Soumith Chintala

cc: @smessmer @ezyang

you cannot printf a cuda pointer, it will segfault. Maybe you can lightly read the CUDA programming guide: docs.nvidia.com/cuda/cuda-c-programming-guide

@Maratyszcza can you review

I reproduced this on docker today, and fixed the issue with this commit https://github.com/pytorch/extension-cpp/commit/1031028f3b048fdea84372f3b81498db53d64d98

whoops, this is my bad. let me re-setup the environment and see what I can do about this.

@jayenashar it used to a long time ago, but the scripts now don't. you should be using the `pytorch` conda channel.

landed #16, ball is in your court now.

because we want to pin to specific cudnn versions and only bump the cudnn versions when we validate them.

@Spandan-Madan this is basically flaking on an ABI incompatibility. (gcc > 5.1 binaries have different std::string ABI than gccc

Good Question. We have clarity on most of the answer, and are refining some of it. What should be in the `research` category? - artifacts of a published or an...