Qml界面渲染机制



  • 渲染层次:

    0_1531798264848_48a56a99-a2a7-4a1e-8d0e-f12d889d74a1-image.png

    渲染循环分三种,basic, windows, and threaded,其中Basic和Windows是单线程的,threaded在单独线程里面渲染场景,qt会根据平台自动选择一种,我电脑上已经验证是threaded

    qt.scenegraph.general: threaded render loop qt.scenegraph.general: Using sg animation driver qt.scenegraph.general: Animation Driver: using vsync: 16.67 ms

    同时可以知道画面垂直同步是16.67 ms,理论60Hz

    线程渲染循环中如何完成一帧

    0_1531798407892_935d0c1f-554a-4cbb-999a-a0966484548d-image.png

    1. A change occurs in the QML scene, causing QQuickItem::update() to be called. This can be the result of for instance an animation or user input. An event is posted to the render thread to initiate a new frame.
    2. The render thread prepares to draw a new frame and makes the OpenGL context current and initiates a block on the GUI thread.
    3. While the render thread is preparing the new frame, the GUI thread calls QQuickItem::updatePolish() to do final touch-up of items before they are rendered.
    4. GUI thread is blocked.
    5. The QQuickWindow::beforeSynchronizing() signal is emitted. Applications can make direct connections (using Qt::DirectConnection) to this signal to do any preparation required before calls to QQuickItem::updatePaintNode().
    6. Synchronization of the QML state into the scene graph. This is done by calling the QQuickItem::updatePaintNode() function on all items that have changed since the previous frame. This is the only time the QML items and the nodes in the scene graph interact.
    7. GUI thread block is released.
    8. The scene graph is rendered:
    9. The QQuickWindow::beforeRendering() signal is emitted. Applications can make direct connections (using Qt::DirectConnection) to this signal to use custom OpenGL calls which will then stack visually beneath the QML scene.
    10. Items that have specified QSGNode::UsePreprocess, will have their QSGNode::preprocess() function invoked.
    11. The renderer processes the nodes and calls OpenGL functions.
    12. The QQuickWindow::afterRendering() signal is emitted. Applications can make direct connections (using Qt::DirectConnection) to this signal to use custom OpenGL calls which will then stack visually over the QML scene.
    13. The rendered frame is swapped and QQuickWindow::frameSwapped() is emitted.
    14. While the render thread is rendering, the GUI is free to advance animations, process events, etc.
      The threaded renderer is currently used by default on Windows with opengl32.dll, Linux with non-Mesa based drivers, macOS, mobile platforms, and Embedded Linux with EGLFS but this is subject to change. It is possible to force use of the threaded renderer by setting QSG_RENDER_LOOP=threaded in the environment.

    非线程渲染

    0_1531798434302_91612780-f2cf-4fdb-b5a1-aeb004a1e7a2-image.png

    查看渲染方式:

    QLoggingCategory::setFilterRules(QStringLiteral("qt.scenegraph.general=true")); qSetMessagePattern("%{category}: %{message}");



  • @青山白云 对,Qt Quick内部实现还是根据屏幕的刷新率来的。不这么做的话,在不同平台上使用Animator做动画速度会出现不一致的现象。所以Qt的Animation Driver也是按照这样的思路规定16.67ms。


Log in to reply
 

走马观花

最近的回复

  • C

    Qt for MCU需要商业授权的

    read more
  • Qt for MCUs

    搭建Qt for MCUs PC端开发环境。qt for mcus提供了一个完整的图形框架和工具包,包含了在MCUs上设计、开发和部署gui所需的一切。它允许您在裸机或实时操作系统上运行应用程序。

    先决条件

    开发主机环境支持仅限于Windows 10

    MSVC compiler v19.16 (Visual Studio 2017 15.9.9 or newer) x64

    CMake v3.13 or newer (you can install it using the Qt Online installer) x64

    使用Qt联机安装程序安装Qt for MCUs,该安装程序可通过Qt帐户下载

    安装Qt 5.14和Qt Creator 4.11 or higher

    安装链接

    › Qt: https://account.qt.io/downloads
    › CMake: https://cmake.org/download/
    › Python 2.7 32-bit: https://www.python.org/downloads/release/python-2716/
    › Arm GCC: https://developer.arm.com/tools-and-software/open-source-software/developer-tools/gnutoolchain/gnu-rm/downloads
    › J-Link Software Pack: https://www.segger.com/downloads/jlink/JLink_Windows.exe
    › J-Link OpenSDA Firmware: https://www.segger.com/downloads/jlink/OpenSDA_MIMXRT1050-EVKHyperflash
    › STM32CubeProgrammer: https://www.st.com/en/development-tools/stm32cubeprog.html
    › STM32 ST-LINK Utility: https://www.st.com/en/development-tools/stsw-link004.html​​​​​​​

