[BugFix][GCU] post_process call sites still use the renamed sampler_output kwarg - #8119
Open
Anai-Guo wants to merge 1 commit into
Open
[BugFix][GCU] post_process call sites still use the renamed sampler_output kwarg#8119Anai-Guo wants to merge 1 commit into
Anai-Guo wants to merge 1 commit into
Conversation
…st_process calls Signed-off-by: Anai-Guo <antai12232931@outlook.com>
|
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Motivation
fastdeploy/worker/gcu_model_runner.pycallspost_processwithsampler_output=, butpre_and_post_process.post_processnames that parametersampler_or_pooler_output:so both GCU call sites raise on every step that reaches them:
fastdeploy/worker/gcu_model_runner.py:837(_dummy_run)fastdeploy/worker/gcu_model_runner.py:1073(execute_model)The parameter was renamed when pooling support was added, and the GCU runner is the only backend that was not updated. Every sibling call site already uses the new name:
gpu_model_runner.pysampler_or_pooler_output=✅metax_model_runner.pysampler_or_pooler_output=✅gcu_model_runner.pysampler_output=❌Modifications
Rename the keyword at the two GCU call sites. No other change.
Note: the GCU runner also does not forward some of the newer optional
post_processkwargs that the GPU runner does (sampling_metadata,async_output_queue,think_end_id,splitwise_role_is_decode,enable_entropy,routing_replay_manager). Those all have defaults so they do not crash, and wiring them up is a behaviour change I cannot test on GCU hardware — deliberately out of scope here.Verification
No Enflame GCU device available, so I verified at the signature level rather than by running the runner: parsed
pre_and_post_process.pywithast, rebuiltpost_process'sinspect.Signature, and replayed everypost_processcall site in the three runners throughSignature.bind.Before:
After: all ten bind cleanly.
black(25.1.0 config, line-length 119) andruff check --config pyproject.tomlboth report no changes for the touched file.🤖 Generated with Claude Code