libvpxenc.c 24.7 KB
Newer Older
James Zern's avatar
James Zern committed
1 2 3
/*
 * Copyright (c) 2010, Google, Inc.
 *
4
 * This file is part of Libav.
James Zern's avatar
James Zern committed
5
 *
6
 * Libav is free software; you can redistribute it and/or
James Zern's avatar
James Zern committed
7 8 9 10
 * 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.
 *
11
 * Libav is distributed in the hope that it will be useful,
James Zern's avatar
James Zern committed
12 13 14 15 16
 * 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
17
 * License along with Libav; if not, write to the Free Software
James Zern's avatar
James Zern committed
18 19 20 21 22 23 24 25 26 27 28 29 30 31
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

/**
 * @file
 * VP8 encoder support via libvpx
 */

#define VPX_DISABLE_CTRL_TYPECHECKS 1
#define VPX_CODEC_DISABLE_COMPAT    1
#include <vpx/vpx_encoder.h>
#include <vpx/vp8cx.h>

#include "avcodec.h"
32
#include "internal.h"
James Zern's avatar
James Zern committed
33
#include "libavutil/base64.h"
34
#include "libavutil/common.h"
35
#include "libavutil/mathematics.h"
36
#include "libavutil/opt.h"
James Zern's avatar
James Zern committed
37 38 39 40 41 42

/**
 * Portion of struct vpx_codec_cx_pkt from vpx_encoder.h.
 * One encoded frame returned from the library.
 */
struct FrameListData {
Benoit Fouet's avatar
Benoit Fouet committed
43 44 45
    void *buf;                       /**< compressed data buffer */
    size_t sz;                       /**< length of compressed data */
    int64_t pts;                     /**< time stamp to show frame
James Zern's avatar
James Zern committed
46
                                          (in timebase units) */
Benoit Fouet's avatar
Benoit Fouet committed
47
    unsigned long duration;          /**< duration to show frame
James Zern's avatar
James Zern committed
48
                                          (in timebase units) */
Benoit Fouet's avatar
Benoit Fouet committed
49
    uint32_t flags;                  /**< flags for this frame */
James Zern's avatar
James Zern committed
50 51 52 53
    struct FrameListData *next;
};

typedef struct VP8EncoderContext {
54
    AVClass *class;
James Zern's avatar
James Zern committed
55 56 57 58 59
    struct vpx_codec_ctx encoder;
    struct vpx_image rawimg;
    struct vpx_fixed_buf twopass_stats;
    unsigned long deadline; //i.e., RT/GOOD/BEST
    struct FrameListData *coded_frame_list;
60 61 62 63 64 65 66
    int cpu_used;
    int auto_alt_ref;
    int arnr_max_frames;
    int arnr_strength;
    int arnr_type;
    int lag_in_frames;
    int error_resilient;
67
    int crf;
James Zern's avatar
James Zern committed
68 69 70
} VP8Context;

/** String mappings for enum vp8e_enc_control_id */
71
static const char *const ctlidstr[] = {
James Zern's avatar
James Zern committed
72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
    [VP8E_UPD_ENTROPY]           = "VP8E_UPD_ENTROPY",
    [VP8E_UPD_REFERENCE]         = "VP8E_UPD_REFERENCE",
    [VP8E_USE_REFERENCE]         = "VP8E_USE_REFERENCE",
    [VP8E_SET_ROI_MAP]           = "VP8E_SET_ROI_MAP",
    [VP8E_SET_ACTIVEMAP]         = "VP8E_SET_ACTIVEMAP",
    [VP8E_SET_SCALEMODE]         = "VP8E_SET_SCALEMODE",
    [VP8E_SET_CPUUSED]           = "VP8E_SET_CPUUSED",
    [VP8E_SET_ENABLEAUTOALTREF]  = "VP8E_SET_ENABLEAUTOALTREF",
    [VP8E_SET_NOISE_SENSITIVITY] = "VP8E_SET_NOISE_SENSITIVITY",
    [VP8E_SET_SHARPNESS]         = "VP8E_SET_SHARPNESS",
    [VP8E_SET_STATIC_THRESHOLD]  = "VP8E_SET_STATIC_THRESHOLD",
    [VP8E_SET_TOKEN_PARTITIONS]  = "VP8E_SET_TOKEN_PARTITIONS",
    [VP8E_GET_LAST_QUANTIZER]    = "VP8E_GET_LAST_QUANTIZER",
    [VP8E_SET_ARNR_MAXFRAMES]    = "VP8E_SET_ARNR_MAXFRAMES",
    [VP8E_SET_ARNR_STRENGTH]     = "VP8E_SET_ARNR_STRENGTH",
    [VP8E_SET_ARNR_TYPE]         = "VP8E_SET_ARNR_TYPE",
88
    [VP8E_SET_CQ_LEVEL]          = "VP8E_SET_CQ_LEVEL",
James Zern's avatar
James Zern committed
89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216
};

