mpv/video/decode/vd_lavc.c

730 lines
22 KiB
C
Raw Normal View History

/*
* This file is part of MPlayer.
*
* MPlayer is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation; either version 2 of the License, or
* (at your option) any later version.
*
* MPlayer 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 General Public License for more details.
*
* You should have received a copy of the GNU General Public License along
* with MPlayer; if not, write to the Free Software Foundation, Inc.,
* 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
*/
#include <stdio.h>
#include <stdlib.h>
#include <assert.h>
#include <time.h>
#include <stdbool.h>
#include <sys/types.h>
#include <libavutil/common.h>
#include <libavutil/opt.h>
#include <libavutil/intreadwrite.h>
#include <libavutil/pixdesc.h>
#include "talloc.h"
#include "config.h"
#include "common/msg.h"
#include "options/options.h"
#include "misc/bstr.h"
#include "common/av_common.h"
#include "common/codecs.h"
#include "video/fmt-conversion.h"
#include "vd.h"
#include "video/img_format.h"
#include "video/filter/vf.h"
#include "video/decode/dec_video.h"
#include "demux/stheader.h"
#include "demux/packet.h"
#include "video/csputils.h"
#include "lavc.h"
#if AVPALETTE_SIZE != MP_PALETTE_SIZE
#error palette too large, adapt video/mp_image.h:MP_PALETTE_SIZE
#endif
#include "options/m_option.h"
static void init_avctx(struct dec_video *vd, const char *decoder,
struct vd_lavc_hwdec *hwdec);
static void uninit_avctx(struct dec_video *vd);
static int get_buffer2_hwdec(AVCodecContext *avctx, AVFrame *pic, int flags);
static enum AVPixelFormat get_format_hwdec(struct AVCodecContext *avctx,
const enum AVPixelFormat *pix_fmt);
#define OPT_BASE_STRUCT struct vd_lavc_params
struct vd_lavc_params {
int fast;
int show_all;
int skip_loop_filter;
int skip_idct;
int skip_frame;
int framedrop;
int threads;
int bitexact;
int check_hw_profile;
char **avopts;
};
static const struct m_opt_choice_alternatives discard_names[] = {
{"none", AVDISCARD_NONE},
{"default", AVDISCARD_DEFAULT},
{"nonref", AVDISCARD_NONREF},
{"bidir", AVDISCARD_BIDIR},
{"nonkey", AVDISCARD_NONKEY},
{"all", AVDISCARD_ALL},
{0}
};
#define OPT_DISCARD(name, field, flags) \
OPT_GENERAL(int, name, field, flags, .type = CONF_TYPE_CHOICE, \
.priv = (void *)discard_names)
const struct m_sub_options vd_lavc_conf = {
.opts = (const m_option_t[]){
OPT_FLAG("fast", fast, 0),
OPT_FLAG("show-all", show_all, 0),
OPT_DISCARD("skiploopfilter", skip_loop_filter, 0),
OPT_DISCARD("skipidct", skip_idct, 0),
OPT_DISCARD("skipframe", skip_frame, 0),
OPT_DISCARD("framedrop", framedrop, 0),
OPT_INTRANGE("threads", threads, 0, 0, 16),
OPT_FLAG("bitexact", bitexact, 0),
OPT_FLAG("check-hw-profile", check_hw_profile, 0),
OPT_KEYVALUELIST("o", avopts, 0),
{0}
},
.size = sizeof(struct vd_lavc_params),
.defaults = &(const struct vd_lavc_params){
.show_all = 0,
.check_hw_profile = 1,
.skip_loop_filter = AVDISCARD_DEFAULT,
.skip_idct = AVDISCARD_DEFAULT,
.skip_frame = AVDISCARD_DEFAULT,
.framedrop = AVDISCARD_NONREF,
},
};
const struct vd_lavc_hwdec mp_vd_lavc_vdpau;
video: add vda decode support (with hwaccel) and direct rendering Decoding H264 using Video Decode Acceleration used the custom 'vda_h264_dec' decoder in FFmpeg. The Good: This new implementation has some advantages over the previous one: - It works with Libav: vda_h264_dec never got into Libav since they prefer client applications to use the hwaccel API. - It is way more efficient: in my tests this implementation yields a reduction of CPU usage of roughly ~50% compared to using `vda_h264_dec` and ~65-75% compared to h264 software decoding. This is mainly because `vo_corevideo` was adapted to perform direct rendering of the `CVPixelBufferRefs` created by the Video Decode Acceleration API Framework. The Bad: - `vo_corevideo` is required to use VDA decoding acceleration. - only works with versions of ffmpeg/libav new enough (needs reference refcounting). That is FFmpeg 2.0+ and Libav's git master currently. The Ugly: VDA was hardcoded to use UYVY (2vuy) for the uploaded video texture. One one end this makes the code simple since Apple's OpenGL implementation actually supports this out of the box. It would be nice to support other output image formats and choose the best format depending on the input, or at least making it configurable. My tests indicate that CPU usage actually increases with a 420p IMGFMT output which is not what I would have expected. NOTE: There is a small memory leak with old versions of FFmpeg and with Libav since the CVPixelBufferRef is not automatically released when the AVFrame is deallocated. This can cause leaks inside libavcodec for decoded frames that are discarded before mpv wraps them inside a refcounted mp_image (this only happens on seeks). For frames that enter mpv's refcounting facilities, this is not a problem since we rewrap the CVPixelBufferRef in our mp_image that properly forwards CVPixelBufferRetain/CvPixelBufferRelease calls to the underying CVPixelBufferRef. So, for FFmpeg use something more recent than `b3d63995` for Libav the patch was posted to the dev ML in July and in review since, apparently, the proposed fix is rather hacky.
