GitHub-Proxy/PrusaSlicer
1
0
Fork 0
mirror of https://git.mirrors.martin98.com/https://github.com/prusa3d/PrusaSlicer.git synced 2025-08-20 12:39:12 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity

Refactoring in GLTexture

Browse Source
This commit is contained in:
enricoturri1966 2021-11-16 10:23:27 +01:00
parent 107f1baa32
commit 67ddcb5c83
5 changed files with 154 additions and 225 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
src/slic3r/GUI/3DBed.cpp
View File

@ -485,7 +485,7 @@ void Bed3D::render_texture(bool bottom, GLCanvas3D& canvas) const
else if (boost::algorithm::iends_with(m_texture_filename, ".png")) {
// generate a temporary lower resolution texture to show while no main texture levels have been compressed
if (temp_texture->get_id() == 0 || temp_texture->get_source() != m_texture_filename) {
if (!temp_texture->load_from_file(m_texture_filename, false, GLTexture::None, false)) {
if (!temp_texture->load_from_file(m_texture_filename, false, GLTexture::ECompressionType::None, false)) {
render_default(bottom, false);
return;
}
@ -493,7 +493,7 @@ void Bed3D::render_texture(bool bottom, GLCanvas3D& canvas) const
}
// starts generating the main texture, compression will run asynchronously
if (!texture->load_from_file(m_texture_filename, true, GLTexture::MultiThreaded, true)) {
if (!texture->load_from_file(m_texture_filename, true, GLTexture::ECompressionType::MultiThreaded, true)) {
render_default(bottom, false);
return;
}

