[Openmp-dev] OpenMP offload implicitly using streams

Ye Luo via Openmp-dev openmp-dev at lists.llvm.org
Sun Mar 17 12:26:46 PDT 2019

How to turn on streams when using OpenMP offload?
When different host threads individually start target regions (even not
using nowait). The offloaded computation goes to different CUDA streams and
may execute concurrently. This is currently available in XL.
With Clang, nvprof shows only the run only uses the default stream.
Is there a way to do that with Clang?
On the other hand,
nvcc has option --*default*-*stream per*-*thread*
I'm not familar with clang CUDA, is there a similar option?
Ye Luo, Ph.D.
Computational Science Division & Leadership Computing Facility
Argonne National Laboratory
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/openmp-dev/attachments/20190317/48af0a7e/attachment.html>

More information about the Openmp-dev mailing list