2 * This file is part of the GROMACS molecular simulation package.
4 * Copyright (c) 1991-2000, University of Groningen, The Netherlands.
5 * Copyright (c) 2001-2012, The GROMACS Development Team
6 * Copyright (c) 2012,2013, by the GROMACS development team, led by
7 * David van der Spoel, Berk Hess, Erik Lindahl, and including many
8 * others, as listed in the AUTHORS file in the top-level source
9 * directory and at http://www.gromacs.org.
11 * GROMACS is free software; you can redistribute it and/or
12 * modify it under the terms of the GNU Lesser General Public License
13 * as published by the Free Software Foundation; either version 2.1
14 * of the License, or (at your option) any later version.
16 * GROMACS is distributed in the hope that it will be useful,
17 * but WITHOUT ANY WARRANTY; without even the implied warranty of
18 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
19 * Lesser General Public License for more details.
21 * You should have received a copy of the GNU Lesser General Public
22 * License along with GROMACS; if not, see
23 * http://www.gnu.org/licenses, or write to the Free Software Foundation,
24 * Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
26 * If you want to redistribute modifications to GROMACS, please
27 * consider that scientific software is very special. Version
28 * control is crucial - bugs must be traceable. We will be happy to
29 * consider code for inclusion in the official distribution, but
30 * derived work must not be called official GROMACS. Details are found
31 * in the README & COPYING files - if they are missing, get the
32 * official version at http://www.gromacs.org.
34 * To help us fund GROMACS development, we humbly ask that you cite
35 * the research papers on the package. Check out http://www.gromacs.org.
38 /* The macros in this file are intended to be used for writing
39 * architecture-independent SIMD intrinsics code.
40 * To support a new architecture, adding macros here should be (nearly)
44 /* This file contains vector operation functions using SIMD intrinsics.
45 * gmx_simd_macros.h should be included before including this file.
48 #ifndef _gmx_simd_vec_h_
49 #define _gmx_simd_vec_h_
51 #ifndef _gmx_simd_macros_h_
52 #error "gmx_simd_macros.h was not included before including gmx_simd_vec.h"
57 static gmx_inline gmx_mm_pr
58 gmx_calc_rsq_pr(gmx_mm_pr x
, gmx_mm_pr y
, gmx_mm_pr z
)
60 return gmx_madd_pr(z
, z
, gmx_madd_pr(y
, y
, gmx_mul_pr(x
, x
)));
63 /* inner-product of multiple vectors */
64 static gmx_inline gmx_mm_pr
65 gmx_iprod_pr(gmx_mm_pr ax
, gmx_mm_pr ay
, gmx_mm_pr az
,
66 gmx_mm_pr bx
, gmx_mm_pr by
, gmx_mm_pr bz
)
70 ret
= gmx_mul_pr(ax
, bx
);
71 ret
= gmx_madd_pr(ay
, by
, ret
);
72 ret
= gmx_madd_pr(az
, bz
, ret
);
77 /* norm squared of multiple vectors */
78 static gmx_inline gmx_mm_pr
79 gmx_norm2_pr(gmx_mm_pr ax
, gmx_mm_pr ay
, gmx_mm_pr az
)
83 ret
= gmx_mul_pr(ax
, ax
);
84 ret
= gmx_madd_pr(ay
, ay
, ret
);
85 ret
= gmx_madd_pr(az
, az
, ret
);
90 /* cross-product of multiple vectors */
91 static gmx_inline
void
92 gmx_cprod_pr(gmx_mm_pr ax
, gmx_mm_pr ay
, gmx_mm_pr az
,
93 gmx_mm_pr bx
, gmx_mm_pr by
, gmx_mm_pr bz
,
94 gmx_mm_pr
*cx
, gmx_mm_pr
*cy
, gmx_mm_pr
*cz
)
96 *cx
= gmx_mul_pr(ay
, bz
);
97 *cx
= gmx_nmsub_pr(az
, by
, *cx
);
99 *cy
= gmx_mul_pr(az
, bx
);
100 *cy
= gmx_nmsub_pr(ax
, bz
, *cy
);
102 *cz
= gmx_mul_pr(ax
, by
);
103 *cz
= gmx_nmsub_pr(ay
, bx
, *cz
);
106 /* a + b + c + d (not really a vector operation, but where else put this?) */
107 static gmx_inline gmx_mm_pr
108 gmx_sum4_pr(gmx_mm_pr a
, gmx_mm_pr b
, gmx_mm_pr c
, gmx_mm_pr d
)
110 return gmx_add_pr(gmx_add_pr(a
, b
), gmx_add_pr(c
, d
));
114 #endif /* _gmx_simd_vec_h_ */