Merge branch 'gpus/half_kernels' into 'master'
Introduce half-precision conversion and gemm kernels for GPUs See merge request !395
No related branches found
No related tags found
Showing
- CMakeLists.txt 3 additions, 3 deletionsCMakeLists.txt
- ChangeLog 5 additions, 0 deletionsChangeLog
- cmake_modules/local_subs.py 9 additions, 2 deletionscmake_modules/local_subs.py
- cmake_modules/morse_cmake 1 addition, 1 deletioncmake_modules/morse_cmake
- control/auxiliary.c 9 additions, 5 deletionscontrol/auxiliary.c
- control/control.c 3 additions, 2 deletionscontrol/control.c
- control/descriptor.c 19 additions, 8 deletionscontrol/descriptor.c
- control/descriptor.h 3 additions, 3 deletionscontrol/descriptor.h
- gpucublas/compute/CMakeLists.txt 38 additions, 2 deletionsgpucublas/compute/CMakeLists.txt
- gpucublas/compute/cuda_dlag2h.cu 290 additions, 0 deletionsgpucublas/compute/cuda_dlag2h.cu
- gpucublas/compute/cuda_gemmex.c 43 additions, 0 deletionsgpucublas/compute/cuda_gemmex.c
- gpucublas/compute/cuda_hgemm.c 42 additions, 0 deletionsgpucublas/compute/cuda_hgemm.c
- gpucublas/compute/cuda_zlag2c.cu 304 additions, 0 deletionsgpucublas/compute/cuda_zlag2c.cu
- gpucublas/include/gpucublas.h 53 additions, 2 deletionsgpucublas/include/gpucublas.h
- gpuhipblas/compute/CMakeLists.txt 3 additions, 2 deletionsgpuhipblas/compute/CMakeLists.txt
- gpuhipblas/compute/hip_hgemm.c 41 additions, 0 deletionsgpuhipblas/compute/hip_hgemm.c
- gpuhipblas/include/gpuhipblas.h 12 additions, 2 deletionsgpuhipblas/include/gpuhipblas.h
- include/chameleon.h 3 additions, 3 deletionsinclude/chameleon.h
- include/chameleon/constants.h 67 additions, 8 deletionsinclude/chameleon/constants.h
- include/chameleon/runtime_struct.h 3 additions, 2 deletionsinclude/chameleon/runtime_struct.h
Loading
Please register or sign in to comment