Peer Stritzinger

The question is if controlling this shell only, don't boot Erlang approach should be configured only in grisp.ini Maybe we need another configurable wait time for keypress when the application...

For configuration section I'd rather introduce a [rtems] shell = true This is about configuring something from RTEMS and we can lump in more RTEMS specific things maybe later. It...

> I am working again on this PR at the moment and I use a `grisp/grisp_base/sys/erl_main.c` locally to overwrite the default `init.c`. I am trying to implement a "switch on/off"...

Hilariously I'm now contradicting my earlier comment and favouring the solution which I called over engineered before ;-) Explanation: back then @Laymer had a pressing problem we needed the RTEMS...

Known problem, since changing RTEMS versions changes paths we probably should template the file its going in

You found the correct position in the code, thank you thats already very useful. So now the next questions are: * Does RTEMS have a field `ifr_hwaddr` in the struct...

We got the information from @sebhub that as we assumed it is supposed to work on RTEMS/libbsd exactly as on FreeBSD. So I built OTP on FreeBSD 12 and checked...

There are 4 mechanisms OTP uses to discover the hardware address: https://github.com/grisp/otp/blob/fe2b1323a3866ed0a9712e9d12e1f8f84793ec47/erts/emulator/drivers/common/inet_drv.c (Search for INET_IFOPT_HWADDR, that's the key used to write the hwaddr field towards Erlang side) We figured that...

So that leaves the forth one: https://github.com/grisp/otp/blob/fe2b1323a3866ed0a9712e9d12e1f8f84793ec47/erts/emulator/drivers/common/inet_drv.c#L5346 config.h: ``` /* Define to 1 if you have the `getifaddrs' function. */ #define HAVE_GETIFADDRS 1 ``` and /usr/include/sys/socket.h: ``` /usr/include/sys/socket.h:#define AF_LINK 18...

So for RTEMS4 based toolchains we already know that HAVE_GETIFADDRS is not defined in config.h Need to figure out the failing test case