frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

error message for instantiating CUDA Stream if CUDA not available (#159868)

Browse Source 
Fixes #159744
Summary:
```
import torch

# Generate input data
input_tensor = torch.randn(3, 3)
stream = torch.cuda.Stream()

# Call the API
input_tensor.record_stream(stream)
```

⚠️ will now show an error message
`torch.cuda.Stream requires CUDA support`

Pull Request resolved: https://github.com/pytorch/pytorch/pull/159868
Approved by: https://github.com/malfet, https://github.com/isuruf
This commit is contained in:
Raman Kumar
2025-10-11 23:21:35 +00:00
committed by PyTorch MergeBot
parent 8d49cd5b26
commit df26c51478
2 changed files with 4 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/test_torch.py
View File

@ -10495,7 +10495,7 @@ tensor([[[1.+1.j, 1.+1.j, 1.+1.j, ..., 1.+1.j, 1.+1.j, 1.+1.j],
def test_no_cuda_monkeypatch(self): def test_no_cuda_monkeypatch(self):
# Note that this is not in test_cuda.py as this whole file is skipped when cuda # Note that this is not in test_cuda.py as this whole file is skipped when cuda
# is not available. # is not available.
with self.assertRaisesRegex(RuntimeError, "Tried to instantiate dummy base class Stream"): with self.assertRaisesRegex(RuntimeError, "torch.cuda.Stream requires CUDA support"):
torch.cuda.Stream() torch.cuda.Stream()
with self.assertRaisesRegex(RuntimeError, "Tried to instantiate dummy base class Event"): with self.assertRaisesRegex(RuntimeError, "Tried to instantiate dummy base class Event"):

3
torch/cuda/streams.py
View File

@ -32,6 +32,9 @@ class Stream(torch._C._CudaStreamBase):
""" """
def __new__(cls, device=None, priority=0, **kwargs): def __new__(cls, device=None, priority=0, **kwargs):
# Check CUDA availability
if not torch.backends.cuda.is_built():
raise RuntimeError("torch.cuda.Stream requires CUDA support")
# setting device manager is expensive, so we avoid it unless necessary # setting device manager is expensive, so we avoid it unless necessary
if device is None or ("stream_id" in kwargs and "device_index" in kwargs): if device is None or ("stream_id" in kwargs and "device_index" in kwargs):
return super().__new__(cls, priority=priority, **kwargs) return super().__new__(cls, priority=priority, **kwargs)