Skip to content
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

Implement SYCL Random #3849

Merged
merged 4 commits into from
Mar 15, 2021
Merged

Implement SYCL Random #3849

merged 4 commits into from
Mar 15, 2021

Conversation

masterleinad
Copy link
Contributor

We just run through the whole lock array starting from zero on each thread which is likely prohibitively expensive but should at least work.

@masterleinad masterleinad marked this pull request as ready for review March 13, 2021 16:22
core/src/SYCL/Kokkos_SYCL_Instance.cpp Outdated Show resolved Hide resolved
algorithms/unit_tests/TestSYCL.cpp Outdated Show resolved Hide resolved
algorithms/src/Kokkos_Random.hpp Show resolved Hide resolved
Copy link
Member

@crtrott crtrott left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks okish, one question is the the compute units. If it actually returns what most people understand under compute units it would be CUs, i.e. only 80 for V100, instead of 80*2048.

core/src/SYCL/Kokkos_SYCL_Instance.cpp Outdated Show resolved Hide resolved
@masterleinad
Copy link
Contributor Author

is max_compute_units correct? Or does that provide the number of CUs (which would only be 80, instead of 80*2048).

The concurrency computed here is 1024*2*80 for NVIDIA GPUs.

@crtrott crtrott merged commit 8a41982 into kokkos:develop Mar 15, 2021
@masterleinad masterleinad deleted the sycl_random branch March 15, 2021 23:32
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

4 participants