Cyclic dependency created between torch 2.0 and triton

[ashishu007]

Issue description

Installing torch 2.0 creates cyclic dependency between torch and triton. However, earlier torch versions (1.x) work fine.

Code example

$ python -m pipdeptree

Warning!! Cyclic dependencies found:
* triton => torch => triton
* torch => triton => torch
------------------------------------------------------------------------
pip==23.0.1
pipdeptree==2.3.3

System Info

• PyTorch or Caffe2: PyTorch
• How you installed PyTorch (conda, pip, from source): pip
• OS: Linux
• PyTorch version: 2.0.0
• Python version: 3.10.10
• Versions of any other relevant libraries: triton==2.0.0

cc @ezyang @soumith @msaroufim @wconstab @ngimel @bdhirsh @voznesenskym @penguinwu @anijain2305 @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @Xia-Weiwen @wenzhe-nrv @jiayisunx @peterbell10 @desertfire

[yanboliang]

cc @jansel @ngimel

[ezyang]

Unfortunately this is unavoidable. As long as the two libraries don't try to use each other on import (which they shouldn't) this is fine and won't break anything.

[milutter]

Can this cyclic dependency not be removed, or solved differently?

For some environments like pip, the cyclic dependency is not a problem but for other enviroments it is. For example, Bazel cannot handle the cyclic dependency in the pip package.

[ezyang]

The easiest fix will be to split pytorch into two packages, one which is traditional eager mode pytorch, and the other which has inductor (which is what has the triton dep). I'm not familiar enough with Bazel to say more about how to do this.

[jansel]

The easiest fix will be to split pytorch into two packages, one which is traditional eager mode pytorch, and the other which has inductor (which is what has the triton dep). I'm not familiar enough with Bazel to say more about how to do this.

Wouldn't you still have a cycle of pytorch->inductor->triton->pytorch?

IMO this is a bug in Bazel.

[ezyang]

the idea is inductor would depend on pytorch but not vice versa

[cphillippi-stripe]

I saw a change to Triton here that moved the torch dependency only to the "test" rule: triton-lang/triton#1389. This is not in the current release, but would just updating the release solve this?

[ngimel]

We cannot update pytorch 2.0.0 to next triton release, but on nightlies triton has been updated

[milutter]

I just tried the nightly version and the cyclic dependency is gone with the new triton release. So everything should be good with the 2.1.0 version

[groodt]

Will this fix make it into torch 2.0.1?

[ngimel]

No, we cannot update triton version used for 2.0.1 as 2.0.1 is just a bugfix release, and updating triton version is more risk and requires more significant changes.

[michael-quinlan]

Can this cyclic dependency not be removed, or solved differently?

For some environments like pip, the cyclic dependency is not a problem but for other enviroments it is. For example, Bazel cannot handle the cyclic dependency in the pip package.

@milutter what was your solution for bazel? Using 2.1 nightly is tricky with some of the package constraints (senetence-transformers etc)