Sway 1.5 started sending more pointer motion events to mpv which broke
the autohiding behavior. The cursor would appear again if you
fullscreened. Sway had a good reason to do this because certain
applications had inconsistencies between hardware cursor and software
cursor without rebasing on state changes[1]. So mpv needs to take this
special case into consideration.
Initially, simply checking mouse coordinates for changes was considered,
but this doesn't work. All coordinates are surface-local in wayland so
something can appear to move in the local coordinate space but not
globally. You're not allowed to know global mouse coordinates in
wayland, and we don't care about local coordinate changes in mpv so this
approach isn't viable.
Instead, let's just keep track of a local state change. If the toplevel
surface changes in some way (fullscreen, maximized, etc.), then just set
a bool that lets us ignore the mp_input_set_mouse_pos function. This
keeps the cursor from appearing simply because the state was changed
(i.e. fullscreening). For compositors that don't send pointer motion
events on a state change, this does technically mean that the initial
mp_input_set_mouse_pos is never set. In practice, this isn't a
noticeable difference though because moving a mouse generates a ton of
motion events so you'll immediately see it on the second motion event.
[1] https://github.com/swaywm/sway/issues/5594
wl_display_dispatch is dangerous because it will block forever if the
event queue is empty. Any direct calls to this function should just be
replaced with wl_display_dispatch_pending which accomplishes the same
thing for mpv's purposes without any chance of blocking.
The other potential trap is wl_display_roundtrip. It can internally call
wl_display_dispatch which in certain circumstances could potentially
block. There are cases where we need the server to finish processing
client requests before doing anything else so this can not be cleanly
avoided. The dangerous call is the usage of wl_display_roundtrip in
vo_wayland_wait_frame. In the majority of cases, this shouldn't be a
problem because the previous wl_display_read_events should always queue
up some events on the fd for wl_display_roundtrip to send. However, the
compositor could potentially send us an error in the display queue that
could lead to bad behavior when wl_display_roundtrip is called.
The wl_display_roundtrip can't be removed because we are relying on its
semi-blocking capabilities, but the logic can be slightly adjusted to be
safer. The obvious thing to do is to make sure we check the pollfd for
any errors. If one is returned, then we call wl_display_cancel_read and
try again. The less obvious trick is to call wl_display_dispatch_pending
and move wl_display_roundtrip outside of the blocking + timeout loop.
This change has some subtle but important differences. Previously,
vo_wayland_wait_frame would read an event and wait on the server to
process it one-by-one. With this change, the events are dispatched as
soon as possible to the server and then we wait on all of those
(potentially multiple) events to be processed after we have either
received frame callback or the loop times out.
After that is done, we can then check for if there are any errors on the
display. If it's all clear, we can run wl_display_roundtrip without any
worries. If some error happens, then don't execute the function at all.
Previously, the compositor was signaled that a drag-and-drop ended with
wl_data_offer_finish in check_dnd_fd. This is, however, erroneous
because it is outside of the data_device_listener and in some cases
caused errors with certain compositors. check_dnd_fd itself does not
need to know or care about anything that happens in wayland. It just
needs to read data from an fd. The simple fix is to just always signal
the end of a drag-and-drop in data_device_handle_drop. check_dnd_fd can
free memory and close the fd later, but it should not talk to the
compositor. Fixes#7954.
The read of the wayland display fd in vo_wayland_wait_events was
incorrect and technically vulnerable to race conditions. The correct
usage as per the client api is to use wl_display_prepare_read as well as
wl_display_read_events.
Rather than after tone-mapping. This prevents overflow when the
pre-tonemapped signal contains inputs exceeding sig_peak. I also
realized that with this clipping in place, post-clipping no longer needs
to be done, so this isn't even particularly slower.
The only two exceptions to the rule are "clip" and "linear", which
relied on the post-clipping to do their tone mapping properly.
Fixes#7929
This normally gets printed by libplacebo itself when initializing the
context, but due to the way our code is structured (for convenience) we
don't have the log hook enabled by the time this function call is
relevant. So instead just print it manually as an easier work-around
than restructuring the code.
This probably makes it much faster (I wouldn't know, I didn't run any
benchmarks ). Seems to work as well (although I'm not sure, it's not
like I'd perform rigorous tests).
The scale_zimg test seems to mysteriously treat color in fully
transparent alpha differently, which makes no sense, and isn't visible
(but makes the test fail). I can't be bothered with investigating this
more. What do you do with failing tests? Correct, you disable them. Or
rather, you disable whatever appears to cause them to fail, which is the
threading in this case.
This change follows mostly the tile_example.cpp. The slice size uses a
minimum of 64, which was suggested by the zimg author. Some of this
commit is a bit inelegant and weird, such as recomputing the scale
factor for every slice, or the way slice_h is managed. Too lazy to make
this more elegant.
zimg git had a regressio around active_region (which is needed by the
slicing), which was fixed in commit 83071706b2e6bc634. Apparently, the
bug was never released, so just add a warning to the manpage.
The intention is to add slice-threading to the wrapper. For that
purpose, move all zimg related state to a new struct mp_zimg_state.
There is now an array of instances of this struct. The intention is to
have one instance for each thread. As of this commit, this is hardcoded
to 1 thread; the following commit extends this.
It turns out that the AMD driver doesn't bother to set the size
field in the descriptor for an exported VA surface. I guess they
assume the caller can always use lseek() and don't bother. So, we
need to use lseek() in these situations.
Modified-by: Niklas Haas <git@haasn.xyz>
Guarded this behind PL_API_VER >= 88 to prevent it from exploding on
older libplacebo versions, where vaapi support does not yet work
properly on AMD due to lack of DRM modifiers.
we properly set the unfs window size on live resize end. due to a race
condition in the fullscreen events, which is also a live resize, the
unfs window size is incorrectly set to a fullscreen size. this happens
when the end fs screen event triggers before the end of live resize one.
this just adds a second condition to not be un fullscreen when updating
the unfs window size.
Mess this into the --geometry option, because I like to be
irresponsible. I considered adding a separate option, but at least this
allows me to defer the question how the hell this should work as
property (geometry simply and inherently does not).
Tested on IceWM only. Option equality test and string output not tested.
Probably did not cause any practical problems, but it sure seems
unclean. sws_utils users might also rely on these fields being exactly
the same as the actual input/output. It's better to avoid this.
