[PATCH 1/4] ptrace: kill __ptrace_detach(), fix ->exit_state check

[PATCH 1/4] ptrace: kill __ptrace_detach(), fix ->exit_state check

[Oleg Nesterov]

Move the code from __ptrace_detach() to its single caller and kill this
helper.

Also, fix the ->exit_state check, we shouldn't wake up EXIT_DEAD tasks.
Actually, I think task_is_stopped_or_traced() makes more sense, but this
needs another patch.

Signed-off-by: Oleg Nesterov <oleg@redhat.com>

--- 6.29-rc3/kernel/ptrace.c~1_KILL_DETACH	2009-01-29 02:30:00.000000000 +0100
+++ 6.29-rc3/kernel/ptrace.c	2009-01-29 03:24:51.000000000 +0100
@@ -235,16 +235,6 @@ out:
 	return retval;
 }
 
-static inline void __ptrace_detach(struct task_struct *child, unsigned int data)
-{
-	child->exit_code = data;
-	/* .. re-parent .. */
-	__ptrace_unlink(child);
-	/* .. and wake it up. */
-	if (child->exit_state != EXIT_ZOMBIE)
-		wake_up_process(child);
-}
-
 int ptrace_detach(struct task_struct *child, unsigned int data)
 {
 	if (!valid_signal(data))
@@ -254,10 +244,16 @@ int ptrace_detach(struct task_struct *ch
 	ptrace_disable(child);
 	clear_tsk_thread_flag(child, TIF_SYSCALL_TRACE);
 
-	write_lock_irq(&tasklist_lock);
 	/* protect against de_thread()->release_task() */
-	if (child->ptrace)
-		__ptrace_detach(child, data);
+	write_lock_irq(&tasklist_lock);
+	if (child->ptrace) {
+		child->exit_code = data;
+
+		__ptrace_unlink(child);
+
+		if (!child->exit_state)
+			wake_up_process(child);
+	}
 	write_unlock_irq(&tasklist_lock);
 
 	return 0;

Re: [PATCH 1/4] ptrace: kill __ptrace_detach(), fix ->exit_state check

[Roland McGrath]

> Move the code from __ptrace_detach() to its single caller and kill this
> helper.
> 
> Also, fix the ->exit_state check, we shouldn't wake up EXIT_DEAD tasks.

Yes, though that is probably impossible anyway.  Perhaps in a de_thread
race, but probably harmlessly does nothing.  Anyway, certainly fine to
change the test to match all others of the same ilk.

> Actually, I think task_is_stopped_or_traced() makes more sense, but this
> needs another patch.

Agreed.  It is one of those nasty cases where the old semantics are
"clearly" wrong, but it's very hard not to worry that some obscure
combination of circumstances means that some common-enough case actually
depends on the "wrong" wakeup.

> Signed-off-by: Oleg Nesterov <oleg@redhat.com>

Acked-by: Roland McGrath <roland@redhat.com>


Thanks,
Roland