This is intended for cases when --hwdec needs to override the decoder
implementation in use, like for example on the RPI.
It does two things:
1. Allow the hwdec to indicate a decoder suffix. libavcodec by
convention adds a suffix to all wrapper decoders, and here we start
relying on it. While not necessarily the best idea, it's the only
thing we got. libavcodec's hwaccel list is useless, because it only
has the codec ID, not the associated decoder's name.
2. Make --hwdec=auto work properly. It shouldn't fail anymore, and hwdec
probing should reliably work, even if a different decoder is selected
with --vd. The semantics of --hwdec should dictate that it overrides
the default decoder.
Until now, we have made the assumption that a driver will use only 1
hardware surface format. the format is dictated by the driver (you
don't create surfaces with a specific format - you just pass a
rt_format and get a surface that will be in a specific driver-chosen
format).
In particular, the renderer created a dummy surface to probe the format,
and hoped the decoder would produce the same format. Due to a driver
bug this required a workaround to actually get the same format as the
driver did.
Change this so that the format is determined in the decoder. The format
is then passed down as hw_subfmt, which allows the renderer to configure
itself with the correct format. If the hardware surface changes its
format midstream, the renderer can be reconfigured using the normal
mechanisms.
This calls va_surface_init_subformat() each time after the decoder
returns a surface. Since libavcodec/AVFrame has no concept of sub-
formats, this is unavoidable. It creates and destroys a derived
VAImage, but this shouldn't have any bad performance effects (at
least I didn't notice any measurable effects).
Note that vaDeriveImage() failures are silently ignored as some
drivers (the vdpau wrapper) support neither vaDeriveImage, nor EGL
interop. In addition, we still probe whether we can map an image
in the EGL interop code. This is important as it's the only way
to determine whether EGL interop is supported at all. With respect
to the driver bug mentioned above, it doesn't matter which format
the test surface has.
In vf_vavpp, also remove the rt_format guessing business. I think the
existing logic was a bit meaningless anyway. It's not even a given
that vavpp produces the same rt_format for output.
The underlying intention of this code is to make changing
--videotoolbox-format at runtime work. For this reason, the format can't
just be statically setup, but must be read from the option at runtime.
This means the format is not fixed anymore, and we have to make sure the
renderer is property reinitialized if the format changes. There is
currently no way to trigger reinit on this level, which is why the
mp_image_params.hw_subfmt field was introduced.
One sketchy thing remains: normally, the renderer is supposed to be
involved with VO format negotiation, which would ensure that the VO
can take the format at all. Since the hw_subfmt is not part of this
format negotiation, it's implied the get_vt_fmt() callback only
returns formats supported by the renderer. This is not necessarily
clear because vo_opengl checks this with converted_imgfmt separately.
None of this matters in practice though, because we know all formats
are always supported.
(This still requires somehow triggering decoder reinit to make the
change effective.)
Until now, the presence of the process_image() callback was used to set
a delay queue with a hardcoded size. Change this to a vd_lavc_hwdec
field instead, so the decoder can explicitly set this if it's really
needed.
Do this so process_image() can be used in the VideoToolbox glue code for
something entirely unrelated.
Some functions which expected a codec name (i.e. the name of the video
format itself) were passed a decoder name. Most "native" libavcodec
decoders have the same name as the codec, so this was never an issue.
This should mean that e.g. using "--vd=lavc:h264_mmal --hwdec=mmal"
should now actually enable native surface mode (instead of doing copy-
back).
AVFormatContext.codec is deprecated now, and you're supposed to use
AVFormatContext.codecpar instead.
Handle this for all of the normal playback code.
Encoding mode isn't touched.
This commit adds the d3d11va-copy hwdec mode using the ffmpeg d3d11va
api. Functions in common with dxva2 are handled in a separate decode/d3d.c
file. A future commit will rewrite decode/dxva2.c to share this code.
The mp_set_av_packet()/mp_pts_from_av() functions check whether the
timebase is set at all (i.e. AVRational.num!=0), so there's no need to
fiddle with pointers.
