This affects only the display-sync code path, as for normal timing the
wakeup_pts stuff handles proper wakeup. It's probably mostly a
theoretical issue.
Discontinuities (like toggling fullscreen) can cause multiple frames to
be dropped in succession, which sounds very weird. It's better to drop
some video frames instead to compensate for larger desyncs.
We roughly base it on the maximum allowed speed changes (audio change is
"additional" to the video change to account for deviations when playing
at max. video speed change).
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.)
ao_alsa: attempt to fix 7.1 over HDMI
The last 2 channels of 7.1 (RLC/RRC in ALSA) were exported as sdl/sdr
instead of sl/sr (I don't even know why I chose sdl/sdr, but SL/SR
and RLC/RRC are different in the ALSA API). libsw/avresample do not
move the sl/sr channels to sdl/sdr when rematrixing, so silence was
sent for 2 channels. If my selection of sdl/sdr is essentially API
abuse, there's no reason why they should do this differently.
The mess here is really that ALSa doesn't map the HDMI layouts cleanly.
Most ALSA drivers export 7.1 in a way compatible to our expectations,
but Intel HDA/HDMI does not:
mpv/ffmpeg: fl-fr-fc-lfe-bl-br-sl-sr
ALSA/generic: FL FR FC LFE RL RR SL SR [1]
ALSA/HDMI: FL FR LFE FC RL RR RLC RRC [2]
The HDMI layout is layout 0x13 (going by CEA-861-B). The comment in
the kernel code has to be correct too. The early standard defines only
1 other layout, which replaces RLC/RRC with FRC/FLC - this probably
corresponds to what we call "7.1(wide)".
So it appears when ALSA requests RLC/RRC, we should feed it sl/sr.
To make it more complicated, Kodi/xbmc apparently also have to deal with
ALSA being special, but instead of sending sl/sr to RLC/RRC, they swap
the last two pairs of the layout, and send sl/sr to RL/RR and bl/br to
RLC/RRC. Or I might have misunderstood their code. I don't have a
7.1-capable A/V receiver, so I can't test this.
For now, go with the simpler solution, and wait until someone tests it.
If the speakers end up swapped, a completely different solution will be
needed.
[1] https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/sound/core/pcm_lib.c?id=refs/tags/v4.3#n2434
[2] https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/sound/pci/hda/patch_hdmi.c?id=refs/tags/v4.3#n307
These calls actually can leave the ALSA configuration space empty (how
very useful), which is why snd_pcm_hw_params() can fail. An earlier
change intended to make this non-fatal, but it didn't work for this
reason.
Backup the old parameters, so we can retry with the non-empty
configuration space. (It has to be non-empty, because the previous
setters didn't fail.)
Note that the buffer settings are not very important to us. They're
a leftover from MPlayer, which needed to write enough data to the
audio device to not underrun while decoding and displaying a video
frame. In mpv, most of these things happen asynchronously, _and_
there is a dedicated thread just for feeding the audio device, so
we should be pretty imune even against extreme buffer settings. But
I suppose it's still useful to prevent PulseAudio from making the
buffer too large, so still keep this code.
Again, this could have bad access, is unlikely, and has no bad
consequences. It's noteworthy that vlc and the ALSA PCM example both do
this first, even if they set the sample rate later.
I'm worried that not restricting the access type before restricting the
format will cause problems. While it's unlikely, it might prevent
failures in some corner cases. Also, since we by default always use
interleaved access (buggy ALSA plugins), this will have no effects at
all.
If the API doesn't list padded channel maps, but the final device
channel map is padded, and if unpadded output is not possible (unlike in
the somewhat similar dmix case), then we shouldn't apply the channel
count mismatch fallback in the beginning. Do it after channel map
negotiation instead.
Doesn't matter much; effectively this prevents just log spam in some
cases where the map is legitimately padded. Normally this is really
only needed for the dmix ALSA case. (See git blame for details.)
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.
Quoting MSDN: "Notifies the Desktop Window Manager (DWM) to opt in to or
out of Multimedia Class Schedule Service (MMCSS) scheduling while the
calling process is alive.". Whatever this means. (An application can
change the scheduling priority of the window manager?)
Does this improve anything? I have no idea. Certainly this is a program
that does multimedia and graphics, so we seem to be a good match for
this.
Is it bad if we enable this even while playback is inactive or paused? I
have no idea either.
Is there a magic cargo cult function that will mark our renderer thread
as multimedia thing? I have no idea. (We use a function to enable MMCSS
for our audio thread in ao_wasapi.)
Enable it by default, but not unconditionally. Add an "auto" mode, which
disable DwmFlush if the compositor is (probably) inactive. Let's see how
this goes.
Since I accidentally enabled DwmFlush always by default (more or less)
in a previous commit touching this code, this is probably mostly just
cargo-culting, and it's uncertain whether it does anything.
