[Misc] Misc code cleanup/simplification (#23304)

Signed-off-by: Nick Hill <nhill@redhat.com>

vllm/v1/sample/sampler.py

@@ -91,7 +91,7 @@ class Sampler(nn.Module):
         logits = self.apply_bad_words(logits, sampling_metadata)
 
         # Apply logits processors which can impact greedy sampling
-        for processor in (sampling_metadata.logitsprocs.non_argmax_invariant):
+        for processor in sampling_metadata.logitsprocs.non_argmax_invariant:
             logits = processor.apply(logits)
 
         # Apply penalties (e.g., min_tokens, freq_penalties).

vllm/v1/worker/gpu_input_batch.py

@@ -442,10 +442,11 @@ class InputBatch:
         # LoRA
         lora_id = self.request_lora_mapping[req_index]
         if lora_id != 0:
-            self.lora_id_to_request_ids[lora_id].discard(req_id)
+            lora_req_ids = self.lora_id_to_request_ids[lora_id]
-            if len(self.lora_id_to_request_ids[lora_id]) == 0:
+            lora_req_ids.discard(req_id)
-                self.lora_id_to_request_ids.pop(lora_id)
+            if not lora_req_ids:
-                self.lora_id_to_lora_request.pop(lora_id)
+                del self.lora_id_to_request_ids[lora_id]
+                del self.lora_id_to_lora_request[lora_id]
             self.request_lora_mapping[req_index] = 0
 
         self.has_allowed_token_ids.discard(req_id)

vllm/v1/worker/gpu_model_runner.py

@@ -358,6 +358,7 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
         if num_pooling_reqs == 0:
             return model_kwargs
 
+        # This does nontrivial work.
         pooling_params = self.input_batch.pooling_metadata.pooling_params
 
         assert num_pooling_reqs == num_reqs
@@ -465,7 +466,7 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
         for req_id in unscheduled_req_ids:
             self.input_batch.remove_request(req_id)
 
-        req_ids_to_add: list[str] = []
+        reqs_to_add: list[CachedRequestState] = []
         # Add new requests to the cached states.
         for new_req_data in scheduler_output.scheduled_new_reqs:
             req_id = new_req_data.req_id
@@ -480,14 +481,14 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
                 generator = None
 
             if pooling_params:
-                assert (task := pooling_params.task) is not None, (
+                task = pooling_params.task
-                    "You did not set `task` in the API")
+                assert task is not None, "You did not set `task` in the API"
 
                 model = cast(VllmModelForPooling, self.get_model())
                 to_update = model.pooler.get_pooling_updates(task)
                 to_update.apply(pooling_params)
 
-            self.requests[req_id] = CachedRequestState(
+            req_state = CachedRequestState(
                 req_id=req_id,
                 prompt_token_ids=new_req_data.prompt_token_ids,
                 mm_kwargs=new_req_data.mm_kwargs,
@@ -501,6 +502,8 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
                 lora_request=new_req_data.lora_request,
             )
 
+            self.requests[req_id] = req_state
+
             # Only relevant for models using M-RoPE (e.g, Qwen2-VL)
             if self.uses_mrope:
                 image_grid_thw = []
@@ -508,29 +511,25 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
                 second_per_grid_ts = []
                 audio_feature_lengths = []
                 use_audio_in_video = False
-                for mm_item in self.requests[req_id].mm_kwargs:
+                for mm_item in req_state.mm_kwargs:
                     mm_input = mm_item.get_data()
-                    if mm_input.get("image_grid_thw") is not None:
+                    if (t := mm_input.get("image_grid_thw")) is not None:
-                        image_grid_thw.append(
+                        image_grid_thw.append(t.tolist())
-                            mm_input["image_grid_thw"].tolist())
+                    if (t := mm_input.get("video_grid_thw")) is not None:
-                    if mm_input.get("video_grid_thw") is not None:
+                        video_grid_thw.append(t.tolist())
-                        video_grid_thw.append(
+                    if (t := mm_input.get("second_per_grid_ts")) is not None:
-                            mm_input["video_grid_thw"].tolist())
+                        second_per_grid_ts.append(t)
-                    if mm_input.get("second_per_grid_ts") is not None:
+                    if (t :=
-                        second_per_grid_ts.append(
+                            mm_input.get("audio_feature_lengths")) is not None:
-                            mm_input["second_per_grid_ts"])
+                        audio_feature_lengths.append(t)
-                    if mm_input.get("audio_feature_lengths") is not None:
-                        audio_feature_lengths.append(
-                            mm_input["audio_feature_lengths"])
                     if mm_input.get("use_audio_in_video") is True:
                         use_audio_in_video = True
 
                 hf_config = self.model_config.hf_config
 
-                self.requests[req_id].mrope_positions, \
+                req_state.mrope_positions, req_state.mrope_position_delta = \
-                    self.requests[req_id].mrope_position_delta = \
                     MRotaryEmbedding.get_input_positions_tensor(
-                        self.requests[req_id].prompt_token_ids,
+                        req_state.prompt_token_ids,
                         hf_config=hf_config,
                         image_grid_thw=image_grid_thw,
                         video_grid_thw=video_grid_thw,
@@ -539,7 +538,7 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
                         use_audio_in_video=use_audio_in_video,
                     )
 
-            req_ids_to_add.append(req_id)
+            reqs_to_add.append(req_state)
 
         # Update the states of the running/resumed requests.
         is_last_rank = get_pp_group().is_last_rank
@@ -587,7 +586,7 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
                 # The request is not in the persistent batch.
                 # The request was either preempted and resumed later, or was not
                 # scheduled in the previous step and needs to be added again.
-                req_ids_to_add.append(req_id)
+                reqs_to_add.append(req_state)
                 continue
 
             # Update the persistent batch.
@@ -624,9 +623,8 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
 
         # Add the new or resumed requests to the persistent batch.
         # The smaller empty indices are filled first.
-        for req_id in req_ids_to_add:
+        for request in reqs_to_add:
-            req_state = self.requests[req_id]
+            self.input_batch.add_request(request)
-            self.input_batch.add_request(req_state)
 
         # Condense the batched states if there are gaps left by removed requests
         self.input_batch.condense()
@@ -639,38 +637,32 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
         self,
         scheduler_output: "SchedulerOutput",
     ) -> BatchedTensorInputs:
-        if self.is_multimodal_raw_input_supported:  # noqa: SIM102
+        if not self.is_multimodal_raw_input_supported or not scheduler_output:  # noqa: SIM102
-            if scheduler_output:
+            return {}
-                mm_kwargs = list[MultiModalKwargsItem]()
-                for req in scheduler_output.scheduled_new_reqs:
-                    req_mm_kwargs = req.mm_kwargs
-                    if not isinstance(req_mm_kwargs, list):
-                        req_mm_kwargs = list(req_mm_kwargs)
-                    mm_kwargs.extend(req_mm_kwargs)
 
-                # Input all modalities at once
+        mm_kwargs = list[MultiModalKwargsItem]()
-                mm_kwargs_combined: BatchedTensorInputs = {}
+        for req in scheduler_output.scheduled_new_reqs:
-                for _, _, mm_kwargs_group in group_mm_kwargs_by_modality(
+            mm_kwargs.extend(req.mm_kwargs)
-                        mm_kwargs,
-                        device=self.device,
-                        pin_memory=self.pin_memory,
-                ):
-                    mm_kwargs_combined.update(mm_kwargs_group)
 
-                return mm_kwargs_combined
+        # Input all modalities at once
+        mm_kwargs_combined: BatchedTensorInputs = {}
+        for _, _, mm_kwargs_group in group_mm_kwargs_by_modality(
+                mm_kwargs,
+                device=self.device,
+                pin_memory=self.pin_memory,
+        ):
+            mm_kwargs_combined.update(mm_kwargs_group)
 
-        return {}
+        return mm_kwargs_combined
 
     def _dummy_mm_kwargs(self, num_seqs: int) -> BatchedTensorInputs:
-        if self.is_multimodal_raw_input_supported:
+        if not self.is_multimodal_raw_input_supported:
-            mm_budget = self.mm_budget
+            return {}
-            assert mm_budget is not None
+        mm_budget = self.mm_budget
+        assert mm_budget is not None
 
-            dummy_modality = mm_budget.get_modality_with_max_tokens()
+        dummy_modality = mm_budget.get_modality_with_max_tokens()
-
+        return self._get_mm_dummy_batch(dummy_modality, num_seqs)
-            return self._get_mm_dummy_batch(dummy_modality, num_seqs)
-
-        return {}
 
     def _get_cumsum_and_arange(
         self,
@@ -1612,6 +1604,7 @@ class GPUModelRunner(LoRAModelRunnerMixin, KVConnectorModelRunnerMixin):
                 batch_descriptor=batch_descriptor,
         ), self.maybe_get_kv_connector_output(
                 scheduler_output) as kv_connector_output:
+
             model_output = self.model(
                 input_ids=input_ids,
                 positions=positions,

vllm/worker/worker_base.py

@@ -544,7 +544,7 @@ class WorkerWrapperBase:
         Arguments are passed to the worker class constructor.
         """
         kwargs = all_kwargs[self.rpc_rank]
-        self.vllm_config = kwargs.get("vllm_config", None)
+        self.vllm_config = kwargs.get("vllm_config")
         assert self.vllm_config is not None, (
             "vllm_config is required to initialize the worker")
         enable_trace_function_call_for_thread(self.vllm_config)