Optimize multicontact kernel parallelization for MI300X#41
Open
nilapate wants to merge 2 commits intoamd-integrationfrom
Open
Optimize multicontact kernel parallelization for MI300X#41nilapate wants to merge 2 commits intoamd-integrationfrom
nilapate wants to merge 2 commits intoamd-integrationfrom
Conversation
nilapate
force-pushed
the
nilapate/multicontact-kernel-opt
branch
4 times, most recently
from
1f18a3b to
1ec057e
Compare
Collaborator
|
/run-ci |
chien-an-chen
approved these changes
Apr 28, 2026
nilapate
force-pushed
the
nilapate/multicontact-kernel-opt
branch
from
3e7df9b to
c99cc64
Compare
Collaborator
|
Reviewed the changes - latest CI throughput - |
rtmadduri
approved these changes
Apr 28, 2026
yaoliu13
requested changes
Apr 29, 2026
Collaborator
yaoliu13
left a comment
yaoliu13
left a comment
There was a problem hiding this comment.
This is no longer based on the latest amd-integration: nilapate/multicontact-kernel-opt...ROCm:Genesis:amd-integration
Please read Confluence page "Genesis PR Review Process"
…elization multiplier from 64 to 128 Changes: - Modified genesis/engine/solvers/rigid/collider/collider.py (lines 218-230) - Added logic to use mult=128 for multicontact kernel on HIP devices Tested on MI300X hardware.
Testing showed mult=256 is the optimal sweet spot. Higher values (mult=384) show diminishing returns. Signed-off-by: Patel, Nilaykumar K <nilapate@amd.com>
nilapate
force-pushed
the
nilapate/multicontact-kernel-opt
branch
from
c99cc64 to
65492b1
Compare
Author
Rebased on latest main |
Collaborator
|
/run-ci |
1 similar comment
Collaborator
|
/run-ci |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Optimize the multicontact narrowphase kernel by increasing the parallelization multiplier from 64 to 128.
Changes:
genesis/engine/solvers/rigid/collider/collider.py(lines 218-230)How Has This Been Tested?
Test Configuration: