cpuset,mm: update tasks' mems_allowed in time

Fix allocating page cache/slab object on the unallowed node when memory
spread is set by updating tasks' mems_allowed after its cpuset's mems is
changed.

In order to update tasks' mems_allowed in time, we must modify the code of
memory policy.  Because the memory policy is applied in the process's
context originally.  After applying this patch, one task directly
manipulates anothers mems_allowed, and we use alloc_lock in the
task_struct to protect mems_allowed and memory policy of the task.

But in the fast path, we didn't use lock to protect them, because adding a
lock may lead to performance regression.  But if we don't add a lock,the
task might see no nodes when changing cpuset's mems_allowed to some
non-overlapping set.  In order to avoid it, we set all new allowed nodes,
then clear newly disallowed ones.

[lee.schermerhorn@hp.com:
  The rework of mpol_new() to extract the adjusting of the node mask to
  apply cpuset and mpol flags "context" breaks set_mempolicy() and mbind()
  with MPOL_PREFERRED and a NULL nodemask--i.e., explicit local
  allocation.  Fix this by adding the check for MPOL_PREFERRED and empty
  node mask to mpol_new_mpolicy().

  Remove the now unneeded 'nodes = NULL' from mpol_new().

  Note that mpol_new_mempolicy() is always called with a non-NULL
  'nodes' parameter now that it has been removed from mpol_new().
  Therefore, we don't need to test nodes for NULL before testing it for
  'empty'.  However, just to be extra paranoid, add a VM_BUG_ON() to
  verify this assumption.]
[lee.schermerhorn@hp.com:

  I don't think the function name 'mpol_new_mempolicy' is descriptive
  enough to differentiate it from mpol_new().

  This function applies cpuset set context, usually constraining nodes
  to those allowed by the cpuset.  However, when the 'RELATIVE_NODES flag
  is set, it also translates the nodes.  So I settled on
  'mpol_set_nodemask()', because the comment block for mpol_new() mentions
  that we need to call this function to "set nodes".

  Some additional minor line length, whitespace and typo cleanup.]
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Paul Menage <menage@google.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

include/linux/cpuset.h

@@ -18,7 +18,6 @@
 
 extern int number_of_cpusets;	/* How many cpusets are defined in system? */
 
-extern int cpuset_init_early(void);
 extern int cpuset_init(void);
 extern void cpuset_init_smp(void);
 extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask);
@@ -27,7 +26,6 @@ extern void cpuset_cpus_allowed_locked(struct task_struct *p,
 extern nodemask_t cpuset_mems_allowed(struct task_struct *p);
 #define cpuset_current_mems_allowed (current->mems_allowed)
 void cpuset_init_current_mems_allowed(void);
-void cpuset_update_task_memory_state(void);
 int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask);
 
 extern int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask);
@@ -92,9 +90,13 @@ extern void rebuild_sched_domains(void);
 
 extern void cpuset_print_task_mems_allowed(struct task_struct *p);
 
+static inline void set_mems_allowed(nodemask_t nodemask)
+{
+	current->mems_allowed = nodemask;
+}
+
 #else /* !CONFIG_CPUSETS */
 
-static inline int cpuset_init_early(void) { return 0; }
 static inline int cpuset_init(void) { return 0; }
 static inline void cpuset_init_smp(void) {}
 
@@ -116,7 +118,6 @@ static inline nodemask_t cpuset_mems_allowed(struct task_struct *p)
 
 #define cpuset_current_mems_allowed (node_states[N_HIGH_MEMORY])
 static inline void cpuset_init_current_mems_allowed(void) {}
-static inline void cpuset_update_task_memory_state(void) {}
 
 static inline int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask)
 {
@@ -188,6 +189,10 @@ static inline void cpuset_print_task_mems_allowed(struct task_struct *p)
 {
 }
 
+static inline void set_mems_allowed(nodemask_t nodemask)
+{
+}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */

include/linux/sched.h

@@ -1318,7 +1318,8 @@ struct task_struct {
 /* Thread group tracking */
    	u32 parent_exec_id;
    	u32 self_exec_id;
-/* Protection of (de-)allocation: mm, files, fs, tty, keyrings */
+/* Protection of (de-)allocation: mm, files, fs, tty, keyrings, mems_allowed,
+ * mempolicy */
 	spinlock_t alloc_lock;
 
 #ifdef CONFIG_GENERIC_HARDIRQS
@@ -1386,8 +1387,7 @@ struct task_struct {
 	cputime_t acct_timexpd;	/* stime + utime since last update */
 #endif
 #ifdef CONFIG_CPUSETS
-	nodemask_t mems_allowed;
-	int cpuset_mems_generation;
+	nodemask_t mems_allowed;	/* Protected by alloc_lock */
 	int cpuset_mem_spread_rotor;
 #endif
 #ifdef CONFIG_CGROUPS
@@ -1410,7 +1410,7 @@ struct task_struct {
 	struct list_head perf_counter_list;
 #endif
 #ifdef CONFIG_NUMA
-	struct mempolicy *mempolicy;
+	struct mempolicy *mempolicy;	/* Protected by alloc_lock */
 	short il_next;
 #endif
 	atomic_t fs_excl;	/* holding fs exclusive resources */

init/main.c

@@ -670,7 +670,6 @@ asmlinkage void __init start_kernel(void)
 		initrd_start = 0;
 	}
 #endif
-	cpuset_init_early();
 	page_cgroup_init();
 	enable_debug_pagealloc();
 	cpu_hotplug_init();
@@ -867,6 +866,11 @@ static noinline int init_post(void)
 static int __init kernel_init(void * unused)
 {
 	lock_kernel();
+
+	/*
+	 * init can allocate pages on any node
+	 */
+	set_mems_allowed(node_possible_map);
 	/*
 	 * init can run on any cpu.
 	 */

kernel/kthread.c

@@ -9,6 +9,7 @@
 #include <linux/kthread.h>
 #include <linux/completion.h>
 #include <linux/err.h>
+#include <linux/cpuset.h>
 #include <linux/unistd.h>
 #include <linux/file.h>
 #include <linux/module.h>
@@ -236,6 +237,7 @@ int kthreadd(void *unused)
 	ignore_signals(tsk);
 	set_user_nice(tsk, KTHREAD_NICE_LEVEL);
 	set_cpus_allowed_ptr(tsk, cpu_all_mask);
+	set_mems_allowed(node_possible_map);
 
 	current->flags |= PF_NOFREEZE | PF_FREEZER_NOSIG;
 

mm/mempolicy.c

@@ -182,13 +182,54 @@ static int mpol_new_bind(struct mempolicy *pol, const nodemask_t *nodes)
 	return 0;
 }
 
-/* Create a new policy */
+/*
+ * mpol_set_nodemask is called after mpol_new() to set up the nodemask, if
+ * any, for the new policy.  mpol_new() has already validated the nodes
+ * parameter with respect to the policy mode and flags.  But, we need to
+ * handle an empty nodemask with MPOL_PREFERRED here.
+ *
+ * Must be called holding task's alloc_lock to protect task's mems_allowed
+ * and mempolicy.  May also be called holding the mmap_semaphore for write.
+ */
+static int mpol_set_nodemask(struct mempolicy *pol, const nodemask_t *nodes)
+{
+	nodemask_t cpuset_context_nmask;
+	int ret;
+
+	/* if mode is MPOL_DEFAULT, pol is NULL. This is right. */
+	if (pol == NULL)
+		return 0;
+
+	VM_BUG_ON(!nodes);
+	if (pol->mode == MPOL_PREFERRED && nodes_empty(*nodes))
+		nodes = NULL;	/* explicit local allocation */
+	else {
+		if (pol->flags & MPOL_F_RELATIVE_NODES)
+			mpol_relative_nodemask(&cpuset_context_nmask, nodes,
+					       &cpuset_current_mems_allowed);
+		else
+			nodes_and(cpuset_context_nmask, *nodes,
+				  cpuset_current_mems_allowed);
+		if (mpol_store_user_nodemask(pol))
+			pol->w.user_nodemask = *nodes;
+		else
+			pol->w.cpuset_mems_allowed =
+						cpuset_current_mems_allowed;
+	}
+
+	ret = mpol_ops[pol->mode].create(pol,
+				nodes ? &cpuset_context_nmask : NULL);
+	return ret;
+}
+
+/*
+ * This function just creates a new policy, does some check and simple
+ * initialization. You must invoke mpol_set_nodemask() to set nodes.
+ */
 static struct mempolicy *mpol_new(unsigned short mode, unsigned short flags,
 				  nodemask_t *nodes)
 {
 	struct mempolicy *policy;
-	nodemask_t cpuset_context_nmask;
-	int ret;
 
 	pr_debug("setting mode %d flags %d nodes[0] %lx\n",
 		 mode, flags, nodes ? nodes_addr(*nodes)[0] : -1);
@@ -210,7 +251,6 @@ static struct mempolicy *mpol_new(unsigned short mode, unsigned short flags,
 			if (((flags & MPOL_F_STATIC_NODES) ||
 			     (flags & MPOL_F_RELATIVE_NODES)))
 				return ERR_PTR(-EINVAL);
-			nodes = NULL;	/* flag local alloc */
 		}
 	} else if (nodes_empty(*nodes))
 		return ERR_PTR(-EINVAL);
@@ -221,30 +261,6 @@ static struct mempolicy *mpol_new(unsigned short mode, unsigned short flags,
 	policy->mode = mode;
 	policy->flags = flags;
 
