LVC21F-208 Floating-point support in Trusted Firmware–M

Session Abstract

Level: Intermediate  Nowadays MCUs are widely used in applications including IoT, network system, industrial and medical systems. Although many of these applications can be implemented using integer arithmetic, there are still high demand of ability to process with floating point numbers, especially for the calculation of signal processing, artificial intelligence. Calculations for floating-point numbers can be accelerated by using a Floating-point unit (FPU) integrated in the processor. This topic focuses on floating point support in Trusted Firmware M as following. How to enable floating point support in Trusted Firmware M. Guidelines of system configuration for security purpose. Approaches about secure floating-point context protection for scenarios of enabled/disabled automatic lazy context save. Security analysis about how to avoid secure floating-point information leakage and how to avoid non-secure floating-point data being impacted. Also address some points should be noticed when enable floating-point in secure side only, non-secure side only or both sides. The main audience takeaways will be general understanding of how to enable floating point support and how to protect secure floating point context in system with secure firmware inside.

Session Speakers

feder liang

Staff software engineer (Arm)

Working on embedded software development for more than 15 years. Involved as application engineer for different embedded products such as optical drive, digital TV, IOT platform. Now working on secure firmware development for cortex-m processor in Arm.

Level: Intermediate 

Nowadays MCUs are widely used in applications including IoT, network system, industrial and medical systems. Although many of these applications can be implemented using integer arithmetic, there are still high demand of ability to process with floating point numbers, especially for the calculation of signal processing, artificial intelligence. Calculations for floating-point numbers can be accelerated by using a Floating-point unit (FPU) integrated in the processor. This topic focuses on floating point support in Trusted Firmware M as following. How to enable floating point support in Trusted Firmware M. Guidelines of system configuration for security purpose. Approaches about secure floating-point context protection for scenarios of enabled/disabled automatic lazy context save. Security analysis about how to avoid secure floating-point information leakage and how to avoid non-secure floating-point data being impacted. Also address some points should be noticed when enable floating-point in secure side only, non-secure side only or both sides. The main audience takeaways will be general understanding of how to enable floating point support and how to protect secure floating point context in system with secure firmware inside.

comments powered by Disqus

Recent Posts

Other Posts

Sign up. Receive Updates. Stay informed.

Sign up to our mailing list to receive updates on the latest Linaro Connect news!