static av_cold void log_encoder_error(AVCodecContext *avctx, const char *desc)
{
    VP8Context *ctx = avctx->priv_data;
    const char *error  = vpx_codec_error(&ctx->encoder);
    const char *detail = vpx_codec_error_detail(&ctx->encoder);

    av_log(avctx, AV_LOG_ERROR, "%s: %s\n", desc, error);
    if (detail)
        av_log(avctx, AV_LOG_ERROR, "  Additional information: %s\n", detail);
}

static av_cold void dump_enc_cfg(AVCodecContext *avctx,
                                 const struct vpx_codec_enc_cfg *cfg)
{
    int width = -30;
    int level = AV_LOG_DEBUG;

    av_log(avctx, level, "vpx_codec_enc_cfg\n");
    av_log(avctx, level, "generic settings\n"
           "  %*s%u\n  %*s%u\n  %*s%u\n  %*s%u\n  %*s%u\n"
           "  %*s{%u/%u}\n  %*s%u\n  %*s%d\n  %*s%u\n",
           width, "g_usage:",           cfg->g_usage,
           width, "g_threads:",         cfg->g_threads,
           width, "g_profile:",         cfg->g_profile,
           width, "g_w:",               cfg->g_w,
           width, "g_h:",               cfg->g_h,
           width, "g_timebase:",        cfg->g_timebase.num, cfg->g_timebase.den,
           width, "g_error_resilient:", cfg->g_error_resilient,
           width, "g_pass:",            cfg->g_pass,
           width, "g_lag_in_frames:",   cfg->g_lag_in_frames);
    av_log(avctx, level, "rate control settings\n"
           "  %*s%u\n  %*s%u\n  %*s%u\n  %*s%u\n"
           "  %*s%d\n  %*s%p(%zu)\n  %*s%u\n",
           width, "rc_dropframe_thresh:",   cfg->rc_dropframe_thresh,
           width, "rc_resize_allowed:",     cfg->rc_resize_allowed,
           width, "rc_resize_up_thresh:",   cfg->rc_resize_up_thresh,
           width, "rc_resize_down_thresh:", cfg->rc_resize_down_thresh,
           width, "rc_end_usage:",          cfg->rc_end_usage,
           width, "rc_twopass_stats_in:",   cfg->rc_twopass_stats_in.buf, cfg->rc_twopass_stats_in.sz,
           width, "rc_target_bitrate:",     cfg->rc_target_bitrate);
    av_log(avctx, level, "quantizer settings\n"
           "  %*s%u\n  %*s%u\n",
           width, "rc_min_quantizer:", cfg->rc_min_quantizer,
           width, "rc_max_quantizer:", cfg->rc_max_quantizer);
    av_log(avctx, level, "bitrate tolerance\n"
           "  %*s%u\n  %*s%u\n",
           width, "rc_undershoot_pct:", cfg->rc_undershoot_pct,
           width, "rc_overshoot_pct:",  cfg->rc_overshoot_pct);
    av_log(avctx, level, "decoder buffer model\n"
            "  %*s%u\n  %*s%u\n  %*s%u\n",
            width, "rc_buf_sz:",         cfg->rc_buf_sz,
            width, "rc_buf_initial_sz:", cfg->rc_buf_initial_sz,
            width, "rc_buf_optimal_sz:", cfg->rc_buf_optimal_sz);
    av_log(avctx, level, "2 pass rate control settings\n"
           "  %*s%u\n  %*s%u\n  %*s%u\n",
           width, "rc_2pass_vbr_bias_pct:",       cfg->rc_2pass_vbr_bias_pct,
           width, "rc_2pass_vbr_minsection_pct:", cfg->rc_2pass_vbr_minsection_pct,
           width, "rc_2pass_vbr_maxsection_pct:", cfg->rc_2pass_vbr_maxsection_pct);
    av_log(avctx, level, "keyframing settings\n"
           "  %*s%d\n  %*s%u\n  %*s%u\n",
           width, "kf_mode:",     cfg->kf_mode,
           width, "kf_min_dist:", cfg->kf_min_dist,
           width, "kf_max_dist:", cfg->kf_max_dist);
    av_log(avctx, level, "\n");
}

