llvm dev - Jan 2016 - Executing OpenMP 4.0 code on Nvidia's GPU

Hi,

I was trying to execute an OpenMP 4.0 code on GPU with the offloading
supported now. I went through the steps described here
<https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus>.
But I get these linking errors during compilation:


LIBRARY_PATH=/home/ahmed/ahmed_home/openmp4/llvm/build/bin/../lib clang-3.8
-fopenmp -omptargets=nvptx64sm_30-nvidia-linux -g -O3 -std=c99 example.c -o
example
ptxas warning : Too big maxrregcount value specified 64, will be ignored
nvlink error   : Undefined reference to '__kmpc_kernel_init' in
'/tmp/example-87866c-36d3a6.cubin'
nvlink error   : Undefined reference to '__kmpc_for_static_init_4' in
'/tmp/example-87866c-36d3a6.cubin'
nvlink error   : Undefined reference to '__kmpc_for_static_fini' in
'/tmp/example-87866c-36d3a6.cubin'
clang-3.8: error: nvlink command failed with exit code 255 (use -v to see
invocation)


Any pointer, where the problem could be ?

Thanks a lot.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160119/5ce01ee4/attachment.html>

Hi,
Did you tried it on clang/LLVM built from trunk? Trunk version does not support
offloading yet (it is under development now). Use clang/LLVM from
clang-omp.github.com, as pointed in
<https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus>
https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus

Best regards,
Alexey Bataev
============Software Engineer
Intel Compiler Team

Hi,
Did you tried it on clang/LLVM built from trunk? Trunk version does not support
offloading yet (it is under development now). Use clang/LLVM from
clang-omp.github.com, as pointed in
<https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus>
https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus
20.01.2016 1:42, Ahmed ElTantawy via llvm-dev пишет:
Hi,

I was trying to execute an OpenMP 4.0 code on GPU with the offloading supported
now. I went through the steps described
here<https://parallel-computing.pro/index.php/9-cuda/43-openmp-4-0-on-nvidia-cuda-gpus>.
But I get these linking errors during compilation:


LIBRARY_PATH=/home/ahmed/ahmed_home/openmp4/llvm/build/bin/../lib clang-3.8
-fopenmp -omptargets=nvptx64sm_30-nvidia-linux -g -O3 -std=c99 example.c -o
example
ptxas warning : Too big maxrregcount value specified 64, will be ignored
nvlink error   : Undefined reference to '__kmpc_kernel_init' in
'/tmp/example-87866c-36d3a6.cubin'
nvlink error   : Undefined reference to '__kmpc_for_static_init_4' in
'/tmp/example-87866c-36d3a6.cubin'
nvlink error   : Undefined reference to '__kmpc_for_static_fini' in
'/tmp/example-87866c-36d3a6.cubin'
clang-3.8: error: nvlink command failed with exit code 255 (use -v to see
invocation)


Any pointer, where the problem could be ?

Thanks a lot.

-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.llvm.org/pipermail/llvm-dev/attachments/20160120/5f2f4e4a/attachment.html>