RepoMirrors/mpv
1
0
Fork 0
mirror of https://github.com/mpv-player/mpv synced 2024-12-21 22:30:22 +00:00
Code Issues Releases Wiki Activity
9dba2a52db
mpv/video/decode/dec_video.c
wm4 9dba2a52db player: add a --dump-stats option 
This collects statistics and other things. The option dumps raw data
into a file. A script to visualize this data is included too.

Litter some of the player code with calls that generate these
statistics.

In general, this will be helpful to debug timing dependent issues, such
as A/V sync problems. Normally, one could argue that this is the task of
a real profiler, but then we'd have a hard time to include extra
information like audio/video PTS differences. We could also just
hardcode all statistics collection and processing in the player code,
but then we'd end up with something like mplayer's status line, which
was cluttered and required a centralized approach (i.e. getting the data
to the status line; so it was all in mplayer.c). Some players can
visualize such statistics on OSD, but that sounds even more complicated.
So the approach added with this commit sounds sensible.

The stats-conv.py script is rather primitive at the moment and its
output is semi-ugly. It uses matplotlib, so it could probably be
extended to do a lot, so it's not a dead-end.
2014-04-17 21:47:00 +02:00

479 lines
16 KiB
C
Raw Blame 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 "config.h"
#include "options/options.h"
#include <stdio.h>
#include <stdlib.h>
#include <stdbool.h>
#include <assert.h>
#include "common/msg.h"
#include "osdep/timer.h"
#include "stream/stream.h"
#include "demux/packet.h"
#include "common/codecs.h"
#include "video/out/vo.h"
#include "video/csputils.h"
#include "demux/stheader.h"
#include "video/decode/vd.h"
#include "video/filter/vf.h"
#include "video/decode/dec_video.h"
extern const vd_functions_t mpcodecs_vd_ffmpeg;
/* Please do not add any new decoders here. If you want to implement a new
* decoder, add it to libavcodec, except for wrappers around external
* libraries and decoders requiring binary support. */
const vd_functions_t * const mpcodecs_vd_drivers[] = {
&mpcodecs_vd_ffmpeg,
/* Please do not add any new decoders here. If you want to implement a new
* decoder, add it to libavcodec, except for wrappers around external
* libraries and decoders requiring binary support. */
NULL
};
void video_reset_decoding(struct dec_video *d_video)
{
video_vd_control(d_video, VDCTRL_RESET, NULL);
if (d_video->vfilter && d_video->vfilter->initialized == 1)
vf_seek_reset(d_video->vfilter);
mp_image_unrefp(&d_video->waiting_decoded_mpi);
d_video->num_buffered_pts = 0;
d_video->last_pts = MP_NOPTS_VALUE;
d_video->last_packet_pdts = MP_NOPTS_VALUE;
d_video->decoded_pts = MP_NOPTS_VALUE;
d_video->codec_pts = MP_NOPTS_VALUE;
d_video->codec_dts = MP_NOPTS_VALUE;
d_video->sorted_pts = MP_NOPTS_VALUE;
d_video->unsorted_pts = MP_NOPTS_VALUE;
}
int video_vd_control(struct dec_video *d_video, int cmd, void *arg)
{
const struct vd_functions *vd = d_video->vd_driver;
if (vd)
return vd->control(d_video, cmd, arg);
return CONTROL_UNKNOWN;
}
int video_set_colors(struct dec_video *d_video, const char *item, int value)
{
vf_equalizer_t data;
data.item = item;
data.value = value;
MP_VERBOSE(d_video, "set video colors %s=%d \n", item, value);
if (d_video->vfilter) {
int ret = video_vf_vo_control(d_video, VFCTRL_SET_EQUALIZER, &data);
if (ret == CONTROL_TRUE)
return 1;
}
MP_VERBOSE(d_video, "Video attribute '%s' is not supported by selected vo.\n",
item);
return 0;
}
int video_get_colors(struct dec_video *d_video, const char *item, int *value)
{
vf_equalizer_t data;
data.item = item;
MP_VERBOSE(d_video, "get video colors %s \n", item);
if (d_video->vfilter) {
int ret = video_vf_vo_control(d_video, VFCTRL_GET_EQUALIZER, &data);
if (ret == CONTROL_TRUE) {
*value = data.value;
return 1;
}
}
return 0;
}
void video_uninit(struct dec_video *d_video)
{
mp_image_unrefp(&d_video->waiting_decoded_mpi);
if (d_video->vd_driver) {
MP_VERBOSE(d_video, "Uninit video.\n");
