diff --git a/colossalai/tensor/dist_spec_mgr.py b/colossalai/tensor/dist_spec_mgr.py
index c82524836..a3e229899 100644
--- a/colossalai/tensor/dist_spec_mgr.py
+++ b/colossalai/tensor/dist_spec_mgr.py
@@ -34,7 +34,7 @@ class DistSpecManager:
             chunk_size = divide(tensor.size(dim), dist_spec.num_partitions[i])
             chunk = chunk.narrow(dim, idx // num_parts * chunk_size, chunk_size)
             idx %= num_parts
-        return chunk.detach().contiguous()
+        return chunk.clone().detach().contiguous()
 
     @staticmethod
     def _gather(tensor: torch.Tensor, old_dist_spec: _DistSpec) -> torch.Tensor:
diff --git a/tests/test_tensor/test_dist_spec_mgr.py b/tests/test_tensor/test_dist_spec_mgr.py
index ada77faef..f21790da1 100644
--- a/tests/test_tensor/test_dist_spec_mgr.py
+++ b/tests/test_tensor/test_dist_spec_mgr.py
@@ -33,8 +33,25 @@ def run():
     assert torch.equal(x, DistSpecManager._gather(mat_shard, mat_spec))
 
 
+def check_mem():
+    group = _get_default_group()
+    size = dist.get_world_size()
+    assert torch.cuda.memory_allocated() == 0
+    x = torch.rand(32, 32).cuda()
+    orig_mem = x.numel() * x.element_size()
+    assert torch.cuda.memory_allocated() == orig_mem
+    old_dist_spec = distspec.replicate()
+    row_spec = distspec.shard(group, [0], [size])
+    x.data = DistSpecManager._shard_as(x, old_dist_spec, row_spec)
+    assert x.size(0) == 32 // size and x.size(1) == 32
+    assert torch.cuda.memory_allocated() == orig_mem // size
+    x.data = DistSpecManager._gather(x, row_spec)
+    assert torch.cuda.memory_allocated() == orig_mem
+
+
 def run_dist(rank, world_size, port):
     colossalai.launch(config={}, rank=rank, world_size=world_size, host='localhost', port=port, backend='nccl')
+    check_mem()
     run()