diff options
author | CNugteren <web@cedricnugteren.nl> | 2015-07-12 15:11:50 +0200 |
---|---|---|
committer | CNugteren <web@cedricnugteren.nl> | 2015-07-12 15:11:50 +0200 |
commit | b5d39d9d0c3e1084cb5131e2822d4fb754b0b412 (patch) | |
tree | 163d6dcb5367f619c88002dd4bf772aa8f13cb74 /include | |
parent | 9a929f3fb2081bd2fd8f68efce3b9d93e86bf611 (diff) |
Added the HEMM routine, tester, and client
Diffstat (limited to 'include')
-rw-r--r-- | include/clblast.h | 11 | ||||
-rw-r--r-- | include/internal/routines/xhemm.h | 58 |
2 files changed, 69 insertions, 0 deletions
diff --git a/include/clblast.h b/include/clblast.h index e6c49dbf..80ea1707 100644 --- a/include/clblast.h +++ b/include/clblast.h @@ -130,6 +130,17 @@ StatusCode Symm(const Layout layout, const Side side, const Triangle triangle, cl_mem c_buffer, const size_t c_offset, const size_t c_ld, cl_command_queue* queue, cl_event* event); +// Templated-precision hermitian matrix-matrix multiplication: CHEMM/ZHEMM +template <typename T> +StatusCode Hemm(const Layout layout, const Side side, const Triangle triangle, + const size_t m, const size_t n, + const T alpha, + const cl_mem a_buffer, const size_t a_offset, const size_t a_ld, + const cl_mem b_buffer, const size_t b_offset, const size_t b_ld, + const T beta, + cl_mem c_buffer, const size_t c_offset, const size_t c_ld, + cl_command_queue* queue, cl_event* event); + // Templated-precision rank-K update of a symmetric matrix: SSYRK/DSYRK/CSYRK/ZSYRK template <typename T> StatusCode Syrk(const Layout layout, const Triangle triangle, const Transpose a_transpose, diff --git a/include/internal/routines/xhemm.h b/include/internal/routines/xhemm.h new file mode 100644 index 00000000..1b1a0dfa --- /dev/null +++ b/include/internal/routines/xhemm.h @@ -0,0 +1,58 @@ + +// ================================================================================================= +// This file is part of the CLBlast project. The project is licensed under Apache Version 2.0. This +// project loosely follows the Google C++ styleguide and uses a tab-size of two spaces and a max- +// width of 100 characters per line. +// +// Author(s): +// Cedric Nugteren <www.cedricnugteren.nl> +// +// This file implements the Xhemm routine. It is based on the generalized matrix multiplication +// routine (Xgemm). The implementation is very similar to the Xsymm routine. +// +// ================================================================================================= + +#ifndef CLBLAST_ROUTINES_XHEMM_H_ +#define CLBLAST_ROUTINES_XHEMM_H_ + +#include "internal/routines/xgemm.h" + +namespace clblast { +// ================================================================================================= + +// See comment at top of file for a description of the class +template <typename T> +class Xhemm: public Xgemm<T> { + public: + + // Uses several variables from the Routine class + using Routine::db_; + using Routine::context_; + + // Uses several helper functions from the Routine class + using Routine::RunKernel; + using Routine::ErrorIn; + using Routine::TestMatrixA; + using Routine::GetProgramFromCache; + + // Uses the regular Xgemm routine + using Xgemm<T>::DoGemm; + + // Constructor + Xhemm(CommandQueue &queue, Event &event); + + // Templated-precision implementation of the routine + StatusCode DoHemm(const Layout layout, const Side side, const Triangle triangle, + const size_t m, const size_t n, + const T alpha, + const Buffer &a_buffer, const size_t a_offset, const size_t a_ld, + const Buffer &b_buffer, const size_t b_offset, const size_t b_ld, + const T beta, + const Buffer &c_buffer, const size_t c_offset, const size_t c_ld); +}; + +// ================================================================================================= +} // namespace clblast + +// CLBLAST_ROUTINES_XHEMM_H_ +#endif |