336
src/slic3r/GUI/GLTexture.cpp
View File

@ -4,6 +4,8 @@
#include "3DScene.hpp"
#include "OpenGLManager.hpp"
#include "libslic3r/Utils.hpp"
#include <GL/glew.h>
#include <wx/image.h>
@ -21,8 +23,6 @@
#include "nanosvg/nanosvg.h"
#include "nanosvg/nanosvgrast.h"
#include "libslic3r/Utils.hpp"
namespace Slic3r {
namespace GUI {
@ -74,7 +74,7 @@ void GLTexture::Compressor::send_compressed_data_to_gpu()
glsafe(::glPixelStorei(GL_UNPACK_ALIGNMENT, 1));
glsafe(::glBindTexture(GL_TEXTURE_2D, m_texture.m_id));
// Querying the atomic m_num_levels_compressed value synchronizes processor caches, so that the dat of m_levels modified by the worker thread are accessible to the calling thread.
int num_compressed = (int)m_num_levels_compressed;
const int num_compressed = (int)m_num_levels_compressed;
for (int i = 0; i < num_compressed; ++ i) {
Level& level = m_levels[i];
if (! level.sent_to_gpu && ! level.compressed_data.empty()) {
@ -119,20 +119,6 @@ void GLTexture::Compressor::compress()
GLTexture::Quad_UVs GLTexture::FullTextureUVs = { { 0.0f, 1.0f }, { 1.0f, 1.0f }, { 1.0f, 0.0f }, { 0.0f, 0.0f } };
GLTexture::GLTexture()
: m_id(0)
, m_width(0)
, m_height(0)
, m_source("")
, m_compressor(*this)
{
}
GLTexture::~GLTexture()
{
reset();
}
bool GLTexture::load_from_file(const std::string& filename, bool use_mipmaps, ECompressionType compression_type, bool apply_anisotropy)
{
reset();
@ -167,15 +153,15 @@ bool GLTexture::load_from_svg_files_as_sprites_array(const std::vector<std::stri
return false;
// every tile needs to have a 1px border around it to avoid artifacts when linear sampling on its edges
unsigned int sprite_size_px_ex = sprite_size_px + 1;
const unsigned int sprite_size_px_ex = sprite_size_px + 1;
m_width = 1 + (int)(sprite_size_px_ex * states.size());
m_height = 1 + (int)(sprite_size_px_ex * filenames.size());
int n_pixels = m_width * m_height;
int sprite_n_pixels = sprite_size_px_ex * sprite_size_px_ex;
int sprite_stride = sprite_size_px_ex * 4;
int sprite_bytes = sprite_n_pixels * 4;
const int n_pixels = m_width * m_height;
const int sprite_n_pixels = sprite_size_px_ex * sprite_size_px_ex;
const int sprite_stride = sprite_size_px_ex * 4;
const int sprite_bytes = sprite_n_pixels * 4;
if (n_pixels <= 0) {
reset();
@ -208,7 +194,7 @@ bool GLTexture::load_from_svg_files_as_sprites_array(const std::vector<std::stri
if (image == nullptr)
continue;
float scale = (float)sprite_size_px / std::max(image->width, image->height);
const float scale = (float)sprite_size_px / std::max(image->width, image->height);
// offset by 1 to leave the first pixel empty (both in x and y)
nsvgRasterize(rast, image, 1, 1, scale, sprite_data.data(), sprite_size_px, sprite_size_px, sprite_stride);
@ -216,7 +202,7 @@ bool GLTexture::load_from_svg_files_as_sprites_array(const std::vector<std::stri
// makes white only copy of the sprite
::memcpy((void*)sprite_white_only_data.data(), (const void*)sprite_data.data(), sprite_bytes);
for (int i = 0; i < sprite_n_pixels; ++i) {
int offset = i * 4;
const int offset = i * 4;
if (sprite_white_only_data.data()[offset] != 0)
::memset((void*)&sprite_white_only_data.data()[offset], 255, 3);
}
@ -224,12 +210,12 @@ bool GLTexture::load_from_svg_files_as_sprites_array(const std::vector<std::stri
// makes gray only copy of the sprite
::memcpy((void*)sprite_gray_only_data.data(), (const void*)sprite_data.data(), sprite_bytes);
for (int i = 0; i < sprite_n_pixels; ++i) {
int offset = i * 4;
const int offset = i * 4;
if (sprite_gray_only_data.data()[offset] != 0)
::memset((void*)&sprite_gray_only_data.data()[offset], 128, 3);
}
int sprite_offset_px = sprite_id * (int)sprite_size_px_ex * m_width;
const int sprite_offset_px = sprite_id * (int)sprite_size_px_ex * m_width;
int state_id = -1;
for (const std::pair<int, bool>& state : states) {
++state_id;
@ -246,13 +232,13 @@ bool GLTexture::load_from_svg_files_as_sprites_array(const std::vector<std::stri
::memcpy((void*)output_data.data(), (const void*)src->data(), sprite_bytes);
// applies background, if needed
if (state.second) {
float inv_255 = 1.0f / 255.0f;