2013-08-14 13:47:18 +00:00
const struct vd_lavc_hwdec mp_vd_lavc_vda;
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
const struct vd_lavc_hwdec mp_vd_lavc_vaapi;
const struct vd_lavc_hwdec mp_vd_lavc_vaapi_copy;
const struct vd_lavc_hwdec mp_vd_lavc_dxva2_copy;
static const struct vd_lavc_hwdec *const hwdec_list[] = {
#if HAVE_VDPAU_HWACCEL
&mp_vd_lavc_vdpau,
#endif
#if HAVE_VDA_HWACCEL
&mp_vd_lavc_vda,
video: add vda decode support (with hwaccel) and direct rendering Decoding H264 using Video Decode Acceleration used the custom 'vda_h264_dec' decoder in FFmpeg. The Good: This new implementation has some advantages over the previous one: - It works with Libav: vda_h264_dec never got into Libav since they prefer client applications to use the hwaccel API. - It is way more efficient: in my tests this implementation yields a reduction of CPU usage of roughly ~50% compared to using `vda_h264_dec` and ~65-75% compared to h264 software decoding. This is mainly because `vo_corevideo` was adapted to perform direct rendering of the `CVPixelBufferRefs` created by the Video Decode Acceleration API Framework. The Bad: - `vo_corevideo` is required to use VDA decoding acceleration. - only works with versions of ffmpeg/libav new enough (needs reference refcounting). That is FFmpeg 2.0+ and Libav's git master currently. The Ugly: VDA was hardcoded to use UYVY (2vuy) for the uploaded video texture. One one end this makes the code simple since Apple's OpenGL implementation actually supports this out of the box. It would be nice to support other output image formats and choose the best format depending on the input, or at least making it configurable. My tests indicate that CPU usage actually increases with a 420p IMGFMT output which is not what I would have expected. NOTE: There is a small memory leak with old versions of FFmpeg and with Libav since the CVPixelBufferRef is not automatically released when the AVFrame is deallocated. This can cause leaks inside libavcodec for decoded frames that are discarded before mpv wraps them inside a refcounted mp_image (this only happens on seeks). For frames that enter mpv's refcounting facilities, this is not a problem since we rewrap the CVPixelBufferRef in our mp_image that properly forwards CVPixelBufferRetain/CvPixelBufferRelease calls to the underying CVPixelBufferRef. So, for FFmpeg use something more recent than `b3d63995` for Libav the patch was posted to the dev ML in July and in review since, apparently, the proposed fix is rather hacky.
2013-08-14 13:47:18 +00:00
#endif
#if HAVE_VAAPI_HWACCEL
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
&mp_vd_lavc_vaapi,
&mp_vd_lavc_vaapi_copy,
#endif
#if HAVE_DXVA2_HWACCEL
&mp_vd_lavc_dxva2_copy,
video: add vaapi decode and output support This is based on the MPlayer VA API patches. To be exact it's based on a very stripped down version of commit f1ad459a263f8537f6c from git://gitorious.org/vaapi/mplayer.git. This doesn't contain useless things like benchmarking hacks and the demo code for GLX interop. Also, unlike in the original patch, decoding and video output are split into separate source files (the separation between decoding and display also makes pixel format hacks unnecessary). On the other hand, some features not present in the original patch were added, like screenshot support. VA API is rather bad for actual video output. Dealing with older libva versions or the completely broken vdpau backend doesn't help. OSD is low quality and should be rather slow. In some cases, only either OSD or subtitles can be shown at the same time (because OSD is drawn first, OSD is prefered). Also, libva can't decide whether it accepts straight or premultiplied alpha for OSD sub-pictures: the vdpau backend seems to assume premultiplied, while a native vaapi driver uses straight. So I picked straight alpha. It doesn't matter much, because the blending code for straight alpha I added to img_convert.c is probably buggy, and ASS subtitles might be blended incorrectly. Really good video output with VA API would probably use OpenGL and the GL interop features, but at this point you might just use vo_opengl. (Patches for making HW decoding with vo_opengl have a chance of being accepted.) Despite these issues, decoding seems to work ok. I still got tearing on the Intel system I tested (Intel(R) Core(TM) i3-2350M). It was also tested with the vdpau vaapi wrapper on a nvidia system; however this was rather broken. (Fortunately, there is no reason to use mpv's VAAPI support over native VDPAU.)