Instead of displaying it only on playback start (or after switching
tracks), always display it even after a seek.
This helps with --lavfi-complex. You can now overlay e.g. audio
visualizations over cover art, and it won't break after a seek.
The downside is that this might make seeks with huge cover art slower.
There is also a glitch on seeking: since cover art pictures always
have timestamp 0, the playback time will be 0 for a moment after seek,
and then revert to audio PTS (as video is considered EOF). This is also
due to how lavfi's overlay filter behaves. (I'm not sure how to tell
lavfi that it's just a single frame.)
Deselecting cover art and then reselecting it did not work. The second
time the cover art picture is not displayed again. (This seems to break
every other month...)
The reason is commit 6640b22a. It mutates the input packet. And it is
correct that we don't own d_video->header->attached_picture at this
point. Fix it by creating a new packet reference.
Completely pointless abominations that FFmpeg refuses to remove. They
are ancient, long deprecated API which we can't use anymore. They
confused users as well.
Pretend that they don't exist. Due to the way --vd works, they can't
even be forced anymore. The older hack which explicitly rejects these
can be dropped as well.
Hr-seek was often off by one frame due to rounding issues, which have
been traditionally taken care off by adding a "tolerance". Essentially,
frames very close to the seek target PTS are not dropped, even if they
may strictly are before the seek target.
Commit 0af53353 accidentally removed this by always removing frames even
if they're within the "tolerance". Fix this by "unsharing" the logic and
making sure the segment code is inactive for normal seeks.
Doing --hwdec=auto ends up picking dxva2, creating a decoder, and then
sending D3D frames down the video chain, which immediately fails and
falls back to software.
Consider dxva2 only if the VO provides a context. If this fails,
autoprobing will proceed to try dxva2-copy as usual.
Fixes#2844.
This is in preparation for a hypothetical API change in libavcodec,
which would allow the decoder to return multiple video frames before
accepting a new input packet.
In theory, the body of the if() added to vd_lavc.c could be replaced
with this code:
packet->buffer += ret;
packet->len -= ret;
but currently this is not needed, as libavformat already outputs one
frame per packet. Also, using libavcodec this way could lead to a
"deadlock" if the decoder refuses to consume e.g. garbage padding, so
enabling this now would introduce bugs.
(Adding this now for easier testing, and for symmetry with the audio
code.)
There is some strange code which sets the DTS of the packet to PTS (but
only if it's not AVI), which apparently helps with timestamp
determination with some broken files. This code is annoying because it
tries to avoid mutating the packet (which it logically doesn't own).
Move it to where it does and get rid of the packet_copy mess.
Needed for the following commit.
This tries to determine whether packet PTS values are accurate and can
be used for frame dropping during seeking. Move both checks (PTS is
missing; PTs is non-monotonic) to the earliest place where they can be
done.
Apparently, some drivers require you to allocate all of the decoder d3d surfaces
at once. This commit changes the strategy from allocating surfaces as needed via
mp_image_pool_set_allocator, to allocating all the surfaces in one call to
IDirectXVideoDecoderService_CreateSurface and adding them to the pool with
mp_image_pool_add.
fixes#2822
This uses a different method to piece segments together. The old
approach basically changes to a new file (with a new start offset) any
time a segment ends. This meant waiting for audio/video end on segment
end, and then changing to the new segment all at once. It had a very
weird impact on the playback core, and some things (like truly gapless
segment transitions, or frame backstepping) just didn't work.
The new approach adds the demux_timeline pseudo-demuxer, which presents
an uniform packet stream from the many segments. This is pretty similar
to how ordered chapters are implemented everywhere else. It also reminds
of the FFmpeg concat pseudo-demuxer.
The "pure" version of this approach doesn't work though. Segments can
actually have different codec configurations (different extradata), and
subtitles are most likely broken too. (Subtitles have multiple corner
cases which break the pure stream-concatenation approach completely.)
