[AMD] Redesign stream pipeliner LDS layout selection logic #8053
Conversation
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
0ede830 to
5e05ad2
Compare
5e05ad2 to
3e7959d
Compare
74b01ce to
8fc179f
Compare
AlexAUT
reviewed
Sep 5, 2025
antiagainst
requested changes
Sep 5, 2025
Collaborator
|
Can we also add a new mimimal lit test with different users to check the new logic? |
antiagainst
approved these changes
Sep 8, 2025
antiagainst
approved these changes
Sep 8, 2025
yiqian1
pushed a commit
to yiqian1/triton
that referenced
this pull request
Sep 9, 2025
…ng#8053) This commit adapts the LDS layout selection logic in Stream Pipeliner so that we pick a common swizzled shared memory layout with vecSize = max kWidth of all users.
ZelboK
pushed a commit
to ZelboK/triton
that referenced
this pull request
Sep 9, 2025
…ng#8053) This commit adapts the LDS layout selection logic in Stream Pipeliner so that we pick a common swizzled shared memory layout with vecSize = max kWidth of all users.
jayfurmanek
pushed a commit
to ROCm/triton
that referenced
this pull request
Sep 26, 2025
…ng#8053) This commit adapts the LDS layout selection logic in Stream Pipeliner so that we pick a common swizzled shared memory layout with vecSize = max kWidth of all users.
This was referenced Nov 13, 2025
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 5, 2026
…out selection logic (#8053)' Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 5, 2026
…out selection logic (#8053)' Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 5, 2026
…out selection logic (#8053)' Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 10, 2026
…out selection logic (#8053)' (facebookexperimental#847) Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Reviewed By: stashuk-olek Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 10, 2026
…out selection logic (#8053)' (facebookexperimental#847) Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Reviewed By: stashuk-olek Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 10, 2026
…out selection logic (#8053)' (facebookexperimental#847) Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Reviewed By: stashuk-olek Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 10, 2026
…out selection logic (#8053)' (facebookexperimental#847) Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Reviewed By: stashuk-olek Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 10, 2026
…out selection logic (#8053)' Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Differential Revision: D92336368
agron911
added a commit
to agron911/triton
that referenced
this pull request
Feb 10, 2026
…out selection logic (#8053)' (facebookexperimental#847) Summary: This is a cherry-pick of an upstream PR: triton-lang/triton#8053 Upstream commit message: ``` > [AMD] Redesign stream pipeliner LDS layout selection logic (#8053) > This commit adapts the LDS layout selection logic in Stream Pipeliner so > that we pick a common swizzled shared memory layout with vecSize = max > kWidth of all users. ``` ***Do not remove the following line from this commit*** Reactor Cherry-pick Revision: 8b792c8 --- This diff was generated by running: ``` buck run fbcode//triton/tools/reactor:reactor -- cherrypick --num-commits 3 ``` Reviewed By: stashuk-olek Differential Revision: D92336368
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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.
This commit adapts the LDS layout selection logic in Stream Pipeliner so that we pick a common swizzled shared memory layout with vecSize = max kWidth of all users.
Fixes https://github.com/ROCm/triton-internal/issues/1089.
Fixes https://github.com/ROCm/triton-internal/issues/1155.