static void coded_frame_add(void *list, struct FrameListData *cx_frame)
{
    struct FrameListData **p = list;

    while (*p != NULL)
        p = &(*p)->next;
    *p = cx_frame;
    cx_frame->next = NULL;
}

static av_cold void free_coded_frame(struct FrameListData *cx_frame)
{
    av_freep(&cx_frame->buf);
    av_freep(&cx_frame);
}

static av_cold void free_frame_list(struct FrameListData *list)
{
    struct FrameListData *p = list;

    while (p) {
        list = list->next;
        free_coded_frame(p);
        p = list;
    }
}

static av_cold int codecctl_int(AVCodecContext *avctx,
                                enum vp8e_enc_control_id id, int val)
{
    VP8Context *ctx = avctx->priv_data;
    char buf[80];
    int width = -30;
    int res;

    snprintf(buf, sizeof(buf), "%s:", ctlidstr[id]);
    av_log(avctx, AV_LOG_DEBUG, "  %*s%d\n", width, buf, val);

    res = vpx_codec_control(&ctx->encoder, id, val);
    if (res != VPX_CODEC_OK) {
        snprintf(buf, sizeof(buf), "Failed to set %s codec control",
                 ctlidstr[id]);
        log_encoder_error(avctx, buf);
    }

    return res == VPX_CODEC_OK ? 0 : AVERROR(EINVAL);
}

static av_cold int vp8_free(AVCodecContext *avctx)
{
    VP8Context *ctx = avctx->priv_data;

    vpx_codec_destroy(&ctx->encoder);
    av_freep(&ctx->twopass_stats.buf);
    av_freep(&avctx->coded_frame);
    av_freep(&avctx->stats_out);
    free_frame_list(ctx->coded_frame_list);
    return 0;
}

Luca Barbato's avatar
Luca Barbato committed
217 218
static av_cold int vpx_init(AVCodecContext *avctx,
                            const struct vpx_codec_iface *iface)
James Zern's avatar
James Zern committed
219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239
{
    VP8Context *ctx = avctx->priv_data;
    struct vpx_codec_enc_cfg enccfg;
    int res;

    av_log(avctx, AV_LOG_INFO, "%s\n", vpx_codec_version_str());
    av_log(avctx, AV_LOG_VERBOSE, "%s\n", vpx_codec_build_config());

    if ((res = vpx_codec_enc_config_default(iface, &enccfg, 0)) != VPX_CODEC_OK) {
        av_log(avctx, AV_LOG_ERROR, "Failed to get config: %s\n",
               vpx_codec_err_to_string(res));
        return AVERROR(EINVAL);
    }
    dump_enc_cfg(avctx, &enccfg);

    enccfg.g_w            = avctx->width;
    enccfg.g_h            = avctx->height;
    enccfg.g_timebase.num = avctx->time_base.num;
    enccfg.g_timebase.den = avctx->time_base.den;
    enccfg.g_threads      = avctx->thread_count;

240 241 242
    if (ctx->lag_in_frames >= 0)
        enccfg.g_lag_in_frames = ctx->lag_in_frames;

James Zern's avatar
James Zern committed
243 244 245 246 247 248 249
    if (avctx->flags & CODEC_FLAG_PASS1)
        enccfg.g_pass = VPX_RC_FIRST_PASS;
    else if (avctx->flags & CODEC_FLAG_PASS2)
        enccfg.g_pass = VPX_RC_LAST_PASS;
    else
        enccfg.g_pass = VPX_RC_ONE_PASS;

250 251 252 253 254 255
    if (!avctx->bit_rate)
        avctx->bit_rate = enccfg.rc_target_bitrate * 1000;
    else
        enccfg.rc_target_bitrate = av_rescale_rnd(avctx->bit_rate, 1, 1000,
                                              AV_ROUND_NEAR_INF);

