Replace the "+" with "/". The "+" was supposed to imply that the cache
is the sum of the time (demuxer cache) and the size in bytes (stream
cache). We could not provide something nicer, because we had no idea how
many seconds of media was buffered in the stream cache.
Now the stream cache is done, and both the duration and byte size show
the amount buffered in the demuxer cache. Hopefully "/" is better to
imply this properly. Update the manpage explanations too.
Check if eglGetPlatformDisplayEXT is available and try to
use it to obtain the display connection. Fall back to eglGetDisplay
if eglGetPlatformDisplayEXT is not available or failing.
From PR #5992
When update_prop() successfully fetched a changed property value, it
set prop->changed to true to indicate the success.
mark_property_changed() also always set
prop->changed to true and additionally prop->need_new_value to true
This is the case since 6ac0ef78
If the observed property changes every frame and then due to timing
the next mark_property_changed() is called before
gen_property_change_event() and therefore directly after update_prop(),
prop->need_new_value was again true and indicated that a new value
has to be retrieved with update_prop(). As a result the event for the
last successful update_prop() was never triggered. This meant that
a property change event were never generated for frame-based properties
only for properties that were observed with MPV_FORMAT_NONE or when the
timing was different and gen_property_change_event() was called after
update_prop().
To fix this, mark_property_change() and update_prop() should not use the
same flag to indicate different things and therefore a new flag for
successfully update a property is introduced. But with the now decoupled property
changed and updated the need_new_value flag is redundant and removed completely.
Fixes#4195
With the stream cache gone, this function had almost no use anymore
(other than opening the stream). Improve this by triggering demuxer
cache readahead.
This enables all streams. At this point we can't know yet what streams
the user's options would select (at least not without great additional
effort). Generally this is what you want, and the stream cache would
have read the same amount of data.
In addition, this will work much better for files that e.g. need to seek
to the end when opening (typically mp4, and mkv files produced by newer
mkvmerge versions).
Remove the deselection call from add_stream_track(). This should be
fine, as streams normally start out as deselected anyway. In the
prefetch case, some code in play_current_file() actually deselects it.
Streams that appear during demuxing are disabled by default, so this
doesn't break this logic either.
Fixes: #6753
Before this commit, EDL or CUE files did not properly enable the cache
if they were on "slow" media (stream->streaming==true). This happened
because the stream is unset for demux_timeline, so the streaming flag
could not be queried anymore.
Fix this by adding this flag to struct demuxer, and propagate it exactly
like the is_network flag. is_network is not used for checking the cache
options anymore, and its main function seems to be something else.
Normal http streams set the streaming flag already.
This should fix#6958.
This is mostly for the case when mpv_render_context_free() is called
while video is going on. This is supposed to gracefully stop video and
deinitialize everything properly. (I feel like it would put too much on
the API user to require that video is stopped before calling this
function. Whether video is running or not is a fairly highlevel thing,
and the API user could not do it in a race-free way.)
One problem was that unit() accessed ctx after ctx->in_use was set to
false. The update(ctx) call was basically a racy use-after-free. It
needed that call to wake up the mpv_render_context_free() loop that
waited for VO uninit. Fix this by triggering the wakeup inside the lock,
and then doing "barrier" locking in mpv_render_context_free().
Another problem was that the wait loop didn't really wait properly. IT
seems the had_kill_update field was a botched attempt to do that. It's
indeed quite hairy to do that with update(). Instead make use of the
dispatch queue (infinite timeout, using mp_dispatch_interrupt()), which
handles the problem of having to wait both for dispatch queue updates
and VO uninit at the same time.
This is because dr_helper.c calls pthread_self(). It's used to avoid
deadlocks. This was not a problem internal to mpv (dr_helper.c was first
created for vo_gpu.c), but it accidentally leaked as an unintended
consequence of an implementation detail to libmpv.
This problem existed for MPV_RENDER_PARAM_ADVANCED_CONTROL users, ever
since it was introduced.
Maybe this could be done differently, but it's certainly very tricky.
the pthread_t is used to avoid deadlocks when the caller is on the same
thread as the one which needs to handle the calls.
The critical code is in free_dr_buffer_on_dr_thread(). It's called when
a DR image is free'd. Freeing a DR image requires access to the GL
context, i.e. it just has to run on the "GL" thread. In libmpv's case,
this is done by calling the API user's wakeup call, and the user will
eventually call mpv_render_context_update() on the "GL" thread, which in
turn calls mp_dispatch_queue_process() on the dispatch queue that was
passed to dr_helper_create(), which then calls av_buffer_unref(), which
calls gl_video_dr_free_buffer(). (God who came up with this rube
goldberg shit.)
