Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

When Radiation Meets Linux: Analyzing Soft Errors in Linux on COTS SoCs under Proton Irradiation

Saad Memon, Rafal Graczyk, Tomasz Rajkowski, Jan Swakon, Damian Wrobel, Sebastian Kusyk, Seth Roffe, Mike Papadakis

TL;DR

This work addresses the reliability of unmodified Linux running on commercial SoCs in space by systematically probing soft errors with proton irradiation across three architectures (ARM Cortex‑A53 on 40 nm CMOS, 14 nm FinFET, and a RISC‑V softcore on FPGA). It combines aggressive stress testing with targeted irradiation to quantify Linux‑level SEE vulnerability through Linux‑SEFI cross‑sections and to identify vulnerable kernel subsystems. The study finds that the 14 nm FinFET i.MX 8M Plus achieved 2–3× longer Linux uptime without ECC than the 40 nm CMOS events, while eMMC storage emerges as a major reliability bottleneck, highlighting peripheral impacts on system resilience. The results provide foundational data and concrete mitigation directions (software hardening, MC/DMR/TMR strategies, ECC adoption, and hardware–software co‑design) to inform space mission readiness for COTS‑based Linux systems."

Abstract

The increasing use of Linux on commercial off-the-shelf (COTS) system-on-chip (SoC) in spaceborne computing inherits COTS susceptibility to radiation-induced failures like soft errors. Modern SoCs exacerbate this issue as aggressive transistor scaling reduces critical charge thresholds to induce soft errors and increases radiation effects within densely packed transistors, degrading overall reliability. Linux's monolithic architecture amplifies these risks, as tightly coupled kernel subsystems propagate errors to critical components (e.g., memory management), while limited error-correcting code (ECC) offers minimal mitigation. Furthermore, the lack of public soft error data from irradiation tests on COTS SoCs running Linux hinders reliability improvements. This study evaluates proton irradiation effects (20-50 MeV) on Linux across three COTS SoC architectures: Raspberry Pi Zero 2 W (40 nm CMOS, Cortex-A53), NXP i MX 8M Plus (14 nm FinFET, Cortex-A53), and OrangeCrab (40 nm FPGA, RISC-V). Irradiation results show the 14 nm FinFET NXP SoC achieved 2-3x longer Linux uptime without ECC memory versus both 40 nm CMOS counterparts, partially due to FinFET's reduced charge collection. Additionally, this work presents the first cross-architecture analysis of soft error-prone Linux kernel components in modern SoCs to develop targeted mitigations. The findings establish foundational data on Linux's soft error sensitivity in COTS SoCs, guiding mission readiness for space applications.

When Radiation Meets Linux: Analyzing Soft Errors in Linux on COTS SoCs under Proton Irradiation

TL;DR

This work addresses the reliability of unmodified Linux running on commercial SoCs in space by systematically probing soft errors with proton irradiation across three architectures (ARM Cortex‑A53 on 40 nm CMOS, 14 nm FinFET, and a RISC‑V softcore on FPGA). It combines aggressive stress testing with targeted irradiation to quantify Linux‑level SEE vulnerability through Linux‑SEFI cross‑sections and to identify vulnerable kernel subsystems. The study finds that the 14 nm FinFET i.MX 8M Plus achieved 2–3× longer Linux uptime without ECC than the 40 nm CMOS events, while eMMC storage emerges as a major reliability bottleneck, highlighting peripheral impacts on system resilience. The results provide foundational data and concrete mitigation directions (software hardening, MC/DMR/TMR strategies, ECC adoption, and hardware–software co‑design) to inform space mission readiness for COTS‑based Linux systems."

Abstract

The increasing use of Linux on commercial off-the-shelf (COTS) system-on-chip (SoC) in spaceborne computing inherits COTS susceptibility to radiation-induced failures like soft errors. Modern SoCs exacerbate this issue as aggressive transistor scaling reduces critical charge thresholds to induce soft errors and increases radiation effects within densely packed transistors, degrading overall reliability. Linux's monolithic architecture amplifies these risks, as tightly coupled kernel subsystems propagate errors to critical components (e.g., memory management), while limited error-correcting code (ECC) offers minimal mitigation. Furthermore, the lack of public soft error data from irradiation tests on COTS SoCs running Linux hinders reliability improvements. This study evaluates proton irradiation effects (20-50 MeV) on Linux across three COTS SoC architectures: Raspberry Pi Zero 2 W (40 nm CMOS, Cortex-A53), NXP i MX 8M Plus (14 nm FinFET, Cortex-A53), and OrangeCrab (40 nm FPGA, RISC-V). Irradiation results show the 14 nm FinFET NXP SoC achieved 2-3x longer Linux uptime without ECC memory versus both 40 nm CMOS counterparts, partially due to FinFET's reduced charge collection. Additionally, this work presents the first cross-architecture analysis of soft error-prone Linux kernel components in modern SoCs to develop targeted mitigations. The findings establish foundational data on Linux's soft error sensitivity in COTS SoCs, guiding mission readiness for space applications.

Paper Structure

This paper contains 23 sections, 2 equations, 8 figures, 8 tables.

Table of Contents

  1. Introduction
  2. Testing Methodology
  3. Significance of Stress Testing
  4. Stress Testing Tool: Workload and Test Parameters
  5. Significance of Proton Testing
  6. Beam Configuration
  7. DUT Selection
  8. DUT 1: Raspberry Pi Zero 2 W
  9. DUT 2: NXP i.MX 8M Plus
  10. DUT 3: OrangeCrab v2
  11. Evaluation I: Estimating Cross-Section of Linux Crashes Under SEFIs
  12. Definition of Linux-SEFI
  13. Significance of Cross-Section Analysis in Estimating SEEs
  14. Cross-Section Analysis of Linux-SEFI Crashes across DUTs
  15. DUT 1:Raspberry Pi Zero 2 W (40 nm CMOS)
  16. ...and 8 more sections

Figures (8)

  • Figure 1: Unmitigated radiation-induced critical soft error on CPU 3 in a Linux COTS SoC causing Linux crash
  • Figure 2: Block diagram of ARM Cortex-A53 SoC (left) and VexRISC-V softcore (right), running Linux OS with kernel and memory errors collected under proton irradiation
  • Figure 3: 2D beam profile, normalised intensity
  • Figure 4: Linux Kernel Performance Metrics: Idle vs. Stressed (stress-ng) via Perf Analysis (IF= Increased Factor)
  • Figure 5: DUTs in Proton Beam Testing: Raspberry Pi Zero 2W (Left), NXP i.MX 8M Plus (Center), OrangeCrab FPGA (Right)
  • ...and 3 more figures