New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Half fixes for ATen #2070
Half fixes for ATen #2070
Conversation
@soumith Is there a reason why in torch/lib/ATen/Half.h cuda_runtime.h was included instead of cuda.h? |
@csarofeen i dont remember. Maybe @zdevito remembers. |
@csarofeen GPU contbuilds are failing on this PR: https://build.pytorch.org/job/pytorch-PR-py2-linux/405//console |
No reason, cuda.h is fine. |
I wanted to get rid of cuda_runtime include but turns out I not only needed that but cuda.h as well for the version check on CUDA. |
this was merged into master. |
* Trying fuller cuda_cull * changes * Migrating to conda-forge for openmpi * Adding openmpi * Adding leveldb * Fixing unrelated minor conda bug * Another unrelated fix
Co-authored-by: Naoya Maruyama <nmaruyama@nvidia.com> Co-authored-by: Xiang Gao <qasdfgtyuiop@gmail.com>
Fixes for CUDA 9