The above case will typically happen when e.g. vd_lavc.c (internal mpv
thread) frees images. Allocation works similarly; deallocation is just
trickier because calls to free images are everywhere.
Now consider if mpv_render_context_render() releases a DR image. This
function is always called on the "GL" thread. Going through the dispatch
queue would obviously deadlock, because according to the render API
rules, the user's wakeup callback must not mpv_render_context_update(),
instead it has to signal the "GL" thread, which must wait until
mpv_render_context_render() returns. To avoid this deadlock, dr_helper.c
checks whether the calling thread is the "GL" thread, and if so, allows
a reentrant call (basically gl_video_dr_free_buffer() is called
directly).
While with GL, you usually will stay on the same thread, API users were
in theory allowed to e.g. move the GL context to a different thread. But
this dr_helper issue means this is not possible. Moreover, other APIs
will not have the same thread-locality problems as GL.
FUCK THIS SHIT
In theory, libmpv could provide an API to "move" the context to a
separate thread, but let's not start with even more broken crap like
this. I'm not sure yet whether there is an easy solution. Maybe all
mpv_render() function could be guarded by entry/exit functions, which
set/unset a flag that dr_helper.c should use reentrant calls.
libmpv users which do not set MPV_RENDER_PARAM_ADVANCED_CONTROL were
never affected.
Preparation for the next commit. Until now, it was only needed if DR was
involved. One reason for not always creating it was that you normally
must not use it if advanced_control is not enabled. This is why e.g.
VOCTRL_SCREENSHOT now checks for that variable; it still can't use
ctx->dispatch if the render API user did not enable it.
render api needs to wait for vo to be destroyed before frees the context.
The purpose of kill_cb is to wake up render api after vo is destroyed,
but uninit did that before kill_cb, so kill_cb tries using the freed
memory. Remove kill_cb to fix the issue as uninit is able to do the
work.
Equalizer control was redone in 03cf150ff3 (over 2 years
ago). Ever since, the equalizer control structs and the GET voctrl have
been unused. Only the SET voctrl is still used as notification mechanism
(actually a bad hack to avoid some further option change handling
complexity).
Remove the unused parts.
I was assuming posix_memalign was the most portable function to use, but
MinGW does not provide it for some reason. Switch to C11 aligned_alloc()
which someone suggested was provided by MinGW (but actually isn't,
someone probably confused it with the incompatible _aligned_malloc),
and add a configure check.
Even though it turned out that MinGW doesn't provide it, the function
is slightly more elegant than posix_memalign(), so stay with it.
This field is documented as internal, so an API user should not
access it. However, this is the only way to get some read statistics
without replacing FFmpeg's entire HLS demuxer. (Using custom I/O as
workaround doesn't work: the HLS code uses some weird internal APIs
that cannot be provided by FFmpeg API users; I even made the author
of the relevant patch to provide a public API, but which was shot
down by another FFmpeg developer. So I take this as my right to
access this field.)
Mention this explicitly, as it affects ABI and API compatibility, and
I don't want that anyone claims this was a "mistake". Add some
explanations.
In theory, a 0 size allocation could have made it memset() on a NULL
pointer (with a non-0 size, which makes it crash in addition to
theoretical UB).
This should never happen, since even packets with size 0 should have an
associated allocation, as FFmpeg currently does. But avoiding this makes
the API slightly more orthogonal and less tricky, I guess.
This reverts commit 95636c65e7.
This change shouldn't be needed, and in fact it's wrong. The FFmpeg API
function could do anything it wants with the packet, including changing
the packet data pointer. Likewise, it's not guaranteed that the
referenced packet's fields mirror the current state of the mpv packet
struct (the AVPacket is only kept for the AVBuffer and the side data
stuff).
skip-logo.lua is just what I wanted to have. Explanations are on the top
of that file. As usual, all documentation threatens to remove this stuff
all the time, since this stuff is just for me, and unlike a normal user
I can afford the luxuary of hacking the shit directly into the player.
vf_fingerprint is needed to support this script. It needs to scale down
video frames as part of its operation. For that, it uses zimg. zimg is
much faster than libswscale and generates more correct output. (The
filter includes a runtime fallback, but it doesn't even work because
libswscale fucks up and can't do YUV->Gray with range adjustment.)