I'm tired of dealing with this frequent spawning of xdg-screensaver when
debugging and what not. xdg-screensaver was never a serious tool anyway,
it's more like some self-deprecating joke by FDO folks.
This will affect X11 on GNOME and other DEs. I'm singling out GNOME
though, because they are the ones actively sabotaging any sane
technical solutions and community cooperation.
I have been accused of taking it out on innocent GNOME users, while none
of this will reach GNOME developers. Of course that is not the
intention.
This can be used to make vo_libmpv render video to a memory buffer. It
only adds a new backend API that takes memory surfaces. All the render
API (such as frame rendering control and so on) is reused.
I'm not quite convinced of the usefulness of this, and until now I
always resisted providing something like this. It only seems to
facilitate inefficient implementation. But whatever.
Unfortunately, this duplicates the software rendering glue code yet
again (like it exists in vo_x11, vo_wlshm, vo_drm, and probably more).
But in theory, these could reuse this backend in the future, just like
vo_gpu could reuse the render_gl API.
Fixes: #7852
libplacebo exposes this feature already, because this particular type of
bug is unusually common in practice. Simply make use of it, by exposing
it as an option.
Could probably also bump the libplacebo minimum version to get rid of
the #if, but that would break debian oldoldstable or something.
Fixes#7867.
Using mediump float on GLES causes problems with kernel resampling,
PQ HDR, and possibly others. The issues are fixed by using highp,
which is available when GL_FRAGMENT_PRECISION_HIGH is defined.
--dscale= and --*scale-window= (i.e. an empty string) are respectively
valid settings for their options (and, in fact, the defaults).
This fixes the bug that it was impossible to reset e.g. tscale-window
back to the default "unset" setting after setting it once.
Credit goes to @CounterPillow for locating the cause of this bug.
For testing in VMs I guess?
This features a very broken hack that probably works. Though I didn't
test the packed format case. Again, the mismatch is essentially due to
big endian byte addresses decreasing as bit addresses increase, so you
can't represent a bit position in a byte stream with a single address,
which the mpv metadata does.
OSD is broken because repack.c doesn't support big endian. You'll have
to live with it.
The recent changes to the image format metadata broke big endian, and
that was intentional. Some things are inherent to little endian (like
the idea to coalesce bit and byte offsets into a single bit offset), and
they don't be fixed. But some obvious things can be fixed, such as
marking LE vs. BE formats the right way around on BE hosts.
The metadata is formally still in LE, except that if the LE/BE flag
matches the host endian, the host endian can be used when accessing
packed formats with bit shifts, or when computing byte aligned component
byte offsets. The former may work because formats with LE/BE variants
use the same bit offsets after byte swapping, the latter may work
because little endian is the natural concept for addressing memory. But
it will "subtly" fail to do the right thing in some cases, and code
using this can't know, so have fun.
Many things are broken, but this makes e.g. vo_gpu mostly work.
My general opinion about BE computers is that you should get a better
computer, you can get one for free from any garbage dump.
IMGFMT_RGB30 was added first; FFmpeg added AV_PIX_FMT_X2RGB10 later.
This is exactly the same, so treat them as such. For some reason,
libswscale still seems to output incompatible data - not sure what this
is about, but I'm not going to debug it.
Instead of applying this only to "regular" formats, do it with all
formats.
For some reason, some repackers still have their own endian code. These
could probably be removed, but whatever.
This change attempts to fix detection of how endian swapping is to be
performed. Details can be found in the code comments.
It should not change anything, other than fixing handling of the
X2RGB10BE ffmpeg pixel format. This format was detected incorrectly, and
the component location metadata was discarded due to an internal
consistency check. With this commit, it is handled correctly. At first I
thought the X2RGB10BE ffmpeg pixdesc metadata was wrong, but it appears
to be correct. The problem with this format is that it's the first
packed RGB format that requires bit shift to access, and where the
endian word size is larger than the (rounded up) component size, all
while pixdesc would "require" you to perform 3 memory accesses (instead
of 1), and the code tries to reverse this.
It appears that trying to use the pixdesc metadata is much, much more
work than just duplicating it in a saner form. To be fair, most problems
are with big endian, and the mpv internal format does not care much
about endian _hosts_.
For cases in which the requirements of the GPU API prevent directly
uploading a texture with a given stride, we need to fix the stride
manually in host memory. This incurs an extra memcpy, but there's not
much we can do about it. (Even in `ra_gl` land, the driver will just
hide this memcpy from the user)
Note: This code could be done better. It could only copy as many texels
as needed, and it could pick a stride that's a multiple of
`gpu->limits.align_tex_xfer_stride` for better performance. Patches
welcome (tm)
Fixes#7759
Implementation copy/pasted from:
https://code.videolan.org/videolan/libplacebo/-/commit/f793fc0480f
This brings mpv's tone mapping more in line with industry standard
practices, for a hopefully more consistent result across the board.
Note that we ignore the black point adjustment of the tone mapping
entirely. In theory we could revisit this, if we ever make black point
compensation part of the mpv rendering pipeline.
This standard says we should use a value of 203 nits instead of 100 for
mapping between SDR and HDR.
Code copied from https://code.videolan.org/videolan/libplacebo/-/commit/9d9164773
In particular, that commit also includes a test case to make sure the
implementation doesn't break roundtrips.
Relevant to #4248 and #7357.
glslang accepted this, perhaps erronneously, but mesa does not. It seems
to be incorrect. A caveat is that this means *all* SSBOs are now
coherent, but since we only use SSBOs for peak detection, that's a
non-issue. (And besides, marking something as coherent when we don't
perform any synchronization commands on it should be a no-op anyway)
Fixes#7823
since the title bar catches the mouse up and down events, the underlying
events view doesn't reset the isMoving state and no mouse movements are
signalled to the core. now we also reset the state in mouse up events
on the title bar.
Fixes#7807
The "screenshot window" command (ctrl+s by default) somehow broke video
colors with --gpu-api=vulkan --profile=gpu-hq when playback was paused.
I don't know the cause, but the rest of the code seems to imply
gl_video_reset_surfaces() needs to be called manually to flush some
caches, and it fixes the issue, so I assume there's no great mystery
here.