2013-08-09 12:01:30 +00:00
#endif
NULL
};
static struct vd_lavc_hwdec *find_hwcodec(enum hwdec_type api)
{
for (int n = 0; hwdec_list[n]; n++) {
if (hwdec_list[n]->type == api)
return (struct vd_lavc_hwdec *)hwdec_list[n];
}
return NULL;
}
static bool hwdec_codec_allowed(struct dec_video *vd, const char *codec)
{
bstr s = bstr0(vd->opts->hwdec_codecs);
while (s.len) {
bstr item;
bstr_split_tok(s, ",", &item, &s);
if (bstr_equals0(item, "all") || bstr_equals0(item, codec))
return true;
}
return false;
}
static void hwdec_lock(struct lavc_ctx *ctx)
{
if (ctx->hwdec && ctx->hwdec->lock)
ctx->hwdec->lock(ctx);
}
static void hwdec_unlock(struct lavc_ctx *ctx)
{
if (ctx->hwdec && ctx->hwdec->unlock)
ctx->hwdec->unlock(ctx);
}
// Find the correct profile entry for the current codec and profile.
// Assumes the table has higher profiles first (for each codec).
const struct hwdec_profile_entry *hwdec_find_profile(
struct lavc_ctx *ctx, const struct hwdec_profile_entry *table)
{
assert(AV_CODEC_ID_NONE == 0);
struct vd_lavc_params *lavc_param = ctx->opts->vd_lavc_params;
enum AVCodecID codec = ctx->avctx->codec_id;
int profile = ctx->avctx->profile;
// Assume nobody cares about these aspects of the profile
if (codec == AV_CODEC_ID_H264) {
if (profile == FF_PROFILE_H264_CONSTRAINED_BASELINE)
profile = FF_PROFILE_H264_MAIN;
}
for (int n = 0; table[n].av_codec; n++) {
if (table[n].av_codec == codec) {
if (table[n].ff_profile == FF_PROFILE_UNKNOWN ||
profile == FF_PROFILE_UNKNOWN ||
table[n].ff_profile == profile ||
!lavc_param->check_hw_profile)
return &table[n];
}
}
return NULL;
}
// Check codec support, without checking the profile.
bool hwdec_check_codec_support(const char *decoder,
const struct hwdec_profile_entry *table)
{
enum AVCodecID codec = mp_codec_to_av_codec_id(decoder);
for (int n = 0; table[n].av_codec; n++) {
if (table[n].av_codec == codec)
return true;
}
return false;
}
int hwdec_get_max_refs(struct lavc_ctx *ctx)
{
return ctx->avctx->codec_id == AV_CODEC_ID_H264 ? 16 : 2;
}
void hwdec_request_api(struct mp_hwdec_info *info, const char *api_name)
{
if (info && info->load_api)
info->load_api(info, api_name);
}
static int hwdec_probe(struct vd_lavc_hwdec *hwdec, struct mp_hwdec_info *info,
const char *decoder)
{
int r = 0;
if (hwdec->probe)
r = hwdec->probe(hwdec, info, decoder);
return r;
}
static struct vd_lavc_hwdec *probe_hwdec(struct dec_video *vd, bool autoprobe,
enum hwdec_type api,
const char *decoder)
{
struct vd_lavc_hwdec *hwdec = find_hwcodec(api);
if (!hwdec) {
MP_VERBOSE(vd, "Requested hardware decoder not compiled.\n");
return NULL;
}
int r = hwdec_probe(hwdec, vd->hwdec_info, decoder);
if (r == HWDEC_ERR_EMULATED) {
if (autoprobe)
return NULL;
// User requested this explicitly.
