On 12/23/06, Axel Dörfler <axeld@xxxxxxxxxxxxxxxx> wrote:
There is no dependency - it's just that the first application that is started that needs a desktop will create it. And only then you actually need an input_server - if there is no desktop, you cannot input anything either.
Yeah I sort of realized that after sending this message. Should will still put a "waitfor _input_server_event_loop_" in the Bootscript after starting the syslog_daemon like in R5?
That's a bit strange: the input port, and the cursor semaphore are both owned by the input_server - if that server is gone, both objects will be deleted automatically, and both threads in the app_server will notice that.
I'll play with it more tomorrow. It is funny, I've learned a lot about the shutdown process trying to properly shutdown the input_server (or in the final solution not shut it down, haha) and now I'm learning a lot about the start-up process by trying to get the input_server starting properly. There is a certain elegant symmetry about it all :) All I know is after the shutdown and start-up fixes, other things (besides maybe the kernel) should be a piece of cake. Because testing start-up and shutdown is a pain. Thank goodness for virtual machine software, even if QEMU is super slow. But for now I'm off to bed. Ryan