Details | Last modification | View Log | RSS feed
Rev | Author | Line No. | Line |
---|---|---|---|
6148 | serge | 1 | /* |
2 | * Copyright (c) 2012 Michael Niedermayer |
||
3 | * |
||
4 | * This file is part of FFmpeg. |
||
5 | * |
||
6 | * FFmpeg is free software; you can redistribute it and/or |
||
7 | * modify it under the terms of the GNU Lesser General Public |
||
8 | * License as published by the Free Software Foundation; either |
||
9 | * version 2.1 of the License, or (at your option) any later version. |
||
10 | * |
||
11 | * FFmpeg is distributed in the hope that it will be useful, |
||
12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
||
13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
||
14 | * Lesser General Public License for more details. |
||
15 | * |
||
16 | * You should have received a copy of the GNU Lesser General Public |
||
17 | * License along with FFmpeg; if not, write to the Free Software |
||
18 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
||
19 | */ |
||
20 | |||
21 | #include "libavutil/x86/asm.h" |
||
22 | #include "libavutil/cpu.h" |
||
23 | #include "libswresample/swresample_internal.h" |
||
24 | |||
25 | int swri_resample_int16_mmx2 (struct ResampleContext *c, int16_t *dst, const int16_t *src, int *consumed, int src_size, int dst_size, int update_ctx); |
||
26 | int swri_resample_int16_ssse3(struct ResampleContext *c, int16_t *dst, const int16_t *src, int *consumed, int src_size, int dst_size, int update_ctx); |
||
27 | |||
28 | DECLARE_ALIGNED(16, const uint64_t, ff_resample_int16_rounder)[2] = { 0x0000000000004000ULL, 0x0000000000000000ULL}; |
||
29 | |||
30 | #define COMMON_CORE_INT16_MMX2 \ |
||
31 | x86_reg len= -2*c->filter_length;\ |
||
32 | __asm__ volatile(\ |
||
33 | "movq "MANGLE(ff_resample_int16_rounder)", %%mm0 \n\t"\ |
||
34 | "1: \n\t"\ |
||
35 | "movq (%1, %0), %%mm1 \n\t"\ |
||
36 | "pmaddwd (%2, %0), %%mm1 \n\t"\ |
||
37 | "paddd %%mm1, %%mm0 \n\t"\ |
||
38 | "add $8, %0 \n\t"\ |
||
39 | " js 1b \n\t"\ |
||
40 | "pshufw $0x0E, %%mm0, %%mm1 \n\t"\ |
||
41 | "paddd %%mm1, %%mm0 \n\t"\ |
||
42 | "psrad $15, %%mm0 \n\t"\ |
||
43 | "packssdw %%mm0, %%mm0 \n\t"\ |
||
44 | "movd %%mm0, (%3) \n\t"\ |
||
45 | : "+r" (len)\ |
||
46 | : "r" (((uint8_t*)(src+sample_index))-len),\ |
||
47 | "r" (((uint8_t*)filter)-len),\ |
||
48 | "r" (dst+dst_index)\ |
||
49 | ); |
||
50 | |||
51 | #define COMMON_CORE_INT16_SSSE3 \ |
||
52 | x86_reg len= -2*c->filter_length;\ |
||
53 | __asm__ volatile(\ |
||
54 | "movdqa "MANGLE(ff_resample_int16_rounder)", %%xmm0 \n\t"\ |
||
55 | "1: \n\t"\ |
||
56 | "movdqu (%1, %0), %%xmm1 \n\t"\ |
||
57 | "pmaddwd (%2, %0), %%xmm1 \n\t"\ |
||
58 | "paddd %%xmm1, %%xmm0 \n\t"\ |
||
59 | "add $16, %0 \n\t"\ |
||
60 | " js 1b \n\t"\ |
||
61 | "phaddd %%xmm0, %%xmm0 \n\t"\ |
||
62 | "phaddd %%xmm0, %%xmm0 \n\t"\ |
||
63 | "psrad $15, %%xmm0 \n\t"\ |
||
64 | "packssdw %%xmm0, %%xmm0 \n\t"\ |
||
65 | "movd %%xmm0, (%3) \n\t"\ |
||
66 | : "+r" (len)\ |
||
67 | : "r" (((uint8_t*)(src+sample_index))-len),\ |
||
68 | "r" (((uint8_t*)filter)-len),\ |
||
69 | "r" (dst+dst_index)\ |
||
70 | ); |