This was accidentally included once per sample, as a result of the
anti-ringing code.

See-Also: https://github.com/haasn/libplacebo/issues/217

$id@c can collide with another identifier that has larger $id, add extra
_ to avoid collision.

Fixes: https://github.com/haasn/libplacebo/issues/217

pl_icc_update needs to be called every frame, so if libplacebo is
compiled without lcms support, this leads to spam. There's no need to
warn more than once so just keep track of it in a bool.

These should have gone through a deprecation period, since they are
a public API.

Fixes: dcd5b75b
Fixes: vlc#28417

At least for mpv, if we have a playlist of images, fail opening the ICC
of one image, go forward and then back to the image with the ICC
failure, it'll hit this assertion failure. Not so nice. If there's no
ICC information, it just does nothing so display the image as usual.

By default, allows for correcting drift/jitter of up to 1 ms per frame,
in line with expected jitter from e.g. Matroska sources. This is enough
to cover the mismatch between 23.976 Hz and 24.000 Hz, but not the
mismatch between 24 and 25 Hz (which is just over 1ms per frame).

There are two use cases in which this code will be useful:

1. Displaying 23.976 Hz video files on e.g. 60.000 Hz monitor (or any
   other near-miss), in display-timed mode. Given the usual 3:2
   cadence, after five vsyncs (three source frames), we will have:

   Vsync (wallclock) PTS = 83.33333333333334
   Frame PTS = 83.41675008341676

   This is a difference of 0.1ms, and will therefore be fixed by this
   code, adjusting all future wallclock PTS up by the cumulative drift
   (0.1ms).

2. Displaying any video with some amount of jitter on the PTS
   measurements. This avoids some situation where we have frames with
   weights 0.999 and 0.001 for example. Normally this should get rounded
   away by the frame mixing code anyway, but it's better to have an
   exact match.

Kacper Michajłow authored 1 year ago and Niklas Haas committed 1 year ago

This will never be used uninitialized, guarded by use_dbghelp, but GCC
analysis is too shallow and fails to see that.

What does it even mean? You can share a vulkan device just fine..

Testing an element's truth value will raise an exception in future
versions.

Prevent log spam.

Sanity. Don't advertise support for SSBOs/UBOs/etc. if basic buffer
support is not available.

Avoids triggering assertions added by 3bd1c970.

This is not a very elegant solution, by the way. Actually, it would be
better to query the maximum buffer size specifically for use as UBOs
etc, e.g. by allocating a test buffer and querying the driver for its
allowed memory types, then looking at only those memory types.

Makes this actually check the relevant requirements.

This is an insane case to work around, because no GPU backend enables
emulated format without the corresponding buffer support. I'd rather
pl_assert() it than return it as an error to the caller.

(Note that we already hit the slice_w assertion in this case)

OpenGL has emulated texture formats that don't need texel storage buffer
emulation, so support this case also.

Note that this change does not actually do anything yet, but I plan on
re-using this helper in the (near) future.

We only need to use the max SSBO sizee for emulated transfers.
Otherwise, max_buf_size is sufficient.

This helper was failing to account for the pixel stride alignment
requirements.

This accidentally sharpened the frame mixer when the source fps was
below monitor fps.

This will inflate all numbers, but is also more informative.

By one that doesn't have such strict requirements on the texture size.

Unceremoniously stolen from multigpu-bench.

Turn these defines into enums and group them more appropriately, and
make the queue configuration configurable (for testing).

Instead of linearly desaturating, use a third-order polynomial
approximation.

Re-use an existing helper and remove a misleading comment.

We can early-dispatch this callback as soon as we're done mempcying the
data into a temporary buffer.

Fixes: https://github.com/mpv-player/mpv/issues/12512

Kacper Michajłow authored 1 year ago and Niklas Haas committed 1 year ago

Fixes: https://github.com/haasn/libplacebo/issues/207

Fortunately, the existence of fast scalers masked this bug to some
extent, except in mpv, where the fast gaussian shader was not yet mapped.

Fixes: 364612f7
Fixes: https://github.com/haasn/libplacebo/issues/209

`pt` is already a vec2.

Regression caused by the #pragma GLSL rewrite, where
sh_const_float/SH_FLOAT was unintentionally replaced by ${const float},
despite this being the equivalent of a hard-coded float (%f).

Take care of some minor cosmetic fixes while we're at it.

Fixes: 94722c55

This was accidentally inverted...