_igemm.hpp

Go to the documentation of this file.

/*************************************************************************
 *
 *  Project
 *                         _____ _____  __  __ _____
 *                        / ____|  __ \|  \/  |  __ \
 *  ___  _ __   ___ _ __ | |  __| |__) | \  / | |__) |
 * / _ \| '_ \ / _ \ '_ \| | |_ |  ___/| |\/| |  ___/
 *| (_) | |_) |  __/ | | | |__| | |    | |  | | |
 * \___/| .__/ \___|_| |_|\_____|_|    |_|  |_|_|
 *      | |
 *      |_|
 *
 * Copyright (C) Akiel Aries, <akiel@akiel.org>, et al.
 *
 * This software is licensed as described in the file LICENSE, which
 * you should have received as part of this distribution. The terms
 * among other details are referenced in the official documentation
 * seen here : https://akielaries.github.io/openGPMP/ along with
 * important files seen in this project.
 *
 * You may opt to use, copy, modify, merge, publish, distribute
 * and/or sell copies of the Software, and permit persons to whom
 * the Software is furnished to do so, under the terms of the
 * LICENSE file. As this is an Open Source effort, all implementations
 * must be of the same methodology.
 *
 *
 *
 * This software is distributed on an AS IS basis, WITHOUT
 * WARRANTY OF ANY KIND, either express or implied.
 *
 ************************************************************************/
#ifndef _IGEMM_HPP
#define _IGEMM_HPP
 
#define BLOCK_SZ_M 384  
#define BLOCK_SZ_K 384  
#define BLOCK_SZ_N 4096 
#define BLOCK_SZ_MR 4   
#define BLOCK_SZ_NR 4   
namespace gpmp {
namespace linalg {
 
class IGEMM {
  public:
    static int IGEMM_BUFF_A[BLOCK_SZ_M * BLOCK_SZ_K];
    static int IGEMM_BUFF_B[BLOCK_SZ_K * BLOCK_SZ_N];
    static int IGEMM_BUFF_C[BLOCK_SZ_MR * BLOCK_SZ_NR];
 
    void
    pack_micro_A(int k, const int *A, int incRowA, int incColA, int *buffer);
 
    void pack_buffer_A(int mc,
                       int kc,
                       const int *A,
                       int incRowA,
                       int incColA,
                       int *buffer);
 
    void
    pack_micro_B(int k, const int *B, int incRowB, int incColB, int *buffer);
 
    void pack_buffer_B(int kc,
                       int nc,
                       const int *B,
                       int incRowB,
                       int incColB,
                       int *buffer);
 
    void igemm_micro_kernel(int kc,
                            int alpha,
                            const int *A,
                            const int *B,
                            int beta,
                            int *C,
                            int incRowC,
                            int incColC);
 
    void igeaxpy(int m,
                 int n,
                 int alpha,
                 const int *X,
                 int incRowX,
                 int incColX,
                 int *Y,
                 int incRowY,
                 int incColY);
 
    void igescal(int m, int n, int alpha, int *X, int incRowX, int incColX);
 
    void igemm_macro_kernel(int mc,
                            int nc,
                            int kc,
                            int alpha,
                            int beta,
                            int *C,
                            int incRowC,
                            int incColC);
 
    void igemm_nn(int m,
                  int n,
                  int k,
                  int alpha,
                  const int *A,
                  int incRowA,
                  int incColA,
                  const int *B,
                  int incRowB,
                  int incColB,
                  int beta,
                  int *C,
                  int incRowC,
                  int incColC);
};
 
} // namespace linalg
} // namespace gpmp
 
#endif