On Thu, Mar 21, 2013 at 11:23:39AM +0200, Pekka Jääskeläinen wrote:
> * Can you add some "smoke tests" for the nvptx under 'tests' (similarly as tce
> or cellspu) so we can quickly test we didn't completely break during
> development
> nvptx? I have an older NVIDIA card which I can test it with.
After fixing the endianness of the cuda device, the example1 test
passes. example2 fails due to unsupported __local kernel argument,
and scalarwave fails due to unsupported __constant argument.
Support for __local kernel arguments could probably be implemented
by manually partitioning a dynamically allocated shared memory block
into segments for each __local argument.
For constant memory, the CUDA driver API provides a function
cuModuleGetGlobal() to get the pointer to a module-scope __constant
array. One or multiple such module-scope arrays could be initialised
with cuMemcpyDtoD at kernel execution to provide memory for buffers
passed as __constant arguments.
I am skeptical of an NVPTX/CUDA device driver at this stage.
The usability of this driver depends entirely on the quality of the
NVPTX backend. When serious problems arise with code generation, who
is to say that, once the term OpenCL is mentioned, the responsible
maintainers at NVIDIA with the knowledge to fix these issues won't
react with the same silence surrounding NVIDIA's OpenCL driver?
(If only the AMD open source compute driver was progressing faster…)
Peter
diff --git a/lib/CL/devices/cuda/pocl-cuda.h b/lib/CL/devices/cuda/pocl-cuda.h
index 379ca28..99ff8bb 100644
--- a/lib/CL/devices/cuda/pocl-cuda.h
+++ b/lib/CL/devices/cuda/pocl-cuda.h
@@ -88,7 +88,7 @@ GEN_PROTOTYPES (cuda)
CL_FALSE, /* error_correction_support */ \
CL_TRUE, /* host_unified_memory */ \
0, /* profiling_timer_resolution */ \
- CL_FALSE, /* endian_little */ \
+ CL_TRUE, /* endian_little */ \
CL_TRUE, /* available */ \
CL_TRUE, /* compiler_available */ \
CL_EXEC_KERNEL, /*execution_capabilities */ \
------------------------------------------------------------------------------
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
_______________________________________________
pocl-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/pocl-devel
