Handle tcp.bind_symbolic_shape ops in fusion algorithm
#82
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
added 2 commits
July 15, 2024 12:40
| // CHECK: %[[V5:.+]] = tcp.add %[[V4]], %[[V4]] : tensor<?x?xf32>, tensor<?x?xf32> -> tensor<?x?xf32> | ||
| // CHECK: tcp.yield %[[V5]] : tensor<?x?xf32> | ||
| // CHECK: } : tensor<?x?xf32> | ||
| // CHECK: tcp.bind_symbolic_shape %[[V2]], [%[[V0]], %[[V1]]], affine_map<()[s0, s1] -> (s0, s1)> : tensor<?x?xf32> |
There was a problem hiding this comment.
Does it work the same way when the tcp.group returns more than one output? i.e. the tcp.bind_symbolic_shape ops are outside the groups?
sjain-stanford
approved these changes
Jul 15, 2024
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Outdated suggestions cannot be applied.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.
The current fusion algorithm does not handle
tcp.bind_symbolic_shapeops very well because these ops cause an op to have multiple downstream uses. This change handles these ops "specially". We collect all the uses of an op which only specify the shape of the output into a separate use category. We then move the bind shape ops along with the original op if it was moved to another region.