Fix a bug in fastPath condition #482
Merged
Conversation
This file contains 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
zhanglx13
commented
Jan 24, 2024
Comment on lines
-331
to
-334
| auto stride0 = findConstValue(smemObj.strides[0]); | ||
| auto stride1 = findConstValue(smemObj.strides[1]); | ||
| auto offset0 = findConstValue(smemObj.offsets[0]); | ||
| auto offset1 = findConstValue(smemObj.offsets[1]); |
There was a problem hiding this comment.
To those who are interested in the issue
These are no longer constants due to stream-pipeline pass. Therefore, they cannot be used to check is we are going to the fast path.
zhanglx13
force-pushed
the
fix-fast-path
branch
from
7529e02 to
36c6a58
Compare
binarman
reviewed
Jan 25, 2024
| const MfmaEncodingAttr &dstEncoding) { | ||
| if (srcEncoding.getMaxPhase() > 1) | ||
| return false; | ||
| auto stride0 = findConstValue(smemObj.strides[0]); |
There was a problem hiding this comment.
In this case you can remove findConstValue definition, it is not used in other places
alefimov-amd
approved these changes
Jan 25, 2024
zhanglx13
added a commit
to triton-lang/triton
that referenced
this pull request
Mar 4, 2024
This PR updates SharedToDotOperandMFMA.cpp and MFMA.cpp. - SharedToDotOperandMFMA.cpp is up to date with triton-mlir as of today, which includes changes until ROCm#482 - Fixed issue with opaque pointers - Fixed API for `getMFMAElemsPerInstrForOperands` and `getMFMARepForOperands` - MFMA.cpp is synced with triton-mlir@6bb04d, which includes changes until ROCm#469 Note to @binarman: changes in other files from ROCm#469 are not included in this PR. We can bring up the support for mfma 64x4 and 4x64 later.
binarman
pushed a commit
to binarman/triton
that referenced
this pull request
Apr 2, 2024
This PR updates SharedToDotOperandMFMA.cpp and MFMA.cpp. - SharedToDotOperandMFMA.cpp is up to date with triton-mlir as of today, which includes changes until ROCm#482 - Fixed issue with opaque pointers - Fixed API for `getMFMAElemsPerInstrForOperands` and `getMFMARepForOperands` - MFMA.cpp is synced with triton-mlir@6bb04d, which includes changes until ROCm#469 Note to @binarman: changes in other files from ROCm#469 are not included in this PR. We can bring up the support for mfma 64x4 and 4x64 later.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Fixes the issue in https://github.com/ROCmSoftwarePlatform/frameworks-internal/issues/6819