transpose.py with ati-stream-sdk
by Achim Gottinger
Hi,
I have a question about the transpose.py example. I tried to run it with
the ati-stream-sdk. First I had to remove the unused pycuda dependency
in line 166 (worked fine with nvidia libOpenGL afterwards). If I run the
test with the ati-stream-sdk libOpenCL I get this error.
benchmarking Naive 6304
benchmarking Naive 6880
benchmarking Naive 7488
benchmarking WithLocal 1024
Traceback (most recent call last):
File "transpose.py", line 217, in <module>
benchmark_transpose()
File "transpose.py", line 176, in benchmark_transpose
method = cls(ctx)
File "transpose.py", line 86, in __init__
"""% {"block_size": block_size}).build().transpose
File
"/root/pyopencl-test/pyopencl-0.91.3/build/lib.linux-x86_64-2.5/pyopencl/__init__.py",
line 130, in program_build
"Build on %s:\n\n%s" % (dev, log) for dev, log in build_logs))
pyopencl.RuntimeError: clBuildProgram failed: invalid operation
Build on <pyopencl.Device 'Intel(R) Core(TM) i5 CPU 750 @
2.67GHz' at 0x125fa96>:
/tmp/OCLVoJzNQ.cl(7): error: kernel must return void; pointer argument must
point to addrSpace global, local, or constant
void transpose(
^
1 error detected in the compilation of "/tmp/OCLVoJzNQ.cl".
Does this mean using local memory does not work with the ati-stream-sdk?
regards
achim~
9 years, 10 months
Nvidia GTC: PyCUDA talk, Meetup
by Andreas Klöckner
Hi all,
If you are attending Nvidia's GPU Technology Conference next week, there are
two things I'd like to point out:
- I'll be giving a talk about PyCUDA on Friday, October 2 at 2pm, where I'll
both introduce PyCUDA and talk about some exciting new developments. The talk
will be 50 minutes in length, and I'd be happy to see you there.
- Also, I'd like to propose a PyCUDA meetup on Thursday, October 1 at noon.
(ie. lunchtime) I'll be hanging out by the "Terrace" seminar room around that
time. I'm looking forward to meeting some of you in person.
See you next week,
Andreas
9 years, 10 months
[PATCH] don't ignore OpenCL compiler parameters
by Michal Januszewski
Currently, any OpenCL compiler options passed to the build() method of a
Program object are ignored, as they are not passed down to the C++
wrapper code. Fix that by passing them correctly to _build().
diff --git a/pyopencl/__init__.py b/pyopencl/__init__.py
index e4d3d7b..7a736c9 100644
--- a/pyopencl/__init__.py
+++ b/pyopencl/__init__.py
@@ -114,7 +114,7 @@ def _add_functionality():
def program_build(self, options="", devices=None):
try:
- self._build()
+ self._build(options=options, devices=devices)
except Exception, e:
build_logs = []
for dev in self.devices:
9 years, 10 months