Commit 07b0c18 introduced some build breakages. Some breakages
were fixed on c1fc535 and a1adafe. This one is still remaining.
This commit fixes the following build error:
[153/521] Compiling video/out/vulkan/context_wayland.c
../video/out/vulkan/context_wayland.c:26:10: fatal error: video/out/wayland/presentation-time.h: No such file or directory
26 | #include "video/out/wayland/presentation-time.h"
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
compilation terminated.
Relevant to: #7802
Commit 6a13954d67 lowered the frequency of wakeups with this
condition. But it seems it sometimes the audio sync mode really should
get the wakeup before the frame is rendered. Normally, vo_thread is
supposed to perform this wakeup. Now the wakeup frequency is twice of
what should be needed - whatever, maybe it can be fixed properly once or
if timing is moved to the VO entirely in the future.
Fixes: #7777 (probably, untested)
When the window is mapped, some ICCCM WM_HINTS are set.
The input field is set to true and state is set to NormalState.
To quote the spec, "The input field is used to communicate to the window
manager the input focus model used by the client" and "[c]lients with
the Passive and Locally Active models should set the input flag to True".
mpv falls under the Passive Input model, since it expects keyboard input,
but only listens for key events on its single, top-level window instead
of subordinate windows (Locally Active) or the root window (Globally
Active).
From the end users prospective, all EWMH/ICCCM compliant WMs (especially
the minimalistic ones) will allow the user to focus mpv, which will allow
mpv to receive key events. If the input field is not set, WMs are
allowed to assume that mpv doesn't require focus.
I'm not sure why it only rendered OSD inside the video. Since OSD
rendering was always done on the X image (after software scaling and
color conversion), there was no technical reason for this. Maybe it was
because the code started out this way, and it was annoying to change it.
Possibly, one reason was that it didn't normally have to clear the black
bars in every frame (if video didn't cover the entire window).
Anyway, simply render OSD to the full window. This gets rid of some
rather weird stuff. It seems to look mostly like vo_wlshm now. The
uncovered regions are cleared every frame, which could probably be
avoided by being clever with the OSD renderer code, but this is where
I'm decidedly losing interest.
There was some mysterious code for aligning the image width to 8 pixels.
Replace that by attempting to align it to SIMD alignment (might matter
for libswscale, or if repack.c gets SIMD). Why are there apparently 4
different ways representing a pixel format (depth, VisualID, Visual,
XVisualInfo), but none of them seem to provide the XImage.bits_per_pixel
value (the actual size of a pixel, including padding)? Even after 33
years, X11 still seems overengineered, confusing, and inconvenient. So
just call X11 a heap of shit, and assume the worst case for alignment.
It may be completely useless, and I can't verify it as no known samples
or other known/accessible software using it, but why not?
Putting this together with he 422 code requires making it slightly more
generic. I'm still staying with a "huge" if tree instead of a table to
select the scanline worker callback, because it's actually small and not
huge (although it not being generic still feels slightly painful).
Now everything is super generic and super undebuggable. Some awkwardness
because the new metadata is basically a transposed version of the
mp_regular_imgfmt metadata, which was used for component info before.
The old code ignored many corner cases, and even wrote "blacker than
black" to YUV images. Use the new pixel format metadata and other
recently added gimmicky crap, which should make this more correct. Even
the almighty fuckup of a format AV_PIX_FMT_UYYVYY411 should work,
although that couldn't be tested for obvious reasons.
This doesn't work for "monow", but this is so extremely fringe while at
the same time painful that I just won't care. In theory, it could be
modeled as some sort of inverted gray colorspace or something.
Make sure to accept only native endian mpv formats. Previously, it
didn't check, and simply matched LE, because these are usually defined
before the BE formats.
red_mask etc. are defined as unsigned long, so use that instead of
hardcoding a 32 bit limit.
A repeat of the previous useless commits.
Pondered whether to use separate fields or just a flags integer for
color and component types; the latter won for now.
Functions like mp_imgfmt_get_component_type() are now discouraged, and
mp_imgfmt_desc.flags is back for defining all information. Some days ago
I felt like the opposite would be the better design. Fortunately, it
doesn't matter.
With this, I think all image format properties that mpv needs are
exhaustively defined all in one place.
Useless, but super generic! Actually may add support for other fringe
formats, however vo_x11 in itself is useless, so nothing won here. Also
I didn't bother with big endian support.
I guess I decided to stuff it all into mp_imgfmt_desc (the "old"
struct). This is probably a mistake. At first I was afraid that this
struct would get too fat (probably justified, and hereby happened), but
on the other hand mp_imgfmt_get_desc() (which builds the struct) calls
the former mp_imgfmt_get_layout(), and the separation doesn't make too
much sense anyway. Just merge them.
Still, try to keep out the extra info for packed YUV bullshit. I think
the result is OK, and there's as much information as there was before.
The test output changes a little. There's no independent bits[] array
anymore, so formats which did not previously have set this now show it.
(These formats are mpv-only and are still missing the metadata. To be
added later). Also, the output for the cursed packed formats changes.
Use the new pixfmt metadata to replace the format tables with weird
generic code.
As you can see, this removes the tables that essentially duplicate
information (which is baaaaaaaaaad), in exchange for code which is
probably more fragile and has less of a chance of being understood by
someone new to the code (which is probably even worse from a maintenance
and robustness point of view, but LALALA I CAN'T HEAR YOU).
There are some more formats which can be handled like this (RGB30 and
packed YUV I guess), maybe later.
My previous commit added support for this format, but it was still
broken, and prevented the allocation code from working. It's unknown
whether it's correct now (because this pixfmt is so obscure and useless,
there are no known samples around), but who cares.
Trying to use anything other than CLOCK_MONOTONIC here would be a
disaster. No idea if it's even possible for the clockid here to be
something other than CLOCK_MONOTONIC in this function but it's better
safe than sorry. Closes#7740.
I thought I'd probably want something like this, so the hardcoded stuff
in repack.c can be removed eventually. Of course this has no purpose at
all, and will not have any. (For now, this provides only metadata, and
nothing uses it, apart from the "test" that dumps it as text.)
This adds full support for AV_PIX_FMT_UYYVYY411 (probably out of spite,
because the format is 100% useless). Support for some mpv-only formats
is missing, ironically.
