[hotfix] fixed memory usage of shardformer module replacement (#5122)

colossalai/shardformer/layer/_operation.py

@@ -477,6 +477,7 @@ class _GatherForwardSplitBackward(torch.autograd.Function):
 
 class HookParameter(torch.autograd.Function):
     """In order to be hooked into Gemini's '__torch_function__', adding a view operation to weight and bias. Used in FusedLayerNorm"""
+
     @staticmethod
     def forward(ctx, input, weight, bias):
         ctx.save_for_backward(weight, bias)
@@ -497,7 +498,6 @@ def hook_paramter_in_backward(input, weight=None, bias=None):
     return HookParameter.apply(input, weight, bias)
 
 
-
 def _reduce(input_, process_group):
     # skip if only one rank involved
     if dist.get_world_size(process_group) == 1:
@@ -522,7 +522,7 @@ def _split(input_, dim=-1, process_group=None):
 
     tensor_list = torch.split(input_, dim_size // world_size, dim=dim)
     rank = dist.get_rank(process_group)
-    output = tensor_list[rank].contiguous()
+    output = tensor_list[rank].clone().contiguous()
 
     return output
 

colossalai/tensor/d_tensor/comm_spec.py

@@ -112,7 +112,7 @@ def _split(tensor: torch.Tensor, comm_spec: CommSpec):
     dim = comm_spec.shard_dim
     length = tensor.shape[comm_spec.shard_dim] // dist.get_world_size(process_group)
     start = length * dist.get_rank(process_group)
-    output = torch.narrow(tensor, dim, start, length).contiguous()
+    output = torch.narrow(tensor, dim, start, length).clone().contiguous()
     return output