-
-
Notifications
You must be signed in to change notification settings - Fork 11.9k
[FrontEnd] UNREVERT CompilationConfig overhaul (#20283): deprecate use_inductor in favor of backend, simplify custom_ops #26502
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
[FrontEnd] UNREVERT CompilationConfig overhaul (#20283): deprecate use_inductor in favor of backend, simplify custom_ops #26502
Conversation
There was a problem hiding this 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 👍.
|
original PR: #26113 |
|
revert PR: #26472 |
vllm/config/vllm.py
Outdated
| self.compilation_config.level > 0 | ||
| and self.compilation_config.backend != "eager" | ||
| ): | ||
| if self.compilation_config.backend == "inductor": |
There was a problem hiding this comment.
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?
There was a problem hiding this comment.
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.
|
Can you summarize what are the changes relative to the original PR to fix the issues? |
5d35d77 to
ce2132c
Compare
|
cc @Chenyaaang |
May I ask what's the difference with the original PR? I believe it will still fail for TPU platform. |
|
@Chenyaaang TPU-specific fixes have not been added yet, except the custom ops on by default only if |
Got it, will it be included in this pr? |
ca21ddb to
849a150
Compare
|
This pull request has merge conflicts that must be resolved before it can be |
Signed-off-by: morrison-turnansky <[email protected]>
Co-authored-by: Luka Govedič <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]>
Co-authored-by: Luka Govedič <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]>
Co-authored-by: Luka Govedič <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]>
Co-authored-by: Luka Govedič <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]>
Co-authored-by: Luka Govedič <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]>
Signed-off-by: morrison-turnansky <[email protected]>
Signed-off-by: Luka Govedič <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: 1994 <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: Dhruvil Bhatt <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: bbartels <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: xuebwang-amd <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: xuebwang-amd <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: 0xrushi <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]> Signed-off-by: 0xrushi <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]>
…deprecate use_inductor in favor of backend, simplify custom_ops (vllm-project#26502) Signed-off-by: morrison-turnansky <[email protected]> Signed-off-by: Morrison Turnansky <[email protected]> Signed-off-by: Luka Govedič <[email protected]> Co-authored-by: Luka Govedič <[email protected]> Co-authored-by: Jiangyun Zhu <[email protected]>
Original PR: #26113
Revert of original PR: #26472
Purpose
Test Plan
Test Result
Essential Elements of an Effective PR Description Checklist
supported_models.mdandexamplesfor a new model.