The code goes through _lengths_ to try to make sense out of the FFmpeg
AVPixFmtDescriptor data. Which is even more amazing that the new
metadata basically mirrors pixdesc, and just adds to it. Considering
code complexity and speed issues (it takes time to crunch through all
this shit all the time), and especially the fact that pixdesc is very
_incomplete_, it would probably better to have our own table to all
formats. But then we'd not scramble every time FFmpeg adds a new format,
which would be annoying. On the other hand, by using pixdesc, we get the
excitement to see whether this code will work, or break everything in
catastrophic ways.
The data structure still sucks a lot. Maybe I'll redo it again. The text
dump is weirdly differently formatted than the C struct - because I'm
not happy with the representation. Maybe I'll redo it all over again.
In summary: this commit does nothing.
Remove the vaguely defined plane_bits and component_bits fields from
struct mp_imgfmt_desc. Add weird replacements for existing uses. Remove
the bytes[] field, replace uses with bpp[].
Fix some potential alignment issues in existing code. As a compromise,
split mp_image_pixel_ptr() into 2 functions, because I think it's a bad
idea to implicitly round, but for some callers being slightly less
strict is convenient.
This shouldn't really change anything. In fact, it's a 100% useless
change. I'm just cleaning up what I started almost 8 years ago (see
commit 00653a3eb0). With this I've decided to keep mp_imgfmt_desc,
just removing the weird parts, and keeping the saner parts.
The image w and h members must match params.w and params.h, so
should not be changed directly. The helper function mp_image_set_size
is designed for this purpose, so just use that instead.
This prevents an assertion error with the rewritten draw_bmp.
Fixes#7721.
Added in libplacebo v60, unfortunately with some changes in design that
make it a bit of an awkward fit for the way timers are used in mpv.
Timer queries in libplacebo don't support "start" and "stop"-style
operations, and instead are attached directly to operations. The only
sane way of implementing this in the ra API is to have a single 'active
timer' that gets attached to every pass, taking care to sort distinct
operations into distinct pl_timer queries within that ra_timer.
This design unfortunately doesn't let us have multiple 'active timers'
concurrently, similar to the current such limitation in ra_gl. But it's
also not a big deal.
Render most of the OSD on the CPU, then draw it using a relatively
simple method. Do this for minimum code maintenance overhead. (While it
doesn't matter for vo_direct3d, and the effort spent here is probably
more than this would ever hope, I do hope to simplify the internal OSD
API for all these fringe VOs. Only vo_gpu should be allowed to do more
sophisticated things.)
If your GPU is shit (which it will be if you "want" to use vo_direct3d),
this might actually improve performance... is what I'd say, but out of
laziness a full screen sized texture gets uploaded on every OSD/subtitle
change, so maybe not.
This isn't useful anymore. We have a much better d3d11 renderer in
vo_gpu. D3D11 is available in all supported Windows versions. The
StretchRect path might still be useful for someone (???), and leaving it
at least evades conflict about users who want to keep using this VO for
inexplicable reasons. (Low power usage might be a justified reason, but
still, no.)
Also fuck the win32 platform, it's a heap of stinky shit. Microsoft is
some sort of psycho clown software company. Granted, maybe still better
than much of the rest of Silly Con Valley.
The original solution for #7017 was sort of a hack, but this hack is no
longer needed because c05e5d9d fixed the underlying issue causing this
error to be spammed in the first place. So just remove the "fix" that
apparently introduced about as many issueas it fixed.
Fixes#7719, hopefully.
This will probably make it slower. But since I don't care about
vo_vaapi, that's perfectly OK. It serves mostly as a test for the
previous commit. In addition, this code was pretty bad (custom broken
scaling and not-blending that probably broke in some situation). If that
wasn't enough, some vaapi drivers also provide only a single overlay at
a time, while this code required a bunch.
There also seems to be a Mesa bug: the overlay gets stretched when
src_x/y was not 0. Or maybe I misunderstood how this is supposed to
work. A bug is probably more likely? Nobody cares about this API.
They are sort of confusing, and they hide the fact that they have an
alpha component. Using the actual formats directly is no problem, sicne
these were useful only for big endian systems, something we can't test
anyway.
This was not intended to be committed in 0e3f893606. It disables
the extra wakeup if working==true. I've convinced myself that the wakeup
was really needed at the time, so no idea how I didn't notice this until
someone pointed it out on the commit diff on github (lol).
This resolves prefixes such as "~/" and "~~/" at the caller, instead of
relying on stream_read_file() to do it. One of the following commits
will remove this from stream_read_file() itself.
Untested.
The caller of render_frame() re-iterates without waiting if this
function returns true. That's normally meant for DS, where we draw
frames as fast as possible to let the driver perform waiting.
draw_bmp.c is the software blender for subtitles and OSD. It's used by
encoding mode (burning subtitles), and some VOs, like vo_drm, vo_x11,
vo_xv, and possibly more.
This changes the algorithm from upsampling the video to 4:4:4 and then
blending to downsampling the OSD and then blending directly to video.
This has far-reaching consequences for its internals, and results in an
effective rewrite.
Since I wanted to avoid un-premultiplying, all blending is done with
premultiplied alpha. That's actually the sane thing to do. The old code
just didn't do it, because it's very weird in YUV fixed point.
Essentially, you'd have to compensate for the chroma centering constant
by subtracting src_alpha/255*128. This seemed so hairy (especially with
correct rounding and high bit depths involved) that I went for using
float.
I think it turned out mostly OK, although it's more complex and less
maintainable than before. reinit() is certainly a bit too long. While it
should be possible to optimize the RGB path more (for example by
blending directly instead of doing the stupid float conversion), this is
probably slower. vo_xv users probably lose in this, because it takes the
slowest path (due to subsampling requirements and using YUV).
Why this rewrite? Nobody knows. I simply forgot the reason. But you'll
have it anyway. Whether or not this would have required a full rewrite,
at least it supports target alpha now (you can for example hard sub
transparent PNGs, if you ever wanted to use mpv for this).
Remove the check in vf_sub. The new draw_bmp.c is not as reliant on
libswscale anymore (mostly uses repack.c now), and osd.c shows an
error message on missing support instead now.
Formats with chroma subsampling of 4 are not supported, because FFmpeg
doesn't provide pixfmt definitions for alpha variants. We could provide
those ourselves (relatively trivial), but why bother.
