VPU video decoding output method and system for Feiteng graphics card

Abstract

The invention discloses a VPU video decoding output method and system for a Feiteng display card. The method comprises the steps that display equipment is obtained, linkage with the display equipment of the Feiteng display card is completed, buffer management equipment GBM is established, a DMABUF buffer area is established, and a corresponding file descriptor fd is obtained. An fd importing function of an open multimedia acceleration layer (OMX) is used, fd of each DMABUF buffer area is imported into a video coding and decoding unit (VPU) in the GPU for use, the OMX locks the decoded fd by using a callback function and then adds the fd into a display output queue, the GPU obtains the fd from the display output queue and performs rendering output, and the GPU informs the OMX to release the fd after using the fd. According to the invention, the VPU video decoding output optimization of the Feiteng display card can be realized, so that the CPU occupancy rate of the VPU of the Feiteng display card during decoding output is reduced, the performance and efficiency of the system are improved, and picture tearing is prevented.

Description

technical field [0001] The invention relates to computer video playing technology, in particular to a VPU video decoding output method and system for Feiteng graphics cards. Background technique [0002] Feiteng X100 graphics card is a domestic GPU developed by Feiteng Information Technology Co., Ltd. It includes a graphics display module and a VPU unit (Video Processing Unit, video codec unit). The VPU integrated in the Phytium X100 graphics card is a high-performance multi-standard video IP that can perform the following decoding: H.264 / AVC BP / MP / HP, VC-1 SP / MP / AP, MPEG-1 / 2, MPEG4 SP / ASP, DivX / XviD, AVS, RV-8 / 9 / 10, VP8 and Theora video. [0003] Currently, the VPU video output of the Phytium X100 graphics card only uses software output display, and the video output capability is largely limited by the processing power of the CPU. Such as figure 1 As shown, the current Phytium X100 video card VPU decoding video output methods include: [0004] Step S101: ...

AI Technical Summary

Problems solved by technology

[0009] It can be seen that although the VPU decoding of the Phytium X100 graphics card uses hardware decoding, if the decoded data needs to be output using the GPU, it is necessary to copy the decoded data once, and the current output uses software. way, which leads to a high CPU usage when using the VPU of the Phytium X100 graphics card for video decoding

When playing multiple videos, since the existing solution does not take into account the resources of the graphics card GPU, each additional video playback can only be processed in this way

However, this method has a lot of shortcomings. During the playback process, the CPU usage rate will rise sharply. Often, the video played later will be stuck due to window dragging, the screen will be stuck, the video frame will be severely delayed, or even the screen will be completely stuck. Not moving and other problems, serious will lead to system crash

The existing Phytium X100 video card VPU video decoding output method has brought great consumption to the overall system resources. If the user performs other operations during video playback, such as daily computer office work, the system CPU resources will be insufficient at this time, and other applications The phenomenon of not being able to get a timely response, which brings a very poor experience to users when playing videos

Therefore, when the user is playing multiple videos, the efficiency is low, there is a large delay and freeze, which seriously affects the user experience

Images