3 * MIPS Technologies, Inc., California.
5 * Redistribution and use in source and binary forms, with or without
6 * modification, are permitted provided that the following conditions
8 * 1. Redistributions of source code must retain the above copyright
9 * notice, this list of conditions and the following disclaimer.
10 * 2. Redistributions in binary form must reproduce the above copyright
11 * notice, this list of conditions and the following disclaimer in the
12 * documentation and/or other materials provided with the distribution.
13 * 3. Neither the name of the MIPS Technologies, Inc., nor the names of is
14 * contributors may be used to endorse or promote products derived from
15 * this software without specific prior written permission.
17 * THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND
18 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
19 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
20 * ARE DISCLAIMED. IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE
21 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
22 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
23 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
24 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
25 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
26 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 * Author: Zoran Lukic (zoranl@mips.com)
31 * This file is part of FFmpeg.
33 * FFmpeg is free software; you can redistribute it and/or
34 * modify it under the terms of the GNU Lesser General Public
35 * License as published by the Free Software Foundation; either
36 * version 2.1 of the License, or (at your option) any later version.
38 * FFmpeg is distributed in the hope that it will be useful,
39 * but WITHOUT ANY WARRANTY; without even the implied warranty of
40 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
41 * Lesser General Public License for more details.
43 * You should have received a copy of the GNU Lesser General Public
44 * License along with FFmpeg; if not, write to the Free Software
45 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
48 #include "libavcodec/dsputil.h"
50 static void vector_fmul_window_mips(float *dst, const float *src0,
51 const float *src1, const float *win, int len)
55 * variables used in inline assembler
57 float * dst_i, * dst_j, * dst_i2, * dst_j2;
58 float temp, temp1, temp2, temp3, temp4, temp5, temp6, temp7;
64 for (i = -len, j = len - 1; i < 0; i += 8, j -= 8) {
73 "mul.s %[temp], %[s1], %[wi] \n\t"
74 "mul.s %[temp1], %[s1], %[wj] \n\t"
75 "mul.s %[temp2], %[s11], %[wi1] \n\t"
76 "mul.s %[temp3], %[s11], %[wj1] \n\t"
78 "msub.s %[temp], %[temp], %[s0], %[wj] \n\t"
79 "madd.s %[temp1], %[temp1], %[s0], %[wi] \n\t"
80 "msub.s %[temp2], %[temp2], %[s01], %[wj1] \n\t"
81 "madd.s %[temp3], %[temp3], %[s01], %[wi1] \n\t"
83 "swc1 %[temp], 0(%[dst_i]) \n\t" /* dst[i] = s0*wj - s1*wi; */
84 "swc1 %[temp1], 0(%[dst_j]) \n\t" /* dst[j] = s0*wi + s1*wj; */
85 "swc1 %[temp2], 4(%[dst_i]) \n\t" /* dst[i+1] = s01*wj1 - s11*wi1; */
86 "swc1 %[temp3], -4(%[dst_j]) \n\t" /* dst[j-1] = s01*wi1 + s11*wj1; */
88 "mul.s %[temp4], %[s12], %[wi2] \n\t"
89 "mul.s %[temp5], %[s12], %[wj2] \n\t"
