qsvenc: add the Access Unit Delimiter NAL Unit support
[libav.git] / libavcodec / qsvenc.h
1 /*
2 * Intel MediaSDK QSV encoder utility functions
3 *
4 * copyright (c) 2013 Yukinori Yamazoe
5 *
6 * This file is part of Libav.
7 *
8 * Libav is free software; you can redistribute it and/or
9 * modify it under the terms of the GNU Lesser General Public
10 * License as published by the Free Software Foundation; either
11 * version 2.1 of the License, or (at your option) any later version.
12 *
13 * Libav is distributed in the hope that it will be useful,
14 * but WITHOUT ANY WARRANTY; without even the implied warranty of
15 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
16 * Lesser General Public License for more details.
17 *
18 * You should have received a copy of the GNU Lesser General Public
19 * License along with Libav; if not, write to the Free Software
20 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
21 */
22
23 #ifndef AVCODEC_QSVENC_H
24 #define AVCODEC_QSVENC_H
25
26 #include <stdint.h>
27 #include <sys/types.h>
28
29 #include <mfx/mfxvideo.h>
30
31 #include "libavutil/avutil.h"
32 #include "libavutil/fifo.h"
33
34 #include "avcodec.h"
35 #include "qsv_internal.h"
36
37 #define QSV_HAVE_CO2 QSV_VERSION_ATLEAST(1, 6)
38 #define QSV_HAVE_CO3 QSV_VERSION_ATLEAST(1, 11)
39
40 #define QSV_HAVE_TRELLIS QSV_VERSION_ATLEAST(1, 8)
41 #define QSV_HAVE_MAX_SLICE_SIZE QSV_VERSION_ATLEAST(1, 9)
42 #define QSV_HAVE_BREF_TYPE QSV_VERSION_ATLEAST(1, 8)
43
44 #define QSV_HAVE_LA QSV_VERSION_ATLEAST(1, 7)
45 #define QSV_HAVE_LA_DS QSV_VERSION_ATLEAST(1, 8)
46 #define QSV_HAVE_LA_HRD QSV_VERSION_ATLEAST(1, 11)
47
48 #if defined(_WIN32)
49 #define QSV_HAVE_AVBR QSV_VERSION_ATLEAST(1, 3)
50 #define QSV_HAVE_ICQ QSV_VERSION_ATLEAST(1, 8)
51 #define QSV_HAVE_VCM QSV_VERSION_ATLEAST(1, 8)
52 #define QSV_HAVE_QVBR QSV_VERSION_ATLEAST(1, 11)
53 #else
54 #define QSV_HAVE_AVBR 0
55 #define QSV_HAVE_ICQ 0
56 #define QSV_HAVE_VCM 0
57 #define QSV_HAVE_QVBR 0
58 #endif
59
60 #if !QSV_HAVE_LA_DS
61 #define MFX_LOOKAHEAD_DS_OFF 0
62 #define MFX_LOOKAHEAD_DS_2x 0
63 #define MFX_LOOKAHEAD_DS_4x 0
64 #endif
65
66 #define QSV_COMMON_OPTS \
67 { "async_depth", "Maximum processing parallelism", OFFSET(qsv.async_depth), AV_OPT_TYPE_INT, { .i64 = ASYNC_DEPTH_DEFAULT }, 0, INT_MAX, VE }, \
68 { "avbr_accuracy", "Accuracy of the AVBR ratecontrol", OFFSET(qsv.avbr_accuracy), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, VE }, \
69 { "avbr_convergence", "Convergence of the AVBR ratecontrol", OFFSET(qsv.avbr_convergence), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, INT_MAX, VE }, \
70 { "preset", NULL, OFFSET(qsv.preset), AV_OPT_TYPE_INT, { .i64 = MFX_TARGETUSAGE_BALANCED }, 0, 7, VE, "preset" }, \
71 { "fast", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_TARGETUSAGE_BEST_SPEED }, INT_MIN, INT_MAX, VE, "preset" }, \
72 { "medium", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_TARGETUSAGE_BALANCED }, INT_MIN, INT_MAX, VE, "preset" }, \
73 { "slow", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_TARGETUSAGE_BEST_QUALITY }, INT_MIN, INT_MAX, VE, "preset" }, \
74 { "la_depth", "Number of frames to analyze before encoding.", OFFSET(qsv.la_depth), AV_OPT_TYPE_INT, { .i64 = 9 }, 9, 100, VE, "la_depth" }, \
75 { "unset", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 9 }, INT_MIN, INT_MAX, VE, "la_depth" }, \
76 { "la_ds", "Downscaling factor for the frames saved for the lookahead analysis", OFFSET(qsv.la_ds), AV_OPT_TYPE_INT, \
77 { .i64 = MFX_LOOKAHEAD_DS_UNKNOWN }, MFX_LOOKAHEAD_DS_UNKNOWN, MFX_LOOKAHEAD_DS_4x, VE, "la_ds" }, \