d_video->vd_driver->uninit(d_video);
}
talloc_free(d_video->priv);
vf_destroy(d_video->vfilter);
talloc_free(d_video);
}
static int init_video_codec(struct dec_video *d_video, const char *decoder)
{
if (!d_video->vd_driver->init(d_video, decoder)) {
MP_VERBOSE(d_video, "Video decoder init failed.\n");
return 0;
}
return 1;
}
struct mp_decoder_list *video_decoder_list(void)
{
struct mp_decoder_list *list = talloc_zero(NULL, struct mp_decoder_list);
for (int i = 0; mpcodecs_vd_drivers[i] != NULL; i++)
mpcodecs_vd_drivers[i]->add_decoders(list);
return list;
}
static struct mp_decoder_list *mp_select_video_decoders(const char *codec,
char *selection)
{
struct mp_decoder_list *list = video_decoder_list();
struct mp_decoder_list *new = mp_select_decoders(list, codec, selection);
talloc_free(list);
return new;
}
static const struct vd_functions *find_driver(const char *name)
{
for (int i = 0; mpcodecs_vd_drivers[i] != NULL; i++) {
if (strcmp(mpcodecs_vd_drivers[i]->name, name) == 0)
return mpcodecs_vd_drivers[i];
}
return NULL;
}
bool video_init_best_codec(struct dec_video *d_video, char* video_decoders)
{
assert(!d_video->vd_driver);
video_reset_decoding(d_video);
d_video->has_broken_packet_pts = -10; // needs 10 packets to reach decision
struct mp_decoder_entry *decoder = NULL;
struct mp_decoder_list *list =
mp_select_video_decoders(d_video->header->codec, video_decoders);
mp_print_decoders(d_video->log, MSGL_V, "Codec list:", list);
for (int n = 0; n < list->num_entries; n++) {
struct mp_decoder_entry *sel = &list->entries[n];
const struct vd_functions *driver = find_driver(sel->family);
if (!driver)
continue;
MP_VERBOSE(d_video, "Opening video decoder %s:%s\n",
sel->family, sel->decoder);
d_video->vd_driver = driver;
if (init_video_codec(d_video, sel->decoder)) {
decoder = sel;
break;
}
d_video->vd_driver = NULL;
MP_WARN(d_video, "Video decoder init failed for "
"%s:%s\n", sel->family, sel->decoder);
}
if (d_video->vd_driver) {
d_video->decoder_desc =
talloc_asprintf(d_video, "%s [%s:%s]", decoder->desc, decoder->family,
decoder->decoder);
MP_INFO(d_video, "Selected video codec: %s\n",
d_video->decoder_desc);
} else {
MP_ERR(d_video, "Failed to initialize a video decoder for codec '%s'.\n",
d_video->header->codec ? d_video->header->codec : "<unknown>");
}
talloc_free(list);
return !!d_video->vd_driver;
}
static void add_pts_to_sort(struct dec_video *d_video, double pts)
{
if (pts != MP_NOPTS_VALUE) {
int delay = -1;
video_vd_control(d_video, VDCTRL_QUERY_UNSEEN_FRAMES, &delay);
if (delay >= 0 && delay < d_video->num_buffered_pts)
d_video->num_buffered_pts = delay;
if (d_video->num_buffered_pts ==
sizeof(d_video->buffered_pts) / sizeof(double))
MP_ERR(d_video, "Too many buffered pts\n");
else {
int i, j;
for (i = 0; i < d_video->num_buffered_pts; i++)
if (d_video->buffered_pts[i] < pts)
break;
for (j = d_video->num_buffered_pts; j > i; j--)
d_video->buffered_pts[j] = d_video->buffered_pts[j - 1];
d_video->buffered_pts[i] = pts;
d_video->num_buffered_pts++;
}
}
}
// Return true if pts1 comes before pts2. pts1 can be MP_NOPTS_VALUE, but pts2
// always has to be valid. pts1 can't be equal or larger than pts2.
#define PTS_IS_ORDERED(pts1, pts2) \
((pts2) != MP_NOPTS_VALUE && ((pts1) == MP_NOPTS_VALUE || ((pts1) < (pts2))))
static double retrieve_sorted_pts(struct dec_video *d_video, double codec_pts)
{
struct MPOpts *opts = d_video->opts;
double sorted_pts;
if (d_video->num_buffered_pts) {
d_video->num_buffered_pts--;
sorted_pts = d_video->buffered_pts[d_video->num_buffered_pts];
} else {
MP_ERR(d_video, "No pts value from demuxer to use for frame!\n");
sorted_pts = MP_NOPTS_VALUE;
}
if (!PTS_IS_ORDERED(d_video->sorted_pts, sorted_pts))
d_video->num_sorted_pts_problems++;
d_video->sorted_pts = sorted_pts;
if (!PTS_IS_ORDERED(d_video->unsorted_pts, codec_pts))
d_video->num_unsorted_pts_problems++;
d_video->unsorted_pts = codec_pts;
if (d_video->header->video->avi_dts) {
// Actually, they don't need to be sorted, we just reuse the buffering.
