diff options
author | Cedric Nugteren <web@cedricnugteren.nl> | 2017-03-05 15:06:14 +0100 |
---|---|---|
committer | Cedric Nugteren <web@cedricnugteren.nl> | 2017-03-05 15:06:14 +0100 |
commit | b114ea49a9228ee7a8bf1b00c092324c0ce972c3 (patch) | |
tree | 5aa34777226bdb11c6349c9ce43de160478d9b41 /src/clblast.cpp | |
parent | cdf354f89524ed88d4f9358004e5a8eabd9ce286 (diff) |
Added first naive version of the batched AXPY routine
Diffstat (limited to 'src/clblast.cpp')
-rw-r--r-- | src/clblast.cpp | 28 |
1 files changed, 14 insertions, 14 deletions
diff --git a/src/clblast.cpp b/src/clblast.cpp index 55562419..e9cac664 100644 --- a/src/clblast.cpp +++ b/src/clblast.cpp @@ -2178,8 +2178,8 @@ template StatusCode PUBLIC_API Omatcopy<half>(const Layout, const Transpose, template <typename T> StatusCode AxpyBatched(const size_t n, const T *alphas, - const cl_mem *x_buffers, const size_t x_offset, const size_t x_inc, - cl_mem *y_buffers, const size_t y_offset, const size_t y_inc, + const cl_mem *x_buffers, const size_t x_inc, + cl_mem *y_buffers, const size_t y_inc, const size_t batch_count, cl_command_queue* queue, cl_event* event) { try { @@ -2195,40 +2195,40 @@ StatusCode AxpyBatched(const size_t n, } routine.DoAxpyBatched(n, alphas_cpp, - x_buffers_cpp, x_offset, x_inc, - y_buffers_cpp, y_offset, y_inc, + x_buffers_cpp, x_inc, + y_buffers_cpp, y_inc, batch_count); return StatusCode::kSuccess; } catch (...) { return DispatchException(); } } template StatusCode PUBLIC_API AxpyBatched<float>(const size_t, const float*, - const cl_mem*, const size_t, const size_t, - cl_mem*, const size_t, const size_t, + const cl_mem*, const size_t, + cl_mem*, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode PUBLIC_API AxpyBatched<double>(const size_t, const double*, - const cl_mem*, const size_t, const size_t, - cl_mem*, const size_t, const size_t, + const cl_mem*, const size_t, + cl_mem*, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode PUBLIC_API AxpyBatched<float2>(const size_t, const float2*, - const cl_mem*, const size_t, const size_t, - cl_mem*, const size_t, const size_t, + const cl_mem*, const size_t, + cl_mem*, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode PUBLIC_API AxpyBatched<double2>(const size_t, const double2*, - const cl_mem*, const size_t, const size_t, - cl_mem*, const size_t, const size_t, + const cl_mem*, const size_t, + cl_mem*, const size_t, const size_t, cl_command_queue*, cl_event*); template StatusCode PUBLIC_API AxpyBatched<half>(const size_t, const half*, - const cl_mem*, const size_t, const size_t, - cl_mem*, const size_t, const size_t, + const cl_mem*, const size_t, + cl_mem*, const size_t, const size_t, cl_command_queue*, cl_event*); // ================================================================================================= |