Removing shapex.tie_shape and its use in flow.dispatch.workgroups.#7745
Merged
Removing shapex.tie_shape and its use in flow.dispatch.workgroups.#7745
shapex.tie_shape and its use in flow.dispatch.workgroups.#7745Conversation
benvanik
added
compiler/dialects
This makes it possible to query the original source/target tensor dims from the flow.dispatch.tensor.load/store ops and removes the need to add magic operands that get tied via more magic. Now all shape dimensions that may be captured are passed as with any other operand would be.
benvanik
force-pushed
the
benvanik-flow-ergo-1
branch
from
a283611 to
6091f98
Compare
ScottTodd
approved these changes
Nov 29, 2021
Merged
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
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
This makes it possible to query the original source/target tensor dims
from the flow.dispatch.tensor.load/store ops and removes the need to
add magic operands that get tied via more magic. Now all shape dimensions
that may be captured are passed as any other operand would be meaning
that the order is now not load-bearing and operand deduplication/elision
all work as expected.
Future changes will clean up the
flow.dispatch.tensortype now that it doesnot need the shape information and remove the remaining shape ops that
get inserted via the ShapeCarryingOpInterface. This is just part 1/N in
decoupling things so we can iterate on the ergonomics.