d_video->pts_assoc_mode = 2;
} else if (opts->user_pts_assoc_mode) {
d_video->pts_assoc_mode = opts->user_pts_assoc_mode;
} else if (d_video->pts_assoc_mode == 0) {
if (codec_pts != MP_NOPTS_VALUE)
d_video->pts_assoc_mode = 1;
else
d_video->pts_assoc_mode = 2;
} else {
int probcount1 = d_video->num_unsorted_pts_problems;
int probcount2 = d_video->num_sorted_pts_problems;
if (d_video->pts_assoc_mode == 2) {
int tmp = probcount1;
probcount1 = probcount2;
probcount2 = tmp;
}
if (probcount1 >= probcount2 * 1.5 + 2) {
d_video->pts_assoc_mode = 3 - d_video->pts_assoc_mode;
MP_WARN(d_video, "Switching to pts association mode %d.\n",
d_video->pts_assoc_mode);
}
}
return d_video->pts_assoc_mode == 1 ? codec_pts : sorted_pts;
}
struct mp_image *video_decode(struct dec_video *d_video,
struct demux_packet *packet,
int drop_frame)
{
struct MPOpts *opts = d_video->opts;
bool sort_pts =
(opts->user_pts_assoc_mode != 1 || d_video->header->video->avi_dts)
&& opts->correct_pts;
double pkt_pts = packet ? packet->pts : MP_NOPTS_VALUE;
double pkt_dts = packet ? packet->dts : MP_NOPTS_VALUE;
double pkt_pdts = pkt_pts == MP_NOPTS_VALUE ? pkt_dts : pkt_pts;
if (pkt_pdts != MP_NOPTS_VALUE)
d_video->last_packet_pdts = pkt_pdts;
if (sort_pts)
add_pts_to_sort(d_video, pkt_pdts);
double prev_codec_pts = d_video->codec_pts;
double prev_codec_dts = d_video->codec_dts;
MP_STATS(d_video, "start decode video");
struct mp_image *mpi = d_video->vd_driver->decode(d_video, packet, drop_frame);
MP_STATS(d_video, "end decode video");
if (!mpi || drop_frame) {
talloc_free(mpi);
return NULL; // error / skipped frame
}
if (opts->field_dominance == 0)
mpi->fields |= MP_IMGFIELD_TOP_FIRST;
else if (opts->field_dominance == 1)
mpi->fields &= ~MP_IMGFIELD_TOP_FIRST;
// Note: the PTS is reordered, but the DTS is not. Both should be monotonic.
double pts = d_video->codec_pts;
double dts = d_video->codec_dts;
if (pts == MP_NOPTS_VALUE) {
d_video->codec_pts = prev_codec_pts;
} else if (pts <= prev_codec_pts) {
d_video->num_codec_pts_problems++;
}
if (dts == MP_NOPTS_VALUE) {
d_video->codec_dts = prev_codec_dts;
} else if (dts <= prev_codec_dts) {
d_video->num_codec_dts_problems++;
}
// If PTS is unset, or non-monotonic, fall back to DTS.
if ((d_video->num_codec_pts_problems > d_video->num_codec_dts_problems ||
pts == MP_NOPTS_VALUE) && dts != MP_NOPTS_VALUE)
pts = dts;
// Alternative PTS determination methods
if (sort_pts)
pts = retrieve_sorted_pts(d_video, pts);
if (!opts->correct_pts || pts == MP_NOPTS_VALUE) {
if (opts->correct_pts)
MP_WARN(d_video, "No video PTS! Making something up.\n");
double frame_time = 1.0f / (d_video->fps > 0 ? d_video->fps : 25);
double base = d_video->last_packet_pdts;
pts = d_video->decoded_pts;
if (pts == MP_NOPTS_VALUE)
pts = base == MP_NOPTS_VALUE ? 0 : base;
pts += frame_time;
}
if (d_video->decoded_pts != MP_NOPTS_VALUE && pts <= d_video->decoded_pts) {
MP_WARN(d_video, "Non-monotonic video pts: %f <= %f\n",
pts, d_video->decoded_pts);
}
if (d_video->has_broken_packet_pts < 0)
d_video->has_broken_packet_pts++;
if (d_video->num_codec_pts_problems || pkt_pts == MP_NOPTS_VALUE)
d_video->has_broken_packet_pts = 1;
mpi->pts = pts;
d_video->decoded_pts = pts;
return mpi;
}
int video_reconfig_filters(struct dec_video *d_video,
const struct mp_image_params *params)
{
struct MPOpts *opts = d_video->opts;
struct mp_image_params p = *params;
struct sh_video *sh = d_video->header->video;
MP_VERBOSE(d_video, "VIDEO: %dx%d %5.3f fps %5.1f kbps (%4.1f kB/s)\n",
p.w, p.h, sh->fps, sh->i_bps * 0.008,
sh->i_bps / 1000.0);
MP_VERBOSE(d_video, "VDec: vo config request - %d x %d (%s)\n",
p.w, p.h, vo_format_name(p.imgfmt));
float decoder_aspect = p.d_w / (float)p.d_h;
if (d_video->initial_decoder_aspect == 0)
d_video->initial_decoder_aspect = decoder_aspect;
// We normally prefer the container aspect, unless the decoder aspect
// changes at least once.
