[theano-users] New GPUArray Related Crash

[John Coolidge]

Theano version 0.90 is crashing python on my system (Ubuntu 14.04) with the
following error message: "Process finished with exit code 139 (interrupted
by signal 11: SIGSEGV)" . This crash happens when I do theano.test().
Theano also similarly crashes when I call theano.function() in some
personal code with device=cuda. The code works fine when device=gpu or
device=cpu. Curiously, running the GPU test code from the documentation
with device=cuda works fine. It may be useful to know that theano seems to
successfully find my graphics card: "Mapped name None to device cuda:
GeForce GTX 870M (0000:01:00.0)." Is there some way to get more detailed
information on the error? Any idea what could be causing it?

[John Coolidge]

I should add, I do not have cudnn installed and pygpu is version 0.6.8.

Post by John Coolidge

Post by John Coolidge
theano.test()

Theano version 0.9.0.dev-c697eeab84e5b8a74908da654b66ec9eca4f1291
theano is installed in
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/theano
NumPy version 1.11.3
NumPy relaxed strides checking option: False
NumPy is installed in
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/numpy
Python version 2.7.12 |Continuum Analytics, Inc.| (default, Jul 2 2016,
17:42:40) [GCC 4.4.7 20120313 (Red Hat 4.4.7-1)]
nose version 1.3.7
Can not use cuDNN on context None: cannot compile with cuDNN. We got this
error:
/tmp/try_flags_I1jSI4.c:4:19: fatal error: cudnn.h: No such file or
directory
#include <cudnn.h>
^
compilation terminated.

Mapped name None to device cuda: GeForce GTX 870M (0000:01:00.0)
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/theano/misc/pycuda_init.py:35:
UserWarning: PyCUDA import failed in theano.misc.pycuda_init
warnings.warn("PyCUDA import failed in theano.misc.pycuda_init")
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/theano/sandbox/gpuarray/__init__.py:10:
UserWarning: theano.sandbox.gpuarray has been moved to theano.gpuarray.
Please update your code and pickles. If the warning persists, clear
theano's cache ('$theano/bin/theano-cache clear').
warnings.warn(message)
EE...........E........./home/john/anaconda2/envs/investing/lib/python2.7/site-packages/numpy/lib/nanfunctions.py:343:
RuntimeWarning: All-NaN slice encountered
warnings.warn("All-NaN slice encountered", RuntimeWarning)
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/numpy/lib/nanfunctions.py:236:
RuntimeWarning: All-NaN axis encountered
warnings.warn("All-NaN axis encountered", RuntimeWarning)
.E......................................../home/john/anaconda2/envs/investing/lib/python2.7/site-packages/theano/gof/vm.py:851:
UserWarning: CVM does not support memory profile, using Stack VM.
'CVM does not support memory profile, using Stack VM.')
...........SS.............0.796523204745
0.796523204745
0.796523204745
0.796523204745
.........................................................................rm:
cannot remove
‘/home/john/.theano/compiledir_Linux-3.13--generic-x86_64-with-debian-jessie-sid-x86_64-2.7.12-64/optimized_graphs.pkl’:
No such file or directory
.................................................................................E/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/theano/gof/vm.py:854:
UserWarning: LoopGC does not support partial evaluation, using Stack VM.
'LoopGC does not support partial evaluation, '
.....EEEESegmentation fault (core dumped)

Post by John Coolidge
Theano version 0.90 is crashing python on my system (Ubuntu 14.04) with
the following error message: "Process finished with exit code 139
(interrupted by signal 11: SIGSEGV)" . This crash happens when I do
theano.test(). Theano also similarly crashes when I call theano.function()
in some personal code with device=cuda. The code works fine when
device=gpu or device=cpu. Curiously, running the GPU test code from the
documentation with device=cuda works fine. It may be useful to know that
theano seems to successfully find my graphics card: "Mapped name None to
device cuda: GeForce GTX 870M (0000:01:00.0)." Is there some way to get
more detailed information on the error? Any idea what could be causing it?

[John Coolidge]

Turns out another poster had the same problem and managed to solve it here
https://groups.google.com/forum/#!topic/theano-users/vMeGLurkSj4 .

Post by John Coolidge
I should add, I do not have cudnn installed and pygpu is version 0.6.8.

Post by John Coolidge

Post by John Coolidge
theano.test()

Theano version 0.9.0.dev-c697eeab84e5b8a74908da654b66ec9eca4f1291
theano is installed in
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/theano
NumPy version 1.11.3
NumPy relaxed strides checking option: False
NumPy is installed in
/home/john/anaconda2/envs/investing/lib/python2.7/site-packages/numpy
Python version 2.7.12 |Continuum Analytics, Inc.| (default, Jul 2 2016,
17:42:40) [GCC 4.4.7 20120313 (Red Hat 4.4.7-1)]
nose version 1.3.7
Can not use cuDNN on context None: cannot compile with cuDNN. We got this
/tmp/try_flags_I1jSI4.c:4:19: fatal error: cudnn.h: No such file or
directory
#include <cudnn.h>
^
compilation terminated.
Mapped name None to device cuda: GeForce GTX 870M (0000:01:00.0)
UserWarning: PyCUDA import failed in theano.misc.pycuda_init
warnings.warn("PyCUDA import failed in theano.misc.pycuda_init")
UserWarning: theano.sandbox.gpuarray has been moved to theano.gpuarray.
Please update your code and pickles. If the warning persists, clear
theano's cache ('$theano/bin/theano-cache clear').
warnings.warn(message)
RuntimeWarning: All-NaN slice encountered
warnings.warn("All-NaN slice encountered", RuntimeWarning)
RuntimeWarning: All-NaN axis encountered
warnings.warn("All-NaN axis encountered", RuntimeWarning)
UserWarning: CVM does not support memory profile, using Stack VM.
'CVM does not support memory profile, using Stack VM.')
...........SS.............0.796523204745
0.796523204745
0.796523204745
0.796523204745
cannot remove
No such file or directory
UserWarning: LoopGC does not support partial evaluation, using Stack VM.
'LoopGC does not support partial evaluation, '
.....EEEESegmentation fault (core dumped)

Post by John Coolidge
Theano version 0.90 is crashing python on my system (Ubuntu 14.04) with
the following error message: "Process finished with exit code 139
(interrupted by signal 11: SIGSEGV)" . This crash happens when I do
theano.test(). Theano also similarly crashes when I call theano.function()
in some personal code with device=cuda. The code works fine when
device=gpu or device=cpu. Curiously, running the GPU test code from the
documentation with device=cuda works fine. It may be useful to know that
theano seems to successfully find my graphics card: "Mapped name None to
device cuda: GeForce GTX 870M (0000:01:00.0)." Is there some way to get
more detailed information on the error? Any idea what could be causing it?