- Fix a race in sched_switch() of sched_4bsd.

In the case of the thread being on a sleepqueue or a turnstile, the
  sched_lock was acquired (without the aid of the td_lock interface) and
  the td_lock was dropped. This was going to break locking rules on other
  threads willing to access to the thread (via the td_lock interface) and
  modify his flags (allowed as long as the container lock was different
  by the one used in sched_switch).
  In order to prevent this situation, while sched_lock is acquired there
  the td_lock gets blocked. [0]
- Merge the ULE's internal function thread_block_switch() into the global
  thread_lock_block() and make the former semantic as the default for
  thread_lock_block(). This means that thread_lock_block() will not
  disable interrupts when called (and consequently thread_unlock_block()
  will not re-enabled them when called). This should be done manually
  when necessary.
  Note, however, that ULE's thread_unblock_switch() is not reaped
  because it does reflect a difference in semantic due in ULE (the
  td_lock may not be necessarilly still blocked_lock when calling this).
  While asymmetric, it does describe a remarkable difference in semantic
  that is good to keep in mind.

[0] Reported by:	Kohji Okuno
			<okuno dot kohji at jp dot panasonic dot com>
Tested by:		Giovanni Trematerra
			<giovanni dot trematerra at gmail dot com>
MFC:			2 weeks

sys/kern/kern_mutex.c

@@ -616,7 +616,6 @@ thread_lock_block(struct thread *td)
 {
 	struct mtx *lock;
 
-	spinlock_enter();
 	THREAD_LOCK_ASSERT(td, MA_OWNED);
 	lock = td->td_lock;
 	td->td_lock = &blocked_lock;
@@ -631,7 +630,6 @@ thread_lock_unblock(struct thread *td, struct mtx *new)
 	mtx_assert(new, MA_OWNED);
 	MPASS(td->td_lock == &blocked_lock);
 	atomic_store_rel_ptr((volatile void *)&td->td_lock, (uintptr_t)new);
-	spinlock_exit();
 }
 
 void

sys/kern/sched_4bsd.c

@@ -920,9 +920,11 @@ sched_sleep(struct thread *td, int pri)
 void
 sched_switch(struct thread *td, struct thread *newtd, int flags)
 {
+	struct mtx *tmtx;
 	struct td_sched *ts;
 	struct proc *p;
 
+	tmtx = NULL;
 	ts = td->td_sched;
 	p = td->td_proc;
 
@@ -931,10 +933,11 @@ sched_switch(struct thread *td, struct thread *newtd, int flags)
 	/* 
 	 * Switch to the sched lock to fix things up and pick
 	 * a new thread.
+	 * Block the td_lock in order to avoid breaking the critical path.
 	 */
 	if (td->td_lock != &sched_lock) {
 		mtx_lock_spin(&sched_lock);
-		thread_unlock(td);
+		tmtx = thread_lock_block(td);
 	}
 
 	if ((td->td_flags & TDF_NOLOAD) == 0)
@@ -1004,7 +1007,7 @@ sched_switch(struct thread *td, struct thread *newtd, int flags)
 			(*dtrace_vtime_switch_func)(newtd);
 #endif
 
-		cpu_switch(td, newtd, td->td_lock);
+		cpu_switch(td, newtd, tmtx != NULL ? tmtx : td->td_lock);
 		lock_profile_obtain_lock_success(&sched_lock.lock_object,
 		    0, 0, __FILE__, __LINE__);
 		/*

sys/kern/sched_ule.c

@@ -301,7 +301,6 @@ static int sched_pickcpu(struct thread *, int);
 static void sched_balance(void);
 static int sched_balance_pair(struct tdq *, struct tdq *);
 static inline struct tdq *sched_setcpu(struct thread *, int, int);
-static inline struct mtx *thread_block_switch(struct thread *);
 static inline void thread_unblock_switch(struct thread *, struct mtx *);
 static struct mtx *sched_switch_migrate(struct tdq *, struct thread *, int);
 static int sysctl_kern_sched_topology_spec(SYSCTL_HANDLER_ARGS);
@@ -1106,9 +1105,11 @@ sched_setcpu(struct thread *td, int cpu, int flags)
 	 * The hard case, migration, we need to block the thread first to
 	 * prevent order reversals with other cpus locks.
 	 */
+	spinlock_enter();
 	thread_lock_block(td);
 	TDQ_LOCK(tdq);
 	thread_lock_unblock(td, TDQ_LOCKPTR(tdq));
+	spinlock_exit();
 	return (tdq);
 }
 
@@ -1714,23 +1715,6 @@ sched_unlend_user_prio(struct thread *td, u_char prio)
 	}
 }
 
-/*
- * Block a thread for switching.  Similar to thread_block() but does not
- * bump the spin count.
- */
-static inline struct mtx *
-thread_block_switch(struct thread *td)
-{
-	struct mtx *lock;
-
-	THREAD_LOCK_ASSERT(td, MA_OWNED);
-	lock = td->td_lock;
-	td->td_lock = &blocked_lock;
-	mtx_unlock_spin(lock);
-
-	return (lock);
-}
-
 /*
  * Handle migration from sched_switch().  This happens only for
  * cpu binding.
@@ -1749,7 +1733,7 @@ sched_switch_migrate(struct tdq *tdq, struct thread *td, int flags)
 	 * not holding either run-queue lock.
 	 */
 	spinlock_enter();
-	thread_block_switch(td);	/* This releases the lock on tdq. */
+	thread_lock_block(td);	/* This releases the lock on tdq. */
 
 	/*
 	 * Acquire both run-queue locks before placing the thread on the new
@@ -1769,7 +1753,8 @@ sched_switch_migrate(struct tdq *tdq, struct thread *td, int flags)
 }
 
 /*
- * Release a thread that was blocked with thread_block_switch().
+ * Variadic version of thread_lock_unblock() that does not assume td_lock
+ * is blocked.
  */
 static inline void
 thread_unblock_switch(struct thread *td, struct mtx *mtx)
@@ -1825,7 +1810,7 @@ sched_switch(struct thread *td, struct thread *newtd, int flags)
 	} else {
 		/* This thread must be going to sleep. */
 		TDQ_LOCK(tdq);
-		mtx = thread_block_switch(td);
+		mtx = thread_lock_block(td);
 		tdq_load_rem(tdq, td);
 	}
 	/*