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

Merge back fixes from ROCm 6.2 #530

Merged
merged 14 commits into from
Aug 9, 2024
Merged

Conversation

lawruble13
Copy link
Collaborator

No description provided.

umfranzw and others added 13 commits April 22, 2024 15:35
…OCm#483)

I ran the benchmark tuning code and scripts on the remaining
architectures in our GPU support matrix. This change updates
the config headers with the results.
…OCm#482)

This change allows device-side generators to be used inside of hipGraphs.
More specifically, you can wrap calls to rocrand_generate_* inside of a hipGraph. There are a few things to be aware of:
  - Generator creation (rocrand_create_generator), initialization (rocrand_initialize_generator), and destruction (rocrand_destroy_generator) must still happen outside the hipGraph.
  - After the generator is created, you may call API functions to set its seed, offset, and order.
  - After the generator is initialized (but before stream capture or manual graph creation begins), use rocrand_set_stream to set the stream the generator will use within the graph.
  - A generator's seed, offset, and stream may not be changed from within the hipGraph. Attempting to do so may result in unpredicable behaviour.
  - API calls for the poisson distribution (eg. rocrand_generate_poisson) are not yet supported inside of hipGraphs.

I've added a note to the changelog that mentions these details.
In addition to the changes necessary to support the behaviour described above, this change also:
- updates the changelog to alert the user to the restrictions mentioned above
- adds new unit test coverage to exercises generators and distributions within hipGraphs.
introduced by generic codepath for host and device generator

Co-authored-by: Nol Moonen <[email protected]>
…erate function (ROCm#495) (ROCm#497)

A compiler change caused the generate function to no longer be inlined which caused a performance regression. Force inlining resolves this.

Co-authored-by: Nara <[email protected]>
* organizing docs into subfolders

* update license and index

* Update LICENSE.txt

Co-authored-by: Nara <[email protected]>

* removing what-is-rocRAND

* Update Read the Docs configuration to use Python 3.10 and latest rocm-docs-core (ROCm#503)

* Bump urllib3 from 2.2.1 to 2.2.2 in /docs/sphinx (ROCm#505)

Bumps [urllib3](https://github.com/urllib3/urllib3) from 2.2.1 to 2.2.2.
- [Release notes](https://github.com/urllib3/urllib3/releases)
- [Changelog](https://github.com/urllib3/urllib3/blob/main/CHANGES.rst)
- [Commits](urllib3/urllib3@2.2.1...2.2.2)

---
updated-dependencies:
- dependency-name: urllib3
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

---------

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: Nara <[email protected]>
Co-authored-by: Sam Wu <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
HOTFIX: Correct CMake for TBB dependency
Copy link
Collaborator

@RobsonRLemos RobsonRLemos left a comment

Choose a reason for hiding this comment

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

@lawruble13 LGTM. Just please check the branch conflicts. Thanks.

@pramenku
Copy link

pramenku commented Aug 1, 2024

When we can expect it to be merged and promoted to Mainline. @lawruble13

@lawruble13
Copy link
Collaborator Author

Conflicts fixed, will merge once tests pass.

@lawruble13 lawruble13 merged commit 8edd91b into ROCm:develop Aug 9, 2024
23 checks passed
@lawruble13 lawruble13 deleted the mergeback-6.2 branch August 9, 2024 20:10
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.

8 participants