if (d_video->initial_decoder_aspect == decoder_aspect) {
if (sh->aspect > 0)
vf_set_dar(&p.d_w, &p.d_h, p.w, p.h, sh->aspect);
} else {
// Even if the aspect switches back, don't use container aspect again.
d_video->initial_decoder_aspect = -1;
}
float force_aspect = opts->movie_aspect;
if (force_aspect > 0.0 && d_video->stream_aspect != 0.0)
force_aspect = d_video->stream_aspect;
if (force_aspect >= 0.0)
vf_set_dar(&p.d_w, &p.d_h, p.w, p.h, force_aspect);
if (abs(p.d_w - p.w) >= 4 || abs(p.d_h - p.h) >= 4) {
MP_VERBOSE(d_video, "Aspect ratio is %.2f:1 - "
"scaling to correct movie aspect.\n", sh->aspect);
MP_SMODE(d_video, "ID_VIDEO_ASPECT=%1.4f\n", sh->aspect);
} else {
p.d_w = p.w;
p.d_h = p.h;
}
// Apply user overrides
if (opts->requested_colorspace != MP_CSP_AUTO)
p.colorspace = opts->requested_colorspace;
if (opts->requested_input_range != MP_CSP_LEVELS_AUTO)
p.colorlevels = opts->requested_input_range;
p.outputlevels = opts->requested_output_range;
// Detect colorspace from resolution.
// Make sure the user-overrides are consistent (no RGB csp for YUV, etc.).
mp_image_params_guess_csp(&p);
// Time to config libvo!
MP_VERBOSE(d_video, "VO Config (%dx%d->%dx%d,0x%X)\n",
p.w, p.h, p.d_w, p.d_h, p.imgfmt);
if (vf_reconfig(d_video->vfilter, &p) < 0) {
MP_WARN(d_video, "FATAL: Cannot initialize video driver.\n");
return -1;
}
d_video->vf_input = p;
return 0;
}
// Send a VCTRL, or if it doesn't work, translate it to a VOCTRL and try the VO.
int video_vf_vo_control(struct dec_video *d_video, int vf_cmd, void *data)
{
if (d_video->vfilter && d_video->vfilter->initialized > 0) {
int r = vf_control_any(d_video->vfilter, vf_cmd, data);
if (r != CONTROL_UNKNOWN)
return r;
}
switch (vf_cmd) {
case VFCTRL_GET_DEINTERLACE:
return vo_control(d_video->vo, VOCTRL_GET_DEINTERLACE, data) == VO_TRUE;
case VFCTRL_SET_DEINTERLACE:
return vo_control(d_video->vo, VOCTRL_SET_DEINTERLACE, data) == VO_TRUE;
case VFCTRL_SET_EQUALIZER: {
vf_equalizer_t *eq = data;
if (!d_video->vo->config_ok)
return CONTROL_FALSE; // vo not configured?
struct voctrl_set_equalizer_args param = {
eq->item, eq->value
};
return vo_control(d_video->vo, VOCTRL_SET_EQUALIZER, &param) == VO_TRUE;
}
case VFCTRL_GET_EQUALIZER: {
vf_equalizer_t *eq = data;
if (!d_video->vo->config_ok)
return CONTROL_FALSE; // vo not configured?
struct voctrl_get_equalizer_args param = {
eq->item, &eq->value
};
return vo_control(d_video->vo, VOCTRL_GET_EQUALIZER, &param) == VO_TRUE;
}
}
return CONTROL_UNKNOWN;
}
Reference in New Issue View Git Blame Copy Permalink