78 { "auto", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_LOOKAHEAD_DS_UNKNOWN }, INT_MIN, INT_MAX, VE, "la_ds" }, \
79 { "off", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_LOOKAHEAD_DS_OFF }, INT_MIN, INT_MAX, VE, "la_ds" }, \
80 { "2x", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_LOOKAHEAD_DS_2x }, INT_MIN, INT_MAX, VE, "la_ds" }, \
81 { "4x", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = MFX_LOOKAHEAD_DS_4x }, INT_MIN, INT_MAX, VE, "la_ds" }, \
82 { "vcm", "Use the video conferencing mode ratecontrol", OFFSET(qsv.vcm), AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, VE }, \
83 { "rdo", "Enable rate distortion optimization", OFFSET(qsv.rdo), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE }, \
84 { "max_frame_size", "Maximum encoded frame size in bytes", OFFSET(qsv.max_frame_size), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, UINT16_MAX, VE }, \
85 { "max_slice_size", "Maximum encoded slice size in bytes", OFFSET(qsv.max_slice_size), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, UINT16_MAX, VE }, \
86 { "bitrate_limit", "Toggle bitrate limitations", OFFSET(qsv.bitrate_limit), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE }, \
87 { "mbbrc", "MB level bitrate control", OFFSET(qsv.mbbrc), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE }, \
88 { "extbrc", "Extended bitrate control", OFFSET(qsv.extbrc), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE }, \
89 { "adaptive_i", "Adaptive I-frame placement", OFFSET(qsv.adaptive_i), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE }, \
90 { "adaptive_b", "Adaptive B-frame placement", OFFSET(qsv.adaptive_b), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE }, \
91 { "b_strategy", "Strategy to choose between I/P/B-frames", OFFSET(qsv.b_strategy), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE },
92
93 typedef struct QSVEncContext {
94 AVCodecContext *avctx;
95
96 QSVFrame *work_frames;
97
98 mfxSession session;
99 mfxSession internal_session;
100
101 int packet_size;
102 int width_align;
103 int height_align;
104
105 mfxVideoParam param;
106 mfxFrameAllocRequest req;
107
108 mfxExtCodingOption extco;
109 #if QSV_HAVE_CO2
110 mfxExtCodingOption2 extco2;
111 #endif
112
113 mfxExtOpaqueSurfaceAlloc opaque_alloc;
114 mfxFrameSurface1 **opaque_surfaces;
115 AVBufferRef *opaque_alloc_buf;
116
117 mfxExtBuffer *extparam_internal[2 + QSV_HAVE_CO2];
118 int nb_extparam_internal;
119
120 mfxExtBuffer **extparam;
121
122 AVFifoBuffer *async_fifo;
123
124 QSVFramesContext frames_ctx;
125
126 // options set by the caller
127 int async_depth;
128 int idr_interval;
129 int profile;
130 int preset;
131 int avbr_accuracy;
132 int avbr_convergence;
133 int la_depth;
134 int la_ds;
135 int vcm;
136 int rdo;
137 int max_frame_size;
138 int max_slice_size;
139
140 int aud;
141
142 int single_sei_nal_unit;
143 int max_dec_frame_buffering;
144 int trellis;
145
146 int bitrate_limit;
147 int mbbrc;
148 int extbrc;
149 int adaptive_i;
150 int adaptive_b;
151 int b_strategy;
152 int cavlc;
153
154 int int_ref_type;
155 int int_ref_cycle_size;
156 int int_ref_qp_delta;
157 int recovery_point_sei;
158
159 char *load_plugins;
160 } QSVEncContext;
161
162 int ff_qsv_enc_init(AVCodecContext *avctx, QSVEncContext *q);
163
164 int ff_qsv_encode(AVCodecContext *avctx, QSVEncContext *q,
165 AVPacket *pkt, const AVFrame *frame, int *got_packet);
166
167 int ff_qsv_enc_close(AVCodecContext *avctx, QSVEncContext *q);
168
169 #endif /* AVCODEC_QSVENC_H */