Note that I still got bad vsync behavior when fullscreening mpv, and
making another window visible on the same screen. This happens even if
forcing DWM.
update_av_diff() works on the timestamps, while time_left is in real
time. When playing at not-1 speed, these are very different, and cause
the A/V difference to jitter. Fix this by scaling the expected A/V
desync to the correct range.
Commit acd5816a broke this. It was stopping playback occasionally.
Another case where the non-display-sync interpolation mode
(in->vsync_timed==true) is causing a lot of subtle issues and will be
removed soon.
PAL8 is the only format that is RGB, has only 1 component, is byte-
aligned. It was accidentally detected by the GBRP case as planar RGB.
(It would have been ok if it were gray; what ruins it is that it's
actually paletted, and the color values do not correspond to colors (but
palette entries).
Pseudo-pal formats are ok; in fact AV_PIX_FMT_GRAY is rightfully marked
as MP_IMGFLAG_YUV_P.
This didn't show up with cases where the frame pattern has a cycle of 1
or 2 like it is the case with 24-on-24 fps, or 24-on-60 fps. It did show
up with 25-on-60 fps. (We don't slow down 25 fps video to 24 on default
settings.)
In this case, we must not add the timing error of the next frame to the
A/V difference estimation of the current frame. Use the previous timing
error instead.
This is another bug resulting from the confusion about whether we
calculate parameters for the currently playing frame, or the one we're
about to queue.
Regression since commit 93db4233. I think the bit that was forgotten
here was to remove the vo_w32_config() return value completely. The VO
failed to init because that function always returned 0. This commit
removes these bits and fixes the VO.
Fixes#2434.
Yet another relatively useless option that tries to make OpenGL's sync
behavior somewhat sane. The results are not too encouraging. With a
value of 1, vsync jitter is gone on nVidia, but there are frame drops
(less than with glfinish). With 2, I get the usual vsync jitter _and_
frame drops.
There's still some hope that it might prevent too deep queuing with some
GPUs, I guess.
The timeout for the wait call is 1 second. The value is pretty
arbitrary; it should just not be too high to freeze the process (if
the GPU is un-nice), and not too low to trigger the timeout in normal
cases, even if the GPU load is very high. So I guess 1 second is ok
as a timeout.
The idea to use fences this way to control the queue depth was stolen
from RetroArch:
df01279cf3/gfx/drivers/gl.c (L1856)
Commit a1315c76 broke this slightly. Frame drops got counted multiple
times, and also vo.c was actually trying to "render" the dropped frame
over and over again (normally not a problem, since frames are always
queued "tightly" in display-sync mode, but could have caused 100% CPU
usage in some rare corner cases).
Do not repeat already dropped frames, but still treat new frames with
num_vsyncs==0 as dropped frames. Also, strictly count dropped frames in
the VO. This means we don't count "soft" dropped frames anymore (frames
that are shown, but for fewer vsyncs than intended). This will be
adjusted in the next commit.
vo_frame.num_vsyncs can be != 1 in some cases in normal sync mode too.
This is not a very exact fix, but in exchange it's robust. (These
vo_frame flags are way too tricky in combination with redrawing and
such.)
There were occasional shader compilation and rendering failures if FBOs
were unavailable. This is caused by the FBO caching code getting active,
even though FBOs are unavailable (i.e. dumb-mode).
Boken by commit 97fc4f.
Fixes#2432.
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.
Bump it to 80, and 2 vsyncs. This is another measure against vsync
jitter. Admittedly this is a bit simplistic (and we should probably
estimate a stable estimated vsync phase instead), but for now this will
do.
It's not needed, because the additional data is not appended, but is the
total size of the audio buffer. The maximum size is the static audio
drop size (or twice, if the audio is duplicated).
Calculate the A/V difference directly in the display sync code, instead
of the awkward current way, which reuses the fields for audio sync.
We still set time_frame, because it makes falling back to audio sync
somewhat smoother.
When dropping or repeating frames, we essentially influence when the
frame after the next frame will be shown, not the next frame. This led
to dropping/repeating frames 2 times, because the A/V difference had a
delay of one frame. Compensate it with the expected value.
av_free_packet() got finally deprecated. Use av_packet_unref() instead,
which has almost the same semantics, has existed for a while, and is
available in all FFmpeg and Libav versions we support.
This was not very reliable.
In the normal vo_opengl case, this didn't deal well enough with vsync
jitter. Vsync timings can jitter quite extremely, up to a whole vsync
duration, in which case the "missed" frame counter keeps growing, even
though nothing is wrong. This behavior also messes up the A/V difference
calculation, but as long as it's within tolerance, it won't provoke
extra frame dropping/repeating. Real misses are harder to detect, and I
might add such detection later.
In the vo_opengl_cb case, this was additionally broken due to the
asynchronity between renderer and VO threads.
wm4
Evgeny Kolesnikov
rr-
Paul B Mahol