[Misc] Remove pad_for_cudagraphs from config by LucasWilkinson · Pull Request #30143 · vllm-project/vllm

LucasWilkinson · 2025-12-05T15:52:38Z

With #28579 we can strip pad_for_cudagraph from the config forcing cudagraph padding through the dispatch consistently making it easier to debug and enable support for separate sizes for FULL and PIECEWISE graphs

gemini-code-assist

Code Review

This pull request refactors the CUDA graph padding logic by removing pad_for_cudagraphs from the configuration and centralizing it within the CudagraphDispatcher. The changes are well-contained and improve the code structure by making the dispatcher solely responsible for padding, which simplifies debugging and future extensions.

My review found one potential issue in vllm/v1/spec_decode/eagle.py where the refactoring might unintentionally change the behavior for EagleProposer when cudagraph_mode is set to FULL. I've left a comment with a suggested fix to ensure the original behavior is preserved. Other than that, the changes look solid and achieve the stated goal.

vllm/v1/spec_decode/eagle.py

mergify · 2025-12-05T15:56:46Z

Hi @LucasWilkinson, the pre-commit checks have failed. Please run:

uv pip install pre-commit
pre-commit install
pre-commit run --all-files

Then, commit the changes and push to your branch.

For future commits, pre-commit will run automatically on changed files before each commit.

chatgpt-codex-connector

💡 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 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

chatgpt-codex-connector · 2025-12-05T15:57:10Z

tests/models/language/generation/test_hybrid.py

    vllm_config = EngineArgs(model=model, trust_remote_code=True).create_engine_config()
-    while len(example_prompts) == vllm_config.pad_for_cudagraph(len(example_prompts)):
+    cudagraph_dispatcher = CudagraphDispatcher(vllm_config)
+    while (
+        len(example_prompts)
+        == cudagraph_dispatcher.dispatch(len(example_prompts))[1].num_tokens
+    ):


Dispatch loop never terminates in hybrid padding test

In test_mamba_cache_cg_padding a CudagraphDispatcher is created and immediately used in the while condition without ever calling initialize_cudagraph_keys, so dispatch returns a BatchDescriptor with num_tokens unchanged when keys_initialized is False. The condition len(example_prompts) == ...num_tokens therefore stays true on every iteration and the loop appends forever, hanging the test before it exercises any logic. This test now never completes under any configuration.

Useful? React with 👍 / 👎.

mergify · 2025-12-05T18:47:01Z

This pull request has merge conflicts that must be resolved before it can be
merged. Please rebase the PR, @LucasWilkinson.

https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/working-with-forks/syncing-a-fork

mergify · 2025-12-11T00:56:03Z

Hi @LucasWilkinson, the pre-commit checks have failed. Please run:

uv pip install pre-commit
pre-commit install
pre-commit run --all-files

Then, commit the changes and push to your branch.

For future commits, pre-commit will run automatically on changed files before each commit.

Tip

Is mypy or markdownlint failing?

mypy and markdownlint are run differently in CI. If the failure is related to either of these checks, please use the following commands to run them locally:

# For mypy (substitute "3.10" with the failing version if needed)
pre-commit run --hook-stage manual mypy-3.10
# For markdownlint
pre-commit run --hook-stage manual markdownlint

mergify · 2025-12-11T05:07:58Z

Hi @LucasWilkinson, the pre-commit checks have failed. Please run:

uv pip install pre-commit
pre-commit install
pre-commit run --all-files

Then, commit the changes and push to your branch.

For future commits, pre-commit will run automatically on changed files before each commit.

Tip

Is mypy or markdownlint failing?

mypy and markdownlint are run differently in CI. If the failure is related to either of these checks, please use the following commands to run them locally:

# For mypy (substitute "3.10" with the failing version if needed)
pre-commit run --hook-stage manual mypy-3.10
# For markdownlint
pre-commit run --hook-stage manual markdownlint

mergify · 2025-12-11T15:59:50Z

Hi @LucasWilkinson, the pre-commit checks have failed. Please run:

uv pip install pre-commit
pre-commit install
pre-commit run --all-files

Then, commit the changes and push to your branch.

For future commits, pre-commit will run automatically on changed files before each commit.

Tip

Is mypy or markdownlint failing?

mypy and markdownlint are run differently in CI. If the failure is related to either of these checks, please use the following commands to run them locally:

# For mypy (substitute "3.10" with the failing version if needed)
pre-commit run --hook-stage manual mypy-3.10
# For markdownlint
pre-commit run --hook-stage manual markdownlint

mergify · 2025-12-12T19:35:31Z

Hi @LucasWilkinson, the pre-commit checks have failed. Please run:

uv pip install pre-commit
pre-commit install
pre-commit run --all-files

Then, commit the changes and push to your branch.

