Preparing Linux Real-Time Kernel and Tuning Robotics Platform with Modern ARM64 SoC
Offered By: Linux Foundation via YouTube
Course Description
Overview
Explore the process of preparing and tuning a Linux Real-Time kernel for a modern ARM64 SoC-based robotics platform in this 30-minute conference talk. Dive into the challenges and solutions encountered while configuring, testing, and deploying a mainline kernel with PREEMPT_RT patches on the Qualcomm Robotics RB5 platform. Learn about kernel configuration, effective testing methods, common bugs and their implications, system tuning techniques using kernel parameters and cgroups, and real-time evaluation strategies. Gain insights into troubleshooting latency spikes through tracing and discover valuable resources for optimizing performance on ARM64-based robotics systems.
Syllabus
Introduction
Test platform RB5
First steps
Kernel build configuration
Checking locking correctness
What can go wrong-disabled preemption
What can go wrong-memory allocation
Evaluation of the system - tools
Measurements - try 1 - busy 60%
Tuning the system - command line
Tuning the system - runtime
idle-basic tuning
Measurements - try 3 - busy 100% - full tuning
Latency spikes - tracing
Resources and references
Measurements-try 3 - busy 60% - full tuning
Taught by
Linux Foundation
Tags
Related Courses
MongoDB for DBAsMongoDB University Optimizing Performance for SQL Based Applications
Microsoft via edX App Deployment, Debugging, and Performance
Google Cloud via Coursera Application Deployment, Debug, Performance 日本語版
Google Cloud via Coursera Optimize TensorFlow Models For Deployment with TensorRT
Coursera Project Network via Coursera