Note on the algorithm: seems almost too simple, but was suggested to me.
It seems to be pretty effective, although long time experience with
false positives is missing. At first I wanted to use dHash [1][2], which
is also pretty simple and effective, but might actually be worse than
the implemented mechanism. dHash has the advantage that the fingerprint
is smaller. But exact matching is too unreliable, and you'd still need
to determine the number of different bits for fuzzier comparison. So
there wasn't really a reason to use it.
[1] https://pypi.org/project/dhash/
[2] http://www.hackerfactor.com/blog/index.php?/archives/529-Kind-of-Like-That.html
Until now they weren't observable and never reported any updates. Apply
a shitty hack to make them mostly-observable. It relies on the "idle"
event, which is basically triggered on every frame displayed, or
similar. This can lead to property change notifications not being sent
quickly enough.
The cleaner solution would be adding a notification mechanisms from
filters, but I'm too lazy for that.
For simplicity, these properties usually query the metadata from the
filter twice, even if it's not technically needed at all. The reason for
this is mostly the horrible (and legacy) sub-path access (which is why
tag_property() is so complex).
But for simple cases, we can easily avoid double querying, so do that.
The benefit is performance (well, won't matter), and supporting filters
that reset information on query (for later).
Generally, using x86 SIMD efficiently (or crash-free) requires aligning
all data on boundaries of 16, 32, or 64 (depending on instruction set
used). 64 bytes is needed or AVX-512, 32 for old AVX, 16 for SSE. Both
FFmpeg and zimg usually require aligned data for this reason.
FFmpeg is very unclear about alignment. Yes, it requires you to align
data pointers and strides. No, it doesn't tell you how much, except
sometimes (libavcodec has a legacy-looking avcodec_align_dimensions2()
API function, that requires a heavy-weight AVCodecContext as argument).
Sometimes, FFmpeg will take a shit on YOUR and ITS OWN alignment. For
example, vf_crop will randomly reduce alignment of data pointers,
depending on the crop parameters. On the other hand, some libavfilter
filters or libavcodec encoders may randomly crash if they get the wrong
alignment. I have no idea how this thing works at all.
FFmpeg usually doesn't seem to signal alignment internal anywhere, and
usually leaves it to av_malloc() etc. to allocate with proper alignment.
libavutil/mem.c currently has a ALIGN define, which is set to 64 if
FFmpeg is built with AVX-512 support, or as low as 16 if built without
any AVX support. The really funny thing is that a normal FFmpeg build
will e.g. align tiny string allocations to 64 bytes, even if the machine
does not support AVX at all.
For zimg use (in a later commit), we also want guaranteed alignment.
Modern x86 should actually not be much slower at unaligned accesses, but
that doesn't help. zimg's dumb intrinsic code apparently randomly
chooses between aligned or unaligned accesses (depending on compiler, I
guess), and on some CPUs these can even cause crashes. So just treat the
requirement to align as a fact of life.
All this means that we should probably make sure our own allocations are
64 bit aligned. This still doesn't guarantee alignment in all cases, but
it's slightly better than before.
This also makes me wonder whether we should always override libavcodec's
buffer pool, just so we have a guaranteed alignment. Currently, we only
do that if --vd-lavc-dr is used (and if that actually works). On the
other hand, it always uses DR on my machine, so who cares.
This matters when talloc allocations set destructors. Before this
commit, destructors were called in the same order as they were added to
the parent allocations. Now it happens in reverse order.
I think this makes more sense. It's reasonable to assume that an
allocation that was added later may depend on any of the previous
allocations, so later additions should be destroyed first. (Of course
other orders are entirely possible too.)
Hopefully this doesn't fix or break anything, but I can't be sure (about
either of those). It's risky. (Then why do it?)
The destructor of a parent allocation is called before its children. It
makes sense and must stay this way, because in most cases, the
destructor wants to access the children.
This is a reason why I don't really like talloc (it wasn't my idea to
use talloc, is my excuse). Quite possible that destructors should be
removed from talloc entirely. Actually, this project should probably be
rewritten in Rust (or a better language), but that would be even more of
a pain; also, I think this is just the right level of suffering and
punishment.
It seems using multiple prefixes for an option isn't supported out of
laziness (and shouldn't, because what the fuck). So assert() on this.