This provides a way to convert YUV in fixed point (or pseudo-fixed
point; probably best to say "uint") to float, or rather coefficients
to perform such a conversion. Things like colorspace conversion is out
of scope, so this is simple and strictly per-component.
This is somewhat similar to mp_get_csp_mul(), but includes proper color
range expansion and correct chroma centering. The old function even
seems to have a bug, and assumes something about shifted range for full
range YCbCr, which is wrong.
vo_gpu should probably use the new function eventually.
Adding all these so I can use them for obscure processing purposes (see
later draw_bmp commit).
There isn't really a reason why they should exist. On the other hand,
they're just labels for formats that can be handled in a generic way,
and this commit adds support for them in the zimg wrapper and vo_gpu
just by making the formats exist. (Well, vo_gpu had to be fixed in the
previous commit.)
This was incorrect at least because the colorspace matrix attempted to
center chroma at (conceptually) 0.5, instead of 0. Also, it tried to
apply the fixed point shift logic for component sizes > 8 bit.
There is no float yuv format in mpv/ffmpeg yet, but see next commit,
which enables zimg to output it. I'm assuming zimg defines this format
such that luma is in range [0,1] and chroma in range [-0.5,0.5], with
the levels flag being ignored. This is consistent with H264/5 Annex E (I
think...), and it sort of seems to look right, so that's it.
Was broken with a zimg wrapper refucktor before the previous commit. In
addition, it seems this didn't match the vo_drm format, or the format
naming convention. So the order actually changes, and the format is
redefined. (The img_format.h comment was probably wrong.)
Change vo_gpu to the new format as well, so we can still test it.
For whatever purpose. If anything, this makes the zimg wrapper cleaner.
The added tests are not particular exhaustive, but nice to have. This
also makes the scale_zimg.c test pretty useless, because it only tests
repacking (going through the zimg wrapper). In theory, the repack_tests
things could also be used on scalers, but I guess it doesn't matter.
Some things are added over the previous zimg wrapper code. For example,
some fringe formats can now be expanded to 8 bit per component for
convenience.
Fallback to drmModeAddFB2 if drmModeAddFB2WithModifiers fails. I've
observed it failing on a pinebook pro running manjaro. We also got "0"
as modifiers from FFmpeg anyway, which might or might not have
something to do with this.
Instead of trying to find the source of the problem, just add this
fallback.
Untested (I don't have a platform that requires modifiers to work
here). Might break something, or might fix something. At least this
looks more intuitive to me.
The current implementation of presentation feedback was designed to be
used with flip model presentation. With the bitblt model,
GetFrameStatistics returns totally different values and it's not clear
if we can use them at all. Previously, this wasn't a problem because
with the bitblt model, GetFrameStatistics only worked in exclusive
fullscreen. Now that mpv supports exclusive fullscreen, we should
explicitly check for a flip model swapchain before using presentation
feedback.
This is really basic for planar image data access; not sure why there
weren't such helpers before.
They also handle trickier formats that use bit-packing, or they would be
mich simpler. (This affects only BGR4/BGR4/MONOW/MONOH, I hope whoever
invented them is proud of triggering so many special cases for so little
gain.)
These were still mapped to MP errors during probing, but they also get
triggered when instance creation fails due to lack of support for e.g.
wayland. Since waylandvk is probed above x11vk, we should probably
suppress these by default.
Closes#7626
For some reason this was never done? Looking through the code, it was
never the case that the frame cache was hit for still frames. I have no
idea why not. It makes a lot of sense to do so.
Notably, this massively improves the performance of updating the OSC
when viewing e.g. large still images, or while paused. (Tested on a
4000x8000 image, the OSC now responds smoothly to user input)
Only _writes_ are aligned, so the assumption doesn't work for reads. But
it's easy to fix by rounding down x0 to the next byte boundary. Writing
pixels outside of the read area is allowed, and we don't go out of
buffer bounds.
Patch by anon32, permission to do anything with it.
This is mostly for testing. It adds passing through the metadata through
the video chain. The metadata can be manipulated with vf_format. Support
for zimg alpha conversion (if built with zimg after it gained alpha
support) is implemented. Support premultiplied input in vo_gpu.
Some things still seem to be buggy.
Used to be used by vo_x11, and some other situations where software
conversion was employed. Haven't seen anyone complain about how software
brightness controls went away (originating from mplayer), so whatever,
it won't be needed again.
In the wayland code, the left mouse click is treated a bit differently.
Dragging the left click allows mpv to request a window move to the
compositor. In some cases, this can also request a window resize if the
osc-windowcontrols are enabled. These functions had the strange side
effect of messing up mpv's deadzone (it seemed to disappear completely).
A harmless enough workaround is to just explictly send an UP event for
left click after the move/resize functions are finished executing. The
xdg_toplevel move and resize functions both finish after the button
press is let go, so we are guarenteed to have the left click in the UP
state here. Sending this event probably unconfuses some calculation
somewhere thus fixing the deadzone bug. It feels a little silly, but
it's safe and works. Fixes#7651.
Move lookup GBRP or planar gray/alpha formats to separate functions in
some cases.
Make setup_regular_rgb_packer() not use a 4:4:4 YUV format to "pass"
RGB. This was used as a "trick" to avoid the stupid GBRP plane
permutation, but it confused severely, so get rid of it. Just do the
reordering, even if the zimg wrapper itself will reorder it back (which
is so stupid that I used the other approach at first). The comment
saying IMGFMT_420P was bogus of course; typically it was IMGFMT_444P.
This was inconsistent for unknown reason. monob was the way we wanted
it, and handling of monow was missing.
See the previous "img_format: add some mpv-only helper formats" commit.
Matters for the zimg wrapper.
mp_imgfmt_get_forced_csp() should be consistent with the MP_CSP_RGB/YUV
flags.
At least the different handling of the XYZ exception was a mess, even if
the result was the same.
Utterly useless, but the intention is to make dealing with corner case
pixel formats (forced upon us by FFmpeg, very rarely) less of a pain.
The zimg wrapper will use them. (It already supports these formats
automatically, but it will help with its internals.)