MP_WARN(vd, "Using emulated hardware decoding API.\n");
r = 0;
}
if (r >= 0) {
return hwdec;
} else if (r == HWDEC_ERR_NO_CODEC) {
MP_VERBOSE(vd, "Hardware decoder '%s' not found in "
"libavcodec.\n", decoder);
} else if (r == HWDEC_ERR_NO_CTX && !autoprobe) {
MP_WARN(vd, "VO does not support requested hardware decoder.\n");
}
return NULL;
}
static void uninit(struct dec_video *vd)
{
uninit_avctx(vd);
talloc_free(vd->priv);
}
static int init(struct dec_video *vd, const char *decoder)
2011-10-22 00:51:37 +00:00
{
vd_ffmpeg_ctx *ctx;
ctx = vd->priv = talloc_zero(NULL, vd_ffmpeg_ctx);
ctx->log = vd->log;
ctx->opts = vd->opts;
ctx->selected_hwdec = vd->opts->hwdec_api;
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
if (bstr_endswith0(bstr0(decoder), "_vdpau")) {
MP_WARN(vd, "VDPAU decoder '%s' was requested. "
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
"This way of enabling hardware\ndecoding is not supported "
"anymore. Use --hwdec=vdpau instead.\nThe --hwdec-codec=... "
"option can be used to restrict which codecs are\nenabled, "
"otherwise all hardware decoding is tried for all codecs.\n",
decoder);
uninit(vd);
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
return 0;
}
struct vd_lavc_hwdec *hwdec = NULL;
if (hwdec_codec_allowed(vd, decoder)) {
if (vd->opts->hwdec_api == HWDEC_AUTO) {
for (int n = 0; hwdec_list[n]; n++) {
hwdec = probe_hwdec(vd, true, hwdec_list[n]->type, decoder);
if (hwdec)
break;
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
}
} else if (vd->opts->hwdec_api != HWDEC_NONE) {
hwdec = probe_hwdec(vd, false, vd->opts->hwdec_api, decoder);
}
} else {
MP_VERBOSE(vd, "Not trying to use hardware decoding: "
"codec %s is blacklisted by user.\n", decoder);
}
if (hwdec) {
ctx->software_fallback_decoder = talloc_strdup(ctx, decoder);
MP_INFO(vd, "Trying to use hardware decoding.\n");
} else if (vd->opts->hwdec_api != HWDEC_NONE) {
MP_INFO(vd, "Using software decoding.\n");
}
init_avctx(vd, decoder, hwdec);
if (!ctx->avctx) {
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
if (ctx->software_fallback_decoder) {
MP_ERR(vd, "Error initializing hardware decoding, "
"falling back to software decoding.\n");
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
decoder = ctx->software_fallback_decoder;
ctx->software_fallback_decoder = NULL;
init_avctx(vd, decoder, NULL);
}
if (!ctx->avctx) {
uninit(vd);
return 0;
}
}
if (ctx->avctx->bit_rate != 0)
vd->bitrate = ctx->avctx->bit_rate;
return 1;
}
static void init_avctx(struct dec_video *vd, const char *decoder,
struct vd_lavc_hwdec *hwdec)
{
vd_ffmpeg_ctx *ctx = vd->priv;
struct vd_lavc_params *lavc_param = vd->opts->vd_lavc_params;
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
bool mp_rawvideo = false;
struct sh_stream *sh = vd->header;
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
assert(!ctx->avctx);
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
if (strcmp(decoder, "mp-rawvideo") == 0) {
mp_rawvideo = true;
decoder = "rawvideo";
}
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
AVCodec *lavc_codec = avcodec_find_decoder_by_name(decoder);
if (!lavc_codec)
return;
video, audio: use lavc decoders without codecs.conf entries Add support for using libavcodec decoders that do not have entries in codecs.conf. This is currently only used with demux_lavf, and the codec selection is based on codec_id returned by libavformat. Also modify codec-related terminal output somewhat to make it use information from libavcodec and avoid excessively long default output. The new any-lavc-codec support is implemented with codecs.conf entries that invoke vd_ffmpeg/ad_ffmpeg without directly specifying any libavcodec codec name. In this mode, the decoders now instead select the libavcodec codec based on codec_id previously set by demux_lavf (if any). These new "generic" codecs.conf entries specify "status buggy", so that they're tried after any specific entries with higher-priority status. Add new directive "anyinput" to codecs.conf syntax. This means the entry will always match regardless of fourcc. This is used for the above new codecs.conf entries (so the driver always gets to decide whether to accept the input, and will fail init() if it can't find a suitable codec in libavcodec). Remove parsing support for the obsolete codecs.conf directive "cpuflags". This directive has not had any effect and has not been used in default codecs.conf since many years ago. Shorten codec-related terminal output. When using libavcodec decoders, show the libavcodec long_name field rather than codecs.conf "info" field as the name of the codec. Stop showing the codecs.conf entry name and "vfm/afm" name by default, as these are rarely needed; they're now in verbose output only. Show "VIDEO:" line at VO initialization rather than at demuxer open. This didn't really belong in demuxer code; the new location may show more accurate values (known after decoder has been opened) and works right if video track is changed after initial demuxer open. The vd.c changes (primarily done for terminal output changes) remove round-to-even behavior from code setting dimensions based on aspect ratio. I hope nothing depended on this; at least the even values were not consistently guaranteed anyway, as the rounding code did not run if the video file did not specify a nonzero aspect value.
