enable float32 and float16 in torch._grouped_mm fallback#162059
Closed
vkuzo wants to merge 3 commits intogh/vkuzo/6/basefrom
Closed
enable float32 and float16 in torch._grouped_mm fallback#162059vkuzo wants to merge 3 commits intogh/vkuzo/6/basefrom
torch._grouped_mm fallback#162059vkuzo wants to merge 3 commits intogh/vkuzo/6/basefrom
Conversation
Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
This was referenced Sep 3, 2025
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/162059
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 8932199 with merge base aed33a8 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
vkuzo
added a commit
that referenced
this pull request
Sep 3, 2025
Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: fdc346e Pull Request resolved: #162059
Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
vkuzo
added a commit
that referenced
this pull request
Sep 3, 2025
Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 6893b58 Pull Request resolved: #162059
ngimel
reviewed
Sep 3, 2025
Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: [ghstack-poisoned]
vkuzo
added a commit
that referenced
this pull request
Sep 4, 2025
Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: ghstack-source-id: 23e9fd6 Pull Request resolved: #162059
Collaborator
|
Should the compute-capability of tests be gated if #161407 is only for sm80+? |
Contributor
Author
currently the high precision grouped_gemm tests are gated with |
ngimel
approved these changes
Sep 4, 2025
eqy
approved these changes
Sep 4, 2025
Contributor
Author
|
@pytorchbot merge |
Collaborator
Merge startedYour change will be merged once all checks pass (ETA 0-4 Hours). Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
markc-614
pushed a commit
to markc-614/pytorch
that referenced
this pull request
Sep 17, 2025
…62059) Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash // on A100 and H100 pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x // on H100 pytest test/test_matmul_cuda.py -s -k test_scaled_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: Pull Request resolved: pytorch#162059 Approved by: https://github.com/ngimel, https://github.com/eqy ghstack dependencies: pytorch#161407, pytorch#161717
mansiag05
pushed a commit
to mansiag05/pytorch
that referenced
this pull request
Sep 22, 2025
…62059) Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash // on A100 and H100 pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x // on H100 pytest test/test_matmul_cuda.py -s -k test_scaled_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: Pull Request resolved: pytorch#162059 Approved by: https://github.com/ngimel, https://github.com/eqy ghstack dependencies: pytorch#161407, pytorch#161717
cleonard530
pushed a commit
to cleonard530/pytorch
that referenced
this pull request
Sep 22, 2025
…62059) Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash // on A100 and H100 pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x // on H100 pytest test/test_matmul_cuda.py -s -k test_scaled_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: Pull Request resolved: pytorch#162059 Approved by: https://github.com/ngimel, https://github.com/eqy ghstack dependencies: pytorch#161407, pytorch#161717
dsashidh
pushed a commit
to dsashidh/pytorch
that referenced
this pull request
Sep 26, 2025
…62059) Summary: Enables `torch.float32` and `torch.float16` options in `torch._grouped_mm`. Note that the fast path is only enabled if `mat_a`, `mat_b`, and `out_dtype` are `torch.bfloat16`. Saving for future PRs: 1. enabling testing on more platforms 2. supporting out_dtype != mat_a.dtype 3. opinfo 4. better compile support Test Plan: ```bash // on A100 and H100 pytest test/test_matmul_cuda.py -s -k test_grouped_gemm -x // on H100 pytest test/test_matmul_cuda.py -s -k test_scaled_grouped_gemm -x ``` Reviewers: Subscribers: Tasks: Tags: Pull Request resolved: pytorch#162059 Approved by: https://github.com/ngimel, https://github.com/eqy ghstack dependencies: pytorch#161407, pytorch#161717
pytorchmergebot
pushed a commit
that referenced
this pull request
Oct 24, 2025
…k was added (#165378) #162059 means we get unexpected successes now on e.g., SM 12.0 Pull Request resolved: #165378 Approved by: https://github.com/Skylion007
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
4 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.
Stack from ghstack (oldest at bottom):
torch._grouped_mmfallback #162059_grouped_mmfallback to composite explicit autograd #161717Summary:
Enables
torch.float32andtorch.float16options intorch._grouped_mm. Note that the fast path is only enabled ifmat_a,mat_b, andout_dtypearetorch.bfloat16.Saving for future PRs:
Test Plan:
Reviewers:
Subscribers:
Tasks:
Tags: