[Misc]feat: adapt to vLLM main (c7aa186d...50acdc5b) - #242
Draft
Meihan-chen wants to merge 36 commits into
Draft
Conversation
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com>
Signed-off-by: Meihan-chen <239345480+Meihan-chen@users.noreply.github.com>
Signed-off-by: Meihan-chen <239345480+Meihan-chen@users.noreply.github.com>
Signed-off-by: Meihan-chen <239345480+Meihan-chen@users.noreply.github.com>
Signed-off-by: Meihan-chen <239345480+Meihan-chen@users.noreply.github.com>
Signed-off-by: Meihan-chen <239345480+Meihan-chen@users.noreply.github.com>
Signed-off-by: Meihan-chen <239345480+Meihan-chen@users.noreply.github.com>
Upstream range: c7aa186d67b6f051680831418e957c67f34ba7a2..16e336491e96041946e2cd379a80034fcac439bd Changes: - Updated mistral_common version requirement from 1.11.0 to 1.11.2 - Refactored Mistral tokenizer: removed adapt_inplace_to_mistral_tool, renamed _prepare_apply_chat_template_tools_and_messages to _validate_apply_chat_template_args - Mistral tool parser updated to use MistralTool.from_openai() No vllm-ascend adaptation needed (no usage of changed functions). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> Signed-off-by: main2main-bot <main2main-bot@users.noreply.github.com>
Upstream range: 16e336491e96041946e2cd379a80034fcac439bd..5d0fd87038b123a11dd9c85a05ce2d258e27ce7b Key upstream changes absorbed: - [Spec Decode] Allow multimodal models with a warning (#41752) - Renamed _raise_if_multimodal to _warn_if_multimodal in base proposer - [Bugfix] Skip PP sampled-token receive on last rank during async scheduling (#40749) - Changed condition from get_pp_group().world_size > 1 to not get_pp_group().is_last_rank vllm-ascend adaptations: - vllm_ascend/spec_decode/dflash_proposer.py: renamed _raise_if_multimodal to _warn_if_multimodal - vllm_ascend/worker/model_runner_v1.py: updated PP async scheduling condition Other upstream changes (no adaptation needed): - CPU-specific improvements (dnnl, FP8 support, OMP binding) - Example/doc reorganization for disaggregated features - Codegen fix for unqualified names - XPU platform CUDA graph fix - ROCm Aiter MoE fix Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> Signed-off-by: main2main-bot <main2main-bot@users.noreply.github.com>
Upstream range: 5d0fd87038b123a11dd9c85a05ce2d258e27ce7b..242afc6bf40d6d088b7b97eda1dd7f00ddcdfa21 Changes: - [MM][Gemma4] Respect max_soft_tokens in encoder budget (#41799) - Added _get_max_soft_tokens helper function - Modified get_mm_max_tokens_per_item to respect configured max_soft_tokens No vllm-ascend adaptation needed (no Gemma4-related code). Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com> Signed-off-by: main2main-bot <main2main-bot@users.noreply.github.com>
MrZ20
force-pushed
the
main
branch
2 times, most recently
from
June 10, 2026 03:25
677396a to
9554149
Compare
zhangxinyuehfad
force-pushed
the
main
branch
5 times, most recently
from
June 16, 2026 04:48
c14c88e to
9768184
Compare
zhangxinyuehfad
force-pushed
the
main
branch
9 times, most recently
from
June 18, 2026 04:04
de24f70 to
60cd7c9
Compare
zhangxinyuehfad
force-pushed
the
main
branch
10 times, most recently
from
June 25, 2026 09:43
ef5e2db to
117f26e
Compare
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
2 participants
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.
Main2Main Summary
Status: completed
Upstream range: c7aa186d67b6f051680831418e957c67f34ba7a2..50acdc5b5cc00f10408d8f98b21fc97efc615173
Reached upstream commit: 50acdc5b5cc00f10408d8f98b21fc97efc615173
Steps: 9/9
CI suite: e2e-main2main
Result
The main2main pipeline successfully absorbed all 30 upstream vLLM commits (c7aa186d67b6f051680831418e957c67f34ba7a2 to 50acdc5b5cc00f10408d8f98b21fc97efc615173) across 9 steps. All CI tests passed for each step. The target commit was reached with verified code.
Completed Steps
Changes Made
_raise_if_multimodalto_warn_if_multimodalinvllm_ascend/spec_decode/dflash_proposer.py.get_pp_group().world_size > 1tonot get_pp_group().is_last_rankinvllm_ascend/worker/model_runner_v1.py.CI Verification
Adaptations Summary
Only 2 code adaptations were required across 30 upstream commits:
_raise_if_multimodal→_warn_if_multimodal(step-2)All other changes (nixl refactor, Gemma4 MTP, KV offload, block table alignment, parser fixes, etc.) were handled automatically by vllm-ascend inheriting from upstream or not using the affected features (TRTLLM, ROCm, TPU, nixl connector, Gemma4).