[MoE][Dist] Fix Qwen MoE accuracy bug in DP senario

vllm_ascend/distributed/communicator.py

@@ -20,6 +20,7 @@
 import torch.distributed as dist
 from vllm.distributed.device_communicators.base_device_communicator import \
     DeviceCommunicatorBase
+from vllm.utils import logger
 
 
 class NPUCommunicator(DeviceCommunicatorBase):
@@ -34,6 +35,12 @@ def __init__(self,
         # init device according to rank
         self.device = torch.npu.current_device()
 
+        if self.use_all2all:
+            from vllm.distributed.device_communicators.all2all import \
+                NaiveAll2AllManager
+            self.all2all_manager = NaiveAll2AllManager(self.cpu_group)
+            logger.info("Using naive all2all manager.")
+
     def all_to_all(self,
                    input_: torch.Tensor,
                    scatter_dim: int = 0,
@@ -73,3 +80,16 @@ def all_to_all(self,
         dist.all_to_all(output_list, input_list, group=self.device_group)
         output_tensor = torch.cat(output_list, dim=gather_dim).contiguous()
         return output_tensor
+
+    def dispatch(
+            self, hidden_states: torch.Tensor,
+            router_logits: torch.Tensor) -> tuple[torch.Tensor, torch.Tensor]:
+        assert self.all2all_manager is not None
+        hidden_states, router_logits = self.all2all_manager.dispatch(
+            hidden_states, router_logits)
+        return hidden_states, router_logits
+
+    def combine(self, hidden_states: torch.Tensor) -> torch.Tensor:
+        assert self.all2all_manager is not None
+        hidden_states = self.all2all_manager.combine(hidden_states)
+        return hidden_states