256 257 258 259
    if (ctx->crf)
        enccfg.rc_end_usage = VPX_CQ;
    else if (avctx->rc_min_rate == avctx->rc_max_rate &&
             avctx->rc_min_rate == avctx->bit_rate)
James Zern's avatar
James Zern committed
260
        enccfg.rc_end_usage = VPX_CBR;
261

262 263 264 265
    if (avctx->qmin > 0)
        enccfg.rc_min_quantizer = avctx->qmin;
    if (avctx->qmax > 0)
        enccfg.rc_max_quantizer = avctx->qmax;
266
    enccfg.rc_dropframe_thresh = avctx->frame_skip_threshold;
James Zern's avatar
James Zern committed
267

268 269 270
    //0-100 (0 => CBR, 100 => VBR)
    enccfg.rc_2pass_vbr_bias_pct           = round(avctx->qcompress * 100);
    enccfg.rc_2pass_vbr_minsection_pct     =
271
        avctx->rc_min_rate * 100LL / avctx->bit_rate;
272 273
    if (avctx->rc_max_rate)
        enccfg.rc_2pass_vbr_maxsection_pct =
274
            avctx->rc_max_rate * 100LL / avctx->bit_rate;
275

276 277
    if (avctx->rc_buffer_size)
        enccfg.rc_buf_sz         =
278
            avctx->rc_buffer_size * 1000LL / avctx->bit_rate;
279 280
    if (avctx->rc_initial_buffer_occupancy)
        enccfg.rc_buf_initial_sz =
281
            avctx->rc_initial_buffer_occupancy * 1000LL / avctx->bit_rate;
282 283
    enccfg.rc_buf_optimal_sz     = enccfg.rc_buf_sz * 5 / 6;

284
    //_enc_init() will balk if kf_min_dist differs from max w/VPX_KF_AUTO
285
    if (avctx->keyint_min >= 0 && avctx->keyint_min == avctx->gop_size)
James Zern's avatar
James Zern committed
286
        enccfg.kf_min_dist = avctx->keyint_min;
287 288
    if (avctx->gop_size >= 0)
        enccfg.kf_max_dist = avctx->gop_size;
James Zern's avatar
James Zern committed
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318

    if (enccfg.g_pass == VPX_RC_FIRST_PASS)
        enccfg.g_lag_in_frames = 0;
    else if (enccfg.g_pass == VPX_RC_LAST_PASS) {
        int decode_size;

        if (!avctx->stats_in) {
            av_log(avctx, AV_LOG_ERROR, "No stats file for second pass\n");
            return AVERROR_INVALIDDATA;
        }

        ctx->twopass_stats.sz  = strlen(avctx->stats_in) * 3 / 4;
        ctx->twopass_stats.buf = av_malloc(ctx->twopass_stats.sz);
        if (!ctx->twopass_stats.buf) {
            av_log(avctx, AV_LOG_ERROR,
                   "Stat buffer alloc (%zu bytes) failed\n",
                   ctx->twopass_stats.sz);
            return AVERROR(ENOMEM);
        }
        decode_size = av_base64_decode(ctx->twopass_stats.buf, avctx->stats_in,
                                       ctx->twopass_stats.sz);
        if (decode_size < 0) {
            av_log(avctx, AV_LOG_ERROR, "Stat buffer decode failed\n");
            return AVERROR_INVALIDDATA;
        }

        ctx->twopass_stats.sz      = decode_size;
        enccfg.rc_twopass_stats_in = ctx->twopass_stats;
    }

319 320 321 322 323
    /* 0-3: For non-zero values the encoder increasingly optimizes for reduced
       complexity playback on low powered devices at the expense of encode
       quality. */
   if (avctx->profile != FF_PROFILE_UNKNOWN)
       enccfg.g_profile = avctx->profile;
James Zern's avatar
James Zern committed
324

325 326
    enccfg.g_error_resilient = ctx->error_resilient;

James Zern's avatar
James Zern committed
327 328 329 330 331 332 333 334 335 336
    dump_enc_cfg(avctx, &enccfg);
    /* Construct Encoder Context */
    res = vpx_codec_enc_init(&ctx->encoder, iface, &enccfg, 0);
    if (res != VPX_CODEC_OK) {
        log_encoder_error(avctx, "Failed to initialize encoder");
        return AVERROR(EINVAL);
    }

