diff options
| author | jsmall-nvidia <jsmall@nvidia.com> | 2021-04-23 11:32:07 -0400 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2021-04-23 11:32:07 -0400 |
| commit | 79e722338cd59aab74b4c57600c5ac6bce3bcd25 (patch) | |
| tree | af77066235c4038bd15c6297ef4f48d3e562171d /tools/gfx/cuda/render-cuda.cpp | |
| parent | a47e7751c2738543e872452debc7494369c9fb35 (diff) | |
Preliminary CUDA Half support (#1808)
* #include an absolute path didn't work - because paths were taken to always be relative.
* WIP CUDA half support.
* Working support for half on CUDA - requires cuda_fp16.h and associated files can be found.
* Fix for win32 for unused funcs.
* Fix for Clang.
* Hack to disable unused local function warning.
Diffstat (limited to 'tools/gfx/cuda/render-cuda.cpp')
| -rw-r--r-- | tools/gfx/cuda/render-cuda.cpp | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/tools/gfx/cuda/render-cuda.cpp b/tools/gfx/cuda/render-cuda.cpp index f871e6246..9d6b08122 100644 --- a/tools/gfx/cuda/render-cuda.cpp +++ b/tools/gfx/cuda/render-cuda.cpp @@ -1356,6 +1356,11 @@ public: SLANG_CUDA_RETURN_WITH_REPORT_ON_FAIL( cuCtxCreate(&m_context->m_context, 0, m_device), reportType); + // Not clear how to detect half support on CUDA. For now we'll assume we have it + { + m_features.add("half"); + } + // Initialize DeviceInfo { m_info.deviceType = DeviceType::CUDA; |