2012-07-24 06:01:47 +00:00
ctx->hwdec_info = vd->hwdec_info;
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
ctx->pix_fmt = AV_PIX_FMT_NONE;
ctx->hwdec = hwdec;
ctx->hwdec_fmt = 0;
ctx->avctx = avcodec_alloc_context3(lavc_codec);
AVCodecContext *avctx = ctx->avctx;
2014-12-12 16:28:22 +00:00
if (!ctx->avctx)
return;
avctx->bit_rate = 0;
avctx->opaque = vd;
avctx->codec_type = AVMEDIA_TYPE_VIDEO;
avctx->codec_id = lavc_codec->id;
avctx->refcounted_frames = 1;
ctx->pic = av_frame_alloc();
2014-12-12 16:28:22 +00:00
if (!ctx->pic)
return;
if (ctx->hwdec) {
avctx->thread_count = 1;
avctx->get_format = get_format_hwdec;
avctx->get_buffer2 = get_buffer2_hwdec;
if (ctx->hwdec->init(ctx) < 0) {
uninit_avctx(vd);
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
return;
}
} else {
mp_set_avcodec_threads(avctx, lavc_param->threads);
}
avctx->flags |= lavc_param->bitexact ? CODEC_FLAG_BITEXACT : 0;
avctx->flags2 |= lavc_param->fast ? CODEC_FLAG2_FAST : 0;
if (lavc_param->show_all) {
#ifdef CODEC_FLAG2_SHOW_ALL
avctx->flags2 |= CODEC_FLAG2_SHOW_ALL; // ffmpeg only?
#endif
#ifdef CODEC_FLAG_OUTPUT_CORRUPT
avctx->flags |= CODEC_FLAG_OUTPUT_CORRUPT; // added with Libav 10
#endif
}
avctx->skip_loop_filter = lavc_param->skip_loop_filter;
avctx->skip_idct = lavc_param->skip_idct;
avctx->skip_frame = lavc_param->skip_frame;
mp_set_avopts(vd->log, avctx, lavc_param->avopts);
// Do this after the above avopt handling in case it changes values
ctx->skip_frame = avctx->skip_frame;
avctx->codec_tag = sh->format;
avctx->coded_width = sh->video->disp_w;
avctx->coded_height = sh->video->disp_h;
avctx->bits_per_coded_sample = sh->video->bits_per_coded_sample;
if (sh->video->coded_width && sh->video->coded_height) {
avctx->coded_width = sh->video->coded_width;
avctx->coded_height = sh->video->coded_height;
}
mp_lavc_set_extradata(avctx, sh->video->extradata, sh->video->extradata_len);
if (mp_rawvideo) {
avctx->pix_fmt = imgfmt2pixfmt(sh->format);
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
avctx->codec_tag = 0;
if (avctx->pix_fmt == AV_PIX_FMT_NONE && sh->format)
MP_ERR(vd, "Image format %s not supported by lavc.\n",
mp_imgfmt_to_name(sh->format));
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
}
if (sh->lav_headers)
mp_copy_lav_codec_headers(avctx, sh->lav_headers);
/* open it */
if (avcodec_open2(avctx, lavc_codec, NULL) < 0) {
MP_ERR(vd, "Could not open codec.\n");
uninit_avctx(vd);
return;
}
}
static void uninit_avctx(struct dec_video *vd)
2011-10-22 00:51:37 +00:00
{
vd_ffmpeg_ctx *ctx = vd->priv;
AVCodecContext *avctx = ctx->avctx;
if (avctx && avcodec_is_open(avctx))
avcodec_flush_buffers(avctx);
if (ctx->hwdec && ctx->hwdec->uninit)
ctx->hwdec->uninit(ctx);
if (avctx) {
if (avctx->codec && avcodec_close(avctx) < 0)
MP_ERR(vd, "Could not close codec.\n");
av_freep(&avctx->extradata);
av_freep(&avctx->slice_offset);
}
av_freep(&ctx->avctx);
av_frame_free(&ctx->pic);
}
static void update_image_params(struct dec_video *vd, AVFrame *frame,
struct mp_image_params *out_params)
{
vd_ffmpeg_ctx *ctx = vd->priv;
struct MPOpts *opts = ctx->opts;
int width = frame->width;
int height = frame->height;
float aspect = av_q2d(frame->sample_aspect_ratio) * width / height;
int pix_fmt = frame->format;
if (pix_fmt != ctx->pix_fmt) {
ctx->pix_fmt = pix_fmt;
ctx->best_csp = pixfmt2imgfmt(pix_fmt);
if (!ctx->best_csp)
MP_ERR(vd, "lavc pixel format %s not supported.\n",
av_get_pix_fmt_name(pix_fmt));
}
int d_w, d_h;
vf_set_dar(&d_w, &d_h, width, height, aspect);
*out_params = (struct mp_image_params) {
.imgfmt = ctx->best_csp,
.w = width,
.h = height,
.d_w = d_w,
.d_h = d_h,
.colorspace = avcol_spc_to_mp_csp(ctx->avctx->colorspace),
.colorlevels = avcol_range_to_mp_csp_levels(ctx->avctx->color_range),
.primaries = avcol_pri_to_mp_csp_prim(ctx->avctx->color_primaries),
.chroma_location =
avchroma_location_to_mp(ctx->avctx->chroma_sample_location),
.rotate = vd->header->video->rotate,
.stereo_in = vd->header->video->stereo_mode,
};
if (opts->video_rotate < 0) {