const float inv_255 = 1.0f / 255.0f;
// offset by 1 to leave the first pixel empty (both in x and y)
for (unsigned int r = 1; r <= sprite_size_px; ++r) {
unsigned int offset_r = r * sprite_size_px_ex;
const unsigned int offset_r = r * sprite_size_px_ex;
for (unsigned int c = 1; c <= sprite_size_px; ++c) {
unsigned int offset = (offset_r + c) * 4;
float alpha = (float)output_data.data()[offset + 3] * inv_255;
const unsigned int offset = (offset_r + c) * 4;
const float alpha = (float)output_data.data()[offset + 3] * inv_255;
output_data.data()[offset + 0] = (unsigned char)(output_data.data()[offset + 0] * alpha);
output_data.data()[offset + 1] = (unsigned char)(output_data.data()[offset + 1] * alpha);
output_data.data()[offset + 2] = (unsigned char)(output_data.data()[offset + 2] * alpha);
@ -261,7 +247,7 @@ bool GLTexture::load_from_svg_files_as_sprites_array(const std::vector<std::stri
}
}
int state_offset_px = sprite_offset_px + state_id * sprite_size_px_ex;
const int state_offset_px = sprite_offset_px + state_id * sprite_size_px_ex;
for (int j = 0; j < (int)sprite_size_px_ex; ++j) {
::memcpy((void*)&data.data()[(state_offset_px + j * m_width) * 4], (const void*)&output_data.data()[j * sprite_stride], sprite_stride);
}
@ -354,65 +340,144 @@ void GLTexture::render_sub_texture(unsigned int tex_id, float left, float right,
bool GLTexture::load_from_png(const std::string& filename, bool use_mipmaps, ECompressionType compression_type, bool apply_anisotropy)
{
bool compression_enabled = (compression_type != None) && GLEW_EXT_texture_compression_s3tc;
reset();
const bool compression_enabled = (compression_type != ECompressionType::None) && GLEW_EXT_texture_compression_s3tc;
// Load a PNG with an alpha channel.
wxImage image;
if (!image.LoadFile(wxString::FromUTF8(filename.c_str()), wxBITMAP_TYPE_PNG)) {
reset();
if (!image.LoadFile(wxString::FromUTF8(filename.c_str()), wxBITMAP_TYPE_PNG))
return false;
}
m_width = image.GetWidth();
m_height = image.GetHeight();
bool requires_rescale = false;
if (compression_enabled && compression_type == MultiThreaded) {
// the stb_dxt compression library seems to like only texture sizes which are a multiple of 4
int width_rem = m_width % 4;
int height_rem = m_height % 4;
if (width_rem != 0) {
m_width += (4 - width_rem);
requires_rescale = true;
}
if (height_rem != 0) {
m_height += (4 - height_rem);
requires_rescale = true;
}
}
if (compression_enabled && compression_type == ECompressionType::MultiThreaded)
requires_rescale = adjust_size_for_compression();
if (requires_rescale)
image = image.ResampleBicubic(m_width, m_height);
int n_pixels = m_width * m_height;
const int n_pixels = m_width * m_height;
if (n_pixels <= 0) {
reset();
return false;
}
std::vector<unsigned char> data;
// Get RGB & alpha raw data from wxImage, pack them into an array.
auto copy_data = [this](wxImage& image, std::vector<unsigned char>& data, int n_pixels) {
unsigned char* img_rgb = image.GetData();
if (img_rgb == nullptr) {
reset();
return false;
}
unsigned char* img_alpha = image.GetAlpha();
std::vector<unsigned char> data(n_pixels * 4, 0);
data.resize(n_pixels * 4);
for (int i = 0; i < n_pixels; ++i) {
int data_id = i * 4;
int img_id = i * 3;
const int data_id = i * 4;
const int img_id = i * 3;
data[data_id + 0] = img_rgb[img_id + 0];
data[data_id + 1] = img_rgb[img_id + 1];
data[data_id + 2] = img_rgb[img_id + 2];
data[data_id + 3] = (img_alpha != nullptr) ? img_alpha[i] : 255;
}
};
// sends data to gpu
copy_data(image, data, n_pixels);
send_to_gpu(data, use_mipmaps, compression_type, apply_anisotropy, [&image, copy_data](int lod_w, int lod_h, std::vector<unsigned char>& data) {
image = image.ResampleBicubic(lod_w, lod_h);
copy_data(image, data, lod_w * lod_h);
});
m_source = filename;
if (compression_enabled && compression_type == ECompressionType::MultiThreaded)
// start asynchronous compression
m_compressor.start_compressing();
return true;
}
bool GLTexture::load_from_svg(const std::string& filename, bool use_mipmaps, bool compress, bool apply_anisotropy, unsigned int max_size_px)
{
reset();
const bool compression_enabled = compress && GLEW_EXT_texture_compression_s3tc;