    //codec control failures are currently treated only as warnings
    av_log(avctx, AV_LOG_DEBUG, "vpx_codec_control\n");
337 338 339 340 341 342 343 344 345 346
    if (ctx->cpu_used != INT_MIN)
        codecctl_int(avctx, VP8E_SET_CPUUSED,          ctx->cpu_used);
    if (ctx->auto_alt_ref >= 0)
        codecctl_int(avctx, VP8E_SET_ENABLEAUTOALTREF, ctx->auto_alt_ref);
    if (ctx->arnr_max_frames >= 0)
        codecctl_int(avctx, VP8E_SET_ARNR_MAXFRAMES,   ctx->arnr_max_frames);
    if (ctx->arnr_strength >= 0)
        codecctl_int(avctx, VP8E_SET_ARNR_STRENGTH,    ctx->arnr_strength);
    if (ctx->arnr_type >= 0)
        codecctl_int(avctx, VP8E_SET_ARNR_TYPE,        ctx->arnr_type);
James Zern's avatar
James Zern committed
347
    codecctl_int(avctx, VP8E_SET_NOISE_SENSITIVITY, avctx->noise_reduction);
348
    codecctl_int(avctx, VP8E_SET_TOKEN_PARTITIONS,  av_log2(avctx->slices));
349
    codecctl_int(avctx, VP8E_SET_STATIC_THRESHOLD,  avctx->mb_threshold);
350
    codecctl_int(avctx, VP8E_SET_CQ_LEVEL,          ctx->crf);
James Zern's avatar
James Zern committed
351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375

    //provide dummy value to initialize wrapper, values will be updated each _encode()
    vpx_img_wrap(&ctx->rawimg, VPX_IMG_FMT_I420, avctx->width, avctx->height, 1,
                 (unsigned char*)1);

    avctx->coded_frame = avcodec_alloc_frame();
    if (!avctx->coded_frame) {
        av_log(avctx, AV_LOG_ERROR, "Error allocating coded frame\n");
        vp8_free(avctx);
        return AVERROR(ENOMEM);
    }
    return 0;
}

static inline void cx_pktcpy(struct FrameListData *dst,
                             const struct vpx_codec_cx_pkt *src)
{
    dst->pts      = src->data.frame.pts;
    dst->duration = src->data.frame.duration;
    dst->flags    = src->data.frame.flags;
    dst->sz       = src->data.frame.sz;
    dst->buf      = src->data.frame.buf;
}

/**
376
 * Store coded frame information in format suitable for return from encode2().
James Zern's avatar
James Zern committed
377
 *
378 379 380
 * Write information from @a cx_frame to @a pkt
 * @return packet data size on success
 * @return a negative AVERROR on error
James Zern's avatar
James Zern committed
381 382
 */
static int storeframe(AVCodecContext *avctx, struct FrameListData *cx_frame,
383
                      AVPacket *pkt, AVFrame *coded_frame)
James Zern's avatar
James Zern committed
384
{
385 386 387 388
    int ret = ff_alloc_packet(pkt, cx_frame->sz);
    if (ret >= 0) {
        memcpy(pkt->data, cx_frame->buf, pkt->size);
        pkt->pts = pkt->dts    = cx_frame->pts;
James Zern's avatar
James Zern committed
389 390 391
        coded_frame->pts       = cx_frame->pts;
        coded_frame->key_frame = !!(cx_frame->flags & VPX_FRAME_IS_KEY);

392
        if (coded_frame->key_frame) {
393
            coded_frame->pict_type = AV_PICTURE_TYPE_I;
394 395
            pkt->flags            |= AV_PKT_FLAG_KEY;
        } else
396
            coded_frame->pict_type = AV_PICTURE_TYPE_P;
James Zern's avatar
James Zern committed
397 398
    } else {
        av_log(avctx, AV_LOG_ERROR,
399 400
               "Error getting output packet of size %zu.\n", cx_frame->sz);
        return ret;
James Zern's avatar
James Zern committed
401
    }
402
    return pkt->size;
James Zern's avatar
James Zern committed
403 404 405 406 407 408 409 410 411 412
}

