[TritonCPU] Support execution on non-AMX hardware - #128
Merged
Merged
Conversation
jopperm
force-pushed
the
triton-cpu-non-amx
branch
3 times, most recently
from
September 7, 2026 11:59
15b74ca to
7731d27
Compare
jopperm
force-pushed
the
triton-cpu-non-amx
branch
3 times, most recently
from
September 7, 2026 13:09
40e384a to
c79ef1f
Compare
jopperm
force-pushed
the
triton-cpu-non-amx
branch
2 times, most recently
from
September 8, 2026 20:19
b311507 to
9d027d4
Compare
jopperm
force-pushed
the
triton-cpu-non-amx
branch
2 times, most recently
from
September 16, 2026 12:13
c1916a3 to
212d260
Compare
jopperm
marked this pull request as ready for review
September 16, 2026 12:35
Contributor
Author
|
@adam-smnk I still have to improve the API surface area of the matmul kernel; this PR doesn't try to. |
adam-smnk
approved these changes
Sep 16, 2026
jopperm
force-pushed
the
triton-cpu-non-amx
branch
from
September 17, 2026 13:41
212d260 to
5b75747
Compare
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Introduce a block size selection based on CPU capabilities, and hide it behind a new
SFCMatmulHelperclass, tying together the weight pre-packing and actual matmul (which must use the same block sizes). The rest of the PR is a mechanical/AI-assisted rewrite of the benchmark kernels.