repositories / deb_ffmpeg.git / blame
| Commit | Line | Data |
|---|---|---|
| 2ba45a60 DM |
1 | /* |
| 2 | * SIMD-optimized pixel operations | |
| 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/attributes.h" | |
| 22 | #include "libavcodec/pixblockdsp.h" | |
| 23 | #include "asm.h" | |
| 24 | ||
| 25 | static void get_pixels_mvi(int16_t *restrict block, | |
| 26 | const uint8_t *restrict pixels, ptrdiff_t line_size) | |
| 27 | { | |
| 28 | int h = 8; | |
| 29 | ||
| 30 | do { | |
| 31 | uint64_t p; | |
| 32 | ||
| 33 | p = ldq(pixels); | |
| 34 | stq(unpkbw(p), block); | |
| 35 | stq(unpkbw(p >> 32), block + 4); | |
| 36 | ||
| 37 | pixels += line_size; | |
| 38 | block += 8; | |
| 39 | } while (--h); | |
| 40 | } | |
| 41 | ||
| 42 | static void diff_pixels_mvi(int16_t *block, const uint8_t *s1, const uint8_t *s2, | |
| 43 | int stride) { | |
| 44 | int h = 8; | |
| 45 | uint64_t mask = 0x4040; | |
| 46 | ||
| 47 | mask |= mask << 16; | |
| 48 | mask |= mask << 32; | |
| 49 | do { | |
| 50 | uint64_t x, y, c, d, a; | |
| 51 | uint64_t signs; | |
| 52 | ||
| 53 | x = ldq(s1); | |
| 54 | y = ldq(s2); | |
| 55 | c = cmpbge(x, y); | |
| 56 | d = x - y; | |
| 57 | a = zap(mask, c); /* We use 0x4040404040404040 here... */ | |
| 58 | d += 4 * a; /* ...so we can use s4addq here. */ | |
| 59 | signs = zap(-1, c); | |
| 60 | ||
| 61 | stq(unpkbw(d) | (unpkbw(signs) << 8), block); | |
| 62 | stq(unpkbw(d >> 32) | (unpkbw(signs >> 32) << 8), block + 4); | |
| 63 | ||
| 64 | s1 += stride; | |
| 65 | s2 += stride; | |
| 66 | block += 8; | |
| 67 | } while (--h); | |
| 68 | } | |
| 69 | ||
| 70 | av_cold void ff_pixblockdsp_init_alpha(PixblockDSPContext *c, AVCodecContext *avctx, | |
| 71 | unsigned high_bit_depth) | |
| 72 | { | |
| 73 | if (amask(AMASK_MVI) == 0) { | |
| 74 | if (!high_bit_depth) | |
| 75 | c->get_pixels = get_pixels_mvi; | |
| 76 | c->diff_pixels = diff_pixels_mvi; | |
| 77 | } | |
| 78 | } |