Fail-Safe Embedded Linux: Designing for Power Resilience
summary
Provides concrete mitigations: add input protection and a power management IC plus a supercapacitor that supplies 20 Ws for 1.5s so software can sync and shut down; choose industrial eMMC with power loss detection, enable PSLC/enhanced area (halves usable capacity) or vendor write‑reliability (costs ~20% perf); use partitioning with read‑only root, A/B duplicates, 10–20% overprovision left unformatted; pick ext4 (ordered/journal modes) or F2FS (checkpointing), use noatime, atomic temp+fsync+rename writes, and monitor unintended boots.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 5:23Hardware mitigations
- 16:50OS reliability strategies
- 26:03File system choices
- 33:39F2FS and UBFS
- 40:57Application-level resilience techniques
- 43:26Kernel recovery
- 45:13Recovery and summary
related talks
53mSecure Embedded Linux Product - A Success Story
42mLinux in Space: Fault Detection, Recovery and Fault-Tolerant System Designs
32mELC 2013 Security: Best Practices for Embedded Systems
48mDirty Tricks of Computer Hardware
01h 07mPEW PEW PEW: Designing Secure Boot Securely
55mBuilding a Robust Embedded Linux Platform
14mHow not to secure your embedded system, lessons from automotive testing
32mEmbedded Linux Security: Dr Jekyll and Mr Hyde