yukijoou/kwin
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Allow direct uploads of 16-bpc QImage formats

Browse source 
This commit expands the QImage-to-GL formatTable to include entries
for RGBX64, RGBA64_Premultiplied, and Grayscale16 image formats.
Uploading 16-bit-per-channel formats is supported by OpenGL and, with
the GL_EXT_texture_norm16 extension, GLES.
This commit is contained in:
Manuel Stoeckl 2022-01-30 09:52:28 -05:00 committed by David Edmundson
parent e00d836708
commit a2b7c09a11
2 changed files with 13 additions and 2 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

14
src/libkwineffects/kwingltexture.cpp
View file

@ -36,6 +36,7 @@ bool GLTexturePrivate::s_supportsUnpack = false;
bool GLTexturePrivate::s_supportsTextureStorage = false;
bool GLTexturePrivate::s_supportsTextureSwizzle = false;
bool GLTexturePrivate::s_supportsTextureFormatRG = false;
bool GLTexturePrivate::s_supportsTexture16Bit = false;
uint GLTexturePrivate::s_fbo = 0;
// Table of GL formats/types associated with different values of QImage::Format.
@ -73,6 +74,11 @@ struct {
{ GL_RGB10_A2, GL_BGRA, GL_UNSIGNED_INT_2_10_10_10_REV }, // QImage::Format_A2RGB30_Premultiplied
{ GL_R8, GL_RED, GL_UNSIGNED_BYTE }, // QImage::Format_Alpha8
{ GL_R8, GL_RED, GL_UNSIGNED_BYTE }, // QImage::Format_Grayscale8
{ GL_RGBA16, GL_RGBA, GL_UNSIGNED_SHORT }, // QImage::Format_RGBX64
{ 0, 0, 0 }, // QImage::Format_RGBA64
{ GL_RGBA16, GL_RGBA, GL_UNSIGNED_SHORT }, // QImage::Format_RGBA64_Premultiplied
{ GL_R16, GL_RED, GL_UNSIGNED_SHORT }, // QImage::Format_Grayscale16
{ 0, 0, 0 }, // QImage::Format_BGR888
};
GLTexture::GLTexture(GLenum target)
@ -129,7 +135,8 @@ GLTexture::GLTexture(const QImage& image, GLenum target)
const QImage::Format index = image.format();
if (index < sizeof(formatTable) / sizeof(formatTable[0]) && formatTable[index].internalFormat) {
if (index < sizeof(formatTable) / sizeof(formatTable[0]) && formatTable[index].internalFormat
&& !(formatTable[index].type == GL_UNSIGNED_SHORT && !d->s_supportsTexture16Bit)) {
internalFormat = formatTable[index].internalFormat;
format = formatTable[index].format;
type = formatTable[index].type;
@ -305,6 +312,7 @@ void GLTexturePrivate::initStatic()
s_supportsTextureSwizzle = hasGLVersion(3, 3) || hasGLExtension(QByteArrayLiteral("GL_ARB_texture_swizzle"));
// see https://www.opengl.org/registry/specs/ARB/texture_rg.txt
s_supportsTextureFormatRG = hasGLVersion(3, 0) || hasGLExtension(QByteArrayLiteral("GL_ARB_texture_rg"));
s_supportsTexture16Bit = true;
s_supportsARGB32 = true;
s_supportsUnpack = true;
} else {
@ -313,6 +321,7 @@ void GLTexturePrivate::initStatic()
s_supportsTextureSwizzle = hasGLVersion(3, 0);
// see https://www.khronos.org/registry/gles/extensions/EXT/EXT_texture_rg.txt
s_supportsTextureFormatRG = hasGLVersion(3, 0) || hasGLExtension(QByteArrayLiteral("GL_EXT_texture_rg"));
s_supportsTexture16Bit = hasGLExtension(QByteArrayLiteral("GL_EXT_texture_norm16"));
// QImage::Format_ARGB32_Premultiplied is a packed-pixel format, so it's only
// equivalent to GL_BGRA/GL_UNSIGNED_BYTE on little-endian systems.
@ -369,7 +378,8 @@ void GLTexture::update(const QImage &image, const QPoint &offset, const QRect &s
if (!GLPlatform::instance()->isGLES()) {
const QImage::Format index = image.format();
if (index < sizeof(formatTable) / sizeof(formatTable[0]) && formatTable[index].internalFormat) {
if (index < sizeof(formatTable) / sizeof(formatTable[0]) && formatTable[index].internalFormat
&& !(formatTable[index].type == GL_UNSIGNED_SHORT && !d->s_supportsTexture16Bit)) {
glFormat = formatTable[index].format;
type = formatTable[index].type;
uploadFormat = index;

1
src/libkwineffects/kwingltexture_p.h
View file

@ -68,6 +68,7 @@ public:
static bool s_supportsTextureStorage;
static bool s_supportsTextureSwizzle;
static bool s_supportsTextureFormatRG;
static bool s_supportsTexture16Bit;
static GLuint s_fbo;
private:
friend void KWin::cleanupGL();