To counter this, we do two things:
- Reinit the decoder with each segment. We go as far as allowing
concatenating files with completely different codecs for the sake
of EDL (which also uses the timeline infrastructure). A "lighter"
approach would try to make use of decoder mechanism to update e.g.
the extradata, but that seems fragile.
- Clip decoded data to segment boundaries. This is equivalent to
normal playback core mechanisms like hr-seek, but now the playback
core doesn't need to care about these things.
These two mechanisms are equivalent to what happened in the old
implementation, except they don't happen in the playback core anymore.
In other words, the playback core is completely relieved from timeline
implementation details. (Which honestly is exactly what I'm trying to
do here. I don't think ordered chapter behavior deserves improvement,
even if it's bad - but I want to get it out from the playback core.)
There is code duplication between audio and video decoder common code.
This is awful and could be shareable - but this will happen later.
Note that the audio path has some code to clip audio frames for the
purpose of codec preroll/gapless handling, but it's not shared as
sharing it would cause more pain than it would help.
This is required so that the individual surfaces can pass beyond the dxva2
decoder and be passed to the vo.
This also adds additional data to mp_image->planes[0] for IMGFMT_DXVA2, which is
required for maintaining and releasing the surface even if the decoder code is
uninited.
The IDirectXVideoDecoder itself is encapsulated together with its surface pool
and configuration in a dxva2_decoder structure whose creation and destruction is
managed by talloc.
Until now (and in mplayer traditionally), avi timestamps were handled
with a timestamp FIFO. AVI timestamps are essentially just strictly
increasing frame numbers and are not reordered like normal timestamps.
Limiting the FIFO is required because frames can be dropped. To make
it worse, frame dropping can't be distinguished from the decoder not
returning output due to increasing the buffering required for B-frames.
("Measuring" the buffering at playback start seems like an interesting
idea, but won't work as the buffering could be increased mid-playback.)
Another problem are skipped frames (packets with data, but which do
not contain a video frame).
Besides dropped and skipped frames, there is the problem that we can't
always know the delay. External decoders like MMAL are not going to
tell us. (And later perhaps others, like direct VideoToolbox usage.)
In general, this works not-well enough that I prefer the solution of
passing through AVI timestamps as DTS. This is slightly incorrect,
because most decoders treat DTS as mpeg-style timestamps, which
already include a b-frame delay, and thus will be shifted by a few
frames. This means there will be a problem with A/V sync in some
situations.
Note that the FFmpeg AVI demuxer shifts timestamps by an additional
amount (which increases after the first seek!?!?), which makes the
situation worse. It works well with VfW-muxed Matroska files, though.
On RPI, the first X timestamps are broken until the MMAL decoder "locks
on".
fd339e3f53 introduced a regression that caused
segfault while uniniting dxva2 decoder (and possibly vdpau too). The problem was
that it freed the avctx earlier, before calling the backend-specific uninit
which referenced it.
Revert some of the changes of that commit, and avoid calling flush by
checking whether the codec is open instead.
(Based on a PR by Kevin Mitchell.)
Signed-off-by: wm4 <wm4@nowhere>
It can be "dangerous". In particular, the decoder might have failed to
initialize, and is now in a broken state. avcodec_flush_buffers() is not
expected to be called in this state, and could trigger undefined
behavior.
Avoids "problems". In particular, it makes MMAL output a NOPTS timestamp
if the input timestamp was NOPTS.
Don't do it for other decoders. Ideally, we will at some point in the
future switch to integer fractions for timestamps at least up until the
filter layer. But this would be a larger change, and for now I'd prefer
keeping the not-rounded demuxer timestamps (if we have them).
Will be helpful for the coming filter support. I planned on merging
audio/video decoding, but this will have to wait a bit longer, so only
remove the duplicate status codes.
Commit b53cb8de added a delay queue for decoded frames. This is supposed
to be used with copy-back decoders like dxva2-copy and vaapi-copy.
