This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

External Abort and Transition between Non Secure & Secure world

Hi !

My question is about the behavior of external aborts: I had to strange behavior on a board that I can only explain by the fact the Linux triggered an external abort (trying to access data/registers protected by a TZASC or CSU for example) just before an FIQ triggered a switch from Linux to the Monitor.

Once in the monitor, External Abort are masked, so I got the actual External Abort fault in my secure kernel.

 

1) Is this scenario possible ?

2) If that's the case, I'd like to test if there is an external abort "pending" in the monitor, is this possible ?

 

Best,

Vincent