Y1 is considered RGB, even though gray formats are generally treated as
YUV for various reasons. mpv will default all YUV formats to limited
range internally, which makes no sense for a 1 bit format, so this is a
problem. I wanted to avoid that mp_image_params_guess_csp() (which
applies the default) explicitly checks for an image format, so although
a bit janky, this seems to be a good solution, especially because I
really don't give a shit about these formats, other than having to
handle them. It's notable that AV_PIX_FMT_MONOBLACK (also 1 bit gray,
just packed) already explicitly marked itself as RGB.
When I added mp_regular_imgfmt, I made the chroma subsampling use the
actual chroma division factor, instead of a shift (log2 of the actual
value). I had some ideas about how this was (probably?) more intuitive
and general. But nothing ever uses non-power of 2 subsampling (except
jpeg in rare cases apparently, because the world is a bad place).
Change the fields back to use shifts and rename them to avoid mistakes.
Make this slightly less ad-hoc. Also correct the missing alpha flag for
yap8/yap16.
Despite reduced redundancy, the LOC is going up anyway... whatever.
On FreeBSD and DragonFly kernel checks if `frsig` is valid and aborts
with `EINVAL` if not. However, `frsig` was never implemented.
$ build/mpv --gpu-context=drm /path/to/video.mkv
[...]
[vo/gpu] VT_SETMODE failed: Invalid argument
[vo/gpu/opengl] Failed to set up VT switcher. Terminal switching will be unavailable.
[...]
$ ./waf configure
Checking for vt.h : no
Checking for DRM : vt.h not found
[...]
../test.c:1:10: fatal error: 'sys/vt.h' file not found
#include <sys/vt.h>
^~~~~~~~~~
$ build/mpv --gpu-context=drm /path/to/video.mkv
Error parsing option gpu-context (option parameter could not be parsed)
Setting commandline option --gpu-context=drm failed.
Exiting... (Fatal error)
One not-so-nice hack in the wayland code is the assumption of when a
window is hidden (out of view from the compositor) and an arbitrary
delay for enabling/disabling the usage of presentation time. Since you
do not receive any presentation feedback when a window is hidden on
wayland (a feature or misfeature depending on who you ask), the ust is
updated based on the refresh_nsec statistic gathered from the previous
feedback event.
The flaw with this is that refresh_nsec basically just reports back the
display's refresh rate (1 / refresh_rate * 10^9). It doesn't tell you
how long the vsync interval really was. So as a video is left playing
out of view, the wl->last_queue_display_time becomes increasingly
inaccurate. This led to a vsync spike when bringing the mpv window back
into sight after it was hidden for a period of time. The hack for
working around this is to just wait a while before enabling presentation
time again. The discrepancy between the "bogus"
wl->last_queue_display_time and the actual value you get from the
feedback only happens initially after a switch. If you just discard
those values, you avoid the dramatic vsync spike.
It turns out that there's a smarter way to do this. Just use mp_time_us
deltas. The whole reason for these hacks is because
wl->last_queue_display_time wasn't close enough to how long it would
take for a frame to actually display if it wasn't hidden. Instead, mpv's
internal timer can be used, and the difference between wayland_sync_swap
calls is a close enough proxy for the vsync interval (certainly better
than using the monitor's refresh rate). This avoids the entire conundrum
of massive vsync spikes when bringing the player back into view, and it
means we can get rid of extra crap like wl->hidden.
In theory this mostly happens automatically, especially after the 5
vsync limit disables this already. But if we uninit before 5 vsyncs are
rendered, this can get left in a dangling 'enabled' state, which leaks a
debug report callback.
Always explicitly disable it just to be on the safe side.
I must have messed this up when I actually added the Y210 format
(because that one is correct). So my comment in the commit adding this
about the FFmpeg pixfmt doxygen being wrong was wrong.
I'd like to use this opportunity to complain once more about the
existence of these terrible pixel formats.
Again worthless, slow, and only for libswscale parity.
With this, we support all formats libswscale supports, except bayer
input, and rgb4/bgr4 output. We even support some formats libswscale
doesn't.
It's possible that the zimg wrapper isn't always as fast as libswscale.
But there is optimization potential: the inner repack loops are
self-contained enough that they could be reasonably be implemented in
assembler (probably), and doing everything slice-wise should reduce the
overhead of the separate pack/unpack stages.
Just lazily tested.
The comment on AV_PIX_FMT_Y210LE seems to be wrong. It claims it's "like
YUYV422", bit it seems more like YVYU422, at last the way libswscale
input treats it. Maybe Intel pays its developers too much?
The repacker inner lop is probably rather inefficient. In theory we
could optimize it by reading the packed pixels as words, doing the
component reshuffling using compile time values etc., but I'd rather
keep the code size small. It's already bad enough that we have to
support 16 bit per component variants, just because this one Intel guy
couldn't keep it in his pants. In general, I can't be bothered to spend
time on optimizing it; I'm only doing this for fun (i.e. masochistic
obligation).
This covers 8 and 16 bit packed RGB formats. It doesn't really help with
any actual use-cases, other than giving the finger to libswscale.
One problem is with different color depths. For example, rgb565 provides
1 bit more resolution to the green channel. zimg can only dither to a
uniform depth. I tried dithering to the highest depth and shifting away
1 bit for the lower channels, but that looked ugly (or I messed up
somewhere), so instead it dithers to the lowest depth, and adjusts the
value range if needed. Testing with bgr4_byte (extreme case with 1/2/1
depths), it looks more "grainy" (ordered dithering artifacts) than
libswscale, but it also looks cleaner and smoother. It doesn't have
libswscale's weird red-shift. So I call it a success.
Big endian formats need to be handled explicitly; the generic big endian
swapper code assumes byte-aligned components.
Unpacking is done with shifts and 3 LUTs. This is symmetric to the
packer. Using a generated palette might be better, but I preferred to
keep the symmetry, and not having to mess with a generated palette and
the pal8 code.
This uses FFmepg pixfmts constants directly. I would have preferred
keeping zimg completely separate. But neither do I want to add an IMGFMT
alias for every of these formats, nor do I want to extend our imgfmt
code such that it can provide a complete description of each packed RGB
format (similar to FFmpeg pixdesc).
It also appears that FFmpeg pixdesc as well as the FFmpeg pixfmt doxygen
have an error regarding RGB8: the R/B bit depths are swapped. libswscale
appears to be handling them differently. Not completely sure, as this is
the only packed format case with R/B havuing different depths (instead
of G, the middle component, where things are symmetric).
One of the extremely annoying dumb things in ffmpeg is that most pixel
formats are available as little endian and big endian variants. (The
sane way would be having native endian formats only.) Usually, most of
the real codecs use native formats only, while non-native formats are
used by fringe raw codecs only. But the PNG encoders and decoders
unfortunately use big endian formats, and since PNG it such a popular
format, this causes problems for us. In particular, the current zimg
wrapper will refuse to work (and mpv will fall back to sws) when writing
non-8 bit PNGs.
So add non-native endian support to zimg. This is done in a fairly
"generic" way (which means lots of potential for bugs). If input is a
"regular" format (and just byte-swapped), the rest happens
automatically, which happens to cover all interesting formats.
Some things could be more efficient; for example, unpacking is done on
the data before it's passed to the unpacker. You could make endian
swapping part of the actual unpacking process, which might be slightly
faster. You could avoid copying twice in some cases (such as when
there's no actual repacker, or if alignment needs to be corrected). But
I don't really care. It's reasonably fast for the normal case.
Not entirely sure whether this is correct. Some (but not many) formats
are covered by the tests, some I tested manually. Some I can't even
test, because libswscale doesn't support them (like nv20*).
This sucks, but is helpful for testing.
Obviously, it would be much nicer if there were a way to specify _all_
scaler options per filter (if the user wanted), instead of always using
the global options. But this is "too hard" for now. For testing, it is
extremely convenient to select the scaler backend, so add this option,
but make clear that it could go away. We'd delete it once there is a
better mechanism for this.
In display-sync mode, the core doesn't need to woken up every vsync, but
only every time a new actual video frame needs to be queued. So don't
wake up if there are still frames to repeat.
In audio-sync mode, the wakeup is simply redundant, since there's a
separate timer (in->wakeup_pts) to control when to queue a new frame. I
think.
This finally brings the required playloop iterations down to almost the
number of video frames. (As originally intended, really.)
Also a fairly risky change.
The wakeup at the end of VO frame rendering seems redundant, because
after rendering almost no state changes. The player core can queue a new
frame once frame rendering begins, and there's a separate wakeup for
this. The only thing that actually changes is in->rendering. The only
thing that seems to depend on it and can trigger a wakeup is the
vo_still_displaying() function. Change it so that it needs an explicit
call to a new API function, so we can avoid wakeups in the common case.
The vo_still_displaying() code is mostly just moved around due to
locking and for avoiding forward declarations.
Also a somewhat risky change (tasty new bugs).
This was optional, with the intention that normally such options require
a valid format. But there is no reason for this (at least not anymore),
and it's actually more logical to accept "no" in all situations this
option type is used. This also gets rid of the weird min field special
use.
Add an infrastructure for collecting performance-related data, use it in
some places. Add rendering of them to stats.lua.
There were two main goals: minimal impact on the normal code and normal
playback. So all these stats_* function calls either happen only during
initialization, or return immediately if no stats collection is going
on. That's why it does this lazily adding of stats entries etc. (a first
iteration made each stats entry an API thing, instead of just a single
stats_ctx, but I thought that was getting too intrusive in the "normal"
code, even if everything gets worse inside of stats.c).
You could get most of this information from various profilers (including
the extremely primitive --dump-stats thing in mpv), but this makes it
easier to see the most important information at once (at least in
theory), partially because we know best about the context of various
things.
Not very happy with this. It's all pretty primitive and dumb. At this
point I just wanted to get over with it, without necessarily having to
revisit it later, but with having my stupid statistics.
Somehow the code feels terrible. There are a lot of meh decisions in
there that could be better or worse (but mostly could be better), and it
just sucks but it's also trivial and uninteresting and does the job. I
guess I hate programming. It's so tedious and the result is always shit.
Anyway, enjoy.
Until now, it used only coordinates clipped to the screen for this,
which meant no negative margins were ever reported to libass. This broke
proper rendering of explicitly positioned ASS events (libass simply
could not know the real video size in this case.)
Fix this by reporting margins even if they're negative. This makes it
apparently work correctly with vo_gpu at least.
Note that I'm not really sure if anything in the rendering chain
required non-negative margins. If so, and that code implicitly assumed
it, I suppose crashes and such are possible.
The "rule" is that a fallback warning message should be shown only shown
if software decoding was used before, or in other words when either
hwdec was enabled before, but the stream suddenly falls back, or it was
attempted to enable it at runtime, and it didn't work.
The message wasn't printed the first time in the latter case, because
hwdec_notified was not set in forced software decoding mode. Fix it with
this commit. Fortunately, the logic becomes simpler.
Change all OPT_* macros such that they don't define the entire m_option
initializer, and instead expand only to a part of it, which sets certain
fields. This requires changing almost every option declaration, because
they all use these macros. A declaration now always starts with
{"name", ...
followed by designated initializers only (possibly wrapped in macros).
The OPT_* macros now initialize the .offset and .type fields only,
sometimes also .priv and others.
I think this change makes the option macros less tricky. The old code
had to stuff everything into macro arguments (and attempted to allow
setting arbitrary fields by letting the user pass designated
initializers in the vararg parts). Some of this was made messy due to
C99 and C11 not allowing 0-sized varargs with ',' removal. It's also
possible that this change is pointless, other than cosmetic preferences.
Not too happy about some things. For example, the OPT_CHOICE()
indentation I applied looks a bit ugly.
Much of this change was done with regex search&replace, but some places
required manual editing. In particular, code in "obscure" areas (which I
didn't include in compilation) might be broken now.
In wayland_common.c the author of some option declarations confused the
flags parameter with the default value (though the default value was
also properly set below). I fixed this with this change.
Previously, the vo wasn't always informed if something about the output
changed during playback. For instance, changing a display's refresh rate
during playback would not update mpv's display fps. Fix this by simply
using VO_EVENT_WIN_STATE in output_handle_done which executes whenever
something about the output is changed.
Allow the --window-maximized and --window-minimized flags to actually
work when the player is started. since macOS doesn't like using both at
the same time the minimized state takes precedence over the maximized
state.
Before this commit, option declarations used M_OPT_MIN/M_OPT_MAX (and
some other identifiers based on these) to signal whether an option had
min/max values. Remove these flags, and make it use a range implicitly
on the condition if min<max is true.
This requires care in all cases when only M_OPT_MIN or M_OPT_MAX were
set (instead of both). Generally, the commit replaces all these
instances with using DBL_MAX/DBL_MIN for the "unset" part of the range.
This also happens to fix some cases where you could pass over-large
values to integer options, which were silently truncated, but now cause
an error.
This commit has some higher potential for regressions.
This was mostly unused, and has certain problems. Just get rid of it.
It was still used in CDDA (--cdda-span) and a debug option for OpenGL
(--opengl-check-pattern). Replace both of these with 2 options, where
each sets the start/end values of the former span. Both were
undocumented somehow (normally we require all options to be documented),
so I'm not caring about compatibility, and not bothering to add it to
the API changelog.
We have this cap now thanks to e2976e662, but we don't actually make
sure our FBOs are storable before we blindly attempt using them with
compute shaders.
There's no more need to unconditionally set `storage_dst = true` as long
as we make sure to include an extra condition on the `fbo_format`
selection to prevent users from accidentally enabling
compute-shader-only features with non-storable FBOs, alongside some
other miscellaneous adjustments to eliminate instances of "assumed
storability" from vo_gpu.
This simply makes the "is the destination FBO format bad?" check a tiny
bit less awful, by making sure we prefer storable FBO formats over
non-storable FBO formats. I'd love to make this also conditional on
whether or not we actually *need* a storable FBO format, but that logic
is decided later, in `pass_draw_to_screen`, and I don't want to
replicate the logic.
Fixes#7017.
Previously if the --fs-screen option was set, it would only use the
screen if mpv was launched with --fs and only on startup. During
runtime, the toggle would ignore it. Rework the logic here so that mpv's
fullscreen always uses --fs-screen if it is set. Additionally, cleanup
some unneeded cruft in vo_wayland_reconfig and make find_output more
useful.
This commit fixes a bug where handle for a framebuffer gets double
freed.
It seems to happen that the same prime fd gets two framebuffers.
As the prime fd is the same the resulting prime handle is also the
same.
This means one handle but 2 framebuffers and can lead to the following
chain:
1. The first framebuffer gets deleted the handle gets also freed via
the ioctl.
2. In startup phase not all 4 dumb buffers for overlay drawing
are set up. It can happen that the last dumb buffer gets the
handle we freed above.
3. The second framebuffer gets freed and the handle will be
freed again resulting that the 4's dumb buffer handle is not
backed by a buffer.
4. Drm prime continues to assign handles to its prime fds an
will lead to have this handle which was just freed to
reassign again but to an prime buffer.
5.Now the overlay should be drawn into dumb buffer 4 which
still has the same handle but is backed by the wrong buffer.
This leads to two different behaviors:
- MPV crashes as the drm prime buffers size als calculated
by the decoder output format. The overlay output format
differs and it takes more space. SO the size check
in kernel fails.
- MPV is continuing play. This happens when the decoders
allocates a bigger buffer than needed for the overlay.
For example overlay is Full HD and decoder output is 4k.
This leads to the behavior das the overlay wil be drawn
into the wrong buffer as its a drm prime buffer and results
in a flicker every fourth step.
the actual character that made mpv crash is IDEOGRAPHIC COMMA
(U+3001, UTF-8: E3 80 81, 、) and that only in some specific
circumstances that could be reliably reproduced on my end.
using an NSString instead of the Swift String actually fixes that issues
even though they should technically do the exact same thing. i tested
all the other String initialisers, but they all had had the same issue.
this is kinda only a workaround till i can find a different way of
handling it.
This was changed 6 years ago (444e583b6) and seemed to work fine. But it
does seem to cause issues with IceWM sometimes, while with StaticGravity
the problem is gone. Comparing both gravity values, reading the confused
source code comment, and reading the referenced commit message, I can't
determine what it even does, I just remove it.
Reproduction:
- start mpv in windowed mode, with 2 videos of different size
- switch to second video
- switch window with alt+tab
- switch back to mpv with alt+tab
- window moves to X=0
There's probably a better way to fix this. Please send a patch.
Libav seems rather dead: no release for 2 years, no new git commits in
master for almost a year (with one exception ~6 months ago). From what I
can tell, some developers resigned themselves to the horrifying idea to
post patches to ffmpeg-devel instead, while the rest of the developers
went on to greener pastures.
Libav was a better project than FFmpeg. Unfortunately, FFmpeg won,
because it managed to keep the name and website. Libav was pushed more
and more into obscurity: while there was initially a big push for Libav,
FFmpeg just remained "in place" and visible for most people. FFmpeg was
slowly draining all manpower and energy from Libav. A big part of this
was that FFmpeg stole code from Libav (regular merges of the entire
Libav git tree), making it some sort of Frankenstein mirror of Libav,
think decaying zombie with additional legs ("features") nailed to it.
"Stealing" surely is the wrong word; I'm just aping the language that
some of the FFmpeg members used to use. All that is in the past now, I'm
probably the only person left who is annoyed by this, and with this
commit I'm putting this decade long problem finally to an end. I just
thought I'd express my annoyance about this fucking shitshow one last
time.
The most intrusive change in this commit is the resample filter, which
originally used libavresample. Since the FFmpeg developer refused to
enable libavresample by default for drama reasons, and the API was
slightly different, so the filter used some big preprocessor mess to
make it compatible to libswresample. All that falls away now. The
simplification to the build system is also significant.
Resizing the window while preserving the aspect ratio actually kind of
sucked. The window size could make big dramatic changes which was pretty
unintuitive with respect to where the mouse was actually located.
Instead, let's just do some math to ensure that the window size is
always contained inside the width/height reported by
handle_toplevel_config while preserving the aspect ratio. Fixes#7426.
Fixes#7441. Just set screenrc to be equal to current_output's geometry.
Also remove some pointless/extra variables and print a warning/fallback
to screen 0 if a bad id is passed to --fs-screen.
Obviously, we don't want to lose fractions, and the zimg active_region
fields in fact have the type double. The integer division was wrong.
Also, always set active_region.width/height. It appears zimg behavior
does not change if they're set to the normal integer values, so the
extra check to not set them in this case was worthless.