out_params->rotate = 0;
} else {
out_params->rotate = (out_params->rotate + opts->video_rotate) % 360;
}
out_params->stereo_out = opts->video_stereo_mode;
}
static enum AVPixelFormat get_format_hwdec(struct AVCodecContext *avctx,
const enum AVPixelFormat *fmt)
{
struct dec_video *vd = avctx->opaque;
vd_ffmpeg_ctx *ctx = vd->priv;
MP_VERBOSE(vd, "Pixel formats supported by decoder:");
for (int i = 0; fmt[i] != AV_PIX_FMT_NONE; i++)
MP_VERBOSE(vd, " %s", av_get_pix_fmt_name(fmt[i]));
MP_VERBOSE(vd, "\n");
assert(ctx->hwdec);
if (ctx->hwdec->image_format) {
for (int i = 0; fmt[i] != AV_PIX_FMT_NONE; i++) {
if (ctx->hwdec->image_format == pixfmt2imgfmt(fmt[i])) {
// There could be more reasons for a change, and it's possible
// that we miss some. (Might also depend on the hwaccel type.)
bool change =
ctx->hwdec_w != avctx->width ||
ctx->hwdec_h != avctx->height ||
ctx->hwdec_fmt != ctx->hwdec->image_format ||
ctx->hwdec_profile != avctx->profile;
ctx->hwdec_w = avctx->width;
ctx->hwdec_h = avctx->height;
ctx->hwdec_fmt = ctx->hwdec->image_format;
ctx->hwdec_profile = avctx->profile;
if (ctx->hwdec->init_decoder && change) {
if (ctx->hwdec->init_decoder(ctx, ctx->hwdec_fmt,
ctx->hwdec_w, ctx->hwdec_h) < 0)
{
ctx->hwdec_fmt = 0;
break;
}
}
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
return fmt[i];
}
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
}
}
return AV_PIX_FMT_NONE;
}
static struct mp_image *get_surface_hwdec(struct dec_video *vd, AVFrame *pic)
{
vd_ffmpeg_ctx *ctx = vd->priv;
/* Decoders using ffmpeg's hwaccel architecture (everything except vdpau)
* can fall back to software decoding automatically. However, we don't
* want that: multithreading was already disabled. ffmpeg's fallback
* isn't really useful, and causes more trouble than it helps.
*
* Instead of trying to "adjust" the thread_count fields in avctx, let
* decoding fail hard. Then decode_with_fallback() will do our own software
* fallback. Fully reinitializing the decoder is saner, and will probably
* save us from other weird corner cases, like having to "reroute" the
* get_buffer callback.
*/
int imgfmt = pixfmt2imgfmt(pic->format);
if (!IMGFMT_IS_HWACCEL(imgfmt) || !ctx->hwdec)
return NULL;
// Using frame->width/height is bad. For non-mod 16 video (which would
// require alignment of frame sizes) we want the decoded size, not the
// aligned size. At least vdpau needs this: the video mixer is created
// with decoded size, and the video surfaces must have matching size.
int w = ctx->avctx->width;
int h = ctx->avctx->height;
if (ctx->hwdec->init_decoder) {
if (imgfmt != ctx->hwdec_fmt && w != ctx->hwdec_w && h != ctx->hwdec_h)
return NULL;
}
struct mp_image *mpi = ctx->hwdec->allocate_image(ctx, imgfmt, w, h);
if (mpi) {
for (int i = 0; i < 4; i++)
pic->data[i] = mpi->planes[i];
}
return mpi;
}
static void free_mpi(void *opaque, uint8_t *data)
{
struct mp_image *mpi = opaque;
talloc_free(mpi);
}
static int get_buffer2_hwdec(AVCodecContext *avctx, AVFrame *pic, int flags)
{
struct dec_video *vd = avctx->opaque;
struct mp_image *mpi = get_surface_hwdec(vd, pic);
if (!mpi)
return -1;
pic->buf[0] = av_buffer_create(NULL, 0, free_mpi, mpi, 0);
return 0;
}
static int decode(struct dec_video *vd, struct demux_packet *packet,
int flags, struct mp_image **out_image)
{
2011-10-22 00:51:37 +00:00
int got_picture = 0;
int ret;
vd_ffmpeg_ctx *ctx = vd->priv;
AVCodecContext *avctx = ctx->avctx;
struct vd_lavc_params *lavc_param = ctx->opts->vd_lavc_params;
AVPacket pkt;
if (flags) {
// hr-seek framedrop vs. normal framedrop
avctx->skip_frame = flags == 2 ? AVDISCARD_NONREF : lavc_param->framedrop;
} else {
// normal playback
avctx->skip_frame = ctx->skip_frame;
}
mp_set_av_packet(&pkt, packet, NULL);
hwdec_lock(ctx);
ret = avcodec_decode_video2(avctx, ctx->pic, &got_picture, &pkt);
hwdec_unlock(ctx);
if (ret < 0) {
MP_WARN(vd, "Error while decoding frame!\n");
return -1;
}
2013-11-24 11:22:25 +00:00
// Skipped frame, or delayed output due to multithreaded decoding.
