Skip to content

Conversation

@morrison-turnansky
Copy link
Contributor

@morrison-turnansky morrison-turnansky commented Oct 9, 2025

Original PR: #26113
Revert of original PR: #26472

Purpose

Test Plan

Test Result


Essential Elements of an Effective PR Description Checklist
  • The purpose of the PR, such as "Fix some issue (link existing issues this PR will resolve)".
  • The test plan, such as providing test command.
  • The test results, such as pasting the results comparison before and after, or e2e results
  • (Optional) The necessary documentation update, such as updating supported_models.md and examples for a new model.
  • (Optional) Release notes update. If your change is user facing, please update the release notes draft in the Google Doc.

Copy link

@chatgpt-codex-connector chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

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

💡 Codex Review

Here are some automated review suggestions for this pull request.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

  • Open a pull request for review
  • Mark a draft as ready
  • Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

@morrison-turnansky
Copy link
Contributor Author

morrison-turnansky commented Oct 9, 2025

original PR: #26113

@morrison-turnansky
Copy link
Contributor Author

revert PR: #26472

self.compilation_config.level > 0
and self.compilation_config.backend != "eager"
):
if self.compilation_config.backend == "inductor":
Copy link
Member

Choose a reason for hiding this comment

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

Can we add a comment to say why we disable custom ops with inductor?

Copy link
Contributor Author

Choose a reason for hiding this comment

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

I reverted this to the logic from main.

@mgoin
Copy link
Member

mgoin commented Oct 9, 2025

Can you summarize what are the changes relative to the original PR to fix the issues?

@morrison-turnansky morrison-turnansky force-pushed the issue-20283-compilation-config branch from 5d35d77 to ce2132c Compare October 9, 2025 19:26
@ProExpertProg ProExpertProg added the ready ONLY add when PR is ready to merge/full CI is needed label Oct 9, 2025
@ProExpertProg
Copy link
Collaborator

cc @Chenyaaang

@Chenyaaang
Copy link
Contributor

cc @Chenyaaang

May I ask what's the difference with the original PR? I believe it will still fail for TPU platform.

@ProExpertProg
Copy link
Collaborator

@Chenyaaang TPU-specific fixes have not been added yet, except the custom ops on by default only if backend == "inductor" (before it was if backend != "eager")

@Chenyaaang
Copy link
Contributor

@Chenyaaang TPU-specific fixes have not been added yet, except the custom ops on by default only if backend == "inductor" (before it was if backend != "eager")

Got it, will it be included in this pr?

@mergify
Copy link

mergify bot commented Oct 13, 2025

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @morrison-turnansky.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

@ProExpertProg
Copy link
Collaborator

Manually kicked off some blackwell and distributed tests just from abundance of caution. Seems like the rest of the tests are all passing!

Copy link
Collaborator

@ProExpertProg ProExpertProg left a comment

Choose a reason for hiding this comment

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

Looks good apart from the one level thing. If you update, you might as well do the other nits 😅

morrison-turnansky and others added 6 commits October 13, 2025 20:08
Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
@ProExpertProg ProExpertProg removed the ready ONLY add when PR is ready to merge/full CI is needed label Oct 13, 2025
Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
@ProExpertProg ProExpertProg enabled auto-merge (squash) October 13, 2025 20:54
@github-actions github-actions bot added the ready ONLY add when PR is ready to merge/full CI is needed label Oct 13, 2025
@ProExpertProg ProExpertProg merged commit e3fdb62 into vllm-project:main Oct 13, 2025
49 checks passed
1994 pushed a commit to 1994/vllm that referenced this pull request Oct 14, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: 1994 <1994@users.noreply.github.com>
Dhruvilbhatt pushed a commit to Dhruvilbhatt/vllm that referenced this pull request Oct 14, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: Dhruvil Bhatt <bhattdbh@amazon.com>
bbartels pushed a commit to bbartels/vllm that referenced this pull request Oct 16, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: bbartels <benjamin@bartels.dev>
lywa1998 pushed a commit to lywa1998/vllm that referenced this pull request Oct 20, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
alhridoy pushed a commit to alhridoy/vllm that referenced this pull request Oct 24, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
xuebwang-amd pushed a commit to xuebwang-amd/vllm that referenced this pull request Oct 24, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: xuebwang-amd <xuebwang@amd.com>
xuebwang-amd pushed a commit to xuebwang-amd/vllm that referenced this pull request Oct 24, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: xuebwang-amd <xuebwang@amd.com>
0xrushi pushed a commit to 0xrushi/vllm that referenced this pull request Oct 26, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: 0xrushi <6279035+0xrushi@users.noreply.github.com>
0xrushi pushed a commit to 0xrushi/vllm that referenced this pull request Oct 26, 2025
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502)

Signed-off-by: morrison-turnansky <mturnans@redhat.com>
Signed-off-by: Morrison Turnansky <mturnans@redhat.com>
Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Jiangyun Zhu <riverclouds.zhu@qq.com>
Signed-off-by: 0xrushi <6279035+0xrushi@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

llama Related to Llama models ready ONLY add when PR is ready to merge/full CI is needed

Projects

None yet

Development

Successfully merging this pull request may close these issues.

[RFC][UX][torch.compile][CUDAGraph]: Overhaul CompilationConfig and improve CLI -O<n>

6 participants