首页> 外国专利> METHOD FOR CONTROLLING A GRAPHIC PROCESSING UNIT IN A CONTROL UNIT, IN PARTICULAR OF A VEHICLE, COMPUTER PROGRAM PRODUCT AND SYSTEM FOR AN EMBEDDED CONTROL UNIT

METHOD FOR CONTROLLING A GRAPHIC PROCESSING UNIT IN A CONTROL UNIT, IN PARTICULAR OF A VEHICLE, COMPUTER PROGRAM PRODUCT AND SYSTEM FOR AN EMBEDDED CONTROL UNIT

机译:用于控制单元中的图形处理单元的控制方法,特别是用于嵌入式控制单元的车辆,计算机程序产品和系统

摘要

The invention concerns a method for controlling a graphic processing unit, GPU, (17) in a control unit (1), in particular of a vehicle, the control unit comprising at least one central processing unit, CPU, the at least one CPU (3) having access to at least one memory (5); the GPU (17) being connected to and controlled by the at least one CPU (3), wherein the GPU (17) is adapted to provide frames to at least one display (13) having a predetermined output display frame rate, the method comprising: receiving (1000), by a GPU server (32) running on the CPU (3), a low priority graphical output from a first application (42), providing, by the GPU server (32), the low priority graphical output to the GPU (17), starting (1010), by the GPU (17), the rendering of the low priority graphical output, receiving (1020), by the GPU server (32), a time critical graphical output from a second application (44); determining a first point in time (t3) before the next frame (n+2) to be provided to the at least one display (13) based on second point in time (t4), when the next frame (n+2) is to be provided to the at least one display (13), and the maximal time needed to display the time critical graphical output on the at least one display (13) using the GPU (17); and forcing (1050), at the determined first point in time (t3), by the GPU server (32), the GPU (17) to render the time critical graphical output, if at the determined first point in time the GPU (17) still renders the low priority graphical output
机译:本发明涉及一种用于控制尤其是车辆的控制单元(1)中的图形处理单元GPU(17)的方法,该控制单元包括至少一个中央处理单元,CPU,至少一个CPU( 3)可以访问至少一个存储器(5);所述GPU(17)连接到所述至少一个CPU(3)并由所述至少一个CPU(3)控制,其中所述GPU(17)适于向至少一个具有预定输出显示帧速率的显示器(13)提供帧。 :由运行在CPU(3)上的GPU服务器(32)接收(1000)来自第一应用程序(42)的低优先级图形输出,由GPU服务器(32)将低优先级图形输出提供给GPU(17)从GPU(17)开始(1010)渲染低优先级图形输出,然后由GPU服务器(32)接收(1020)来自第二应用程序的时间关键图形输出( 44);当下一帧(n + 2)是第二时点(t4)时,基于第二时间点(t4)确定要提供给至少一个显示器(13)的下一帧(n + 2)之前的第一时间点(t3)。提供给至少一个显示器(13),以及使用GPU(17)在至少一个显示器(13)上显示时间关键图形输出所需的最大时间;如果在确定的第一时间点GPU(17),则由GPU服务器(32)在确定的第一时间点(t3)强制GPU(17)渲染时间关键的图形输出(1050)。 )仍呈现低优先级的图形输出

著录项

  • 公开/公告号EP3113169A1

    专利类型

  • 公开/公告日2017-01-04

    原文格式PDF

  • 申请/专利权人 OPENSYNERGY GMBH;

    申请/专利号EP20150175034

  • 发明设计人 TRETTER MICHAEL;BORKAR PARAG;

    申请日2015-07-02

  • 分类号G09G5;G09G5/36;G06F9/44;G06F3/14;

  • 国家 EP

  • 入库时间 2022-08-21 14:02:10

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号