90 "mul.s %[temp6], %[s13], %[wi3] \n\t"
91 "mul.s %[temp7], %[s13], %[wj3] \n\t"
93 "msub.s %[temp4], %[temp4], %[s02], %[wj2] \n\t"
94 "madd.s %[temp5], %[temp5], %[s02], %[wi2] \n\t"
95 "msub.s %[temp6], %[temp6], %[s03], %[wj3] \n\t"
96 "madd.s %[temp7], %[temp7], %[s03], %[wi3] \n\t"
98 "swc1 %[temp4], 8(%[dst_i]) \n\t" /* dst[i+2] = s02*wj2 - s12*wi2; */
99 "swc1 %[temp5], -8(%[dst_j]) \n\t" /* dst[j-2] = s02*wi2 + s12*wj2; */
100 "swc1 %[temp6], 12(%[dst_i]) \n\t" /* dst[i+2] = s03*wj3 - s13*wi3; */
101 "swc1 %[temp7], -12(%[dst_j]) \n\t" /* dst[j-3] = s03*wi3 + s13*wj3; */
102 : [temp]"=&f"(temp), [temp1]"=&f"(temp1), [temp2]"=&f"(temp2),
103 [temp3]"=&f"(temp3), [temp4]"=&f"(temp4), [temp5]"=&f"(temp5),
104 [temp6]"=&f"(temp6), [temp7]"=&f"(temp7)
105 : [dst_j]"r"(dst_j), [dst_i]"r" (dst_i),
106 [s0] "f"(src0[i]), [wj] "f"(win[j]), [s1] "f"(src1[j]),
107 [wi] "f"(win[i]), [s01]"f"(src0[i + 1]),[wj1]"f"(win[j - 1]),
108 [s11]"f"(src1[j - 1]), [wi1]"f"(win[i + 1]), [s02]"f"(src0[i + 2]),
109 [wj2]"f"(win[j - 2]), [s12]"f"(src1[j - 2]),[wi2]"f"(win[i + 2]),
110 [s03]"f"(src0[i + 3]), [wj3]"f"(win[j - 3]), [s13]"f"(src1[j - 3]),
116 "mul.s %[temp], %[s1], %[wi] \n\t"
117 "mul.s %[temp1], %[s1], %[wj] \n\t"
118 "mul.s %[temp2], %[s11], %[wi1] \n\t"
119 "mul.s %[temp3], %[s11], %[wj1] \n\t"
121 "msub.s %[temp], %[temp], %[s0], %[wj] \n\t"
122 "madd.s %[temp1], %[temp1], %[s0], %[wi] \n\t"
123 "msub.s %[temp2], %[temp2], %[s01], %[wj1] \n\t"
124 "madd.s %[temp3], %[temp3], %[s01], %[wi1] \n\t"
126 "swc1 %[temp], 0(%[dst_i2]) \n\t" /* dst[i] = s0*wj - s1*wi; */
127 "swc1 %[temp1], 0(%[dst_j2]) \n\t" /* dst[j] = s0*wi + s1*wj; */
128 "swc1 %[temp2], 4(%[dst_i2]) \n\t" /* dst[i+1] = s01*wj1 - s11*wi1; */
129 "swc1 %[temp3], -4(%[dst_j2]) \n\t" /* dst[j-1] = s01*wi1 + s11*wj1; */
131 "mul.s %[temp4], %[s12], %[wi2] \n\t"
132 "mul.s %[temp5], %[s12], %[wj2] \n\t"
133 "mul.s %[temp6], %[s13], %[wi3] \n\t"
134 "mul.s %[temp7], %[s13], %[wj3] \n\t"
136 "msub.s %[temp4], %[temp4], %[s02], %[wj2] \n\t"
137 "madd.s %[temp5], %[temp5], %[s02], %[wi2] \n\t"
138 "msub.s %[temp6], %[temp6], %[s03], %[wj3] \n\t"
139 "madd.s %[temp7], %[temp7], %[s03], %[wi3] \n\t"
141 "swc1 %[temp4], 8(%[dst_i2]) \n\t" /* dst[i+2] = s02*wj2 - s12*wi2; */
142 "swc1 %[temp5], -8(%[dst_j2]) \n\t" /* dst[j-2] = s02*wi2 + s12*wj2; */
143 "swc1 %[temp6], 12(%[dst_i2]) \n\t" /* dst[i+2] = s03*wj3 - s13*wi3; */
144 "swc1 %[temp7], -12(%[dst_j2]) \n\t" /* dst[j-3] = s03*wi3 + s13*wj3; */
146 [temp1]"=&f"(temp1), [temp2]"=&f"(temp2), [temp3]"=&f"(temp3),
147 [temp4]"=&f"(temp4), [temp5]"=&f"(temp5), [temp6]"=&f"(temp6),
148 [temp7] "=&f" (temp7)
149 : [dst_j2]"r"(dst_j2), [dst_i2]"r"(dst_i2),
150 [s0] "f"(src0[i + 4]), [wj] "f"(win[j - 4]), [s1] "f"(src1[j - 4]),
151 [wi] "f"(win[i + 4]), [s01]"f"(src0[i + 5]),[wj1]"f"(win[j - 5]),
152 [s11]"f"(src1[j - 5]), [wi1]"f"(win[i + 5]), [s02]"f"(src0[i + 6]),
153 [wj2]"f"(win[j - 6]), [s12]"f"(src1[j - 6]),[wi2]"f"(win[i + 6]),
154 [s03]"f"(src0[i + 7]), [wj3]"f"(win[j - 7]), [s13]"f"(src1[j - 7]),
161 av_cold void ff_dsputil_init_mips( DSPContext* c, AVCodecContext *avctx )
163 c->vector_fmul_window = vector_fmul_window_mips;