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.

Sign up for GitHub

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

Jump to bottom

Some clean after remarks in merged #82 #184

Merged
merged 1 commit into from
Feb 1, 2024
Merged

Some clean after remarks in merged #82 #184

merged 1 commit into from
Feb 1, 2024

Conversation

vvchernov
Copy link

@vvchernov vvchernov commented Feb 1, 2024 •
edited
Loading

Transfer get_logprob_infos to model_common.py see details in #82

cc @masahi

@vvchernov vvchernov mentioned this pull request Feb 1, 2024
Merged
@masahi masahi merged commit 583bb4b into octoml:batch-serving Feb 1, 2024
1 check passed
@vvchernov vvchernov deleted the vc/clean-after-82 branch February 1, 2024 09:28
Lunderberg pushed a commit to Lunderberg/mlc-llm that referenced this pull request Feb 27, 2024
@MasterJH5574
[Transform] Lift global buffer allocation in TIR PrimFunc (octoml#184)
4ebfba9 
This PR introduces the pass which can help lift all the global-scope
buffer allocation in all TIR PrimFunc to the Relax function side.

With this PR, we will no longer (in most of the cases where the pass
is able to resolve the symbolic var relation across TIR and Relax
function side) have global alloc_buffers at TIR level.

Confirmed its effect on RedPajama.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@masahi masahi masahi approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@vvchernov @masahi