Modernize bf16 cutlass grouped gemm #3889
Closed
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
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
✅ Deploy Preview for pytorch-fbgemm-docs ready!
To edit notification comments on pull requests, go to your Netlify site configuration. |
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
added a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 27, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64, mxz297 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
c197a45 to
b891570
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
added a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 27, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64, mxz297 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
b891570 to
47449ce
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
added a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 27, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64, mxz297 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
47449ce to
bc3dd8d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
added a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 27, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64, mxz297 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
bc3dd8d to
3eadc7c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
added a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 27, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64, mxz297 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
3eadc7c to
fba1035
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
added a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 27, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64, mxz297 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
fba1035 to
9c8d29d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
force-pushed
the
export-D71920813
branch
from
9c8d29d to
7b4324e
Compare
jwfromm
pushed a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 28, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64 Differential Revision: D71920813
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
pushed a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 28, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
7b4324e to
3e535ec
Compare
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
pushed a commit
to jwfromm/FBGEMM
that referenced
this pull request
Mar 28, 2025
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64 Differential Revision: D71920813
jwfromm
force-pushed
the
export-D71920813
branch
from
3e535ec to
48c9d3d
Compare
Summary: Pull Request resolved: pytorch#3889 X-link: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64 Differential Revision: D71920813
|
This pull request was exported from Phabricator. Differential Revision: D71920813 |
jwfromm
force-pushed
the
export-D71920813
branch
from
48c9d3d to
07feea3
Compare
|
This pull request has been merged in 5e684ba. |
q10
pushed a commit
to q10/FBGEMM
that referenced
this pull request
Apr 10, 2025
Summary: X-link: pytorch#3889 Pull Request resolved: facebookresearch/FBGEMM#982 This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes. Reviewed By: jiawenliu64 Differential Revision: D71920813 fbshipit-source-id: 4928c4299d2b62e1722faf8b2bc1ba278adf23a1
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.
Summary:
X-link: https://github.com/facebookresearch/FBGEMM/pull/982
This diff unifies the API between FP8 and BF16 grouped gemm. Specifically we add the same dynamic, concatenated, and stacked APIs that are used for FP8 across both cutlass and CK. After this change, our tests can also be unified into a single grouped gemm test that covers all the various modes.
Reviewed By: jiawenliu64, mxz297
Differential Revision: D71920813