1 /*
2 * Format Conversion Utils for MIPS
3 *
4 * Copyright (c) 2012
5 * MIPS Technologies, Inc., California.
6 *
7 * Redistribution and use in source and binary forms, with or without
8 * modification, are permitted provided that the following conditions
9 * are met:
10 * 1. Redistributions of source code must retain the above copyright
11 * notice, this list of conditions and the following disclaimer.
12 * 2. Redistributions in binary form must reproduce the above copyright
13 * notice, this list of conditions and the following disclaimer in the
14 * documentation and/or other materials provided with the distribution.
15 * 3. Neither the name of the MIPS Technologies, Inc., nor the names of is
16 * contributors may be used to endorse or promote products derived from
17 * this software without specific prior written permission.
18 *
19 * THIS SOFTWARE IS PROVIDED BY THE MIPS TECHNOLOGIES, INC. ``AS IS'' AND
20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
22 * ARE DISCLAIMED. IN NO EVENT SHALL THE MIPS TECHNOLOGIES, INC. BE LIABLE
23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
29 * SUCH DAMAGE.
30 *
31 * Author: Zoran Lukic (zoranl@mips.com)
32 * Author: Nedeljko Babic (nbabic@mips.com)
33 *
34 * This file is part of FFmpeg.
35 *
36 * FFmpeg is free software; you can redistribute it and/or
37 * modify it under the terms of the GNU Lesser General Public
38 * License as published by the Free Software Foundation; either
39 * version 2.1 of the License, or (at your option) any later version.
40 *
41 * FFmpeg is distributed in the hope that it will be useful,
42 * but WITHOUT ANY WARRANTY; without even the implied warranty of
43 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
44 * Lesser General Public License for more details.
45 *
46 * You should have received a copy of the GNU Lesser General Public
47 * License along with FFmpeg; if not, write to the Free Software
48 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
49 */
50 #include "config.h"
51 #include "libavcodec/avcodec.h"
52 #include "libavcodec/fmtconvert.h"
53 #include "libavutil/mips/asmdefs.h"
54
55 #if HAVE_INLINE_ASM
int32_to_float_fmul_scalar_mips(float * dst,const int * src,float mul,int len)56 static void int32_to_float_fmul_scalar_mips(float *dst, const int *src,
57 float mul, int len)
58 {
59 /*
60 * variables used in inline assembler
61 */
62 float temp1, temp3, temp5, temp7, temp9, temp11, temp13, temp15;
63
64 int rpom1, rpom2, rpom11, rpom21, rpom12, rpom22, rpom13, rpom23;
65 const int *src_end = src + len;
66 /*
67 * loop is 8 times unrolled in assembler in order to achieve better performance
68 */
69 __asm__ volatile (
70 "i32tf_lp%=: \n\t"
71 "lw %[rpom11], 0(%[src]) \n\t"
72 "lw %[rpom21], 4(%[src]) \n\t"
73 "lw %[rpom1], 8(%[src]) \n\t"
74 "lw %[rpom2], 12(%[src]) \n\t"
75 "mtc1 %[rpom11], %[temp1] \n\t"
76 "mtc1 %[rpom21], %[temp3] \n\t"
77 "mtc1 %[rpom1], %[temp5] \n\t"
78 "mtc1 %[rpom2], %[temp7] \n\t"
79
80 "lw %[rpom13], 16(%[src]) \n\t"
81 "lw %[rpom23], 20(%[src]) \n\t"
82 "lw %[rpom12], 24(%[src]) \n\t"
83 "lw %[rpom22], 28(%[src]) \n\t"
84 "mtc1 %[rpom13], %[temp9] \n\t"
85 "mtc1 %[rpom23], %[temp11] \n\t"
86 "mtc1 %[rpom12], %[temp13] \n\t"
87 "mtc1 %[rpom22], %[temp15] \n\t"
88
89 PTR_ADDIU "%[src], 32 \n\t"
90 "cvt.s.w %[temp1], %[temp1] \n\t"
91 "cvt.s.w %[temp3], %[temp3] \n\t"
92 "cvt.s.w %[temp5], %[temp5] \n\t"
93 "cvt.s.w %[temp7], %[temp7] \n\t"
94
95 "cvt.s.w %[temp9], %[temp9] \n\t"
96 "cvt.s.w %[temp11], %[temp11] \n\t"
97 "cvt.s.w %[temp13], %[temp13] \n\t"
98 "cvt.s.w %[temp15], %[temp15] \n\t"
99
100 "mul.s %[temp1], %[temp1], %[mul] \n\t"
101 "mul.s %[temp3], %[temp3], %[mul] \n\t"
102 "mul.s %[temp5], %[temp5], %[mul] \n\t"
103 "mul.s %[temp7], %[temp7], %[mul] \n\t"
104
105 "mul.s %[temp9], %[temp9], %[mul] \n\t"
106 "mul.s %[temp11], %[temp11], %[mul] \n\t"
107 "mul.s %[temp13], %[temp13], %[mul] \n\t"
108 "mul.s %[temp15], %[temp15], %[mul] \n\t"
109
110 "swc1 %[temp1], 0(%[dst]) \n\t" /*dst[i] = src[i] * mul; */
111 "swc1 %[temp3], 4(%[dst]) \n\t" /*dst[i+1] = src[i+1] * mul;*/
112 "swc1 %[temp5], 8(%[dst]) \n\t" /*dst[i+2] = src[i+2] * mul;*/
113 "swc1 %[temp7], 12(%[dst]) \n\t" /*dst[i+3] = src[i+3] * mul;*/
114
115 "swc1 %[temp9], 16(%[dst]) \n\t" /*dst[i+4] = src[i+4] * mul;*/
116 "swc1 %[temp11], 20(%[dst]) \n\t" /*dst[i+5] = src[i+5] * mul;*/
117 "swc1 %[temp13], 24(%[dst]) \n\t" /*dst[i+6] = src[i+6] * mul;*/
118 "swc1 %[temp15], 28(%[dst]) \n\t" /*dst[i+7] = src[i+7] * mul;*/
119 PTR_ADDIU "%[dst], 32 \n\t"
120 "bne %[src], %[src_end], i32tf_lp%= \n\t"
121 : [temp1]"=&f"(temp1), [temp11]"=&f"(temp11),
122 [temp13]"=&f"(temp13), [temp15]"=&f"(temp15),
123 [temp3]"=&f"(temp3), [temp5]"=&f"(temp5),
124 [temp7]"=&f"(temp7), [temp9]"=&f"(temp9),
125 [rpom1]"=&r"(rpom1), [rpom2]"=&r"(rpom2),
126 [rpom11]"=&r"(rpom11), [rpom21]"=&r"(rpom21),
127 [rpom12]"=&r"(rpom12), [rpom22]"=&r"(rpom22),
128 [rpom13]"=&r"(rpom13), [rpom23]"=&r"(rpom23),
129 [dst]"+r"(dst), [src]"+r"(src)
130 : [mul]"f"(mul), [src_end]"r"(src_end)
131 : "memory"
132 );
133 }
134 #endif /* HAVE_INLINE_ASM */
135
ff_fmt_convert_init_mips(FmtConvertContext * c)136 av_cold void ff_fmt_convert_init_mips(FmtConvertContext *c)
137 {
138 #if HAVE_INLINE_ASM
139 c->int32_to_float_fmul_scalar = int32_to_float_fmul_scalar_mips;
140 #endif
141 }
142