Surfaces returned by them can't be referenced after uninitializing the
decoders, so they have to be released before destroying the decoder.
Move the flush_all() call above decoder uninit accordingly. Also move
the destruction of the AVFrame used for decoding (just for being
defensive - normally it doesn't hold any reference).
We just need to provide an entrypoint for it, and move the main init
code to a separate function. This gets rid of the messy video chain full
reinit in command.c, which completely destroyed and recreated the video
state for the purpose of mid-stream hw/sw switching.
Don't give the "software_fallback_decoder" field special meaning. Alwass
set it, and rename it to "decoder". Whether hw decoding is used is
determined by the "hwdec" field already.
This codes tries to deal with broken PTS timestamps, but since commit
271cabe6 it didn't always overwrite the previous timestamp as it should
have. This mattered only if there were broken timestamps in the video
stream.
Also remove the pointless prev_codec_pts variables, since the decoder
doesn't overwrite these fields anymore.
Commit b53cb8de increased this by the number of additionally delayed
surfaces. But since this is only enabled in copy-back mode (which is
what process_image is about), the other additional surfaces accounted
for the direct rendering case can be ignored.
Another fix for the crazy and insane nvidia preemption behavior.
This time, the situation is that we are using vo_opengl with vdpau
interop, and that vdpau got preempted in the background while mpv was
sitting idly. This can be e.g. reproduced by using:
--force-window=immediate --idle --hwdec=vdpau
and switching VTs. Then after switching back, load a video file.
This will not let mp_vdpau_handle_preemption() perform preemption
recovery, simply because it will do so only once vdp_decoder_create()
has been called. There are some other API calls which trigger
preemption, but many don't.
Due to the way the libavcodec API works, vdp_decoder_create() is way too
late. It does so when get_format returns. It notices creating the
decoder fails, and continues calling get_format without the vdpau
format. We could perhaps force it to reinit again (by adding a call to
vdpau.c, that checks for preemption, and sets hwdec_request_reinit), but
this seems too much of a mess.
Solve it by calling API in mp_vdpau_handle_preemption() that empirically
does trigger preemption: output_surface_put_bits_native(). This call is
useless, and in fact should be doing nothing (empty update VdpRect).
There's the slight chance that in theory it will slow down operation,
but in practice it's bound to be harmless. It's the likely cheapest and
simplest API call I've found that can trigger the fallback this way.
(The driver is closed source, so it was up to trial & error.)
Also, when initializing decoding, allow initial preemption recovery,
which is needed to pass the test mention above.
vo_chain_uninit() isn't supposed to care much about the decoder
(although decoders and outputs still go strictly together, so there is
not much of an actual difference now).
Also unset track.d_video correctly.
Remove a stale declaration from dec_video.h as well.
This covers source files which were added in mplayer2 and mpv times
only, and where all code is covered by LGPL relicensing agreements.
There are probably more files to which this applies, but I'm being
conservative here.
A file named ao_sdl.c exists in MPlayer too, but the mpv one is a
complete rewrite, and was added some time after the original ao_sdl.c
was removed. The same applies to vo_sdl.c, for which the SDL2 API is
radically different in addition (MPlayer supports SDL 1.2 only).
common.c contains only code written by me. But common.h is a strange
case: although it originally was named mp_common.h and exists in MPlayer
too, by now it contains only definitions written by uau and me. The
exceptions are the CONTROL_ defines - thus not changing the license of
common.h yet.
codec_tags.c contained once large tables generated from MPlayer's
codecs.conf, but all of these tables were removed.
From demux_playlist.c I'm removing a code fragment from someone who was
not asked; this probably could be done later (see commit 15dccc37).
misc.c is a bit complicated to reason about (it was split off mplayer.c
and thus contains random functions out of this file), but actually all
functions have been added post-MPlayer. Except get_relative_time(),
which was written by uau, but looks similar to 3 different versions of
something similar in each of the Unix/win32/OSX timer source files. I'm
not sure what that means in regards to copyright, so I've just moved it
into another still-GPL source file for now.
screenshot.c once had some minor parts of MPlayer's vf_screenshot.c, but
they're all gone.
