Merge tag 'v0.7.2' into v0.7.2-dev

tests/core/test_num_computed_tokens_update.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import pytest
 

tests/core/test_scheduler.py

+# SPDX-License-Identifier: Apache-2.0
+
 import time
 from collections import deque
 from typing import List, Set, Tuple

tests/core/test_scheduler_encoder_decoder.py

+# SPDX-License-Identifier: Apache-2.0
+
 from typing import List
 
 import pytest  # noqa

tests/core/test_serialization.py

+# SPDX-License-Identifier: Apache-2.0
+
 import msgspec
 
 from vllm.executor.msgspec_utils import decode_hook, encode_hook

tests/core/utils.py

+# SPDX-License-Identifier: Apache-2.0
+
 import time
 from collections import defaultdict
 from typing import Any, Dict, List, Optional

tests/distributed/test_ca_buffer_sharing.py

+# SPDX-License-Identifier: Apache-2.0
+
 # can only run on machines with p2p access across GPUs
 # can only run with torchrun:
 # torchrun --nproc_per_node=2 tests/distributed/test_ca_buffer_sharing.py

tests/distributed/test_comm_ops.py

+# SPDX-License-Identifier: Apache-2.0
 """Test the communication operators.
 
 Run `pytest tests/distributed/test_comm_ops.py`.

tests/distributed/test_custom_all_reduce.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 import random
 

tests/distributed/test_distributed_oot.py

+# SPDX-License-Identifier: Apache-2.0
+
 from ..entrypoints.openai.test_oot_registration import (
     run_and_test_dummy_opt_api_server)
 

tests/distributed/test_multi_node_assignment.py

+# SPDX-License-Identifier: Apache-2.0
 """Make sure ray assigns GPU workers to the correct node.
 
 Run:

tests/distributed/test_pipeline_parallel.py

+# SPDX-License-Identifier: Apache-2.0
 """
 WARNING: This test runs in both single-node (4 GPUs) and multi-node
  (2 node with 2 GPUs each) modes. If the test only uses 2 GPUs, it is

tests/distributed/test_pipeline_partition.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 
 import pytest

tests/distributed/test_pp_cudagraph.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 
 import pytest

tests/distributed/test_pynccl.py

+# SPDX-License-Identifier: Apache-2.0
+
 import multiprocessing
 import os
 from typing import Dict, List

tests/distributed/test_same_node.py

+# SPDX-License-Identifier: Apache-2.0
+
 import os
 
 import torch.distributed as dist

tests/distributed/test_shm_broadcast.py

+# SPDX-License-Identifier: Apache-2.0
+
 import multiprocessing
 import random
 import time

tests/distributed/test_torchrun_example.py

+# SPDX-License-Identifier: Apache-2.0
+
 # unit test for `examples/offline_inference/torchrun_example.py`
 
 import random

tests/distributed/test_utils.py

+# SPDX-License-Identifier: Apache-2.0
+
 import socket
 
 import pytest

tests/encoder_decoder/test_e2e_correctness.py

+# SPDX-License-Identifier: Apache-2.0
 """E2E tests to verify the correctness of the encoder-decoder framework
 
 Run `pytest tests/encoder_decoder/test_e2e_correctness.py`.

tests/engine/output_processor/test_multi_step.py

+# SPDX-License-Identifier: Apache-2.0
+
 import random
 from unittest.mock import MagicMock