experiments/STT-tensorflow
1
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity

Update comments

Browse Source
This commit is contained in:
jerryyin 2019-07-10 14:31:59 +00:00
parent ffe8958169
commit 4f249075c8
1 changed files with 3 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

7
tensorflow/compiler/xla/types.h
View File

@ -46,10 +46,9 @@ using complex128 = std::complex<double>;
using ::Eigen::half;
namespace gpu {
// GpuVersion is used to abstract GPU hardware version.
// On CUDA platform, it comprises CUDA compute capability which comprises a pair
// of integers denoting major verison and minor version.
// On ROCm platform, it comprises AMD GCN ISA version which is an integer.
// GpuVersion is used to abstract Gpu hardware version. On Cuda platform,
// it comprises a pair of integers denoting major and minor version.
// On ROCm platform, it comprises one integer for AMD GCN ISA version.
using GpuVersion = absl::variant<std::pair<int, int>, int>;
} // namespace gpu
} // namespace xla