Commit message (Collapse) | Author | Age | Files | Lines | ||
---|---|---|---|---|---|---|
... | ||||||
* | | Merge branch 'backport-clang-format-cuda' into clang-format-cuda | Brad King | 2017-02-14 | 1 | -0/+9 | |
|\ \ | |/ | ||||||
| * | clang-format.bash: Format CUDA source files too | Brad King | 2017-02-14 | 1 | -0/+9 | |
| | | | | | | | | | | | | Add `.clang-format` configuration files for Cuda test directories that use `Standard: Cpp11`. Otherwise clang-format splits the triple angle brackets used for CUDA kernels. | |||||
* | | Merge topic 'cmake_provide_include_dir_for_toolkit' | Brad King | 2017-02-14 | 3 | -0/+20 | |
|\ \ | |/ | | | | | | | 44f3acb2 CUDA: Detect the toolkit include directories | |||||
| * | CUDA: Detect the toolkit include directories | Robert Maynard | 2017-02-10 | 3 | -0/+20 | |
| | | | | | | | | | | The `nvcc -v` output provides what include directories need to be added to use the CUDA toolkit from other languages ( C/C++ ). | |||||
* | | Tests/Cuda: Select a CUDA device supporting compute 3.0 | Pierre Moreau | 2017-02-10 | 2 | -0/+46 | |
| | | ||||||
* | | Tests/Cuda: Return a non-zero code if errors occurred | Pierre Moreau | 2017-02-10 | 2 | -6/+8 | |
| | | ||||||
* | | Tests/Cuda: Use memory allocated on the GPU in the kernels | Pierre Moreau | 2017-02-10 | 2 | -10/+45 | |
| | | ||||||
* | | Tests/Cuda: Add missing separable compilation property | Pierre Moreau | 2017-02-10 | 1 | -0/+2 | |
| | | | | | | | | | | | | | | This resulted in `mixed_kernel()` returning an "invalid device function" at runtime for `file1_func()`. Suggested-by: Robert Maynard | |||||
* | | Tests/Cuda: Fix missing CUDA static library at runtime on macOS | Pierre Moreau | 2017-02-10 | 1 | -0/+6 | |
| | | | | | | | | Suggested-by: Robert Maynard | |||||
* | | Tests/Cuda: Output error messages to std::cerr instead of std::cout | Pierre Moreau | 2017-02-10 | 1 | -2/+2 | |
| | | ||||||
* | | Tests/Cuda: Print asynchronous error messages, if any | Pierre Moreau | 2017-02-10 | 3 | -0/+20 | |
| | | | | | | | | | | | | | | | | | | | | As kernel launches are asynchronous, a `cudaGetLastError()` right after the kernel launch might be executed while the kernel is still running. Synchronizing the device will ensure that all the work is completed before progressing further on, and allows to catch errors that were previously missed. The `cudaGetLastError()` after the `cudaDeviceSynchronize()` is there to reset the error variable to `cudaSuccess`. | |||||
* | | Tests/Cuda: Print error message if mixed_kernel failed | Pierre Moreau | 2017-02-10 | 1 | -0/+7 | |
| | | ||||||
* | | Tests/Cuda: Add identifiers to error messages | Pierre Moreau | 2017-02-10 | 2 | -2/+4 | |
| | | ||||||
* | | Tests/Cuda: Print error message if an error occurred | Pierre Moreau | 2017-02-10 | 2 | -2/+2 | |
|/ | | | | Fixes c59811a2 "CUDA: Tests now state why they are failing when no CUDA card is found." | |||||
* | CUDA: Test that CUDA flags are used when device linking executables. | Robert Maynard | 2017-01-12 | 5 | -0/+48 | |
| | ||||||
* | CUDA: Port test cases to Windows with MSVC host compiler | Brad King | 2017-01-12 | 5 | -9/+43 | |
| | ||||||
* | CUDA: ConsumeCompileFeatures use cxx_nullptr for wider compiler support | Robert Maynard | 2017-01-11 | 1 | -1/+1 | |
| | | | | | We need to use a C++11 feature that is supported by the widest range of compilers, so we chose nullptr instead of constexpr. | |||||
* | CUDA: Tests now state why they are failing when no CUDA card is found. | Robert Maynard | 2016-12-15 | 3 | -11/+15 | |
| | ||||||
* | CUDA: Fix Cuda.Complex test case extern function signatures | Brad King | 2016-12-09 | 1 | -2/+2 | |
| | | | | | Make the return types of the extern function declared in `main.cpp` match those in the method implementations. | |||||
* | CUDA: Add tests to verify CUDA compiler works properly. | Robert Maynard | 2016-11-14 | 20 | -0/+316 | |