(Unfortunately this prefix nonsense is still needed. Especially AO and
VO options use this through the options_prefix field.)
A dumb thing that the cursed property-option bridge accidentally did.
Normal deprecated options on the other hand are fine in the property
list, because they're wanted for compatibility.
A previous commit changed m_config so that it always creates the shadow
thing, and the function's only remaining purpose was to initialize
mpv_global. It makes much more sense to do that at the caller, and it's
only 1 line of code too.
m_config has a m_config_option array, that is used for all option
access. The code maintaining shadow copies also tried to make use of it,
and did so by "cleverly" assigning each m_sub_options run a slice of
that array. But actually it's much simpler to, you know, directly access
the damn options.
This helps separation m_config and the general option code slightly.
Still seems to work after a superficial test, good enough.
This is good because a private thing is not so public anymore, and it's
also preparation for further changes.
Some tricky memory management issues: m_config_data (i.e. config->data)
now depends on m_config_shadow, instead of m_config. In particular,
free_option_data() accesses the m_config_shadow.groups array. Obviously
it must be freed before m_config_shadow.
Unused now. The old stream cache used it, but it was removed.
On a side note, the demuxer cache uses mp_mkostemps(). It looks like our
Windows open() emulation handles this correctly by using CREATE_NEW, so
no functionality gets lost by the "new" approach. On the other hand, the
demuxer cache does not set FILE_FLAG_DELETE_ON_CLOSE, but instead tries
to delete the file after opening (POSIX style), which probably won't
work on Windows. But I'm not sure how to make it use the DELETE_ON_CLOSE
flag, so whatever.
Move the comments documenting exported functions to the header. It looks
like the header is the preferred place for that (although I don't really
appreciate headers where you lose the overview because of all the
documentation comments). Add comments to some undocumented prototypes.
This was one of those "shouldn't exist" type of functions that could
access internals that were supposed to be isolated away, but some code
needed to access it anyway.
It looks like the last use of it went away in 2016, shortly after it was
introduced.
Dear diary,
today I fixed a shitty bug that was all my fault because I made a
horrible mess. (Except it was a horrible mess before I even touched
this shit, but let's not blame others.)
Sometimes, updates to VO option that control video sizing (like panscan)
didn't update the screen correctly. They were delayed until the next
option change or so.
It turns out that if the option update happens at the "same" time as a
VOCTRL, update_opts() doesn't actually notify the vo_driver of the
change. This in turn happened because run_control() called
m_config_cache_update(). The latter function returns true if the options
changed since the last call, and update_opts() also calls it (on the
same config cache) for the same purpose. The update_opts() call, which
is triggered by a third mechanism, comes later, but the cache update
call will return false (as it should). Basically, given the config API,
you can't act differently on multiple update calls and expect it to
work. The skipped handling in update_opts() meant that the notification
required to apply the changed option wasn't run.
Fix this by simply calling update_opts() directly instead. Now there's
only 1 m_config_cache_update() call on this specific instance. Fix the
call in run_reconfig() too, so the previous sentence isn't a lie (but it
probably doesn't make a difference in practice due to certain details).
I'm not sure how I even ran into this sort-of race condition. The VOCTRL
that messed up the option update was VOCTRL_UPDATE_PLAYBACK_STATE, which
happens semi-regularly.
Why this config cache shit and all the other shit? Rediscovering this
crap wasn't pleasant. It's a bunch of hacks that became necessary when
the ancient MPlayer architecture made it hard to move the VO to a
separate thread.
All the VO code typically accesses vo->opts (whose fields all used to be
global variables in MPlayer). The frontend changes these on user input.
Putting locking around all the options would be a nightmare, and keeping
a copy of the options in the thread was much simpler. You need a way to
propagate option changes, notify the thread, and update the local copy
too. And the result of these thoughts was the config cache mechanism.
In this specific case, the relevant cache update call in update_opts()
triggers a VOCTRL_SET_PANSCAN to the VO driver, which isn't related to
its former function anymore. Instead, it causes the VO driver to update
the video sizing/placing options, which the generic VO code can't do.
(Mostly because the VO driver includes the windowing stuff and is
responsible for resizing etc. itself.)
VOCTRLs sent by the frontend are even worse. MPlayer had no real runtime
option change mechanism. Some options were vaguely duplicated by
properties, so you could effectively change those options at runtime.
Each of these options had its own VOCTRL, which still exist today, e.g.
