I would like to know if TrustZone will ever be implemented on Cortex R:
I was able to find in your documentation that:
“TrustZone technology for ARMv8-M is also different from the virtualization approach as supported in the ARMv8-R architecture.In systems with virtualization, each of the virtualized software environments (Virtual Machines, or VMs) is isolated from each other, and there is no direct interaction between VMs other than going through the hypervisor, interrupts, or shared memories.As a result, interaction between software in different VMs requires additional execution cycles and software overhead, and while this is perfectly acceptable in the high-performance ARM Cortex-R processors, it is not ideal for resource constrained ARM Cortex-M applications.”
Do you think it means that there are no plans to implement TrustZone in CortexR?
Thank you :)
ARMv8-R has hyp mode. Not as strict as TZ. But together with a system MPU might be sufficient to separate host and guest.
But do you think that in some scenarios TZ could make sense also in CortexR (and so it is reasonable to expect it to be available in the future) or it’s simply not necessary because of hyp mode? Thanks by the way :)
IMHO, TZ is the stronger separation as it is on the chip level. So, yes, a Cortex-R with the same TZ features as Cortex-A would be great. But I doubt it will come.
View all questions in Cortex-R / R-Profile forum