NSVGimage* image = nsvgParseFromFile(filename.c_str(), "px", 96.0f);
if (image == nullptr)
return false;
float scale = (float)max_size_px / std::max(image->width, image->height);
m_width = (int)(scale * image->width);
m_height = (int)(scale * image->height);
if (compression_enabled)
adjust_size_for_compression();
const int n_pixels = m_width * m_height;
if (n_pixels <= 0) {
reset();
nsvgDelete(image);
return false;
}
NSVGrasterizer* rast = nsvgCreateRasterizer();
if (rast == nullptr) {
nsvgDelete(image);
reset();
return false;
}
// creates the temporary buffer only once, with max size, and reuse it for all the levels, if generating mipmaps
std::vector<unsigned char> data(n_pixels * 4, 0);
nsvgRasterize(rast, image, 0, 0, scale, data.data(), m_width, m_height, m_width * 4);
send_to_gpu(data, use_mipmaps, compress ? ECompressionType::MultiThreaded : ECompressionType::None, apply_anisotropy, [&scale, rast, image](int lod_w, int lod_h, std::vector<unsigned char>& data) {
scale *= 0.5f;
data.resize(lod_w * lod_h * 4);
nsvgRasterize(rast, image, 0, 0, scale, data.data(), lod_w, lod_h, lod_w * 4);
});
m_source = filename;
if (compression_enabled)
// start asynchronous compression
m_compressor.start_compressing();
nsvgDeleteRasterizer(rast);
nsvgDelete(image);
return true;
}
bool GLTexture::adjust_size_for_compression()
{
bool ret = false;
// the stb_dxt compression library seems to like only texture sizes which are a multiple of 4
const int width_rem = m_width % 4;
const int height_rem = m_height % 4;
if (width_rem != 0) {
m_width += (4 - width_rem);
ret = true;
}
if (height_rem != 0) {
m_height += (4 - height_rem);
ret = true;
}
return ret;
}
void GLTexture::send_to_gpu(std::vector<unsigned char>& data, bool use_mipmaps, ECompressionType compression_type, bool apply_anisotropy,
std::function<void(int, int, std::vector<unsigned char>&)> resampler)
{
glsafe(::glPixelStorei(GL_UNPACK_ALIGNMENT, 1));
glsafe(::glGenTextures(1, &m_id));
glsafe(::glBindTexture(GL_TEXTURE_2D, m_id));
@ -423,8 +488,9 @@ bool GLTexture::load_from_png(const std::string& filename, bool use_mipmaps, ECo
glsafe(::glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAX_ANISOTROPY_EXT, max_anisotropy));
}
bool compression_enabled = (compression_type != ECompressionType::None) && GLEW_EXT_texture_compression_s3tc;
if (compression_enabled) {
if (compression_type == SingleThreaded)
if (compression_type == ECompressionType::SingleThreaded)
glsafe(::glTexImage2D(GL_TEXTURE_2D, 0, GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, (GLsizei)m_width, (GLsizei)m_height, 0, GL_RGBA, GL_UNSIGNED_BYTE, (const void*)data.data()));
else {
// initializes the texture on GPU
@ -446,27 +512,12 @@ bool GLTexture::load_from_png(const std::string& filename, bool use_mipmaps, ECo
lod_w = std::max(lod_w / 2, 1);
lod_h = std::max(lod_h / 2, 1);
n_pixels = lod_w * lod_h;
image = image.ResampleBicubic(lod_w, lod_h);
data.resize(n_pixels * 4);
img_rgb = image.GetData();
img_alpha = image.GetAlpha();
for (int i = 0; i < n_pixels; ++i) {
int data_id = i * 4;
int img_id = i * 3;
data[data_id + 0] = img_rgb[img_id + 0];
data[data_id + 1] = img_rgb[img_id + 1];
data[data_id + 2] = img_rgb[img_id + 2];
data[data_id + 3] = (img_alpha != nullptr) ? img_alpha[i] : 255;
}
resampler(lod_w, lod_h, data);
if (compression_enabled) {
if (compression_type == SingleThreaded)
glsafe(::glTexImage2D(GL_TEXTURE_2D, level, GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, (GLsizei)m_width, (GLsizei)m_height, 0, GL_RGBA, GL_UNSIGNED_BYTE, (const void*)data.data()));
if (compression_type == ECompressionType::SingleThreaded)
glsafe(::glTexImage2D(GL_TEXTURE_2D, level, GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, (GLsizei)lod_w, (GLsizei)lod_h, 0, GL_RGBA, GL_UNSIGNED_BYTE, (const void*)data.data()));
else {
// initializes the texture on GPU
glsafe(::glTexImage2D(GL_TEXTURE_2D, level, GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, (GLsizei)lod_w, (GLsizei)lod_h, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0));