VOCTRL_FULLSCREEN, or VOCTRL_ONTOP. I tried to make all options runtime
changeable, and to unify properties with options. But I couldn't be
bothered with updating all VO drivers to listen to option changes
directly, because that would be pretty tedious. So the property code is
still all there and sends the old VOCTRLs. But of course you need to
sync up the options, which is why the run_control() code did that.
(Unrelated: VO_EVENT_FULLSCREEN_STATE is the worst shithack of them all.
Currently, only the frontend can actually write to options (for awful
reasons), so if the fullscreen state changes due to outside interaction,
the VO driver can't update the corresponding option fields. So the VO
notifies the frontend with said VO_EVENT_, and the frontend then sends
VOCTRL_GET_FULLSCREEN, and updates the global copy of the option with
the value returned by that. I still like to think the situation is not
that bad considering the monstrous effort of converting single-threaded
code that had hundreds of options in global variables to multi-threaded
code with no global variables at all.)
Helper for the ab-loop-dump-cache command, see manpage additions.
This is kind of shit. Not only is this a very "special" feature, but it
also vomits more messy code into the big and already bloated demux.c,
and the implementation is sort of duplicated with the dump-cache code.
(Except it's different.) In addition, the results sort of depend what a
video player would do with the dump-cache output, or what the user wants
(for example, a user might be more interested in the range of output
audio, instead of the video).
But hey, I don't actually need to justify it. I'm only justifying it for
fun.
This is the muxer used by all 3 stream recording features (why are there
so many?). It tried hard to avoid writing broken files. In particular,
it buffered packets until it new there was a keyframe packet (which, in
mpv's/FFmpeg's definition, mean seek points from which decoding can
resume), or final EOF. The danger that was probably considered here was
that due to video frame reordering, not muxing some trailing, missing
packets of a keyframe range could lead to broken decoding or skipped
frames, so better discard packets belonging to an incomplete range.
Sounds like a good idea so far.
Unfortunately, this will drop an entire keyframe range even if the
current packet run is complete and mp_recorder_mark_discontinuity() is
called, simply because recorder.c can not know that the next packet
would have been a keyframe.
It seems better to mux all packets to avoid losing valid data, even if
it means that sometimes packets/frames will be missing from the file. It
benefits especially the dump-cache command, which will call the function
to signal a discontinuity after every range. Before this commit, it
discarded the last packets, even if they were perfectly fine.
(An alternative solution for dump-cache would have been a second
discontinuity marker function, that communicates that the current packet
range is complete. But this commit's solution is simpler and overall
more robust, at the danger of producing more semi-broken files.)
This may make some of the complex buffering/waiting logic in recorder.c
pointless.
Untested (in this final form).
But don't tell the reader which those APIs are. Hope the user will just
search for "async" in the Lua section (lua.rst). But of course, nobody
will ever care about anything related to this.
That's right, and it's probably not the end of it. I'll just claim that
I have no idea how to create a proper user interface for this, so I'm
creating multiple partially-orthogonal, of which some may work better in
each of its special use cases.
Until now, there was --record-file. You get relatively good control
about what is muxed, and it can use the cache. But it sucks that it's
bound to playback. If you pause while it's set, muxing stops. If you
seek while it's set, the output will be sort-of trashed, and that's by
design.
Then --stream-record was added. This is a bit better (especially for
live streams), but you can't really control well when muxing stops or
ends. In particular, it can't use the cache (it just dumps whatever the
underlying demuxer returns).
Today, the idea is that the user should just be able to select a time
range to dump to a file, and it should not affected by the user seeking
around in the cache. In addition, the stream may still be running, so
there's some need to continue dumping, even if it's redundant to
--stream-record.
One notable thing is that it uses the async command shit. Not sure
whether this is a good idea. Maybe not, but whatever. Also, a user can
always use the "async" prefix to pretend it doesn't.
Much of this was barely tested (especially the reinterleaving crap),
let's just hope it mostly works. I'm sure you can tolerate the one or
other crash?
The screenshot command has this weird behavior that it shows messages
both on terminal and OSD by default, but that a command prefix can be
used to disable the OSD message.
Move this mechanism to common code, and make this available to other
commands too (although as of this commit only the screenshot commands
use it).
This gets rid of the weird screenshot_ctx.osd field too, which was sort
of set on a command, and sometimes inconsistently restored after the
command.
It makes some slight sense and helps with one of the following commits.
Also rename that other function to make it sound less similar to
find_seek_target().
