1 /*
2 * Copyright © 2018-2022, VideoLAN and dav1d authors
3 * Copyright © 2018-2022, Two Orioles, LLC
4 * All rights reserved.
5 *
6 * Redistribution and use in source and binary forms, with or without
7 * modification, are permitted provided that the following conditions are met:
8 *
9 * 1. Redistributions of source code must retain the above copyright notice, this
10 * list of conditions and the following disclaimer.
11 *
12 * 2. Redistributions in binary form must reproduce the above copyright notice,
13 * this list of conditions and the following disclaimer in the documentation
14 * and/or other materials provided with the distribution.
15 *
16 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
17 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
18 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
19 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR
20 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
21 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
22 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
23 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
24 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
25 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
26 */
27
28 #include "src/cpu.h"
29 #include "src/filmgrain.h"
30
31 #define decl_fg_fns(ext) \
32 decl_generate_grain_y_fn(BF(dav1d_generate_grain_y, ext)); \
33 decl_generate_grain_uv_fn(BF(dav1d_generate_grain_uv_420, ext)); \
34 decl_generate_grain_uv_fn(BF(dav1d_generate_grain_uv_422, ext)); \
35 decl_generate_grain_uv_fn(BF(dav1d_generate_grain_uv_444, ext)); \
36 decl_fgy_32x32xn_fn(BF(dav1d_fgy_32x32xn, ext)); \
37 decl_fguv_32x32xn_fn(BF(dav1d_fguv_32x32xn_i420, ext)); \
38 decl_fguv_32x32xn_fn(BF(dav1d_fguv_32x32xn_i422, ext)); \
39 decl_fguv_32x32xn_fn(BF(dav1d_fguv_32x32xn_i444, ext))
40
41 decl_fg_fns(ssse3);
42 decl_fg_fns(avx2);
43 decl_fg_fns(avx512icl);
44
bitfn(dav1d_film_grain_dsp_init_x86)45 COLD void bitfn(dav1d_film_grain_dsp_init_x86)(Dav1dFilmGrainDSPContext *const c) {
46 const unsigned flags = dav1d_get_cpu_flags();
47
48 if (!(flags & DAV1D_X86_CPU_FLAG_SSSE3)) return;
49
50 c->generate_grain_y = BF(dav1d_generate_grain_y, ssse3);
51 c->generate_grain_uv[DAV1D_PIXEL_LAYOUT_I420 - 1] = BF(dav1d_generate_grain_uv_420, ssse3);
52 c->fgy_32x32xn = BF(dav1d_fgy_32x32xn, ssse3);
53 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I420 - 1] = BF(dav1d_fguv_32x32xn_i420, ssse3);
54 c->generate_grain_uv[DAV1D_PIXEL_LAYOUT_I422 - 1] = BF(dav1d_generate_grain_uv_422, ssse3);
55 c->generate_grain_uv[DAV1D_PIXEL_LAYOUT_I444 - 1] = BF(dav1d_generate_grain_uv_444, ssse3);
56 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I422 - 1] = BF(dav1d_fguv_32x32xn_i422, ssse3);
57 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I444 - 1] = BF(dav1d_fguv_32x32xn_i444, ssse3);
58
59 #if ARCH_X86_64
60 if (!(flags & DAV1D_X86_CPU_FLAG_AVX2)) return;
61
62 c->generate_grain_y = BF(dav1d_generate_grain_y, avx2);
63 c->generate_grain_uv[DAV1D_PIXEL_LAYOUT_I420 - 1] = BF(dav1d_generate_grain_uv_420, avx2);
64 c->generate_grain_uv[DAV1D_PIXEL_LAYOUT_I422 - 1] = BF(dav1d_generate_grain_uv_422, avx2);
65 c->generate_grain_uv[DAV1D_PIXEL_LAYOUT_I444 - 1] = BF(dav1d_generate_grain_uv_444, avx2);
66
67 if (!(flags & DAV1D_X86_CPU_FLAG_SLOW_GATHER)) {
68 c->fgy_32x32xn = BF(dav1d_fgy_32x32xn, avx2);
69 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I420 - 1] = BF(dav1d_fguv_32x32xn_i420, avx2);
70 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I422 - 1] = BF(dav1d_fguv_32x32xn_i422, avx2);
71 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I444 - 1] = BF(dav1d_fguv_32x32xn_i444, avx2);
72 }
73
74 if (!(flags & DAV1D_X86_CPU_FLAG_AVX512ICL)) return;
75
76 c->fgy_32x32xn = BF(dav1d_fgy_32x32xn, avx512icl);
77 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I420 - 1] = BF(dav1d_fguv_32x32xn_i420, avx512icl);
78 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I422 - 1] = BF(dav1d_fguv_32x32xn_i422, avx512icl);
79 c->fguv_32x32xn[DAV1D_PIXEL_LAYOUT_I444 - 1] = BF(dav1d_fguv_32x32xn_i444, avx512icl);
80 #endif
81 }
82