Hi Antti! El 24/05/15 a les 16:00, Antti Kantee ha escrit:
Can you briefly explain what problem this patch is intending to solve? I'm not sure I understand why you want it.
I'm actually not sure. When I wrote this, I was chasing a race condition caused by another bug (which is now fixed). Then I noticed kpreempt_disable() was a NO-OP stub, and I thought implementing it might help. It turned out not to, but I figured it might be useful to someone else.
In any case, changing the rumpuser interface is not acceptable, since it screws all who implement that interface, including 3rd parties, so we'll have to find some other way to solve your problem.
I seem to recall code in sys/rump was built with "kernel" namespace, unlike code in librumpuser which used "user" namespace, is that correct? I guess we need to #include <sched.h> and call sched_setscheduler() from somewhere else then. Any suggestion? -- Robert Millan