/**
 * Queue multiple output frames from the encoder, returning the front-most.
 * In cases where vpx_codec_get_cx_data() returns more than 1 frame append
 * the frame queue. Return the head frame if available.
 * @return Stored frame size
 * @return AVERROR(EINVAL) on output size error
 * @return AVERROR(ENOMEM) on coded frame queue data allocation error
 */
413
static int queue_frames(AVCodecContext *avctx, AVPacket *pkt_out,
James Zern's avatar
James Zern committed
414 415 416 417 418 419 420 421 422 423
                        AVFrame *coded_frame)
{
    VP8Context *ctx = avctx->priv_data;
    const struct vpx_codec_cx_pkt *pkt;
    const void *iter = NULL;
    int size = 0;

    if (ctx->coded_frame_list) {
        struct FrameListData *cx_frame = ctx->coded_frame_list;
        /* return the leading frame if we've already begun queueing */
424
        size = storeframe(avctx, cx_frame, pkt_out, coded_frame);
James Zern's avatar
James Zern committed
425
        if (size < 0)
426
            return size;
James Zern's avatar
James Zern committed
427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
        ctx->coded_frame_list = cx_frame->next;
        free_coded_frame(cx_frame);
    }

    /* consume all available output from the encoder before returning. buffers
       are only good through the next vpx_codec call */
    while ((pkt = vpx_codec_get_cx_data(&ctx->encoder, &iter))) {
        switch (pkt->kind) {
        case VPX_CODEC_CX_FRAME_PKT:
            if (!size) {
                struct FrameListData cx_frame;

                /* avoid storing the frame when the list is empty and we haven't yet
                   provided a frame for output */
                assert(!ctx->coded_frame_list);
                cx_pktcpy(&cx_frame, pkt);
443
                size = storeframe(avctx, &cx_frame, pkt_out, coded_frame);
James Zern's avatar
James Zern committed
444
                if (size < 0)
445
                    return size;
James Zern's avatar
James Zern committed
446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490
            } else {
                struct FrameListData *cx_frame =
                    av_malloc(sizeof(struct FrameListData));

                if (!cx_frame) {
                    av_log(avctx, AV_LOG_ERROR,
                           "Frame queue element alloc failed\n");
                    return AVERROR(ENOMEM);
                }
                cx_pktcpy(cx_frame, pkt);
                cx_frame->buf = av_malloc(cx_frame->sz);

                if (!cx_frame->buf) {
                    av_log(avctx, AV_LOG_ERROR,
                           "Data buffer alloc (%zu bytes) failed\n",
                           cx_frame->sz);
                    return AVERROR(ENOMEM);
                }
                memcpy(cx_frame->buf, pkt->data.frame.buf, pkt->data.frame.sz);
                coded_frame_add(&ctx->coded_frame_list, cx_frame);
            }
            break;
        case VPX_CODEC_STATS_PKT: {
            struct vpx_fixed_buf *stats = &ctx->twopass_stats;
            stats->buf = av_realloc(stats->buf,
                                    stats->sz + pkt->data.twopass_stats.sz);
            if (!stats->buf) {
                av_log(avctx, AV_LOG_ERROR, "Stat buffer realloc failed\n");
                return AVERROR(ENOMEM);
            }
            memcpy((uint8_t*)stats->buf + stats->sz,
                   pkt->data.twopass_stats.buf, pkt->data.twopass_stats.sz);
            stats->sz += pkt->data.twopass_stats.sz;
            break;
        }
        case VPX_CODEC_PSNR_PKT: //FIXME add support for CODEC_FLAG_PSNR
        case VPX_CODEC_CUSTOM_PKT:
            //ignore unsupported/unrecognized packet types
            break;
        }
    }

    return size;
}

491 492
static int vp8_encode(AVCodecContext *avctx, AVPacket *pkt,
                      const AVFrame *frame, int *got_packet)
James Zern's avatar
James Zern committed
493 494 495 496 497
{
    VP8Context *ctx = avctx->priv_data;
    struct vpx_image *rawimg = NULL;
    int64_t timestamp = 0;
    int res, coded_size;
498
    vpx_enc_frame_flags_t flags = 0;
James Zern's avatar
James Zern committed
499 500 501 502 503 504 505 506 507 508

    if (frame) {
        rawimg                      = &ctx->rawimg;
        rawimg->planes[VPX_PLANE_Y] = frame->data[0];
        rawimg->planes[VPX_PLANE_U] = frame->data[1];
        rawimg->planes[VPX_PLANE_V] = frame->data[2];
        rawimg->stride[VPX_PLANE_Y] = frame->linesize[0];
        rawimg->stride[VPX_PLANE_U] = frame->linesize[1];
        rawimg->stride[VPX_PLANE_V] = frame->linesize[2];
        timestamp                   = frame->pts;
509 510
        if (frame->pict_type == AV_PICTURE_TYPE_I)
            flags |= VPX_EFLAG_FORCE_KF;
James Zern's avatar
James Zern committed
511 512 513
    }

    res = vpx_codec_encode(&ctx->encoder, rawimg, timestamp,
514
                           avctx->ticks_per_frame, flags, ctx->deadline);
James Zern's avatar
James Zern committed
515 516 517 518
    if (res != VPX_CODEC_OK) {
        log_encoder_error(avctx, "Error encoding frame");
        return AVERROR_INVALIDDATA;
    }
519
    coded_size = queue_frames(avctx, pkt, avctx->coded_frame);
James Zern's avatar
James Zern committed
520 521

    if (!frame && avctx->flags & CODEC_FLAG_PASS1) {
522
        unsigned int b64_size = AV_BASE64_SIZE(ctx->twopass_stats.sz);
James Zern's avatar
James Zern committed
523 524 525 526 527 528 529 530 531 532

        avctx->stats_out = av_malloc(b64_size);
        if (!avctx->stats_out) {
            av_log(avctx, AV_LOG_ERROR, "Stat buffer alloc (%d bytes) failed\n",
                   b64_size);
            return AVERROR(ENOMEM);
        }
        av_base64_encode(avctx->stats_out, b64_size, ctx->twopass_stats.buf,
                         ctx->twopass_stats.sz);
    }
533 534 535

    *got_packet = !!coded_size;
    return 0;
James Zern's avatar
James Zern committed
536 537
}

538 539 540
#define OFFSET(x) offsetof(VP8Context, x)
#define VE AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM
static const AVOption options[] = {
541
    { "cpu-used",        "Quality/Speed ratio modifier",           OFFSET(cpu_used),        AV_OPT_TYPE_INT, {.i64 = INT_MIN}, INT_MIN, INT_MAX, VE},
542
    { "auto-alt-ref",    "Enable use of alternate reference "
543
                         "frames (2-pass only)",                   OFFSET(auto_alt_ref),    AV_OPT_TYPE_INT, {.i64 = -1},      -1,      1,       VE},
544
    { "lag-in-frames",   "Number of frames to look ahead for "
545 546 547 548
                         "alternate reference frame selection",    OFFSET(lag_in_frames),   AV_OPT_TYPE_INT, {.i64 = -1},      -1,      INT_MAX, VE},
    { "arnr-maxframes",  "altref noise reduction max frame count", OFFSET(arnr_max_frames), AV_OPT_TYPE_INT, {.i64 = -1},      -1,      INT_MAX, VE},
    { "arnr-strength",   "altref noise reduction filter strength", OFFSET(arnr_strength),   AV_OPT_TYPE_INT, {.i64 = -1},      -1,      INT_MAX, VE},
    { "arnr-type",       "altref noise reduction filter type",     OFFSET(arnr_type),       AV_OPT_TYPE_INT, {.i64 = -1},      -1,      INT_MAX, VE, "arnr_type"},
549 550 551
    { "backward",        NULL, 0, AV_OPT_TYPE_CONST, {.i64 = 1}, 0, 0, VE, "arnr_type" },
    { "forward",         NULL, 0, AV_OPT_TYPE_CONST, {.i64 = 2}, 0, 0, VE, "arnr_type" },
    { "centered",        NULL, 0, AV_OPT_TYPE_CONST, {.i64 = 3}, 0, 0, VE, "arnr_type" },
552
    { "deadline",        "Time to spend encoding, in microseconds.", OFFSET(deadline),      AV_OPT_TYPE_INT, {.i64 = VPX_DL_GOOD_QUALITY}, INT_MIN, INT_MAX, VE, "quality"},
553 554 555
    { "best",            NULL, 0, AV_OPT_TYPE_CONST, {.i64 = VPX_DL_BEST_QUALITY}, 0, 0, VE, "quality"},
    { "good",            NULL, 0, AV_OPT_TYPE_CONST, {.i64 = VPX_DL_GOOD_QUALITY}, 0, 0, VE, "quality"},
    { "realtime",        NULL, 0, AV_OPT_TYPE_CONST, {.i64 = VPX_DL_REALTIME},     0, 0, VE, "quality"},
556
    { "error-resilient", "Error resilience configuration", OFFSET(error_resilient), AV_OPT_TYPE_FLAGS, {.i64 = 0}, INT_MIN, INT_MAX, VE, "er"},
557
#ifdef VPX_ERROR_RESILIENT_DEFAULT
558
    { "default",         "Improve resiliency against losses of whole frames", 0, AV_OPT_TYPE_CONST, {.i64 = VPX_ERROR_RESILIENT_DEFAULT}, 0, 0, VE, "er"},
559 560 561
    { "partitions",      "The frame partitions are independently decodable "
                         "by the bool decoder, meaning that partitions can be decoded even "
                         "though earlier partitions have been lost. Note that intra predicition"
562
                         " is still done over the partition boundary.",       0, AV_OPT_TYPE_CONST, {.i64 = VPX_ERROR_RESILIENT_PARTITIONS}, 0, 0, VE, "er"},
563
#endif
564
    { "crf",              "Select the quality for constant quality mode", offsetof(VP8Context, crf), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 63, VE },
565 566 567
    { NULL }
};