This moves some code related to decoding from video.c to dec_video.c,
and also removes some accesses to dec_video.c from the filtering code.
dec_video.ch is starting to make sense, and simply returns video frames
from a demuxer stream. The API exposed is also somewhat intended to be
easily changeable to move decoding to a separate thread, if we ever want
this (due to libavcodec already being threaded, I don't see much of a
reason, but it might still be helpful).
The aspect ratio calculations are cached (mainly so that aspect ratio
related messages are not logged on every frame). The cache is not clared
anymore when video filters are reconfigured, but changing the
video-aspect-ratio property relied on it. Make it explicit.
Fixes#2714.
Lots of noise to remove the vfilter/vo fields from dec_video.
From now on, video filtering and output will still be done together,
summarized under struct vo_chain.
There is the question where exactly the vf_chain should go in such a
decoupled architecture. The end goal is being able to place a "complex"
filter between video decoders and output (which will culminate in
natural integration of A->V filters for natural integration of
libavfilter audio visualizations). The vf_chain is still useful for
"final" processing, such as format conversions and deinterlacing. Also,
there's only 1 VO and 1 --vf option. So having 1 vf_chain for a VO seems
ideal, since otherwise there would be no natural way to handle all these
existing options and mechanisms.
There is still some work required to truly decouple decoding.
Instead of handling this on filter chain reinit, do it directly after
the decoder. This makes the code less entangled. In particular, this
gets rid of the really weird "override params" concept in the video
filter code.
The last_format/fixed_formats have some redundance with decoder_output,
but unfortunately the latter has a slightly different use.
This is mainly a refactor. I'm hoping it will make some things easier
in the future due to cleanly separating codec metadata and stream
metadata.
Also, declare that the "codec" field can not be NULL anymore. demux.c
will set it to "" if it's NULL when added. This gets rid of a corner
case everything had to handle, but which rarely happened.
MPlayer traditionally always used the display aspect ratio, e.g. 16:9,
while FFmpeg uses the sample (aka pixel) aspect ratio.
Both have a bunch of advantages and disadvantages. Actually, it seems
using sample aspect ratio is generally nicer. The main reason for the
change is making mpv closer to how FFmpeg works in order to make life
easier. It's also nice that everything uses integer fractions instead
of floats now (except --video-aspect option/property).
Note that there is at least 1 user-visible change: vf_dsize now does
not set the display size, only the display aspect ratio. This is
because the image_params d_w/d_h fields did not just set the display
aspect, but also the size (except in encoding mode).
If reinit after a fallback from hardware fails, this field can be NULL.
The check in control() was broken due to a typo (found by Coverity), and
decode() lacked the check entirely.
Approximately reverts commit 3ccac74d. This failed with some avi files,
which do pseudo-VFR by sending packets with empty frames (or repeat
frames, depending on point of view). Specifically, these packets are not
0 bytes, so they don't get skipped by libavformat, as with the usual VFR
avi hack. Instead, the packet contains a VOP with vop_coded=0, so
libavcodec will just return no frame. We could probably distinguish such
skipped frames and delayed frames by explicitly measuring the codec
delay by counting how long it takes to get the very first frame (and
then treat skips as explicit drops), but we may as well simply reinstate
the old code.
To appease to at least one semi-broken case, do not enable this logic on
the RPI, as the FFmpeg MMAL wrapper has arbitrary buffering (and MMAL
itself is asynchronous).
10 bit HEVC would require DXVA2_ModeHEVC_VLD_Main10, and most a
different surface type (judging by lavfsplitter source code, both
P010 and P016 would work). Since I'm unable to test this stuff,
exclude 10 bit for now.
See #2516.
Until now, we've relied on the following things:
- you can send flush packets to the decoder even if it's fully flushed,
- you can send new packets to a flushed decoder,
- you can send new packers to a partially flushed decoder.
