1. 1. 1. 🐛 Describe the bug

```python
import torch
torch.set_default_device('cuda')
@torch.compile
def random_func(x: torch.Tensor, seed: int) -> torch.Tensor:
torch.manual_seed(seed)

rnd = torch.randint(0, 2**32, size=x.shape, dtype=torch.uint32)
return x + rnd

random_func(torch.randn(()), 0)
```

1. 1. 1. Error logs

```
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_dynamo/eval_frame.py", line 929, in _fn
return fn(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_functorch/aot_autograd.py", line 1241, in forward
return compiled_fn(full_args)
^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 384, in runtime_wrapper
all_outs = call_func_at_runtime_with_args(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/utils.py", line 126, in call_func_at_runtime_with_args
out = normalize_as_list(f(args))
^^^^^^^
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_functorch/_aot_autograd/runtime_wrappers.py", line 556, in wrapper
return compiled_fn(runtime_args)
^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/inductor/output_code.py", line 584, in __call_
return self.current_callable(inputs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_inductor/utils.py", line 2716, in run
out = model(new_inputs)
^^^^^^^^^^^^^^^^^
File "/tmp/torchinductor_horace/lw/clwlxnkwrtchdr2hdokhjj2io3tm2vrqtmmy7usmxkqmrchztr6h.py", line 97, in call
triton_poi_fused_add_randint_0.run(arg0_1, buf0, buf1, 0, 1, stream=stream0)
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_inductor/runtime/triton_heuristics.py", line 1182, in run
return launcher(
^^^^^^^^^
File "<string>", line 5, in launcher
File "/tmp/horace/uv/install/6fbae45f965d50cba20076240d49c0af606f091cbea5976d1cc16bb095b35930/lib/python3.12/site-packages/torch/_inductor/runtime/static_cuda_launcher.py", line 227, in run
_StaticCudaLauncher._launch_kernel(
RuntimeError: CUDA driver error: invalid argument
```

1. 1. 1. Versions

pytorch 2.8.0

cc @chauhang @penguinwu @voznesenskym @EikanWang @jgong5 @Guobing-Chen @XiaobingSuper @zhuhaozhe @blzheng @wenzhe-nrv @jiayisunx @ipiszy @chenyang78 @kadeng @muchulee8 @amjames @aakhundov @coconutruben @jataylo