@ -491,131 +542,6 @@ bool GLTexture::load_from_png(const std::string& filename, bool use_mipmaps, ECo
glsafe(::glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR));
glsafe(::glBindTexture(GL_TEXTURE_2D, 0));
m_source = filename;
if (compression_enabled && compression_type == MultiThreaded)
// start asynchronous compression
m_compressor.start_compressing();
return true;
}
bool GLTexture::load_from_svg(const std::string& filename, bool use_mipmaps, bool compress, bool apply_anisotropy, unsigned int max_size_px)
{
bool compression_enabled = compress && GLEW_EXT_texture_compression_s3tc;
NSVGimage* image = nsvgParseFromFile(filename.c_str(), "px", 96.0f);
if (image == nullptr) {
reset();
return false;
}
float scale = (float)max_size_px / std::max(image->width, image->height);
m_width = (int)(scale * image->width);
m_height = (int)(scale * image->height);
if (compression_enabled) {
// the stb_dxt compression library seems to like only texture sizes which are a multiple of 4
int width_rem = m_width % 4;
int height_rem = m_height % 4;
if (width_rem != 0)
m_width += (4 - width_rem);
if (height_rem != 0)
m_height += (4 - height_rem);
}
int n_pixels = m_width * m_height;
if (n_pixels <= 0) {
reset();
nsvgDelete(image);
return false;
}
NSVGrasterizer* rast = nsvgCreateRasterizer();
if (rast == nullptr) {
nsvgDelete(image);
reset();
return false;
}
// creates the temporary buffer only once, with max size, and reuse it for all the levels, if generating mipmaps
std::vector<unsigned char> data(n_pixels * 4, 0);
nsvgRasterize(rast, image, 0, 0, scale, data.data(), m_width, m_height, m_width * 4);
// sends data to gpu
glsafe(::glPixelStorei(GL_UNPACK_ALIGNMENT, 1));
glsafe(::glGenTextures(1, &m_id));
glsafe(::glBindTexture(GL_TEXTURE_2D, m_id));
if (apply_anisotropy) {
GLfloat max_anisotropy = OpenGLManager::get_gl_info().get_max_anisotropy();
if (max_anisotropy > 1.0f)
glsafe(::glTexParameterf(GL_TEXTURE_2D, GL_TEXTURE_MAX_ANISOTROPY_EXT, max_anisotropy));
}
if (compression_enabled) {
// initializes the texture on GPU
glsafe(::glTexImage2D(GL_TEXTURE_2D, 0, GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, (GLsizei)m_width, (GLsizei)m_height, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0));
// and send the uncompressed data to the compressor
m_compressor.add_level((unsigned int)m_width, (unsigned int)m_height, data);
}
else
glsafe(::glTexImage2D(GL_TEXTURE_2D, 0, GL_RGBA, (GLsizei)m_width, (GLsizei)m_height, 0, GL_RGBA, GL_UNSIGNED_BYTE, (const void*)data.data()));
if (use_mipmaps) {
// we manually generate mipmaps because glGenerateMipmap() function is not reliable on all graphics cards
int lod_w = m_width;
int lod_h = m_height;
GLint level = 0;
while (lod_w > 1 || lod_h > 1) {
++level;
lod_w = std::max(lod_w / 2, 1);
lod_h = std::max(lod_h / 2, 1);
scale /= 2.0f;
data.resize(lod_w * lod_h * 4);
nsvgRasterize(rast, image, 0, 0, scale, data.data(), lod_w, lod_h, lod_w * 4);
if (compression_enabled) {
// initializes the texture on GPU
glsafe(::glTexImage2D(GL_TEXTURE_2D, level, GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, (GLsizei)lod_w, (GLsizei)lod_h, 0, GL_RGBA, GL_UNSIGNED_BYTE, 0));
// and send the uncompressed data to the compressor
m_compressor.add_level((unsigned int)lod_w, (unsigned int)lod_h, data);
}
else
glsafe(::glTexImage2D(GL_TEXTURE_2D, level, GL_RGBA, (GLsizei)lod_w, (GLsizei)lod_h, 0, GL_RGBA, GL_UNSIGNED_BYTE, (const void*)data.data()));
}
if (!compression_enabled) {
glsafe(::glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAX_LEVEL, level));
glsafe(::glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR_MIPMAP_LINEAR));
}
}
else {
glsafe(::glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MIN_FILTER, GL_LINEAR));
glsafe(::glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAX_LEVEL, 0));
}
glsafe(::glTexParameteri(GL_TEXTURE_2D, GL_TEXTURE_MAG_FILTER, GL_LINEAR));
glsafe(::glBindTexture(GL_TEXTURE_2D, 0));
m_source = filename;
if (compression_enabled)
// start asynchronous compression
m_compressor.start_compressing();
nsvgDeleteRasterizer(rast);
nsvgDelete(image);
return true;
}
} // namespace GUI

