-
Notifications
You must be signed in to change notification settings - Fork 74k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
OpenMP/OpenACC support for tensorflow #12434
Comments
@andydavis1 and @benoitsteiner looked at it a while back. There were two issues
|
Thank you, Yaroslav. Closing for now. |
In the exploration of this issue, was attention paid to using Eigen in multi-threaded applications? Please see the note on calling Eigen::initParallel() before creating application threads, in the following: |
@yaroslavvb Could OpenACC be used to optimize parallel tasks like matmuls and convolutions (IDK if you use a library) on a wider range of accelerators (like AMD GPUs), that the CUDA kernels and CPU-oriented code are not targeting? I think that this could complement the existing Tensorflow primitives for parallelization. |
…d test. Imported from GitHub PR openxla/xla#12434 Copybara import of the project: -- 723c9bb29adfcc33c015b74f90ce8024c2f79255 by Ilia Sergachev <isergachev@nvidia.com>: [GPU] Fix OSS compilation problems in a previously disabled test. Merging this change closes #12434 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#12434 from openxla:fix_triton_test 723c9bb29adfcc33c015b74f90ce8024c2f79255 PiperOrigin-RevId: 633509548
…d test. Imported from GitHub PR openxla/xla#12434 Copybara import of the project: -- 723c9bb29adfcc33c015b74f90ce8024c2f79255 by Ilia Sergachev <isergachev@nvidia.com>: [GPU] Fix OSS compilation problems in a previously disabled test. Merging this change closes #12434 FUTURE_COPYBARA_INTEGRATE_REVIEW=openxla/xla#12434 from openxla:fix_triton_test 723c9bb29adfcc33c015b74f90ce8024c2f79255 PiperOrigin-RevId: 633509548
…d test. Imported from GitHub PR openxla/xla#12434 Copybara import of the project: -- 723c9bb29adfcc33c015b74f90ce8024c2f79255 by Ilia Sergachev <isergachev@nvidia.com>: [GPU] Fix OSS compilation problems in a previously disabled test. Merging this change closes #12434 PiperOrigin-RevId: 633533757
Is there any effort in porting the Tensorflow using OpenMP or OpenACC?
The text was updated successfully, but these errors were encountered: