[inductor] Reduce cold compilation time caused by duplicated user-defined Triton kernels#168292
Closed
desertfire wants to merge 2 commits intogh/desertfire/611/basefrom
Closed
[inductor] Reduce cold compilation time caused by duplicated user-defined Triton kernels#168292desertfire wants to merge 2 commits intogh/desertfire/611/basefrom
desertfire wants to merge 2 commits intogh/desertfire/611/basefrom
Conversation
…ined Triton kernels Summary: Similar to #167132, but the previous PR didn't consider user-defined Triton kernels. When cudagraphs-partition is enabled in Inductor, different partitions can use the same user-defined Triton kernels. Each user-defined Trition kernel should only be defined and compiled once. Local measure shoes this PR can reduce Qwen/Qwen3-VL-235B-A22B-Instruct's cold compilation time from 243.65s to 114.69s. [ghstack-poisoned]
desertfire
added a commit
that referenced
this pull request
Nov 20, 2025
…ined Triton kernels Summary: Similar to #167132, but the previous PR didn't consider user-defined Triton kernels. When cudagraphs-partition is enabled in Inductor, different partitions can use the same user-defined Triton kernels. Each user-defined Trition kernel should only be defined and compiled once. Local measure shoes this PR can reduce Qwen/Qwen3-VL-235B-A22B-Instruct's cold compilation time from 243.65s to 114.69s. ghstack-source-id: 5556193 Pull Request resolved: #168292
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/168292
Note: Links to docs will display an error until the docs builds have been completed. This comment was automatically generated by Dr. CI and updates every 15 minutes. |
eellison
approved these changes
Nov 21, 2025
…ed user-defined Triton kernels" Summary: Similar to #167132, but the previous PR didn't consider user-defined Triton kernels. When cudagraphs-partition is enabled in Inductor, different partitions can use the same user-defined Triton kernels. Each user-defined Trition kernel should only be defined and compiled once. Local measure shoes this PR can reduce Qwen/Qwen3-VL-235B-A22B-Instruct's cold compilation time from 243.65s to 114.69s. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy chenyang78 kadeng muchulee8 amjames chauhang aakhundov coconutruben mlazos [ghstack-poisoned]
desertfire
added a commit
that referenced
this pull request
Nov 21, 2025
…ined Triton kernels Summary: Similar to #167132, but the previous PR didn't consider user-defined Triton kernels. When cudagraphs-partition is enabled in Inductor, different partitions can use the same user-defined Triton kernels. Each user-defined Trition kernel should only be defined and compiled once. Local measure shoes this PR can reduce Qwen/Qwen3-VL-235B-A22B-Instruct's cold compilation time from 243.65s to 114.69s. ghstack-source-id: 2ffe739 Pull Request resolved: #168292
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 |
JacobSzwejbka
pushed a commit
that referenced
this pull request
Dec 8, 2025
…ined Triton kernels (#168292) Summary: Similar to #167132, but the previous PR didn't consider user-defined Triton kernels. When cudagraphs-partition is enabled in Inductor, different partitions can use the same user-defined Triton kernels. Each user-defined Trition kernel should only be defined and compiled once. Local measure shoes this PR can reduce Qwen/Qwen3-VL-235B-A22B-Instruct's cold compilation time from 243.65s to 114.69s. Pull Request resolved: #168292 Approved by: https://github.com/eellison ghstack dependencies: #168281
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
3 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):
Summary: Similar to #167132, but the previous PR didn't consider user-defined Triton kernels. When cudagraphs-partition is enabled in Inductor, different partitions can use the same user-defined Triton kernels. Each user-defined Trition kernel should only be defined and compiled once.
Local measure shoes this PR can reduce Qwen/Qwen3-VL-235B-A22B-Instruct's cold compilation time from 243.65s to 114.69s.
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @chauhang @aakhundov @coconutruben @mlazos