[TritonIntelGPUToLLVM] Extend sub-group transposition support #2521
Conversation
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
…near layout Detect sub-group transpose cases as those in which warp and lane dimensions get swapped and no transfer within block-groups is needed. Use sub-group write operations to store the contents in local memory and vector operations to write back. These will be translated to non-transposed and transposed store and loads respectively. As data is moved within sub-groups, no barriers are needed. For now, handle only the case of a `single sub_group_size^2` block being transposed. This may be split in the future by performing `N*M` iterations for matrices of size `N*sub_group_sizexM*sub_group_size`. Signed-off-by: victor-eds <[email protected]>
Extend sub-group transposition support allowing `N*sub_group_size` elements per thread. As per block load semantics (matrix of `sub_group_size` columns), we need `N` vector loads to load the transposed matrix from local memory. Signed-off-by: victor-eds <[email protected]>
victor-eds
changed the title
Contributor
Author
|
Review only 602175b |
Contributor
Author
|
Part of #2266. |
Contributor
Author
|
Requires #2511 to be merged. |
etiotto
reviewed
Oct 22, 2024
etiotto
approved these changes
Oct 23, 2024
jopperm
approved these changes
Oct 24, 2024
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 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Outdated suggestions cannot be applied.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.
Extend sub-group transposition support allowing
N*sub_group_sizeelements per thread.As per block load semantics (matrix of
sub_group_sizecolumns), we needNvector loads to load the transposed matrix from local memory.