[ROCm][inductor] autotune support for persistent reduction kernels#163908
Closed
naromero77amd wants to merge 9 commits intopytorch:mainfrom
Closed
[ROCm][inductor] autotune support for persistent reduction kernels#163908naromero77amd wants to merge 9 commits intopytorch:mainfrom
naromero77amd wants to merge 9 commits intopytorch:mainfrom
Conversation
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/163908
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 90f2fe2 with merge base 382b015 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
naromero77amd
added
release notes: inductor
and removed
module: rocm
Collaborator
Author
|
This is a redo of this PR: #162056 But this time branch was created in ROCm repo. |
jeffdaily
added
ciflow/inductor
ciflow/inductor-rocm
naromero77amd
force-pushed
the
rocm_persistent_reduction_autotune
branch
from
f894ae7 to
e549d1c
Compare
pytorch-bot
bot
removed
ciflow/inductor
ciflow/inductor-rocm
naromero77amd
changed the title
Collaborator
Author
|
We demonstrate that the updated heuristics are generally applicable to other models such as the HuggingFace (HF) Inductor Dashboard benchmark suite. Here is a high-level outline of the steps that were taken. We ran the HF benchmark suite from Inductor Dashboard using the command below: Then we extract the relevant kernels and benchmark with and without the PR (baseline also used TORCHINDCTOR_MAX_AUTOTUNE_POINTWISE=1). There is a geomean speed-up of 1.08 X on over 400 kernels. |
jeffdaily
added
release notes: rocm
PaulZhang12
approved these changes
Oct 17, 2025
pytorch-bot
bot
removed
ciflow/inductor
ciflow/rocm
jeffdaily
added
ciflow/inductor
ciflow/rocm
Collaborator
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 |
Chao1Han
pushed a commit
to Chao1Han/pytorch
that referenced
this pull request
Oct 21, 2025
…ytorch#163908) After the removal of want_no_x_dim for persistent reduction kernels, we can improve the autotuning setup for persistent reduction kernels. Currently even with tuning enable, filtering will only try a single config in many cases. Avoid filtering with autotune mode, and override MAX_BLOCK limit. Also we always include tiny_config when autotuning is enabled. Contributions from several members of the AMD Inductor and Triton teams: @jataylo @iupaikov-amd @AmdSampsa @xiaohuguo2023 Pull Request resolved: pytorch#163908 Approved by: https://github.com/jansel, https://github.com/PaulZhang12
zhudada0120
pushed a commit
to zhudada0120/pytorch
that referenced
this pull request
Oct 22, 2025
…ytorch#163908) After the removal of want_no_x_dim for persistent reduction kernels, we can improve the autotuning setup for persistent reduction kernels. Currently even with tuning enable, filtering will only try a single config in many cases. Avoid filtering with autotune mode, and override MAX_BLOCK limit. Also we always include tiny_config when autotuning is enabled. Contributions from several members of the AMD Inductor and Triton teams: @jataylo @iupaikov-amd @AmdSampsa @xiaohuguo2023 Pull Request resolved: pytorch#163908 Approved by: https://github.com/jansel, https://github.com/PaulZhang12
This was referenced Nov 4, 2025
pruthvistony
pushed a commit
to ROCm/pytorch
that referenced
this pull request
Nov 26, 2025
…ports (#2807) These are backports based on these upstream PRs. Cherrypicks were performed when they where possible. pytorch#163908 (persistent reduction autotune) pytorch#161280 (reduction) pytorch#162053 (foreach) pytorch#163197 (pointwise) pytorch#166470 (pointwise config for atomic add) Also included are some additional customer-specific configs which were not upstreamed but are in this backport to 2.9 #2723 Did not backport filter functions such as ` _maybe_filter_configs_for_tma_restrictions` https://github.com/ROCm/pytorch/blob/release/2.9/torch/_inductor/runtime/triton_heuristics.py#L2614 --------- Co-authored-by: Jack Taylor <jack.taylor@amd.com> Co-authored-by: Jack Taylor <108682042+jataylo@users.noreply.github.com> Co-authored-by: Sampsa Riikonen <sriikone@amd.com> Co-authored-by: AmdSampsa <sampsa.riikonen@amd.com>
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
6 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.
After the removal of want_no_x_dim for persistent reduction kernels, we can improve the autotuning setup for persistent reduction kernels.
Currently even with tuning enable, filtering will only try a single config in many cases. Avoid filtering with autotune mode, and override MAX_BLOCK limit. Also we always include tiny_config when autotuning is enabled.
Contributions from several members of the AMD Inductor and Triton teams: @jataylo @iupaikov-amd @AmdSampsa @xiaohuguo2023
cc @mlazos @jeffdaily @sunway513 @jithunnair-amd @pruthvistony @ROCmSupport @dllehr-amd @jataylo @hongxiayang @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben