[Core][Bugfix] Use correct device to initialize GPU data during CUDA-graph-capture (#11233)

Signed-off-by: Yan Burman <yanburman@users.noreply.github.com>
Signed-off-by: Ido Asraff <idoa@atero.ai>

tests/distributed/test_custom_all_reduce.py

@@ -50,7 +50,7 @@ def graph_allreduce(tp_size, pp_size, rank, distributed_init_port):
 
     for sz in test_sizes:
         for dtype in [torch.float32, torch.float16, torch.bfloat16]:
-            with graph_capture() as graph_capture_context:
+            with graph_capture(device=device) as graph_capture_context:
                 # use integers so result matches NCCL exactly
                 inp1 = torch.randint(1,
                                      16, (sz, ),

tests/distributed/test_pynccl.py

@@ -107,7 +107,7 @@ def multiple_allreduce_with_vllm_worker_fn():
     device = torch.device(f"cuda:{torch.distributed.get_rank()}")
     ensure_model_parallel_initialized(2, 2)
     tensor = torch.ones(16, 1024, 1024, dtype=torch.float32, device=device)
-    with graph_capture():
+    with graph_capture(device=device):
         # two tp groups can communicate independently
         if torch.distributed.get_rank() in [0, 1]:
             tensor = tensor_model_parallel_all_reduce(tensor)

vllm/distributed/parallel_state.py

@@ -920,7 +920,7 @@ def get_kv_transfer_group() -> kv_transfer.KVTransferAgent:
 
 
 @contextmanager
-def graph_capture():
+def graph_capture(device: torch.device):
     """
     `graph_capture` is a context manager which should surround the code that
     is capturing the CUDA graph. Its main purpose is to ensure that the
@@ -934,8 +934,9 @@ def graph_capture():
     in order to explicitly distinguish the kernels to capture
     from other kernels possibly launched on background in the default stream.
     """
-    with get_tp_group().graph_capture() as context, get_pp_group(
-    ).graph_capture(context):
+    context = GraphCaptureContext(torch.cuda.Stream(device=device))
+    with get_tp_group().graph_capture(context), get_pp_group().graph_capture(
+            context):
         yield context
 
 

vllm/v1/worker/gpu_model_runner.py

@@ -836,7 +836,7 @@ class GPUModelRunner:
         # Trigger CUDA graph capture for specific shapes.
         # Capture the large shapes first so that the smaller shapes
         # can reuse the memory pool allocated for the large shapes.
-        with graph_capture():
+        with graph_capture(device=self.device):
             for num_tokens in reversed(self.cudagraph_batch_sizes):
                 for _ in range(self.vllm_config.compilation_config.
                                cudagraph_num_of_warmups):

vllm/worker/model_runner.py

@@ -1426,10 +1426,15 @@ class GPUModelRunnerBase(ModelRunnerBase[TModelInputForGPU]):
 
         # Prepare dummy inputs. These will be reused for all batch sizes.
         max_batch_size = self.max_batchsize_to_capture
-        input_tokens = torch.zeros(max_batch_size, dtype=torch.long).cuda()
-        input_positions = torch.zeros(max_batch_size, dtype=torch.long).cuda()
+        input_tokens = torch.zeros(max_batch_size,
+                                   dtype=torch.long,
+                                   device=self.device)
+        input_positions = torch.zeros(max_batch_size,
+                                      dtype=torch.long,
+                                      device=self.device)
         if self.model_config.uses_mrope:
-            input_positions = torch.tile(input_positions, (3, 1))
+            input_positions = torch.tile(input_positions,
+                                         (3, 1)).cuda(device=self.device)
         # Prepare dummy previous_hidden_states only if needed by the model.
         # This is used by draft models such as EAGLE.
         previous_hidden_states = None
@@ -1448,8 +1453,8 @@ class GPUModelRunnerBase(ModelRunnerBase[TModelInputForGPU]):
                 dtype=self.model_config.dtype,
                 device=self.device)
 
-        with self.attn_state.graph_capture(
-                max_batch_size), graph_capture() as graph_capture_context:
+        with self.attn_state.graph_capture(max_batch_size), graph_capture(
+                self.device) as graph_capture_context:
             # NOTE: Capturing the largest batch size first may help reduce the
             # memory usage of CUDA graph.
             for virtual_engine in range(
@@ -1549,10 +1554,12 @@ class GPUModelRunnerBase(ModelRunnerBase[TModelInputForGPU]):
         """
         # During the decode phase encoder_input_ids and encoder_positions are
         # unset. Do the same thing for graph capture.
-        capture_inputs["encoder_input_ids"] = torch.tensor(
-            [], dtype=torch.long).cuda()
-        capture_inputs["encoder_positions"] = torch.tensor(
-            [], dtype=torch.long).cuda()
+        capture_inputs["encoder_input_ids"] = torch.tensor([],
+                                                           dtype=torch.long,
+                                                           device=self.device)
+        capture_inputs["encoder_positions"] = torch.tensor([],
+                                                           dtype=torch.long,
+                                                           device=self.device)
 
     @property
     def vocab_size(self) -> int: