d/freebsd-skq
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Move the schedlock owner state update following the context

Browse Source 
switch in fork_exit() to before anything else is done (but keep
schedlock for the deadthread check).  This means one less
nasty bug if ever in the future whatever might have been called
before the update played with schedlock or critical sections.

Discussed with: tjr
This commit is contained in:
bmilekic 2004-07-27 03:46:31 +00:00
parent 7c33e30743
commit 1c3958ce88
1 changed files with 15 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

28
sys/kern/kern_fork.c
View File

@ -782,6 +782,21 @@ fork_exit(callout, arg, frame)
struct proc *p;
struct thread *td;
/*
* Finish setting up thread glue so that it begins execution in a
* non-nested critical section with sched_lock held but not recursed.
*/
td = curthread;
p = td->td_proc;
td->td_oncpu = PCPU_GET(cpuid);
KASSERT(p->p_state == PRS_NORMAL, ("executing process is still new"));
sched_lock.mtx_lock = (uintptr_t)td;
mtx_assert(&sched_lock, MA_OWNED | MA_NOTRECURSED);
cpu_critical_fork_exit();
CTR3(KTR_PROC, "fork_exit: new thread %p (pid %d, %s)", td, p->p_pid,
p->p_comm);
/*
* Processes normally resume in mi_switch() after being
* cpu_switch()'ed to, but when children start up they arrive here
@ -793,19 +808,6 @@ fork_exit(callout, arg, frame)
thread_stash(td);
}
td = curthread;
p = td->td_proc;
td->td_oncpu = PCPU_GET(cpuid);
KASSERT(p->p_state == PRS_NORMAL, ("executing process is still new"));
/*
* Finish setting up thread glue so that it begins execution in a
* non-nested critical section with sched_lock held but not recursed.
*/
sched_lock.mtx_lock = (uintptr_t)td;
mtx_assert(&sched_lock, MA_OWNED | MA_NOTRECURSED);
cpu_critical_fork_exit();
CTR3(KTR_PROC, "fork_exit: new thread %p (pid %d, %s)", td, p->p_pid,
p->p_comm);
mtx_unlock_spin(&sched_lock);
/*