From: Thomas Gleixner <tglx@linutronix.de>
Date: Sat, 25 Jun 2011 09:21:04 +0200
Subject: sched: Add saved_state for tasks blocked on sleeping locks
Git-repo: git://git.kernel.org/pub/scm/linux/kernel/git/rt/linux-rt-devel.git
Git-commit: b4e7212b0ca1acdcb10620cb93e9adf8107e7003
Patch-mainline: Queued in subsystem maintainer repository
References: SLE Realtime Extension

Spinlocks are state preserving in !RT. RT changes the state when a
task gets blocked on a lock. So we need to remember the state before
the lock contention. If a regular wakeup (not a RTmutex related
wakeup) happens, the saved_state is updated to running. When the lock
sleep is done, the saved state is restored.

Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Mike Galbraith <mgalbraith@suse.de>
---
 include/linux/sched.h |    3 +++
 kernel/sched/core.c   |   31 ++++++++++++++++++++++++++++++-
 kernel/sched/sched.h  |    1 +
 3 files changed, 34 insertions(+), 1 deletion(-)

--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -509,6 +509,8 @@ struct task_struct {
 #endif
 	/* -1 unrunnable, 0 runnable, >0 stopped: */
 	volatile long			state;
+	/* saved state for "spinlock sleepers" */
+	volatile long			saved_state;
 	void				*stack;
 	atomic_t			usage;
 	/* Per task flags (PF_*), defined further below: */
@@ -1432,6 +1434,7 @@ extern struct task_struct *find_task_by_
 
 extern int wake_up_state(struct task_struct *tsk, unsigned int state);
 extern int wake_up_process(struct task_struct *tsk);
+extern int wake_up_lock_sleeper(struct task_struct *tsk);
 extern void wake_up_new_task(struct task_struct *tsk);
 
 #ifdef CONFIG_SMP
--- a/kernel/sched/core.c
+++ b/kernel/sched/core.c
@@ -2024,8 +2024,25 @@ try_to_wake_up(struct task_struct *p, un
 	 */
 	smp_mb__before_spinlock();
 	raw_spin_lock_irqsave(&p->pi_lock, flags);
-	if (!(p->state & state))
+	if (!(p->state & state)) {
+		/*
+		 * The task might be running due to a spinlock sleeper
+		 * wakeup. Check the saved state and set it to running
+		 * if the wakeup condition is true.
+		 */
+		if (!(wake_flags & WF_LOCK_SLEEPER)) {
+			if (p->saved_state & state)
+				p->saved_state = TASK_RUNNING;
+		}
 		goto out;
+	}
+
+	/*
+	 * If this is a regular wakeup, then we can unconditionally
+	 * clear the saved state of a "lock sleeper".
+	 */
+	if (!(wake_flags & WF_LOCK_SLEEPER))
+		p->saved_state = TASK_RUNNING;
 
 	trace_sched_waking(p);
 
@@ -2189,6 +2206,18 @@ int wake_up_process(struct task_struct *
 }
 EXPORT_SYMBOL(wake_up_process);
 
+/**
+ * wake_up_lock_sleeper - Wake up a specific process blocked on a "sleeping lock"
+ * @p: The process to be woken up.
+ *
+ * Same as wake_up_process() above, but wake_flags=WF_LOCK_SLEEPER to indicate
+ * the nature of the wakeup.
+ */
+int wake_up_lock_sleeper(struct task_struct *p)
+{
+	return try_to_wake_up(p, TASK_ALL, WF_LOCK_SLEEPER);
+}
+
 int wake_up_state(struct task_struct *p, unsigned int state)
 {
 	return try_to_wake_up(p, state, 0);
--- a/kernel/sched/sched.h
+++ b/kernel/sched/sched.h
@@ -1292,6 +1292,7 @@ static inline void finish_lock_switch(st
 #define WF_SYNC		0x01		/* waker goes to sleep after wakeup */
 #define WF_FORK		0x02		/* child wakeup after fork */
 #define WF_MIGRATED	0x4		/* internal use, task got migrated */
+#define WF_LOCK_SLEEPER	0x08		/* wakeup spinlock "sleeper" */
 
 /*
  * To aid in avoiding the subversion of "niceness" due to uneven distribution