/*
    DFT++ is a density functional package developed by the research group
    of Professor Tomas Arias

    Copyright 1996-2003 Sohrab Ismail-Beigi

    This file is part of DFT++.

    DFT++ is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version.

    DFT++ is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License
    along with DFT++; if not, write to the Free Software
    Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA

    Please see the file CREDITS for a list of authors.

    For academic users, we request that publications using results obtained with
    this software reference

    "New algebraic formulation of density functional calculation," by Sohrab Ismail-Beigi
    and T.A. Arias, Computer Physics Communications 128:1-2, 1-45 (June 2000).

    and, if using the wavelet basis, further reference

    "Multiresolution analysis of electronic structure: semicardinal and wavelet bases,"
    T.A. Arias, Reviews of Modern Physics 71:1, 267-311 (January 1999).

    and 

    "Robust ab initio calculation of condensed matter: transparent convergence through
    semicardinal multiresolution analysis,'' I.P. Daykov, T.A. Arias, and
    Torkel D. Engeness, Physical Review Letters, 90:21, 216402 (May 2003).

    For your convenience, preprints of the above articles may be obtained from
    http://arXiv.org/abs/cond-mat/9909130, 9805262, and 0204411, respectively.
*/

/*
 * Routines that do various matrix multiplications.  These routines
 * are the computational Kernels for all the matrix multiplies in the code,
 * so optimizing them is the way to improve matrix multiplication
 * performance.
 */
#ifndef DFT_MATRIX_MULT_H
#define DFT_MATRIX_MULT_H

void small_block_matrix_mult(int n1, int n2, int nc,
                             int s1, int s2, int sc,
                             scalar *b1,scalar *b2,scalar *result);

void Y1dagY2_block_matrix_mult(const ColumnBundle &Y1,
                               const ColumnBundle &Y2,
                               Matrix &M,
                               int n1, int n2, int N,
                               int offsetMrow,int offsetMcol,
                               int transpose,int offsetY2rowtranspose);

void Y_M_block_matrix_mult(const ColumnBundle &Y,
                           const Matrix &M,
                           ColumnBundle &YM,
                           int N, int nrM, int ncM,
                           int offsetMrow, int offsetMcol,
                           int transpose,
                           int accum);

void matrix_matrix_block_matrix_mult(const Matrix &m1,
                                     const Matrix &m2,
                                     Matrix &mprod);

#endif // DFT_MATRIX_MULT_H