Skip to content

Conversation

@CongMa13
Copy link
Collaborator

Proposed changes

[CK TILE GEMM] Add bf8 support to tile engine streamk generator

Checklist

Please put an x into the boxes that apply. You can also fill these out after creating the PR. If you're not sure, please don't hesitate to ask.

  • I have added tests relevant to the introduced functionality, and the unit tests are passing locally
  • I have added the test to REGRESSION_TESTS list defined at the top of CMakeLists.txt in tests/CMakeLists.txt, IF the test takes more than 30 seconds to run.
  • I have added inline documentation which enables the maintainers with understanding the motivation
  • I have removed the stale documentation which is no longer relevant after this pull request
  • (If this change is user-facing) I have added release notes which provide the end users with a brief summary of the improvement from this pull request
  • I have run clang-format on all changed files
  • Any dependent changes have been merged

Discussion

If this is a relatively large or complex change, feel free to start a discussion by explaining why you chose the solution you did and what alternatives you considered

"--datatype",
required=True,
choices=["fp16", "fp8", "bf16", "fp32", "fp64"],
choices=["fp16", "fp8", "bf16", "bf8"],
Copy link
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is the replacement of fp32 and fp64 with bf8 intended?
From the title of the PR, it makes me assume we should not remove any existing data type.

"fp8": "ck_tile::fp8_t",
"bf16": "ck_tile::bf16_t",
"bf8": "ck_tile::bf8_t",
"fp32": "float",
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

If we delete the fp32 and fp64, should we keep the dtype to string?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants