Kieffers-MacBook-Pro:test kieffer$ ./testOpenCL.py -i
WARNING:xsdimage:lxml library is probably not part of your python installation: disabling xsdimage format
WARNING:root:Exception No module named fftw3: FFTw3 not available. Falling back on Scipy
INFO:testOpenCL:Testing histogram-based algorithm (forward-integration)
ERROR:testOpenCL:No suitable GPU OpenCL device found
INFO:testOpenCL:I found a suitable device CPU (0, 0): Apple Intel(R) Core(TM)2 Duo CPU T9550 @ 2.66GHz
/Library/Python/2.7/site-packages/pyopencl-2012.2-py2.7-macosx-10.7-intel.egg/pyopencl/init.py:32: CompilerWarning: From-source build succeeded, but resulted in non-empty logs:
Build on <pyopencl.Device 'Intel(R) Core(TM)2 Duo CPU T9550 @ 2.66GHz' on 'Apple' at 0xffffffff> succeeded, but said:

:207:26: warning: comparison of integers of different signs: 'uint' (aka 'unsigned int') and 'int'
if(tid==0 && blockId < Ngroups){
~~~~~~~ ^ ~~~~~~~
:207:26: warning: comparison of integers of different signs: 'uint' (aka 'unsigned int') and 'int'
if(tid==0 && blockId < Ngroups){
~~~~~~~ ^ ~~~~~~~
:357:13: warning: expression result unused
atom_add(&binarray[cbin],convert0);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/System/Library/Frameworks/OpenCL.framework/Resources/cl_kernel.h:2694:23: note: instantiated from:
#define atom_add(X,Y) __CLFN_A2(X, Y, add)
^
/System/Library/Frameworks/OpenCL.framework/Resources/cl_kernel.h:2684:26: note: instantiated from:
#define __CLFN_A2(x,y,R) __builtin_overload(2, x, y,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~
:358:13: warning: expression result unused
atom_add(&histogram[cbin],convert1);
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
/System/Library/Frameworks/OpenCL.framework/Resources/cl_kernel.h:2694:23: note: instantiated from:
#define atom_add(X,Y) __CLFN_A2(X, Y, add)
^
/System/Library/Frameworks/OpenCL.framework/Resources/cl_kernel.h:2684:26: note: instantiated from:
#define __CLFN_A2(x,y,R) __builtin_overload(2, x, y,
^~~~~~~~~~~~~~~~~~~~~~~~~~~~~

warn(text, CompilerWarning)
EINFO:testOpenCL:Testing LUT-based algorithm (backward-integration)
INFO:testOpenCL:I found a suitable device GPU (0, 2): Apple GeForce 9600M GT
INFO:testOpenCL:For image Pilatus1M.edf; speed up is 2.632x; rate is 35.151 Hz
INFO:testOpenCL:For image halfccd.edf; speed up is 2.874x; rate is 16.563 Hz
INFO:testOpenCL:For image Frelon2k.edf; speed up is 3.277x; rate is 8.819 Hz
INFO:testOpenCL:For image Pilatus6M.cbf; speed up is 2.748x; rate is 5.731 Hz
INFO:testOpenCL:I found a suitable device CPU (0, 0): Apple Intel(R) Core(TM)2 Duo CPU T9550 @ 2.66GHz

E

ERROR: test_OpenCL (main.test_mask)

Traceback (most recent call last):
File "./testOpenCL.py", line 81, in test_OpenCL
res = ai.xrpd_OpenCL(data, 1000, devicetype="all", platformid=ids[0], deviceid=ids[1], useFp64=True)
File "/Users/kieffer/workspace/pyFAI/build/lib.macosx-10.7-intel-2.7/pyFAI/azimuthalIntegrator.py", line 548, in xrpd_OpenCL
if integr.loadTth(pos0, delta_pos0, pos0_min, pos0_max):
File "/Users/kieffer/workspace/pyFAI/build/lib.macosx-10.7-intel-2.7/pyFAI/ocl_azim.py", line 355, in loadTth
get_spans = self._cl_program.get_spans(self._queue, self.wdim_data, self.tdim, *self._cl_kernel_args["get_spans"])
File "/Library/Python/2.7/site-packages/pyopencl-2012.2-py2.7-macosx-10.7-intel.egg/pyopencl/init.py", line 498, in kernel_call
global_offset, wait_for, g_times_l=g_times_l)
LogicError: clEnqueueNDRangeKernel failed: invalid work group size

ERROR: test_OpenCL_LUT (main.test_mask)

Traceback (most recent call last):
File "./testOpenCL.py", line 104, in test_OpenCL_LUT
res = ai.xrpd_LUT_OCL(data, 1000, devicetype="all", platformid=ids[0], deviceid=ids[1])
File "/Users/kieffer/workspace/pyFAI/build/lib.macosx-10.7-intel-2.7/pyFAI/azimuthalIntegrator.py", line 801, in xrpd_LUT_OCL
I, J, K = self._ocl_lut_integr.integrate(data, solidAngle=solid_angle_array, solidAngle_checksum=solid_angle_crc, dummy=dummy, delta_dummy=delta_dummy)
File "/Users/kieffer/workspace/pyFAI/build/lib.macosx-10.7-intel-2.7/pyFAI/ocl_azim_lut.py", line 265, in integrate
self._program.corrections(self._queue, ((self.size + 512 - 1) & ~(512 - 1), 1, 1), (512, 1, 1), *self._cl_kernel_args["corrections"]).wait()
File "/Library/Python/2.7/site-packages/pyopencl-2012.2-py2.7-macosx-10.7-intel.egg/pyopencl/init.py", line 498, in kernel_call
global_offset, wait_for, g_times_l=g_times_l)
LogicError: clEnqueueNDRangeKernel failed: invalid work group size

Ran 2 tests in 22.489s

FAILED (errors=2)

from pyfai.

Apple_OpenCL on CPU only accepts work group size =1 (even when it advertises a max work group size = 1024)

from pyfai.

Python 2.7.1 (r271:86832, Jul 31 2011, 19:30:53)
Type "copyright", "credits" or "license" for more information.

IPython 0.14.dev -- An enhanced Interactive Python.
? -> Introduction and overview of IPython's features.
%quickref -> Quick reference.
help -> Python's own help system.
object? -> Details about 'object', use 'object??' for extra details.

In [1]: import pyopencl

In [2]: ctx=pyopencl.create_some_context()
Choose device(s):
[0] <pyopencl.Device 'Intel(R) Core(TM)2 Duo CPU T9550 @ 2.66GHz' on 'Apple' at 0xffffffff>
[1] <pyopencl.Device 'GeForce 9400M' on 'Apple' at 0x1022600>
[2] <pyopencl.Device 'GeForce 9600M GT' on 'Apple' at 0x2022600>
Choice, comma-separated [0]:2
Set the environment variable PYOPENCL_CTX='2' to avoid being asked again.

In [3]: queue=pyopencl.CommandQueue(ctx)

RuntimeError Traceback (most recent call last)
in ()
----> 1 queue=pyopencl.CommandQueue(ctx)

RuntimeError: CommandQueue failed: device not available

from pyfai.