Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | CUDA: Clang separable compilation | Raul Tambre | 2020-09-24 | 1 | -0/+4 |
For NVCC the compiler takes care of device linking when passed the "-dlink" flag. Clang doesn't support such magic and requires the buildsystem to do the work that NVCC does behind the scenes. The implementation is based on Bazel's device linking documentation: https://github.com/tensorflow/tensorflow/blob/7cabcdf073abad8c46e9dda62bb8fa4682d2061e/third_party/nccl/build_defs.bzl.tpl#L259 Closes: #20726 |