diff options
Diffstat (limited to 'libavcodec/arm/vc1dsp_init_neon.c')
| -rw-r--r-- | libavcodec/arm/vc1dsp_init_neon.c | 113 |
1 files changed, 59 insertions, 54 deletions
diff --git a/libavcodec/arm/vc1dsp_init_neon.c b/libavcodec/arm/vc1dsp_init_neon.c index 9ded7a2..bb873e6 100644 --- a/libavcodec/arm/vc1dsp_init_neon.c +++ b/libavcodec/arm/vc1dsp_init_neon.c @@ -1,18 +1,18 @@ /* - * This file is part of Libav. + * This file is part of FFmpeg. * - * Libav is free software; you can redistribute it and/or + * FFmpeg 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, + * FFmpeg 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 + * License along with FFmpeg; if not, write to the Free Software * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA */ @@ -37,40 +37,38 @@ 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); +#define DECL_PUT(X, Y) \ +void ff_put_vc1_mspel_mc##X##Y##_neon(uint8_t *dst, const uint8_t *src, \ + ptrdiff_t stride, int rnd); \ +static void ff_put_vc1_mspel_mc##X##Y##_16_neon(uint8_t *dst, const uint8_t *src, \ + ptrdiff_t stride, int rnd) \ +{ \ + ff_put_vc1_mspel_mc##X##Y##_neon(dst+0, src+0, stride, rnd); \ + ff_put_vc1_mspel_mc##X##Y##_neon(dst+8, src+8, stride, rnd); \ + dst += 8*stride; src += 8*stride; \ + ff_put_vc1_mspel_mc##X##Y##_neon(dst+0, src+0, stride, rnd); \ + ff_put_vc1_mspel_mc##X##Y##_neon(dst+8, src+8, stride, rnd); \ +} + +DECL_PUT(1, 0) +DECL_PUT(2, 0) +DECL_PUT(3, 0) + +DECL_PUT(0, 1) +DECL_PUT(0, 2) +DECL_PUT(0, 3) + +DECL_PUT(1, 1) +DECL_PUT(1, 2) +DECL_PUT(1, 3) + +DECL_PUT(2, 1) +DECL_PUT(2, 2) +DECL_PUT(2, 3) + +DECL_PUT(3, 1) +DECL_PUT(3, 2) +DECL_PUT(3, 3) void ff_put_vc1_chroma_mc8_neon(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y); @@ -81,6 +79,10 @@ void ff_put_vc1_chroma_mc4_neon(uint8_t *dst, uint8_t *src, int stride, int h, void ff_avg_vc1_chroma_mc4_neon(uint8_t *dst, uint8_t *src, int stride, int h, int x, int y); +#define FN_ASSIGN(X, Y) \ + dsp->put_vc1_mspel_pixels_tab[0][X+4*Y] = ff_put_vc1_mspel_mc##X##Y##_16_neon; \ + dsp->put_vc1_mspel_pixels_tab[1][X+4*Y] = ff_put_vc1_mspel_mc##X##Y##_neon + av_cold void ff_vc1dsp_init_neon(VC1DSPContext *dsp) { dsp->vc1_inv_trans_8x8 = ff_vc1_inv_trans_8x8_neon; @@ -92,23 +94,26 @@ av_cold void ff_vc1dsp_init_neon(VC1DSPContext *dsp) 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][ 0] = ff_put_pixels8x8_neon; if (HAVE_AS_DN_DIRECTIVE) { - 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; + FN_ASSIGN(1, 0); + FN_ASSIGN(2, 0); + FN_ASSIGN(3, 0); + + FN_ASSIGN(0, 1); + FN_ASSIGN(1, 1); + FN_ASSIGN(2, 1); + FN_ASSIGN(3, 1); + + FN_ASSIGN(0, 2); + FN_ASSIGN(1, 2); + FN_ASSIGN(2, 2); + FN_ASSIGN(3, 2); + + FN_ASSIGN(0, 3); + FN_ASSIGN(1, 3); + FN_ASSIGN(2, 3); + FN_ASSIGN(3, 3); } dsp->put_no_rnd_vc1_chroma_pixels_tab[0] = ff_put_vc1_chroma_mc8_neon; |
