variable search spaces for gemm autotuning #126220
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
add a switch to change the gemm autotuning search space between the default (the current set of hardcoded configs) and an exhaustive search space that enumerates all block sizes in [16, 32, 64, 128, 256], stages in [1, 2, 3, 4, 5], and warps in [2, 4, 6]
eellison
approved these changes
May 16, 2024
|
@pytorchbot merge |
Merge failedReason: This PR needs a If not, please add the To add a label, you can comment to pytorchbot, for example For more information, see Details for Dev Infra teamRaised by workflow job |
|
@pytorchbot label "topic: not user facing" |
|
@pytorchbot merge |
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 |
ZelboK
pushed a commit
to ZelboK/pytorch
that referenced
this pull request
May 19, 2024
add a switch to change the gemm autotuning search space between the default (the current set of hardcoded configs) and an exhaustive search space that enumerates all block sizes in [16, 32, 64, 128, 256], stages in [1, 2, 3, 4, 5], and warps in [2, 4, 6] Pull Request resolved: pytorch#126220 Approved by: https://github.com/eellison
jerryzh168
added a commit
to jerryzh168/ao
that referenced
this pull request
Dec 9, 2024
Summary: similar to pytorch/pytorch#126220 we added exhaustive option for int8mm and scaled_mm kernels in torchao Note that there seems to be native int8mm and scaled_mm support in pytorch: https://github.com/pytorch/pytorch/blob/0610b9730e27d066e26396a2d655ba0d98c2012d/torch/_inductor/kernel/mm.py#L305 for int8mm and https://github.com/pytorch/pytorch/blob/0610b9730e27d066e26396a2d655ba0d98c2012d/torch/_inductor/kernel/mm_scaled.py#L575 for scaled mm maybe we should use that at some point. Test Plan: ``` cd benchmarks TORCHAO_AUTOTUNER_ENABLE=1 python intmm.py --file_path intmm_shapes.csv TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_SEARCH_SPACE=EXHAUSTIVE TORCHAO_AUTOTUNER_ENABLE=1 python intmm.py --file_path intmm_shapes.csv ``` Reviewers: Subscribers: Tasks: Tags:
jerryzh168
added a commit
to pytorch/ao
that referenced
this pull request
Dec 11, 2024
* Add exhaustive config option to intmm kernel Summary: similar to pytorch/pytorch#126220 we added exhaustive option for int8mm and scaled_mm kernels in torchao Note that there seems to be native int8mm and scaled_mm support in pytorch: https://github.com/pytorch/pytorch/blob/0610b9730e27d066e26396a2d655ba0d98c2012d/torch/_inductor/kernel/mm.py#L305 for int8mm and https://github.com/pytorch/pytorch/blob/0610b9730e27d066e26396a2d655ba0d98c2012d/torch/_inductor/kernel/mm_scaled.py#L575 for scaled mm maybe we should use that at some point. Test Plan: ``` cd benchmarks TORCHAO_AUTOTUNER_ENABLE=1 python intmm.py --file_path intmm_shapes.csv TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_SEARCH_SPACE=EXHAUSTIVE TORCHAO_AUTOTUNER_ENABLE=1 python intmm.py --file_path intmm_shapes.csv ``` Reviewers: Subscribers: Tasks: Tags: * remove unused * enable all autoquant qtensor * guard float8 qtensor subclass * guard exhaustive config torch version
amdfaa
pushed a commit
to pytorch/ao
that referenced
this pull request
Jan 10, 2025
* Add exhaustive config option to intmm kernel Summary: similar to pytorch/pytorch#126220 we added exhaustive option for int8mm and scaled_mm kernels in torchao Note that there seems to be native int8mm and scaled_mm support in pytorch: https://github.com/pytorch/pytorch/blob/0610b9730e27d066e26396a2d655ba0d98c2012d/torch/_inductor/kernel/mm.py#L305 for int8mm and https://github.com/pytorch/pytorch/blob/0610b9730e27d066e26396a2d655ba0d98c2012d/torch/_inductor/kernel/mm_scaled.py#L575 for scaled mm maybe we should use that at some point. Test Plan: ``` cd benchmarks TORCHAO_AUTOTUNER_ENABLE=1 python intmm.py --file_path intmm_shapes.csv TORCHINDUCTOR_MAX_AUTOTUNE_GEMM_SEARCH_SPACE=EXHAUSTIVE TORCHAO_AUTOTUNER_ENABLE=1 python intmm.py --file_path intmm_shapes.csv ``` Reviewers: Subscribers: Tasks: Tags: * remove unused * enable all autoquant qtensor * guard float8 qtensor subclass * guard exhaustive config torch version
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.
add a switch to change the gemm autotuning search space between the default (the current set of hardcoded configs) and an exhaustive search space that enumerates all block sizes in [16, 32, 64, 128, 256], stages in [1, 2, 3, 4, 5], and warps in [2, 4, 6]
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @peterbell10 @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang