Revert "[LinalgExt] Add online_attention op" #17658
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
This reverts commit abf0087.
benvanik
approved these changes
Jun 12, 2024
|
|
|
Cc @Groverkss |
monorimet
added a commit
that referenced
this pull request
Jun 14, 2024
This reverts commit 2ff4102.
Groverkss
added a commit
that referenced
this pull request
Jun 17, 2024
This patch adds a new online_attention op. This op represents a partially reduced attention op which can be tiled along it's k2 reduction dimension. This op also has indexing maps, supports tiling on all dimensions other than k1 dimension, and can decompose based on any given indexing maps. This patch also makes the CPU backend use online attention to decompose and tile reduction dimension, allowing it to be tiled along N and batch dimensions, and tiling using LLVMCPUTile. This is a reland of #17658 , with more conservative tile size selection to not unroll too much.
LLITCHEV
pushed a commit
to LLITCHEV/iree
that referenced
this pull request
Jul 30, 2024
Reverts iree-org#17536 This caused `sdxl-scheduled-unet-3-tank` to hit timeouts when compiling for cpu: https://github.com/iree-org/iree/actions/runs/9484305572/job/26134004282 Signed-off-by: Lubo Litchev <lubol@google.com>
LLITCHEV
pushed a commit
to LLITCHEV/iree
that referenced
this pull request
Jul 30, 2024
This patch adds a new online_attention op. This op represents a partially reduced attention op which can be tiled along it's k2 reduction dimension. This op also has indexing maps, supports tiling on all dimensions other than k1 dimension, and can decompose based on any given indexing maps. This patch also makes the CPU backend use online attention to decompose and tile reduction dimension, allowing it to be tiled along N and batch dimensions, and tiling using LLVMCPUTile. This is a reland of iree-org#17658 , with more conservative tile size selection to not unroll too much. Signed-off-by: Lubo Litchev <lubol@google.com>
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.
Reverts #17536
This caused
sdxl-scheduled-unet-3-tankto hit timeouts when compiling for cpu: https://github.com/iree-org/iree/actions/runs/9484305572/job/26134004282