568 569 570 571 572 573 574 575
static const AVCodecDefault defaults[] = {
    { "qmin",             "-1" },
    { "qmax",             "-1" },
    { "g",                "-1" },
    { "keyint_min",       "-1" },
    { NULL },
};

576 577 578 579 580 581 582 583 584 585 586 587 588 589
#if CONFIG_LIBVPX_VP8_ENCODER
static av_cold int vp8_init(AVCodecContext *avctx)
{
    return vpx_init(avctx, &vpx_codec_vp8_cx_algo);
}

static const AVClass class_vp8 = {
    .class_name = "libvpx encoder",
    .item_name  = av_default_item_name,
    .option     = options,
    .version    = LIBAVUTIL_VERSION_INT,
};

AVCodec ff_libvpx_vp8_encoder = {
590 591
    .name           = "libvpx",
    .type           = AVMEDIA_TYPE_VIDEO,
592
    .id             = AV_CODEC_ID_VP8,
593 594
    .priv_data_size = sizeof(VP8Context),
    .init           = vp8_init,
595
    .encode2        = vp8_encode,
596
    .close          = vp8_free,
597
    .capabilities   = CODEC_CAP_DELAY | CODEC_CAP_AUTO_THREADS,
598
    .pix_fmts       = (const enum AVPixelFormat[]){ AV_PIX_FMT_YUV420P, AV_PIX_FMT_NONE },
599
    .long_name      = NULL_IF_CONFIG_SMALL("libvpx VP8"),
600
    .priv_class     = &class_vp8,
601
    .defaults       = defaults,
James Zern's avatar
James Zern committed
602
};
603
#endif /* CONFIG_LIBVPX_VP8_ENCODER */
Luca Barbato's avatar
Luca Barbato committed
604

605
#if CONFIG_LIBVPX_VP9_ENCODER
Luca Barbato's avatar
Luca Barbato committed
606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631
static av_cold int vp9_init(AVCodecContext *avctx)
{
    return vpx_init(avctx, &vpx_codec_vp9_cx_algo);
}

static const AVClass class_vp9 = {
    .class_name = "libvpx encoder",
    .item_name  = av_default_item_name,
    .option     = options,
    .version    = LIBAVUTIL_VERSION_INT,
};

AVCodec ff_libvpx_vp9_encoder = {
    .name           = "libvpx-vp9",
    .type           = AVMEDIA_TYPE_VIDEO,
    .id             = AV_CODEC_ID_VP9,
    .priv_data_size = sizeof(VP8Context),
    .init           = vp9_init,
    .encode2        = vp8_encode,
    .close          = vp8_free,
    .capabilities   = CODEC_CAP_DELAY | CODEC_CAP_AUTO_THREADS | CODEC_CAP_EXPERIMENTAL,
    .pix_fmts       = (const enum AVPixelFormat[]){ AV_PIX_FMT_YUV420P, AV_PIX_FMT_NONE },
    .long_name      = NULL_IF_CONFIG_SMALL("libvpx VP9"),
    .priv_class     = &class_vp9,
    .defaults       = defaults,
};
632
#endif /* CONFIG_LIBVPX_VP9_ENCODER */