-
Notifications
You must be signed in to change notification settings - Fork 281
[Feature] Optimize forward metadata collection across dp ranks #1857
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
base: main
Are you sure you want to change the base?
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## main #1857 +/- ##
=======================================
Coverage 54.80% 54.80%
=======================================
Files 75 75
Lines 9327 9327
=======================================
Hits 5112 5112
Misses 4215 4215
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
Rebase here: f9dfde0 |
7b605f8
to
2b0c3c1
Compare
vllm_ascend/worker/worker_v1.py
Outdated
if runner.dp_size > 1: | ||
max_num_tokens, with_prefill = runner._get_forward_metadata_across_dp( | ||
max_num_tokens, with_prefill) | ||
if runner.dp_size <= 1: |
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.
Let's do the torchair refactor fist. Then add related code there #1885
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.
Got it.
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.
What are your plans for the model_runner refactoring?
This pull request has conflicts, please resolve those before we can evaluate the pull request. |
Signed-off-by: Jade Zheng <zheng.shoujian@outlook.com>
Resubmit #1593
What this PR does / why we need it?
This PR introduces two optimizations for cases where data parallel size > 1:
set_forward_context
Does this PR introduce any user-facing change?
no
How was this patch tested?
CI passed.