[proces server] implement ping to determine if domain is alive

[abertschi]

• ~new rpc call ping~
• rpc call to signalize death of domain
• new rpc call process info
• shell util ps is aware of which domains are active/inactive
• shell util oncore can spawn a domain in foreground
• ~a newly spawned domain will frequently ping process server if process server is available~
• signalize death if all threads exited

[eikendev]

Is this still WIP or am I supposed to have a look?

Originally, I meant doing one single ping to evaluate if a given domain is still active. That way, we don't have periodic pings, and thus better performance. Switching away from threads introduced quite some latency, and I'd prefer not adding any more overhead on top.

[abertschi]

Is this still WIP or am I supposed to have a look?

Still WIP but you are welcome to comment on it :). I did some more testing and I think we should rather signalize death if a dispatcher calls libc_exit instead. Within the scope of this project I think it is reasonable to just pass the pid and not further authorize the caller to remove itself from the process list.

[abertschi]

servus @eikendev. You are warmly invited to take a look :)

[abertschi]

Merging this to prevent further merge conflicts. Creating rpc connections to process server for each spawned domain makes process server with current nameserver/ urpc approach slow. Pending decision whether feature wants to be enabled or not.