2011-10-22 00:51:37 +00:00
if (!got_picture)
2013-11-24 11:22:25 +00:00
return 0;
struct mp_image_params params;
update_image_params(vd, ctx->pic, &params);
vd->codec_pts = mp_pts_from_av(ctx->pic->pkt_pts, NULL);
vd->codec_dts = mp_pts_from_av(ctx->pic->pkt_dts, NULL);
struct mp_image *mpi = mp_image_from_av_frame(ctx->pic);
av_frame_unref(ctx->pic);
if (!mpi)
return 0; // mpi==NULL, or OOM
assert(mpi->planes[0] || mpi->planes[3]);
mp_image_set_params(mpi, &params);
if (ctx->hwdec && ctx->hwdec->process_image)
video: add vda decode support (with hwaccel) and direct rendering Decoding H264 using Video Decode Acceleration used the custom 'vda_h264_dec' decoder in FFmpeg. The Good: This new implementation has some advantages over the previous one: - It works with Libav: vda_h264_dec never got into Libav since they prefer client applications to use the hwaccel API. - It is way more efficient: in my tests this implementation yields a reduction of CPU usage of roughly ~50% compared to using `vda_h264_dec` and ~65-75% compared to h264 software decoding. This is mainly because `vo_corevideo` was adapted to perform direct rendering of the `CVPixelBufferRefs` created by the Video Decode Acceleration API Framework. The Bad: - `vo_corevideo` is required to use VDA decoding acceleration. - only works with versions of ffmpeg/libav new enough (needs reference refcounting). That is FFmpeg 2.0+ and Libav's git master currently. The Ugly: VDA was hardcoded to use UYVY (2vuy) for the uploaded video texture. One one end this makes the code simple since Apple's OpenGL implementation actually supports this out of the box. It would be nice to support other output image formats and choose the best format depending on the input, or at least making it configurable. My tests indicate that CPU usage actually increases with a 420p IMGFMT output which is not what I would have expected. NOTE: There is a small memory leak with old versions of FFmpeg and with Libav since the CVPixelBufferRef is not automatically released when the AVFrame is deallocated. This can cause leaks inside libavcodec for decoded frames that are discarded before mpv wraps them inside a refcounted mp_image (this only happens on seeks). For frames that enter mpv's refcounting facilities, this is not a problem since we rewrap the CVPixelBufferRef in our mp_image that properly forwards CVPixelBufferRetain/CvPixelBufferRelease calls to the underying CVPixelBufferRef. So, for FFmpeg use something more recent than `b3d63995` for Libav the patch was posted to the dev ML in July and in review since, apparently, the proposed fix is rather hacky.
2013-08-14 13:47:18 +00:00
mpi = ctx->hwdec->process_image(ctx, mpi);
vdpau: split off decoder parts, use "new" libavcodec vdpau hwaccel API Move the decoder parts from vo_vdpau.c to a new file vdpau_old.c. This file is named so because because it's written against the "old" libavcodec vdpau pseudo-decoder (e.g. "h264_vdpau"). Add support for the "new" libavcodec vdpau support. This was recently added and replaces the "old" vdpau parts. (In fact, Libav is about to deprecate and remove the "old" API without deprecation grace period, so we have to support it now. Moreover, there will probably be no Libav release which supports both, so the transition is even less smooth than we could hope, and we have to support both the old and new API.) Whether the old or new API is used is checked by a configure test: if the new API is found, it is used, otherwise the old API is assumed. Some details might be handled differently. Especially display preemption is a bit problematic with the "new" libavcodec vdpau support: it wants to keep a pointer to a specific vdpau API function (which can be driver specific, because preemption might switch drivers). Also, surface IDs are now directly stored in AVFrames (and mp_images), so they can't be forced to VDP_INVALID_HANDLE on preemption. (This changes even with older libavcodec versions, because mp_image always uses the newer representation to make vo_vdpau.c simpler.) Decoder initialization in the new code tries to deal with codec profiles, while the old code always uses the highest profile per codec. Surface allocation changes. Since the decoder won't call config() in vo_vdpau.c on video size change anymore, we allow allocating surfaces of arbitrary size instead of locking it to what the VO was configured. The non-hwdec code also has slightly different allocation behavior now. Enabling the old vdpau special decoders via e.g. --vd=lavc:h264_vdpau doesn't work anymore (a warning suggesting the --hwdec option is printed instead).