28
src/slic3r/GUI/GLTexture.hpp
View File

@ -17,13 +17,13 @@ namespace GUI {
{
struct Level
{
unsigned int w;
unsigned int h;
unsigned int w{ 0 };
unsigned int h{ 0 };
bool sent_to_gpu{ false };
std::vector<unsigned char> src_data;
std::vector<unsigned char> compressed_data;
bool sent_to_gpu;
Level(unsigned int w, unsigned int h, const std::vector<unsigned char>& data) : w(w), h(h), src_data(data), sent_to_gpu(false) {}
Level(unsigned int w, unsigned int h, const std::vector<unsigned char>& data) : w(w), h(h), sent_to_gpu(false), src_data(data) {}
};
GLTexture& m_texture;
@ -55,7 +55,7 @@ namespace GUI {
};
public:
enum ECompressionType : unsigned char
enum class ECompressionType : unsigned char
{
None,
SingleThreaded,
@ -64,8 +64,8 @@ namespace GUI {
struct UV
{
float u;
float v;
float u{ 0.0f };
float v{ 0.0f };
};
struct Quad_UVs
@ -79,15 +79,15 @@ namespace GUI {
static Quad_UVs FullTextureUVs;
protected:
unsigned int m_id;
int m_width;
int m_height;
unsigned int m_id{ 0 };
int m_width{ 0 };
int m_height{ 0 };
std::string m_source;
Compressor m_compressor;
public:
GLTexture();
virtual ~GLTexture();
GLTexture() : m_compressor(*this) {}
virtual ~GLTexture() { reset(); }
bool load_from_file(const std::string& filename, bool use_mipmaps, ECompressionType compression_type, bool apply_anisotropy);
bool load_from_svg_file(const std::string& filename, bool use_mipmaps, bool compress, bool apply_anisotropy, unsigned int max_size_px);
@ -119,6 +119,10 @@ namespace GUI {
bool load_from_png(const std::string& filename, bool use_mipmaps, ECompressionType compression_type, bool apply_anisotropy);
bool load_from_svg(const std::string& filename, bool use_mipmaps, bool compress, bool apply_anisotropy, unsigned int max_size_px);
bool adjust_size_for_compression();
void send_to_gpu(std::vector<unsigned char>& data, bool use_mipmaps, ECompressionType compression_type, bool apply_anisotropy,
std::function<void(int, int, std::vector<unsigned char>&)> resampler);
friend class Compressor;
};

2
src/slic3r/GUI/GLToolbar.cpp
View File

@ -177,7 +177,7 @@ bool GLToolbar::init(const BackgroundTexture::Metadata& background_texture)
bool res = false;
if (!background_texture.filename.empty())
res = m_background_texture.texture.load_from_file(path + background_texture.filename, false, GLTexture::SingleThreaded, false);
res = m_background_texture.texture.load_from_file(path + background_texture.filename, false, GLTexture::ECompressionType::SingleThreaded, false);
if (res)
m_background_texture.metadata = background_texture;

5
src/slic3r/GUI/Gizmos/GLGizmosManager.cpp
View File

@ -86,9 +86,8 @@ bool GLGizmosManager::init()
m_background_texture.metadata.right = 16;
m_background_texture.metadata.bottom = 16;
if (!m_background_texture.metadata.filename.empty())
{
if (!m_background_texture.texture.load_from_file(resources_dir() + "/icons/" + m_background_texture.metadata.filename, false, GLTexture::SingleThreaded, false))
if (!m_background_texture.metadata.filename.empty()) {
if (!m_background_texture.texture.load_from_file(resources_dir() + "/icons/" + m_background_texture.metadata.filename, false, GLTexture::ECompressionType::SingleThreaded, false))
return false;
}