Re-enable lds_barrier on RDNA4 #21922
Merged
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
4c466d6 to
e143d8a
Compare
Member
|
@krzysz00 can you update the PR description? there should be no miscompiles now |
Contributor
Author
|
Updated |
kuhar
approved these changes
Sep 27, 2025
e143d8a to
e3c89b1
Compare
Contributor
Author
|
Going to wait for one more integrate on this one |
e3c89b1 to
7767486
Compare
krzysz00
pushed a commit
that referenced
this pull request
Oct 10, 2025
Reverts #21922. This PR re-introduced matmul failures on rdna4.
krzysz00
added a commit
to krzysz00/iree
that referenced
this pull request
Oct 15, 2025
This reverts commit cdce159. This re-applies iree-org#21922. This reverts iree-org#22278. The changes in iree-org#22198 appear to have solved the problem.
weidel-p
pushed a commit
to weidel-p/iree
that referenced
this pull request
Oct 21, 2025
Now that `amdgpu.lds_barrier` emits the correct fences, we can restore usage of `lds_barrier` and not need the `__syncthreads`-type everything-fence Signed-off-by: Philipp <[email protected]>
weidel-p
pushed a commit
to weidel-p/iree
that referenced
this pull request
Oct 21, 2025
Reverts iree-org#21922. This PR re-introduced matmul failures on rdna4. Signed-off-by: Philipp <[email protected]>
weidel-p
pushed a commit
to weidel-p/iree
that referenced
this pull request
Oct 21, 2025
…2326) This reverts commit cdce159. This re-applies iree-org#21922. This reverts iree-org#22278. The changes in iree-org#22198 appear to have solved the problem. ci-extra:test_torch Signed-off-by: Philipp <[email protected]>
mischirmer
pushed a commit
to mischirmer/iree
that referenced
this pull request
Nov 24, 2025
Reverts iree-org/iree#21922. This PR re-introduced matmul failures on rdna4.
pstarkcdpr
pushed a commit
to pstarkcdpr/iree
that referenced
this pull request
Nov 28, 2025
Now that `amdgpu.lds_barrier` emits the correct fences, we can restore usage of `lds_barrier` and not need the `__syncthreads`-type everything-fence
pstarkcdpr
pushed a commit
to pstarkcdpr/iree
that referenced
this pull request
Nov 28, 2025
Reverts iree-org#21922. This PR re-introduced matmul failures on rdna4.
pstarkcdpr
pushed a commit
to pstarkcdpr/iree
that referenced
this pull request
Nov 28, 2025
…2326) This reverts commit cdce159. This re-applies iree-org#21922. This reverts iree-org#22278. The changes in iree-org#22198 appear to have solved the problem. ci-extra:test_torch
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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.
Now that
amdgpu.lds_barrieremits the correct fences, we can restore usage oflds_barrierand not need the__syncthreads-type everything-fence