For future commits, pre-commit will run automatically on changed files before each commit.

Tip

Is mypy or markdownlint failing?

mypy and markdownlint are run differently in CI. If the failure is related to either of these checks, please use the following commands to run them locally:

# For mypy (substitute "3.10" with the failing version if needed)
pre-commit run --hook-stage manual mypy-3.10
# For markdownlint
pre-commit run --hook-stage manual markdownlint

ProExpertProg

Nice cleanup!

…nfig

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com> Signed-off-by: Matthew Bonanni <mbonanni@redhat.com> Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com> Signed-off-by: Matthew Bonanni <mbonanni@redhat.com> Co-authored-by: Matthew Bonanni <mbonanni@redhat.com> Signed-off-by: dsuhinin <suhinin.dmitriy@gmail.com>

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com> Signed-off-by: Matthew Bonanni <mbonanni@redhat.com> Co-authored-by: Matthew Bonanni <mbonanni@redhat.com> Signed-off-by: mohammad najafi <mohammad.najafi@amd.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com>

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com> Signed-off-by: Matthew Bonanni <mbonanni@redhat.com> Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: momochenchuw <chenchuw@huawei.com>

Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com> Signed-off-by: Matthew Bonanni <mbonanni@redhat.com> Co-authored-by: Matthew Bonanni <mbonanni@redhat.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: zrj026 <zhangrunjiang026@gmail.com>

### What this PR does / why we need it? 1. ✅ Upgrade vllm commit to: 0115 (8471b27df97c3eb79f891802fc0e858f8f7ac6a0) Modify import paths due to the refactors： vllm-project/vllm#32245 vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21034239336/job/60490156965?pr=5913 2. ✅Upgrade vllm commit to: 0119 (9a1f16da1e423ede2c2f52a9850cbfbb39cefe96) Fix `WorkerProc.__init__() missing 1 required positional argument: 'is_driver_worker'` due to vllm-project/vllm#28506 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21156263050/job/60841668755?5569 3. ✅Upgrade vllm commit to: 0120(148117ea2e689cd43df4be6892671a17cdae5833) 1. Add `skip_compiled` param in `set_forward_context` due to vllm-project/vllm#30385 2. Modify `tests/ut/spec_decode/test_eagle_proposer.py` due to vllm-project/vllm#24322 change `self.max_num_tokens = vllm_config.scheduler_config.max_num_batched_tokens + max_batch_size` 3. Modify UT import paths due to the refactors：vllm-project/vllm#32060 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21204851770/job/60999046946 4. ✅Upgrade vllm commit to: 0121(f23fb5a7c1b61350c5c40ca1115d3bf8cf2b8cc9) 1. vLLM switched `uses_mrope` from target to draft model config, making `positions`/`mrope_positions` mutually exclusive, breaking vllm-ascend's direct self.positions access and tests missing `draft_model_config.uses_mrope`. vllm-project/vllm#32048 2. Moved bs_to_padded_graph_size from CompilationConfig to CudagraphDispatcher due to the refactor vllm-project/vllm#30143 3. Remove unused `maybe_setup_kv_connector` due to vllm-project/vllm#32077 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21217728738/job/61043738834 6. ✅Upgrade vllm commit to: 0122(8ebf271bb6d1e7e9b1a55be73d755ef1a57dbbe5) Updating FusedMoEParallelConfig (added enable_eplb) and FusedMoEConfig due to vllm-project/vllm#32414 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21249922546/job/61148613054 8. ✅Upgrade vllm commit to: 0123(dc917cceb877dfd13f98c538c4c96158047d98bd) Setting temperature=0.0 due to the removal of the default temperature value in vllm-project/vllm#32723 Test result: https://github.com/vllm-project/vllm-ascend/actions/runs/21280796875 ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? - vLLM version: v0.14.0 - vLLM main: vllm-project/vllm@d682094 --------- Signed-off-by: wjunLu <wjunlu217@gmail.com> Signed-off-by: Meihan-chen <jcccx.cmh@gmail.com> Co-authored-by: wjunLu <wjunlu217@gmail.com>

