[inductor] add support for TRITON_INTERPRET #140841
Conversation
[ghstack-poisoned]
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/140841
Note: Links to docs will display an error until the docs builds have been completed. ❗ 1 Active SEVsThere are 1 currently active SEVs. If your PR is affected, please view them below: ✅ You can merge normally! (1 Unrelated Failure)As of commit 3c8bd5f with merge base 98e441f ( FLAKY - The following job failed but was likely due to flakiness present on trunk:
This comment was automatically generated by Dr. CI and updates every 15 minutes. |
Was debugging the issue lower in the stack and found this to be helpful / quick enough to add. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov [ghstack-poisoned]
eellison
changed the title
| _compile_start() | ||
| _set_triton_ptxas_path() | ||
|
|
||
| if os.environ.get("TRITON_INTERPRET", "0") == "1": |
There was a problem hiding this comment.
Might be better to be in a config so people can see it as an option.
There was a problem hiding this comment.
You need it to be set as an envvar otherwise triton wont return the interpreted function. So setting the config value wouldnt be enough wed need to thread it to triton as well.. going to leave like this for now
Was debugging the issue lower in the stack and found this to be helpful / quick enough to add. Fix for #123956 cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov [ghstack-poisoned]
|
@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 |
Merge failedReason: 2 mandatory check(s) failed. The first few are:
Dig deeper by viewing the failures on hud |
Was debugging the issue lower in the stack and found this to be helpful / quick enough to add. Fix for #123956 cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov [ghstack-poisoned]
|
@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 |
Was debugging the issue lower in the stack and found this to be helpful / quick enough to add. Fix for #123956 cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang aakhundov [ghstack-poisoned]
|
@pytorchbot merge |
|
The merge job was canceled or timed out. This most often happen if two merge requests were issued for the same PR, or if merge job was waiting for more than 6 hours for tests to finish. In later case, please do not hesitate to reissue the merge command |
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 |
This torch.cat kernel has multiple subblocks which load from the same input. We were incorrectly reusing the mask vars from the first load for the second load. Pull Request resolved: #140838 Approved by: https://github.com/jansel ghstack dependencies: #140841
Was debugging the issue lower in the stack and found this to be helpful / quick enough to add. Fix for pytorch#123956 Pull Request resolved: pytorch#140841 Approved by: https://github.com/exclamaforte
This torch.cat kernel has multiple subblocks which load from the same input. We were incorrectly reusing the mask vars from the first load for the second load. Pull Request resolved: pytorch#140838 Approved by: https://github.com/jansel ghstack dependencies: pytorch#140841
3 participants
Stack from ghstack (oldest at bottom):
Was debugging the issue lower in the stack and found this to be helpful / quick enough to add.
Fix for #123956
cc @voznesenskym @penguinwu @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @yf225 @chenyang78 @kadeng @muchulee8 @ColinPeppler @amjames @desertfire @chauhang @aakhundov