HipKittens MXFP8 GEMM Support#566
Open
alextmagro wants to merge 33 commits into
Open
Conversation
alextmagro
requested review from
ipanfilo,
wangye805 and
wenchenvincent
as code owners
wangye805
requested changes
May 1, 2026
aris134
reviewed
May 6, 2026
aris134
reviewed
May 6, 2026
aris134
reviewed
May 6, 2026
aris134
reviewed
May 6, 2026
ipanfilo
requested changes
May 8, 2026
| [](const testing::TestParamInfo<DqGEMMTestSuite::ParamType>& info) { | ||
| return MKN(std::get<0>(info.param)) + "x" + TN(std::get<3>(info.param)); | ||
| return MKN(std::get<0>(info.param)) + "x" + | ||
| std::to_string(std::get<1>(info.param)) + "x" + |
Collaborator
There was a problem hiding this comment.
What is a point, they are set to false only
|
|
||
| return torch.empty(get_cublas_workspace_size_bytes(), dtype=torch.uint8, device=device) | ||
| key = (device, ub, grouped_gemm) | ||
| ws = _workspace_cache.get(key) |
Collaborator
There was a problem hiding this comment.
Why we don't rely on torch memory caching?
Contributor
Author
There was a problem hiding this comment.
I have made this change. I will need to run an E2E run to make sure that performance isn't affected, but should be ok given my understanding of torch.empty()
Collaborator
There was a problem hiding this comment.
It doesn't seem changed.
alextmagro
added
ci-level 3
ipanfilo
reviewed
May 15, 2026
| if (use_hipkittens) { | ||
| auto param = CanonicalizeGemmInput(*inputA, transa, *inputB, transb, m, n, k); | ||
|
|
||
| hipStream_t s = use_service_stream ? ss_ctl.stream : stream; |
Collaborator
There was a problem hiding this comment.
the same like with is_mxfp8, no point of having it defined for one branch only
| @@ -743,12 +786,15 @@ MAKE_DQ_GEMM_TEST(Testfp8xfp8xfp16, fp8, fp8, fp16) | |||
|
|
|||
| INSTANTIATE_TEST_SUITE_P(OperatorTest, DqGEMMTestSuite, | |||
Collaborator
There was a problem hiding this comment.
If you end up with having separate prefix for MXFP8, it has to be use for this suite for consistency
ipanfilo
reviewed
May 15, 2026
ipanfilo
reviewed
May 29, 2026
ipanfilo
reviewed
Jun 4, 2026
ipanfilo
reviewed
Jun 18, 2026
ipanfilo
reviewed
Jun 19, 2026
ipanfilo
left a comment
ipanfilo
left a comment
Collaborator
There was a problem hiding this comment.
Some comments in rocm_gemm and test_cublas_gemm are still open
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
5 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.
Creates an MXFP8 GEMM with HipKittens that outperforms hipBLASlt, and offers additional epilogues such as BIAS and GELU AUX
Requires a workspace sized relative to the model. Often larger than hipBLASlt, but with significant performance improvements. Only builds for gfx950, and requires M / 256 and N / 256.
Adds hipKittens header library as a submodule.