Vincenzo Nicosia authored 5 years ago and Thomas Guillem committed 5 years ago

It looks like by reverting and reapplying the gopher patch we have actually
scrapped the two lines in modules/access/Makefile.am.

Signed-off-by: Thomas Guillem <thomas@gllm.fr>

When using 0xXX notation the issue we also have to adjust the width.

Fixes compilation with gcc 9 for Raspberry:

compile:  arm-linux-gnueabihf-g++ -DHAVE_CONFIG_H -Icontrib/arm-linux-gnueabihf/include -Icontrib/arm-linux-gnueabihf/include -g -O2 -DHAVE_UNIX -I../../include -I../../include/sidplay -c xsid.cpp -fPIC -DPIC -o xsid.o
xsid.cpp:101:1: error: narrowing conversion of ‘'\200'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
  101 | };
      | ^
xsid.cpp:101:1: error: narrowing conversion of ‘'\224'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
xsid.cpp:101:1: error: narrowing conversion of ‘'\251'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
xsid.cpp:101:1: error: narrowing conversion of ‘'\274'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
xsid.cpp:101:1: error: narrowing conversion of ‘'\316'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
xsid.cpp:101:1: error: narrowing conversion of ‘'\341'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
xsid.cpp:101:1: error: narrowing conversion of ‘'\362'’ from ‘char’ to ‘int8_t’ {aka ‘signed char’} [-Wnarrowing]
make[4]: *** [Makefile:409: xsid.lo] Error 1

Versions above 0.51.1 don't add the -lpthread properly in libplacebo.pc.

There are 3 choices between Hardware Video Scaler (default, best), ISP and
resizer (slower but uses less memory).

Corresponds to the preexisting enum filter_resizer_t.

Creating data structures can fail for many reasons. Handle this scenario gracefully.

When starting and quitting VLC very quickly, we might still try to create media library data structures while the interface thread already ended, which was leading to a segfault when attempting to access the library with an invalid interface thread pointer.

Biswapriyo Nath authored 5 years ago and Steve Lhomme committed 5 years ago

* The deleted patch was added in mingw-w64 v7.0.0

Signed-off-by: Biswapriyo Nath <nathbappai@gmail.com>
Signed-off-by: Steve Lhomme <robux4@ycbcr.xyz>

allows resuming playback with start-time

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

Use the global conversion matrix to swap U/V components at the same
time.

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

The conversion matrix combines several transformations, it does not
contain "YUV coefficients".

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

Directly initialize the conversion matrix in place in column-major
order.

Note: We could not just pass GL_TRUE to the transpose parameter of
glUniformMatrix4fv, because it is not supported on OpenGL ES 2:

> GL_INVALID_VALUE is generated if transpose is not GL_FALSE.

<https://www.khronos.org/registry/OpenGL-Refpages/es2.0/xhtml/glUniform.xml>

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

Apply the range conversion via the conversion matrix.

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

Construct chroma conversion matrices programmatically, from the luma
weight of the R, G and B components.

This allows to combine (multiply) matrices and paves the way to properly
support full color range (for now, the range is hardcoded to
COLOR_RANGE_LIMITED to match the previous implementation).

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

Use native matrix multiplication to apply the chroma conversion.

Concretely, the generated fragment shader looked like:

    uniform vec4 Coefficients[4];
    uniform vec4 FillColor;
    void main(void) {
     float val;vec4 colors;
     colors = texture2D(Texture0, TexCoord0);
     val = colors.r;
     vec4 color0 = vec4(val, val, val, 1);
     colors = texture2D(Texture1, TexCoord1);
     val = colors.r;
     vec4 color1 = vec4(val, val, val, 1);
     val = colors.g;
     vec4 color2 = vec4(val, val, val, 1);
     vec4 result = (color0 * Coefficients[0]) + Coefficients[3];
     result = (color1 * Coefficients[1]) + result;
     result = (color2 * Coefficients[2]) + result;
     result = _main_0_0(result);
     gl_FragColor = result * FillColor;
    }

Now, it looks like:

    uniform mat4 ConvMatrix;
    uniform vec4 FillColor;
    void main(void) {
     vec4 texel;
     vec4 pixel = vec4(0.0, 0.0, 0.0, 1.0);
     texel = texture2D(Texture0, TexCoord0);
     pixel[0] = texel.r;
     texel = texture2D(Texture1, TexCoord1);
     pixel[1] = texel.r;
     pixel[2] = texel.g;
     vec4 result = ConvMatrix * pixel;
     result = _main_0_0(result);
     gl_FragColor = result * FillColor;
    }

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

In case the texture format is GL_LUMINANCE, swizzle_per_tex was set to
{NULL, "xa"}. Using NULL instead of an explicit swizzle ("x") was a
small optimization to assign the texel value directly:

    vec4 color0 = texture2D(Texture0, TexCoord0);

instead of using the general code generation:

    colors = texture2D(Texture0, TexCoord0);
    val = colors.x;
    vec4 color0 = vec4(val, val, val, 1);

This was possible because the texture exposes the luminance value in
its three components: (L, L, L, 1).

But this special case will disappear when we will rewrite the chroma
conversion using matrix multiplication directly (instead of computing
its steps separately).

This simplifies the logic of the code, and paves the way to refactor
chroma conversion more easily.

List the supported codecs.

It's now reported via the viewpoint_moved callback.

Farid Hammane authored 5 years ago and Thomas Guillem committed 5 years ago

amem now supports three formats: S16N, S32N and FL32.

Signed-off-by: Farid Hammane <farid.hammane@gmail.com>
Signed-off-by: Thomas Guillem <thomas@gllm.fr>

Avoid double rotation from MediaCodec and interop_android.

This fixes 90° videos that are flipped horizontally with the recent vout_helper
changes.

No functional changes, will be used by the next commit.

Romain Vimont authored 5 years ago and Thomas Guillem committed 5 years ago

The orientation matrix must be applied on the input coordinates, before
the (internal) Android transformation, which depends on how the texture
is actually stored.

For example, if the orientation matrix represents a vertical flip,
"vertical" is only meaningful on the input coordinates (if the Android
transformation rotates by 90°, the vertical flip would result in an
horizontal flip instead).

The first version was actually correct:
<https://mailman.videolan.org/pipermail/vlc-devel/2019-December/130153.html

>

Signed-off-by: Thomas Guillem <thomas@gllm.fr>

John Cox authored 5 years ago and Steve Lhomme committed 5 years ago

Signed-off-by: Steve Lhomme <robux4@ycbcr.xyz>

There can be 3 frames (instead of the default 2) in the decoder memory at once.

ref #23563

Biswapriyo Nath authored 5 years ago and Steve Lhomme committed 5 years ago

* The dxgidebug.patch and processor_format.patch were added in upstream.
* This patch removes those and update wine tarball to version 5.0.

Signed-off-by: Biswapriyo Nath <nathbappai@gmail.com>
Signed-off-by: Steve Lhomme <robux4@ycbcr.xyz>

Instead of caching by library ID, cache by MRL as it may not be unique across multiple library items, for instance regarding album tracks.