6-10 October 2024
Concordia University Conference Centre
America/New_York timezone

GStreamer NVIDIA memory in Jetson: looking for an efficient way to process video frames when handling raw CUDA kernels

7 Oct 2024, 10:50
20m
Room 2

Room 2

Speaker

Diego Nieto Munoz (Fluendo)

Description

Mixing NVIDIA Jetson memory and the raw CUDA one is not straightforward. In this talk, we present the steps we follow to develop GStreamer plugins based on CUDA frame process. In order to work with this kind of memory, several data conversions and mappings are required. These mappings are expensive, so we based our filters on a custom CUDA mapped memory pool to get a noticeable speed-up.

Speaker bio Diego is a senior software engineer at Fluendo. He has been working with GPU programming in different NVIDIA chipsets, as well as a wide range of multimedia projects.
Duration of the talk

Primary author

Presentation Materials

2024 Platinum Sponsor
Collabora
2024 Gold Sponsors
Arm
Google
Microsoft
NVIDIA
2024 Silver Sponsors
AMD
FEX-Emu
Igalia
Qualcomm
The Linux Foundation
2024 Bronze Sponsors
CodeWeavers
Khronos Group
Libre Computer