Fix histc with empty tensor error (#38987)
Summary:
Fix https://github.com/pytorch/pytorch/issues/38979
The error in mentioned https://github.com/pytorch/pytorch/issues/38979 is a [`cudaErrorInvalidConfiguration` error](https://docs.nvidia.com/cuda/cuda-runtime-api/group__CUDART__TYPES.html#group__CUDART__TYPES_1g3f51e3575c2178246db0a94a430e0038):
> This indicates that a kernel launch is requesting resources that can never be satisfied by the current device. Requesting more shared memory per block than the device supports will trigger this error, as will requesting too many threads or blocks. See cudaDeviceProp for more device limitations.
This is because we are trying to launch a kernel with block size 0.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/38987
Differential Revision: D21722993
Pulled By: ezyang
fbshipit-source-id: 2c283e0a9f542b4acb96e895a43b991ccac808fe
