Upgrade to new vllm commit (#3719)

### What this PR does / why we need it?
Upgrade to new vllm commit:
c9461e05a4

- Fix many imports, caused by
https://github.com/vllm-project/vllm/pull/26908
- Fix import ```sha256```, caused by
https://github.com/vllm-project/vllm/pull/27169
- Remove ```SchedulerConfig.send_delta_data```, caused by
https://github.com/vllm-project/vllm/pull/27142
- Fix ```FusedMoE``` because of dual stream execution, caused by
https://github.com/vllm-project/vllm/pull/26440

### Does this PR introduce _any_ user-facing change?
N/A

### How was this patch tested?
CI passed with new added/existing test.


- vLLM version: v0.11.0rc3
- vLLM main:
17c540a993

---------

Signed-off-by: MengqingCao <cmq0113@163.com>
Signed-off-by: Icey <1790571317@qq.com>
Co-authored-by: MengqingCao <cmq0113@163.com>

vllm_ascend/distributed/cpu_offload_manager/metadata.py

@@ -9,11 +9,18 @@ import torch
 import vllm.envs as envs
 import zmq
 from vllm.config import KVTransferConfig, VllmConfig
-from vllm.utils import get_dtype_size, logger, make_zmq_socket
+from vllm.utils import logger
 from vllm.v1.kv_cache_interface import AttentionSpec
 
 from vllm_ascend.distributed.cpu_offload_manager.cpu_kv_cache_manager import \
     CPUKVCacheManager
+from vllm_ascend.utils import vllm_version_is
+
+if vllm_version_is("0.11.0"):
+    from vllm.utils import get_dtype_size, make_zmq_socket
+else:
+    from vllm.utils.network_utils import make_zmq_socket
+    from vllm.utils.torch_utils import get_dtype_size
 
 
 @dataclass

vllm_ascend/distributed/llmdatadist_c_mgr_connector.py

@@ -25,19 +25,25 @@ from vllm.distributed.kv_transfer.kv_connector.v1.base import (
 from vllm.distributed.parallel_state import (get_dcp_group, get_tp_group,
                                              get_world_group)
 from vllm.forward_context import ForwardContext
-from vllm.utils import get_ip, logger
+from vllm.utils import logger
 from vllm.v1.core.kv_cache_manager import KVCacheBlocks
 from vllm.v1.core.sched.output import SchedulerOutput
 from vllm.v1.request import Request, RequestStatus
 
 import vllm_ascend.envs as envs_ascend
 from vllm_ascend.utils import (AscendSocVersion, get_ascend_soc_version,
-                               prefill_context_parallel_enable)
+                               prefill_context_parallel_enable,
+                               vllm_version_is)
 
 if prefill_context_parallel_enable():
     from vllm.distributed.parallel_state import \
         get_prefill_context_model_parallel_rank
 
+if vllm_version_is("0.11.0"):
+    from vllm.utils import get_ip
+else:
+    from vllm.utils.network_utils import get_ip
+
 TORCH_DTYPE_TO_NPU_DTYPE = {
     torch.half: llm_datadist.DataType.DT_FLOAT16,
     torch.float16: llm_datadist.DataType.DT_FLOAT16,

vllm_ascend/distributed/mooncake/mooncake_engine.py

@@ -7,7 +7,7 @@ from typing import Generator, List, Optional, Union
 # Third Party
 import torch
 from vllm.config import VllmConfig
-from vllm.utils import get_kv_cache_torch_dtype, logger
+from vllm.utils import logger
 
 from vllm_ascend.distributed.mooncake.config_data import (
     ChunkedTokenDatabase, LasyerMultiBlockReqMeta, MooncakeConnectorMetadata,
@@ -16,6 +16,12 @@ from vllm_ascend.distributed.mooncake.kv_transfer import (
     KVCacheStoreLayerRecvingThread, KVCacheStoreLayerSendingThread,
     KVCacheStoreRecvingThread, KVCacheStoreSendingThread, KVTransferThread)
 from vllm_ascend.distributed.mooncake.mooncake_store import Mooncakestore
+from vllm_ascend.utils import vllm_version_is
+
+if vllm_version_is("0.11.0"):
+    from vllm.utils import get_kv_cache_torch_dtype
+else:
+    from vllm.utils.torch_utils import get_kv_cache_torch_dtype
 
 
 class MooncakeEngine:

vllm_ascend/distributed/mooncake_connector.py

@@ -26,13 +26,19 @@ from vllm.distributed.kv_transfer.kv_connector.v1.base import (
     KVConnectorBase_V1, KVConnectorMetadata, KVConnectorRole)
 from vllm.distributed.parallel_state import (get_tensor_model_parallel_rank,
                                              get_tp_group)
-from vllm.utils import get_ip, logger, make_zmq_path, make_zmq_socket
+from vllm.utils import logger
 from vllm.v1.core.sched.output import SchedulerOutput
 from vllm.v1.request import RequestStatus
 
 import vllm_ascend.envs as envs_ascend
 from vllm_ascend.ascend_config import get_ascend_config, init_ascend_config
 from vllm_ascend.distributed.mooncake.transfer_engine import get_global_te
+from vllm_ascend.utils import vllm_version_is
+
+if vllm_version_is("0.11.0"):
+    from vllm.utils import get_ip, make_zmq_path, make_zmq_socket
+else:
+    from vllm.utils.network_utils import get_ip, make_zmq_path, make_zmq_socket
 
 if TYPE_CHECKING:
     from vllm.attention.backends.abstract import AttentionMetadata

vllm_ascend/distributed/mooncake_layerwise_connector.py

@@ -26,7 +26,7 @@ from vllm.distributed.kv_transfer.kv_connector.v1.base import (
     KVConnectorBase_V1, KVConnectorMetadata, KVConnectorRole)
 from vllm.distributed.parallel_state import (get_tensor_model_parallel_rank,
                                              get_tp_group, get_world_group)
-from vllm.utils import get_ip, logger, make_zmq_path, make_zmq_socket
+from vllm.utils import logger
 from vllm.v1.core.sched.output import SchedulerOutput
 from vllm.v1.request import RequestStatus
 
@@ -34,6 +34,12 @@ import vllm_ascend.envs as envs_ascend
 from vllm_ascend.ascend_config import get_ascend_config
 from vllm_ascend.distributed.utils import (align_memory,
                                            kv_alltoall_and_rearrange)
+from vllm_ascend.utils import vllm_version_is
+
+if vllm_version_is("0.11.0"):
+    from vllm.utils import get_ip, make_zmq_path, make_zmq_socket
+else:
+    from vllm.utils.network_utils import get_ip, make_zmq_path, make_zmq_socket
 
 if TYPE_CHECKING:
     from vllm.attention.backends.abstract import AttentionMetadata