Skip to content
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

[Relay] Add LowerTEPass into vm lowering proecss #9134

Closed
wants to merge 2 commits into from

Conversation

mikepapadim
Copy link
Contributor

This PR adds the LowerTEPass in the VM compiler.

@mbs-octoml @jroesch @electriclilies

@Mousius
Copy link
Member

Mousius commented Sep 27, 2021

Hi @mikepapadim, my apologies but I don't quite understand what the impact of this change is as it appears to have no effect on any of the test cases - could you add some information to motivate the change?

@electriclilies
Copy link
Contributor

Just catching up after being out for a week -- is "vm_mod" used for anything other than name mangling?
Also is LowerTEPass supposed replacing any existing lowering logic in the VM? (or further downstream?)


targets_ = GetTecTargetMapFromTargetMap(targets);

pass_seqs.push_back(tec::LowerTEPass(targets_, device_map, "vm_mod", [this](Function func) {}));
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

For others: This PR also needs to remove the direct calls to the TECompiler Lower and LowerShapeFunc which are a bit fiddly to change over. I believe Michalis is working on that.

@mikepapadim mikepapadim marked this pull request as draft October 14, 2021 16:56
@masahi
Copy link
Member

masahi commented Jan 9, 2022

What's the status of this PR?

@mbs-octoml
Copy link
Contributor

@masahi this ended up being subsumed by #9613, I think it can be closed.

@masahi masahi closed this Jan 10, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants