diff options
author | Mason Carter <void.main.argc.argv@gmail.com> | 2013-12-14 17:32:56 -0800 |
---|---|---|
committer | Martin Storsjö <martin@martin.st> | 2013-12-20 14:53:39 +0200 |
commit | 832e19063209a5f355af733d1a45f5051f49ce33 (patch) | |
tree | 816403978498ea9fb3b7a3dbc3619784b9ae1485 /libavcodec/arm/vc1dsp_init_neon.c | |
parent | e1b9de4fe15cda08cb18a2821a5729cf95e33d2a (diff) | |
download | ffmpeg-832e19063209a5f355af733d1a45f5051f49ce33.tar.gz |
vc1: arm: Add NEON assembly
For:
ff_vc1_inv_trans_{8,4}x{8,4}_{dc_,}neon
ff_put_pixels8x8_neon
ff_put_vc1_mspel_mc{0,1,2,3}{0,1,2,3}_neon (except for 00)
Based on ARM assembly code in libavcodec/arm by Rob Clark and Mans
Rullgard.
Signed-off-by: Martin Storsjö <martin@martin.st>
Diffstat (limited to 'libavcodec/arm/vc1dsp_init_neon.c')
-rw-r--r-- | libavcodec/arm/vc1dsp_init_neon.c | 100 |
1 files changed, 100 insertions, 0 deletions
diff --git a/libavcodec/arm/vc1dsp_init_neon.c b/libavcodec/arm/vc1dsp_init_neon.c new file mode 100644 index 0000000000..5cc65d7abc --- /dev/null +++ b/libavcodec/arm/vc1dsp_init_neon.c @@ -0,0 +1,100 @@ +/* + * This file is part of Libav. + * + * Libav is free software; you can redistribute it and/or + * modify it under the terms of the GNU Lesser General Public + * License as published by the Free Software Foundation; either + * version 2.1 of the License, or (at your option) any later version. + * + * Libav is distributed in the hope that it will be useful, + * but WITHOUT ANY WARRANTY; without even the implied warranty of + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU + * Lesser General Public License for more details. + * + * You should have received a copy of the GNU Lesser General Public + * License along with Libav; if not, write to the Free Software + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA + */ + +#include <stdint.h> + +#include "libavutil/attributes.h" +#include "libavcodec/vc1dsp.h" +#include "vc1dsp.h" + +void ff_vc1_inv_trans_8x8_neon(int16_t *block); +void ff_vc1_inv_trans_4x8_neon(uint8_t *dest, int linesize, int16_t *block); +void ff_vc1_inv_trans_8x4_neon(uint8_t *dest, int linesize, int16_t *block); +void ff_vc1_inv_trans_4x4_neon(uint8_t *dest, int linesize, int16_t *block); + +void ff_vc1_inv_trans_8x8_dc_neon(uint8_t *dest, int linesize, int16_t *block); +void ff_vc1_inv_trans_4x8_dc_neon(uint8_t *dest, int linesize, int16_t *block); +void ff_vc1_inv_trans_8x4_dc_neon(uint8_t *dest, int linesize, int16_t *block); +void ff_vc1_inv_trans_4x4_dc_neon(uint8_t *dest, int linesize, int16_t *block); + +void ff_put_pixels8x8_neon(uint8_t *block, const uint8_t *pixels, + ptrdiff_t line_size, int rnd); + +void ff_put_vc1_mspel_mc10_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc20_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc30_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); + +void ff_put_vc1_mspel_mc01_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc02_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc03_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); + +void ff_put_vc1_mspel_mc11_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc12_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc13_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); + +void ff_put_vc1_mspel_mc21_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc22_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc23_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); + +void ff_put_vc1_mspel_mc31_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc32_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); +void ff_put_vc1_mspel_mc33_neon(uint8_t *dst, const uint8_t *src, + ptrdiff_t stride, int rnd); + +av_cold void ff_vc1dsp_init_neon(VC1DSPContext *dsp) +{ + dsp->vc1_inv_trans_8x8 = ff_vc1_inv_trans_8x8_neon; + dsp->vc1_inv_trans_4x8 = ff_vc1_inv_trans_4x8_neon; + dsp->vc1_inv_trans_8x4 = ff_vc1_inv_trans_8x4_neon; + dsp->vc1_inv_trans_4x4 = ff_vc1_inv_trans_4x4_neon; + dsp->vc1_inv_trans_8x8_dc = ff_vc1_inv_trans_8x8_dc_neon; + dsp->vc1_inv_trans_4x8_dc = ff_vc1_inv_trans_4x8_dc_neon; + dsp->vc1_inv_trans_8x4_dc = ff_vc1_inv_trans_8x4_dc_neon; + dsp->vc1_inv_trans_4x4_dc = ff_vc1_inv_trans_4x4_dc_neon; + + dsp->put_vc1_mspel_pixels_tab[ 0] = ff_put_pixels8x8_neon; + dsp->put_vc1_mspel_pixels_tab[ 1] = ff_put_vc1_mspel_mc10_neon; + dsp->put_vc1_mspel_pixels_tab[ 2] = ff_put_vc1_mspel_mc20_neon; + dsp->put_vc1_mspel_pixels_tab[ 3] = ff_put_vc1_mspel_mc30_neon; + dsp->put_vc1_mspel_pixels_tab[ 4] = ff_put_vc1_mspel_mc01_neon; + dsp->put_vc1_mspel_pixels_tab[ 5] = ff_put_vc1_mspel_mc11_neon; + dsp->put_vc1_mspel_pixels_tab[ 6] = ff_put_vc1_mspel_mc21_neon; + dsp->put_vc1_mspel_pixels_tab[ 7] = ff_put_vc1_mspel_mc31_neon; + dsp->put_vc1_mspel_pixels_tab[ 8] = ff_put_vc1_mspel_mc02_neon; + dsp->put_vc1_mspel_pixels_tab[ 9] = ff_put_vc1_mspel_mc12_neon; + dsp->put_vc1_mspel_pixels_tab[10] = ff_put_vc1_mspel_mc22_neon; + dsp->put_vc1_mspel_pixels_tab[11] = ff_put_vc1_mspel_mc32_neon; + dsp->put_vc1_mspel_pixels_tab[12] = ff_put_vc1_mspel_mc03_neon; + dsp->put_vc1_mspel_pixels_tab[13] = ff_put_vc1_mspel_mc13_neon; + dsp->put_vc1_mspel_pixels_tab[14] = ff_put_vc1_mspel_mc23_neon; + dsp->put_vc1_mspel_pixels_tab[15] = ff_put_vc1_mspel_mc33_neon; +} |