cpp extension tutorial: not device agnostic?

[c-hofer]

Would the kerne calll in lltm_cuda_forward in the tutorial tutorials/advanced_source/cpp_extension.rst fail on multi gpu systems if the inputs are not on the default device, i.e., device:0?

To my understanding, some "magic" takes care of setting the right context if we add functionality do pytorch via custom kernels, see here.
However, it seems like in the tutorial this machinery is not used.
Explicit usage of at::OptionalDeviceGuard should resolve the issue (?) in the tutorial.

[carljparker]

I believe that this issue is addressed by the comments--especially the final comments--in #1196 . . . Also see the discussion in ClementPinard/Pytorch-Correlation-extension#32