2013-07-27 23:49:45 +00:00
*out_image = mpi;
return 1;
}
static int force_fallback(struct dec_video *vd)
{
vd_ffmpeg_ctx *ctx = vd->priv;
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
if (ctx->software_fallback_decoder) {
uninit_avctx(vd);
MP_ERR(vd, "Error using hardware "
"decoding, falling back to software decoding.\n");
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
const char *decoder = ctx->software_fallback_decoder;
ctx->software_fallback_decoder = NULL;
init_avctx(vd, decoder, NULL);
return ctx->avctx ? CONTROL_OK : CONTROL_ERROR;
}
return CONTROL_FALSE;
}
static struct mp_image *decode_with_fallback(struct dec_video *vd,
struct demux_packet *packet, int flags)
{
vd_ffmpeg_ctx *ctx = vd->priv;
if (!ctx->avctx)
return NULL;
struct mp_image *mpi = NULL;
int res = decode(vd, packet, flags, &mpi);
if (res < 0) {
// Failed hardware decoding? Try again in software.
if (force_fallback(vd) == CONTROL_OK)
decode(vd, packet, flags, &mpi);
}
return mpi;
}
static int control(struct dec_video *vd, int cmd, void *arg)
{
vd_ffmpeg_ctx *ctx = vd->priv;
AVCodecContext *avctx = ctx->avctx;
switch (cmd) {
case VDCTRL_RESET:
avcodec_flush_buffers(avctx);
return CONTROL_TRUE;
case VDCTRL_QUERY_UNSEEN_FRAMES:;
int delay = avctx->has_b_frames;
assert(delay >= 0);
if (avctx->active_thread_type & FF_THREAD_FRAME)
delay += avctx->thread_count - 1;
*(int *)arg = delay;
return CONTROL_TRUE;
case VDCTRL_GET_HWDEC: {
int hwdec = ctx->selected_hwdec;
if (!ctx->software_fallback_decoder)
hwdec = 0;
*(int *)arg = hwdec;
return CONTROL_TRUE;
}
case VDCTRL_FORCE_HWDEC_FALLBACK:
return force_fallback(vd);
}
return CONTROL_UNKNOWN;
}
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
static void add_decoders(struct mp_decoder_list *list)
{
mp_add_lavc_decoders(list, AVMEDIA_TYPE_VIDEO);
mp_add_decoder(list, "lavc", "mp-rawvideo", "mp-rawvideo",
"raw video");
}
const struct vd_functions mpcodecs_vd_ffmpeg = {
core: redo how codecs are mapped, remove codecs.conf Use codec names instead of FourCCs to identify codecs. Rewrite how codecs are selected and initialized. Now each decoder exports a list of decoders (and the codec it supports) via add_decoders(). The order matters, and the first decoder for a given decoder is preferred over the other decoders. E.g. all ad_mpg123 decoders are preferred over ad_lavc, because it comes first in the mpcodecs_ad_drivers array. Likewise, decoders within ad_lavc that are enumerated first by libavcodec (using av_codec_next()) are preferred. (This is actually critical to select h264 software decoding by default instead of vdpau. libavcodec and ffmpeg/avconv use the same method to select decoders by default, so we hope this is sane.) The codec names follow libavcodec's codec names as defined by AVCodecDescriptor.name (see libavcodec/codec_desc.c). Some decoders have names different from the canonical codec name. The AVCodecDescriptor API is relatively new, so we need a compatibility layer for older libavcodec versions for codec names that are referenced internally, and which are different from the decoder name. (Add a configure check for that, because checking versions is getting way too messy.) demux/codec_tags.c is generated from the former codecs.conf (minus "special" decoders like vdpau, and excluding the mappings that are the same as the mappings libavformat's exported RIFF tables). It contains all the mappings from FourCCs to codec name. This is needed for demux_mkv, demux_mpg, demux_avi and demux_asf. demux_lavf will set the codec as determined by libavformat, while the other demuxers have to do this on their own, using the mp_set_audio/video_codec_from_tag() functions. Note that the sh_audio/video->format members don't uniquely identify the codec anymore, and sh->codec takes over this role. Replace the --ac/--vc/--afm/--vfm with new --vd/--ad options, which provide cover the functionality of the removed switched. Note: there's no CODECS_FLAG_FLIP flag anymore. This means some obscure container/video combinations (e.g. the sample Film_200_zygo_pro.mov) are played flipped. ffplay/avplay doesn't handle this properly either, so we don't care and blame ffmeg/libav instead.
2013-02-09 14:15:19 +00:00
.name = "lavc",
.add_decoders = add_decoders,
.init = init,
.uninit = uninit,
.control = control,
.decode = decode_with_fallback,
};