[PromotePadsToL1] Implement pass lowering tensor.pad
#119
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
`tensor.pad`, as we produce them, can be lowered to `start_tensor_copy`. The latter will perform the L1 allocation and copy operations during bufferization. The one not-nice part of the pass is that it does so indiscriminately for all `tensor.pad` operations. I am very unsure about whether there are any guarantees by IREE that'd never encounter `tensor.pad` operations in kernels. This was also the last pass required to compile the last kernel in NsNet2
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Outdated suggestions cannot be applied.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.
tensor.pad, as we produce them, can be lowered tostart_tensor_copy. The latter will perform the L1 allocation and copy operations during bufferization.The one not-nice part of the pass is that it does so indiscriminately for all
tensor.padoperations. I am very unsure about whether there are any guarantees by IREE that'd never encountertensor.padoperations in kernels.This was also the last pass required to compile the last kernel in NsNet2