LucasWilkinson requested review from ProExpertProg, WoosukKwon, benchislett, hmellor, houseroad, luccafong, mgoin, robertgshaw2-redhat, tdoublep, tlrmchlsmth, yewentao256 and youkaichao as code owners December 5, 2025 15:52

mergify bot added nvidia speculative-decoding v1 labels Dec 5, 2025

github-project-automation bot added this to NVIDIA Dec 5, 2025

gemini-code-assist bot reviewed Dec 5, 2025

View reviewed changes

vllm/v1/spec_decode/eagle.py Outdated Show resolved Hide resolved

chatgpt-codex-connector bot reviewed Dec 5, 2025

View reviewed changes

mergify bot added the needs-rebase label Dec 5, 2025

LucasWilkinson mentioned this pull request Dec 9, 2025

[BugFix] Fix DeepSeek-R1 hang with DP and MTP #30119

Merged

LucasWilkinson force-pushed the lwilkinson/remove-pad-for-cudagraphs-from-config branch from 575ad80 to aca162f Compare December 11, 2025 00:53

mergify bot removed the needs-rebase label Dec 11, 2025

LucasWilkinson force-pushed the lwilkinson/remove-pad-for-cudagraphs-from-config branch from 2624e49 to 0167a71 Compare December 12, 2025 19:31

LucasWilkinson requested a review from zou3519 as a code owner December 12, 2025 19:37

LucasWilkinson force-pushed the lwilkinson/remove-pad-for-cudagraphs-from-config branch from a1f70cc to 012f2d9 Compare January 16, 2026 03:54

ProExpertProg approved these changes Jan 16, 2026

View reviewed changes

github-project-automation bot moved this to Ready in NVIDIA Jan 16, 2026

Merge branch 'main' into lwilkinson/remove-pad-for-cudagraphs-from-co…

6415dce

…nfig

MatthewBonanni force-pushed the lwilkinson/remove-pad-for-cudagraphs-from-config branch from db89510 to 6415dce Compare January 19, 2026 21:15

MatthewBonanni added 3 commits January 19, 2026 17:58

Update test

a3930d4

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>

Fix None value error

7e0fe86

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>

Fix hanging test

fa9034b

Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>

LucasWilkinson merged commit 2261340 into vllm-project:main Jan 20, 2026
146 checks passed

github-project-automation bot moved this from Ready to Done in NVIDIA Jan 20, 2026

mawong-amd mentioned this pull request Jan 22, 2026

[Bugfix] Fix potential EAGLE spec decode segfault during graph capture #32818

Merged

5 tasks

Meihan-chen mentioned this pull request Jan 26, 2026

[Main2Main] Upgrade vllm commit to 0123 vllm-project/vllm-ascend#6169

Merged

LucasWilkinson mentioned this pull request Jan 28, 2026

[Misc] Remove missed pad_for_cudagraph #33283

Merged

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

[Misc] Remove pad_for_cudagraphs from config#30143

[Misc] Remove pad_for_cudagraphs from config#30143
LucasWilkinson merged 19 commits intovllm-project:mainfrom
neuralmagic:lwilkinson/remove-pad-for-cudagraphs-from-config

LucasWilkinson commented Dec 5, 2025

Uh oh!

gemini-code-assist bot left a comment

Uh oh!

Uh oh!

mergify bot commented Dec 5, 2025

Uh oh!

chatgpt-codex-connector bot left a comment

Uh oh!

chatgpt-codex-connector bot Dec 5, 2025

Uh oh!

mergify bot commented Dec 5, 2025

Uh oh!

mergify bot commented Dec 11, 2025

Uh oh!

mergify bot commented Dec 11, 2025

Uh oh!

mergify bot commented Dec 11, 2025

Uh oh!

mergify bot commented Dec 12, 2025

Uh oh!

ProExpertProg left a comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Uh oh!

Conversation

LucasWilkinson commented Dec 5, 2025

Uh oh!

gemini-code-assist bot left a comment

Choose a reason for hiding this comment

Code Review

Uh oh!

Uh oh!

mergify bot commented Dec 5, 2025

Uh oh!

chatgpt-codex-connector bot left a comment

Choose a reason for hiding this comment

💡 Codex Review

Uh oh!

chatgpt-codex-connector bot Dec 5, 2025

Choose a reason for hiding this comment

Uh oh!

mergify bot commented Dec 5, 2025

Uh oh!

mergify bot commented Dec 11, 2025

Uh oh!

mergify bot commented Dec 11, 2025

Uh oh!

mergify bot commented Dec 11, 2025

Uh oh!

mergify bot commented Dec 12, 2025

Uh oh!

ProExpertProg left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants