Move internal changes #475
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
christopherbate
force-pushed
the
move_internal_changes
branch
from
80acaf1 to
a925b4e
Compare
## [tensorrt] Fix various edge cases in 'tensorrt-broadcast-elimination' This change fixes several edge cases in 'tensorrt-broadcast-elimination' for a particular pattern that commutes 'tensorrt.broadcast' and 'tensorrt.collapse_rank'. Previously it did not correctly handle multiple collapsed unit dimensions and would produce an erroneous assertion when the collapsed input is dynamic. Several additional regression tests are added. ## [tensorrt] Fix nvinfer1::Dims max rank assertion Fixes an assertion which limited translation of MLIR to TensorRT to programs with tensors of rank < 8 when it should be <= 8 to align with TensorRT op verification and the actual nvinfer API. Adds an additional translation test. ## NFC: apply isort to Python files ## Fix options configuration to enable global flags in "compiler task" API
christopherbate
force-pushed
the
move_internal_changes
branch
from
a925b4e to
95160bc
Compare
This was referenced Jan 13, 2025
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Outdated suggestions cannot be applied.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.
[tensorrt] Fix various edge cases in 'tensorrt-broadcast-elimination'
This change fixes several edge cases in 'tensorrt-broadcast-elimination'
for a particular pattern that commutes 'tensorrt.broadcast' and
'tensorrt.collapse_rank'. Previously it did not correctly handle multiple
collapsed unit dimensions and would produce an erroneous assertion when
the collapsed input is dynamic. Several additional regression tests
are added.
[tensorrt] Fix nvinfer1::Dims max rank assertion
Fixes an assertion which limited translation of MLIR to TensorRT to
programs with tensors of rank < 8 when it should be <= 8 to align
with TensorRT op verification and the actual nvinfer API. Adds an
additional translation test.
NFC: apply isort to Python files
Fix options configuration to enable global flags in "compiler task" API