[PATCH] cpuset oom lock fix

The problem, reported in:

  http://bugzilla.kernel.org/show_bug.cgi?id=5859

and by various other email messages and lkml posts is that the cpuset hook
in the oom (out of memory) code can try to take a cpuset semaphore while
holding the tasklist_lock (a spinlock).

One must not sleep while holding a spinlock.

The fix seems easy enough - move the cpuset semaphore region outside the
tasklist_lock region.

This required a few lines of mechanism to implement.  The oom code where
the locking needs to be changed does not have access to the cpuset locks,
which are internal to kernel/cpuset.c only.  So I provided a couple more
cpuset interface routines, available to the rest of the kernel, which
simple take and drop the lock needed here (cpusets callback_sem).

Signed-off-by: Paul Jackson <pj@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

include/linux/cpuset.h

@@ -48,6 +48,9 @@ extern void __cpuset_memory_pressure_bump(void);
 extern struct file_operations proc_cpuset_operations;
 extern char *cpuset_task_status_allowed(struct task_struct *task, char *buffer);
 
+extern void cpuset_lock(void);
+extern void cpuset_unlock(void);
+
 #else /* !CONFIG_CPUSETS */
 
 static inline int cpuset_init_early(void) { return 0; }
@@ -93,6 +96,9 @@ static inline char *cpuset_task_status_allowed(struct task_struct *task,
 	return buffer;
 }
 
+static inline void cpuset_lock(void) {}
+static inline void cpuset_unlock(void) {}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */

kernel/cpuset.c

@@ -2149,6 +2149,33 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
 	return allowed;
 }
 
+/**
+ * cpuset_lock - lock out any changes to cpuset structures
+ *
+ * The out of memory (oom) code needs to lock down cpusets
+ * from being changed while it scans the tasklist looking for a
+ * task in an overlapping cpuset.  Expose callback_sem via this
+ * cpuset_lock() routine, so the oom code can lock it, before
+ * locking the task list.  The tasklist_lock is a spinlock, so
+ * must be taken inside callback_sem.
+ */
+
+void cpuset_lock(void)
+{
+	down(&callback_sem);
+}
+
+/**
+ * cpuset_unlock - release lock on cpuset changes
+ *
+ * Undo the lock taken in a previous cpuset_lock() call.
+ */
+
+void cpuset_unlock(void)
+{
+	up(&callback_sem);
+}
+
 /**
  * cpuset_excl_nodes_overlap - Do we overlap @p's mem_exclusive ancestors?
  * @p: pointer to task_struct of some other task.
@@ -2158,16 +2185,14 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
  * determine if task @p's memory usage might impact the memory
  * available to the current task.
  *
- * Acquires callback_sem - not suitable for calling from a fast path.
+ * Call while holding callback_sem.
  **/
 
 int cpuset_excl_nodes_overlap(const struct task_struct *p)
 {
 	const struct cpuset *cs1, *cs2;	/* my and p's cpuset ancestors */
 	int overlap = 0;		/* do cpusets overlap? */
 
-	down(&callback_sem);
-
 	task_lock(current);
 	if (current->flags & PF_EXITING) {
 		task_unlock(current);
@@ -2186,8 +2211,6 @@ int cpuset_excl_nodes_overlap(const struct task_struct *p)
 
 	overlap = nodes_intersects(cs1->mems_allowed, cs2->mems_allowed);
 done:
-	up(&callback_sem);
-
 	return overlap;
 }
 

mm/oom_kill.c

@@ -274,6 +274,7 @@ void out_of_memory(gfp_t gfp_mask, int order)
 		show_mem();
 	}
 
+	cpuset_lock();
 	read_lock(&tasklist_lock);
 retry:
 	p = select_bad_process();
@@ -284,6 +285,7 @@ void out_of_memory(gfp_t gfp_mask, int order)
 	/* Found nothing?!?! Either we hang forever, or we panic. */
 	if (!p) {
 		read_unlock(&tasklist_lock);
+		cpuset_unlock();
 		panic("Out of memory and no killable processes...\n");
 	}
 
@@ -293,6 +295,7 @@ void out_of_memory(gfp_t gfp_mask, int order)
 
  out:
 	read_unlock(&tasklist_lock);
+	cpuset_unlock();
 	if (mm)
 		mmput(mm);