("flushing" refers to sending flush packets to the decoder until the
decoder does not return new pictures, not avcodec_flush_buffers().)
All of these are questionable. The libavcodec API probably doesn't
guarantee that these work well or at all, even though most decoders have
no issue with these. But especially with hardware decoding wrappers
(like MMAL), real problems can be expected. Isolate us from these corner
cases by handling them explicitly.
A hw decoder might fail to decode a frame for multiple reasons, and not
always just because decoding is impossible. We can't generally
distinguish these reasons well. Make it more tolerant by accepting
failures of 3 frames, but not more. The threshold can be adjusted by the
repurposed --vd-lavc-software-fallback option.
(This behavior was suggested much earlier in some PR, but at the time
the "proper" hwdec fallback was indistinguishable from decoding error.
With the current situation, "proper" fallback is still instantious.)
The uninit() function was called twice if the uninit() function failed
(once by init(), once by vd_lavc.c code), which caused crashes due to
double-free. (This failure is a corner case, and all other hwdec
backends appear to handle this case gracefully.)
I do not think this code should be able to deal with uninit() being
called other than once. Guarantee that it's called exactly once.
Fixes linker failure. How did this ever work? Apparently it did most of
the time, but apparently we just got the first case where it didn't.
Fixes#2433.
The previous commit moved the av_frame_unref() after the got_picture
check. This accidentally also deferred the software fallback
reinitialization to until a software picture was decoded (instead of the
exact time of the fallback), which is not ideal.
Just rely on the fact that calling av_frame_unref() on a frame is ok
even if nothing was decoded.
Commit 12cd48a8 started setting the hwdec_failed field even if hwdec was
not active, and because it also checked this field even if hwdec was not
active, broke decoding forever.
Fix this, and also avoid a memory leak or API misuse by releasing the
decoded picture. Passing an unreleased frame to the decoder has as far
as I know no defined effects.
The libavcodec h264 decoder contains some idiotic code with unknown
purpose (no sample or explanation known that necessitates its
existence), that causes the AVCodecContext.get_format callback to be
invoked at a time when hwaccels can't be initialized. By definition, the
get_format callback is supposed to initialize hwaccels (another idiotic
thing now part of the API, but different story). This causes hwdec
initialization sometimes to fail (WolfensteinTwitch.mp4): the first
get_format callback will mark it as failed, so the second get_format
(the "proper" normal one) will not bother restoring the state, and hwdec
init fails.
While this should be fixed in libavcodec (good luck with that), it's
quite easy to workaround.
Use the first encountered packet PTS/DTS as base, instead of the last
one. This does not add the amount of frames buffered in the codec to the
PTS offset, and thus is better.
Also, don't add the frame time if there was no decoded frame yet. The
first frame should obviously have the timestamp of the first packet
(going by this heuristic).
While b-frame reordering limits the maximum required number to around
16, the number of additionally buffered frames can be much higher.
Guess when this actually matters? (For the libavcodec MMAL wrapper.)
Useless. Sometimes it might be useful to make some extremely broken
files work, but on the other hand --no-correct-pts is sufficient for
these cases.
While we still need some of the code for AVI, the "auto" mode in
particular inflated the size of the code.
This can't be handled correctly at all. Other cases when the decoder
might drop a frame (such as completely failing to decode a frame) will
shift timestamps by a frame, and it can't be avoided.
While we could maybe find a better way to handle this with libavcodec's
main decoders, this seems to be much harder if it should work with
certain HW decoders, which don't passthrough the DTS field (such as
MMAL). Another problem are .avi files with b-frames. So just leave it
as it is.
This was used only by the timestamp sorting code, which is a fallback
for avi files (as well as avi-muxed mkv files). This was supposed to
prevent accumulating timestamps in case the decoder consumes more
packets than it outputs frames (i.e. frames are dropped). This didn't
work very well (timestamps could be off by a large amount), the
estimation of the delay was fragile, and the interdependencies with the
decoder were annoying, so kill it.
