proc_wakeup() replaces schedule_scp()
authorBarret Rhoden <brho@cs.berkeley.edu>
Wed, 25 Apr 2012 02:17:10 +0000 (19:17 -0700)
committerBarret Rhoden <brho@cs.berkeley.edu>
Wed, 25 Apr 2012 02:17:10 +0000 (19:17 -0700)
commitaf33a870e15116f4f66280725af9fba31c2c7a67
treeae01baed242740dabde7f96ccd0e1c755f236611
parent6a31e762ce90e49543754e9d5f084d20ecf83508
proc_wakeup() replaces schedule_scp()

All places wanting to make sure a proc runs now can spam proc_wakeup().
This commit makes new-proc calls and yield-related calls that aren't
already in the ksched call proc_wakeup().

This should also have cleaned up the last of the deadlockable lock
orderings (ever since we had proc_destroy in the ksched, it was possible
to deadlock during a concurrent sched_scp()).

Finally, this just gets rid of schedule_scp(), which was in no way
related to the __schedule_scp().  Ugh.
kern/include/process.h
kern/include/schedule.h
kern/src/kthread.c
kern/src/monitor.c
kern/src/process.c
kern/src/schedule.c
kern/src/syscall.c
kern/src/testing.c
tests/spawn.c