22]:
>
>
> On 6/23/07, rex <[EMAIL PROTECTED]> wrote:
>
> Stefan van der Walt <[EMAIL PROTECTED]> [2007-06-23 15:06]:
>>
>> On Sat, Jun 23, 2007 at 07:35:35PM +, John Ollinger wrote:
>&g
uo 32-bit
Kernel 2.6.18.2-34-default
gcc version 4.1.2 20061115 (prerelease) (SUSE Linux)
Python 2.5 (r25:51908, Nov 27 2006
print N.__version__
1.0.4.dev3868
python -c 'import numpy; numpy.test(level=1,verbosity=2)'
[...]
Ran 590 tests in 0.473s
OK
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
uctions
from source (both for gcc and Intel software) on the SciPy wiki, and
your corrections are helpful. Thanks.
-rex
--
"Men occasionally stumble over the truth, but most of them pick themselves
up and hurry off as if nothing ever happened."--Winston Churchill
___
ow works:
cc_exe = 'icc -g -fomit-frame-pointer -xT -fast'
After these changes, NumPy and SciPy built successfully, but
scipy.test() returns a number of errors. The error reports are on the
SciPy-dev list at the end of the thread: "Compiling scipy with Intel ifort &
MKL"
Perhaps this will help someone...
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
o do
simple edits in MoinMoin.)
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
rex <[EMAIL PROTECTED]> [2007-06-09 11:02]:
> I changed the cc_exe line in
> numpy-1.0.3/numpy/distutils/intelccompiler.py to:
>
> cc_exe = 'icc -msse3 -xP -fast' #Core 2 Duo
>
> >From the numpy-1.03 directory executed:
>
> python setup.py config -
l/mkl/9.1/lib/64
I don't understand where the '-lmkl_lapack32' and '-lmkl_lapack64'
are coming from. I looked in lapack_litemodule.c, but didn't see
anything about lapack.
Thanks for any pointers,
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
David Cournapeau <[EMAIL PROTECTED]> [2007-06-09 06:35]:
> rex wrote:
> >
> > I've been using SUSE since version
> > 6.4, and it's always a battle to get Numpy/SciPy running.
>
> This is really Suse fault, honnestly. As Fedora, they have mostly broken
&
David Cournapeau <[EMAIL PROTECTED]> [2007-06-08 22:49]:
> rex wrote:
> > I don't know if this is the appropriate place for this, but thanks for
> > any pointers to what the problem is.
> >
> > -rex
> >
> > Traceback (most recent call last
I don't know if this is the appropriate place for this, but thanks for
any pointers to what the problem is.
-rex
deserv:/ # smart install python-numpy
Loading cache...
Updating cache...
## [100%]
Comp
)
MonteCarlo: Mflops: 115.70
Sparse matmult Mflops: 825.81(N=10, nz=100)
LU Mflops: 1010.10(M=1000, N=1000)
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
Andrew Straw <[EMAIL PROTECTED]> [2007-04-18 13:22]:
> rex wrote:
> > If your use is entirely non-commercial you can use Intel's MKL with
> > built-in optimized BLAS and LAPACK and avoid the need for ATLAS.
> >
> Just to clarify, my understanding is that if y
looking for a broad performance test (something like
Scimark, say).
The best compiler flags I've found are: -fast -parallel
In some cases -funroll-loops and -fno-alias helps.
-rex
--
Time flies like wind. Fruit flies like pears.
___
Numpy-discussi
no-prec-div -static)
-funroll-loops
-fno-alias
-parallel
I don't know if there are obviously better choices for the Core 2 Duo.
I'd like to run a more comprehensive benchmark, say Scimark translated
from C to Python/NumPy.
http://math.nist.gov/scimark
a 500x500 grid
numeric took 1.57 seconds
slow (100 iterations) took 154.29 seconds
slow with Psyco (100 iterations) took 119.88 seconds
Python compiled with gcc takes 101 times longer to run this benchmark
than Python/NumPy/icc does.
The C++ version compiled with gcc 4.1.2 runs in .19 seconds.
-r
rex <[EMAIL PROTECTED]> [2007-04-16 15:53]:
> I'm about to build numpy using Intel's MKL 9.1 beta and want to compare
> it with the version I built using MKL 8.1. Is the LINPACK
> benchmark the most appropriate?
I'm buried in responses. Not.
A well-known benchmar
I'm about to build numpy using Intel's MKL 9.1 beta and want to compare
it with the version I built using MKL 8.1. Is the LINPACK
benchmark the most appropriate?
Thanks,
-rex
--
Pollytheism: n., the belief that there are many gods, all of th
crease on Pybench was ~49%, even
before compiling Python with icc. My post about it was on 25 Jan, and
has subject: Compiling Python with icc
I haven't tried the FFT lib.
-rex
--
"Experience is a wonderful thing. It enables you to recognize a mis
to be updated for 1+ gigaflop systems.
http://psyco.sourceforge.net/
-rex
--
"I have always wished that my computer would be as easy to use as my
telephone. My wish has come true. I no longer know how to use my telephone"
--Bjorne Stroustrup (originat
enchmark.
> Congrats on completing the gauntlet,
Thank. It's the 2nd time. I eventually succeed with an earlier version
as well, thanks to Travis.
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
tion [2,6] fraction [lam,lam] largest% total[mean,mean]
10 0.01502 0.03764 3895 3764
10 0.01513 0.03841 3841 3841
10 0.01511 0.03753 3810 3753
10 0.01577 0.03766 3873 3766
10 0.01541 0.0373 3963 3730
10 0.01586 0.03862 3912 3862
10 0.01552 0.03785 3870 3785
10 0.0
Robert Kern <[EMAIL PROTECTED]> [2007-01-24 03:22]:
> rex wrote:
> >>>> import numpy
> >>>> print numpy
> >
> >
> > What am I to make of this? Is it the rpm numpy or is it the numpy I
> > built using the Intel compiler and MKL?
&g
David Cournapeau <[EMAIL PROTECTED]> [2007-01-23 23:40]:
> rex wrote:
> > Robert Kern <[EMAIL PROTECTED]> [2007-01-23 22:18]:
> >> You need to install the development package for Python. Usually it's named
> >> something like python2.5-devel.
> >
t appears the new version is not
being used. I expect a significant speed difference using the Intel
compiler and MKL on a Core 2 Duo.
Why is this so difficult?
Thanks,
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.
r Python2.5 are:
./site-packages/numpy/doc/swig/Makefile
./site-packages/numpy/doc/pyrex/Makefile
Help anyone? I hope I don't have to build Python2.5 from source to build NumPy.
Thanks,
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
Has anyone done any benchmarks to compare Numpy's speed with these two
compilers on a Core 2 Duo?
Thanks,
-rex
___
Numpy-discussion mailing list
Numpy-discussion@scipy.org
http://projects.scipy.org/mailman/listinfo/numpy-discussion
37 matches
Mail list logo
