Currently the scripts check for the presence of an ARM core to determine if a board is a Zynq or not when the client disconnects, in order to determine which reset logic to use.
If the Zynq somehow gets into a bad state this check can fail, as well as the ability to run code on the ARM from SDK, needing a board power cycle.
It should be possible to annotate boards as Zynq (or other types) in the board configuration file, then use this same check when a container is started to find the error during client connection and handle this (e.g. pushing client onto a different board, blacklisting the board, notifying someone of the issue).
Currently the scripts check for the presence of an ARM core to determine if a board is a Zynq or not when the client disconnects, in order to determine which reset logic to use.
If the Zynq somehow gets into a bad state this check can fail, as well as the ability to run code on the ARM from SDK, needing a board power cycle.
It should be possible to annotate boards as Zynq (or other types) in the board configuration file, then use this same check when a container is started to find the error during client connection and handle this (e.g. pushing client onto a different board, blacklisting the board, notifying someone of the issue).