    Qt Creator设置 启用Qt Creator插件 选择“帮助>关于插件”,然后从列表中选择“MCU支持(实验性)”插件,重新启动Qt Creator以应用更改
    替代文字 为MCU创建Qt工具包

    选择工具>选项>设备>MCU

    选择Qt for MCUs-Desktop 32bpp作为目标

    如果尚未设置,请提供Qt for MCUs安装目录的路径。

    单击Apply应用。

    替代文字

    替代文字
    替代文字

    注意:

    编译器要选X64,Qt版本要选64bit,CMake Tool选x64

    打开恒温器项目demo

    选择文件>打开文件或项目。。。

    打开CMakefiles.txt文件来自thermo文件夹的文件。

    选择Qt作为MCU-桌面32bpp套件。

    单击“配置项目”以完成。

    替代文字

    问题

    开发主机环境支持仅限于Windows 10

    C++编译失败,文本大字体.pixelSize.

    文本类型无法正确呈现需要复杂文本布局的unicode序列。对复杂文本使用StaticText

    read more
  • H

    hi 有问题请教你,方便加个联系方式吗

    read more
  • boost.asio是一个很棒的网络库,这回儿我也开始系统地学习起来了。想想当年接触boost,也有八年多了。这次开始接触boost,觉得既熟悉又陌生。熟悉的是小写字母+下划线的命名方式、晦涩的模板、很慢的编译速度以及较大的程序体积,陌生的是asio的各种概念:io服务、接收器、套接字等等:我之前对网络编程不是非常了解。

    于是根据我的理解,参考《Boost.Asio C++网络编程》实现了这样一个简单的客户端和服务端通信的例子,例子非常简单,还不完善,但是幸运的是,可以在本机上互通了。
    下面是客户端的代码:

    #include <iostream> #include <boost/asio.hpp> #include <boost/proto/detail/ignore_unused.hpp> using namespace std; using namespace boost::asio; using namespace boost::system; using namespace boost::proto::detail;// 提供ignore_unused方法 void writeHandler( const boost::system::error_code& ec, size_t bytesTransferred ) { if ( ec ) { cout << "Write data error, code: " << ec.value( ) << "transferred: " << bytesTransferred << endl; } else { cout << "OK! " << bytesTransferred << "bytes written. " << endl; } } int main(int argc, char *argv[]) { ignore_unused( argc ); ignore_unused( argv ); io_service service; ip::tcp::socket sock( service ); ip::tcp::endpoint ep( ip::address::from_string( "127.0.0.1" ), 6545 ); boost::system::error_code ec; sock.connect( ep, ec ); if ( ec ) { cout << "Connect error, code: " << ec.value( ) << ", We will exit." << endl; return ec.value( ); } else { char buf[1024] = "Hello world!"; sock.async_write_some( buffer( buf ), writeHandler ); sock.close( ); } return service.run( ); }

    下面是服务端的代码:

    #include <iostream> #include <boost/asio.hpp> #include <boost/proto/detail/ignore_unused.hpp> using namespace std; using namespace boost::asio; using namespace boost::system; using namespace boost::proto::detail;// 提供ignore_unused方法 void acceptHandle( const boost::system::error_code& code ) { cout << "Accepted." << endl; } int main(int argc, char *argv[]) { ignore_unused( argc ); ignore_unused( argv ); io_service service; ip::tcp::endpoint ep( ip::address::from_string( "127.0.0.1" ), 6545 ); boost::system::error_code ec; ip::tcp::socket sock( service ); ip::tcp::acceptor acceptor( service, ep ); acceptor.async_accept( sock, acceptHandle ); if ( ec ) { cout << "There is an error in server. code: " << ec.value( ) << endl; } return service.run( );// 阻塞运行 }

    运行结果是这样的:
    78448d7b-b3ae-42fc-9e2e-4dd2fbdac2c2-image.png

    我对boost.asio中几个概念的理解:

    io_service,这就是一个类似事件循环的东西,它为io设备提供服务,故名。不管是套接字、文件还是串口设备,都要使用它的服务。它的run()函数相当于启动了一个事件循环。一旦有消息了,即进行响应。这也是实现异步编程的重要基础。 socket,这个类则是套接字,可以处理TCP或者是UDP请求。有同步以及异步的处理方式,也有带异常以及不带异常的处理方式。 acceptor,接收器,仅仅是服务端使用。相当于其余框架中的listener,作接收用的。

    比较浅显,如果有不当之处,敬请指正。

    read more

关注我们

微博
QQ群