New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Weird AST constructor issue with mode="max-autotune" with python 3.11 #125374
Comments
I just setup a pytorch-311 build and can't repro this. Python 3.11.9. Do you have to run multiple times to repro or does it happen consistently? |
@eellison hmm.. this script repros it consistently for me. Like, when you run it, it goes through compilation for all the shapes without erroring? |
That's correct |
I tried to get it to repo on CI in #125331 but it seems to have green / with an unrelated failure. The error is definitely real though, because we see it in AO here: pytorch/ao#197 |
What python version are you specifically ? |
There is an AST Parsing bug that was fixed in 3.11.8: pytest-dev/pytest#11724 (comment). I'm using 3.11.9 here so maybe that explains it. |
Yeah i'm on 3.11.7 |
I guess we can disable precompilation from 3.11.0 to 3.11.7 and print a warning.
|
…karound 311 cpython bug" Fix for #125374. We dont have CI for this specific versions, but I verified locally. THere is a cpython bug from 3.11->3.17 where the ast parsing state is global, and errors with multiple threads. when dust settles a little around the new process based compilation we can look into migrating. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang [ghstack-poisoned]
…karound 311 cpython bug" Fix for #125374. We dont have CI for this specific versions, but I verified locally. THere is a cpython bug from 3.11.0->3.11.7 where the ast parsing state is global, and errors with multiple threads. when dust settles a little around the new process based compilation we can look into migrating. cc voznesenskym penguinwu EikanWang jgong5 Guobing-Chen XiaobingSuper zhuhaozhe blzheng wenzhe-nrv jiayisunx peterbell10 ipiszy yf225 chenyang78 kadeng muchulee8 ColinPeppler amjames desertfire chauhang [ghstack-poisoned]
… cpython bug (#125446) Fix for #125374. We dont have CI for this specific versions, but I verified locally. THere is a cpython bug from 3.11.0->3.11.7 where the ast parsing state is global, and errors with multiple threads. when dust settles a little around the new process based compilation we can look into migrating. Pull Request resolved: #125446 Approved by: https://github.com/Chillee ghstack dependencies: #125289
🐛 Describe the bug
Repro:
I bisected back to this commit: #124030
cc: @eellison
cc: @cpuhrsch
Setting
TORCHINDUCTOR_COMPILE_THREADS=1
also fixes it.Versions
python=3.11
cc @ezyang @msaroufim @bdhirsh @anijain2305 @chauhang
The text was updated successfully, but these errors were encountered: