[PATCH] introduce TASK_DEAD state

I am not sure about this patch, I am asking Ingo to take a decision.

task_struct->state == EXIT_DEAD is a very special case, to avoid a confusion
it makes sense to introduce a new state, TASK_DEAD, while EXIT_DEAD should
live only in ->exit_state as documented in sched.h.

Note that this state is not visible to user-space, get_task_state() masks off
unsuitable states.

Signed-off-by: Oleg Nesterov <oleg@tv-sign.ru>
Cc: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

4 files changed, 7 insertions, 6 deletions

diff --git a/include/linux/sched.h b/include/linux/sched.h
index 9763de334f09..a06fc89cf6e5 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -148,6 +148,7 @@ extern unsigned long weighted_cpuload(const int cpu);
 #define EXIT_DEAD		32
 /* in tsk->state again */
 #define TASK_NONINTERACTIVE	64
+#define TASK_DEAD		128
 
 #define __set_task_state(tsk, state_value)		\
 	do { (tsk)->state = (state_value); } while (0)
diff --git a/kernel/exit.c b/kernel/exit.c
index 9dd5f1336da2..2e4c13cba95a 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -955,7 +955,7 @@ fastcall NORET_TYPE void do_exit(long code)
 
 	preempt_disable();
 	/* causes final put_task_struct in finish_task_switch(). */
-	tsk->state = EXIT_DEAD;
+	tsk->state = TASK_DEAD;
 
 	schedule();
 	BUG();
diff --git a/kernel/sched.c b/kernel/sched.c
index a9405d7cc6ab..74f169ac0773 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1761,10 +1761,10 @@ static inline void finish_task_switch(struct rq *rq, struct task_struct *prev)
 
 	/*
 	 * A task struct has one reference for the use as "current".
-	 * If a task dies, then it sets EXIT_DEAD in tsk->state and calls
+	 * If a task dies, then it sets TASK_DEAD in tsk->state and calls
 	 * schedule one last time. The schedule call will never return, and
 	 * the scheduled task must drop that reference.
-	 * The test for EXIT_DEAD must occur while the runqueue locks are
+	 * The test for TASK_DEAD must occur while the runqueue locks are
 	 * still held, otherwise prev could be scheduled on another cpu, die
 	 * there before we look at prev->state, and then the reference would
 	 * be dropped twice.
@@ -1775,7 +1775,7 @@ static inline void finish_task_switch(struct rq *rq, struct task_struct *prev)
 	finish_lock_switch(rq, prev);
 	if (mm)
 		mmdrop(mm);
-	if (unlikely(prev_state == EXIT_DEAD)) {
+	if (unlikely(prev_state == TASK_DEAD)) {
 		/*
 		 * Remove function-return probe instances associated with this
 		 * task and put them back on the free list.
@@ -5153,7 +5153,7 @@ static void migrate_dead(unsigned int dead_cpu, struct task_struct *p)
 	BUG_ON(p->exit_state != EXIT_ZOMBIE && p->exit_state != EXIT_DEAD);
 
 	/* Cannot have done final schedule yet: would have vanished. */
-	BUG_ON(p->state == EXIT_DEAD);
+	BUG_ON(p->state == TASK_DEAD);
 
 	get_task_struct(p);
 
diff --git a/mm/oom_kill.c b/mm/oom_kill.c
index 202f186a753a..21f0a7e8e514 100644
--- a/mm/oom_kill.c
+++ b/mm/oom_kill.c
@@ -227,7 +227,7 @@ static struct task_struct *select_bad_process(unsigned long *ppoints)
 						p->flags & PF_EXITING;
 		if (releasing) {
 			/* TASK_DEAD tasks have already released their mm */
-			if (p->state == EXIT_DEAD)
+			if (p->state == TASK_DEAD)
 				continue;
 			if (p->flags & PF_EXITING && p == current) {
 				chosen = p;