-	if (nodes) {
-		/*
-		 * cpuset related setup doesn't apply to local allocation
-		 */
-		cpuset_update_task_memory_state();
-		if (flags & MPOL_F_RELATIVE_NODES)
-			mpol_relative_nodemask(&cpuset_context_nmask, nodes,
-					       &cpuset_current_mems_allowed);
-		else
-			nodes_and(cpuset_context_nmask, *nodes,
-				  cpuset_current_mems_allowed);
-		if (mpol_store_user_nodemask(policy))
-			policy->w.user_nodemask = *nodes;
-		else
-			policy->w.cpuset_mems_allowed =
-						cpuset_mems_allowed(current);
-	}
-
-	ret = mpol_ops[mode].create(policy,
-				nodes ? &cpuset_context_nmask : NULL);
-	if (ret < 0) {
-		kmem_cache_free(policy_cache, policy);
-		return ERR_PTR(ret);
-	}
 	return policy;
 }
 
@@ -324,6 +340,8 @@ static void mpol_rebind_policy(struct mempolicy *pol,
 /*
  * Wrapper for mpol_rebind_policy() that just requires task
  * pointer, and updates task mempolicy.
+ *
+ * Called with task's alloc_lock held.
  */
 
 void mpol_rebind_task(struct task_struct *tsk, const nodemask_t *new)
@@ -600,8 +618,9 @@ static void mpol_set_task_struct_flag(void)
 static long do_set_mempolicy(unsigned short mode, unsigned short flags,
 			     nodemask_t *nodes)
 {
-	struct mempolicy *new;
+	struct mempolicy *new, *old;
 	struct mm_struct *mm = current->mm;
+	int ret;
 
 	new = mpol_new(mode, flags, nodes);
 	if (IS_ERR(new))
@@ -615,20 +634,33 @@ static long do_set_mempolicy(unsigned short mode, unsigned short flags,
 	 */
 	if (mm)
 		down_write(&mm->mmap_sem);
-	mpol_put(current->mempolicy);
+	task_lock(current);
+	ret = mpol_set_nodemask(new, nodes);
+	if (ret) {
+		task_unlock(current);
+		if (mm)
+			up_write(&mm->mmap_sem);
+		mpol_put(new);
+		return ret;
+	}
+	old = current->mempolicy;
 	current->mempolicy = new;
 	mpol_set_task_struct_flag();
 	if (new && new->mode == MPOL_INTERLEAVE &&
 	    nodes_weight(new->v.nodes))
 		current->il_next = first_node(new->v.nodes);
+	task_unlock(current);
 	if (mm)
 		up_write(&mm->mmap_sem);
 
+	mpol_put(old);
 	return 0;
 }
 
 /*
  * Return nodemask for policy for get_mempolicy() query
+ *
+ * Called with task's alloc_lock held
  */
 static void get_policy_nodemask(struct mempolicy *p, nodemask_t *nodes)
 {
@@ -674,7 +706,6 @@ static long do_get_mempolicy(int *policy, nodemask_t *nmask,
 	struct vm_area_struct *vma = NULL;
 	struct mempolicy *pol = current->mempolicy;
 
-	cpuset_update_task_memory_state();
 	if (flags &
 		~(unsigned long)(MPOL_F_NODE|MPOL_F_ADDR|MPOL_F_MEMS_ALLOWED))
 		return -EINVAL;
@@ -683,7 +714,9 @@ static long do_get_mempolicy(int *policy, nodemask_t *nmask,
 		if (flags & (MPOL_F_NODE|MPOL_F_ADDR))
 			return -EINVAL;
 		*policy = 0;	/* just so it's initialized */
+		task_lock(current);
 		*nmask  = cpuset_current_mems_allowed;
+		task_unlock(current);
 		return 0;
 	}
 
@@ -738,8 +771,11 @@ static long do_get_mempolicy(int *policy, nodemask_t *nmask,
 	}
 
 	err = 0;
-	if (nmask)
+	if (nmask) {
+		task_lock(current);
 		get_policy_nodemask(pol, nmask);
+		task_unlock(current);
+	}
 
  out:
 	mpol_cond_put(pol);
@@ -979,6 +1015,14 @@ static long do_mbind(unsigned long start, unsigned long len,
 			return err;
 	}
 	down_write(&mm->mmap_sem);
+	task_lock(current);
+	err = mpol_set_nodemask(new, nmask);
+	task_unlock(current);
+	if (err) {
+		up_write(&mm->mmap_sem);
+		mpol_put(new);
+		return err;
+	}
 	vma = check_range(mm, start, end, nmask,
 			  flags | MPOL_MF_INVERT, &pagelist);
 
@@ -1545,8 +1589,6 @@ alloc_page_vma(gfp_t gfp, struct vm_area_struct *vma, unsigned long addr)
 	struct mempolicy *pol = get_vma_policy(current, vma, addr);
 	struct zonelist *zl;
 
-	cpuset_update_task_memory_state();
-
 	if (unlikely(pol->mode == MPOL_INTERLEAVE)) {
 		unsigned nid;
 
@@ -1593,8 +1635,6 @@ struct page *alloc_pages_current(gfp_t gfp, unsigned order)
 {
 	struct mempolicy *pol = current->mempolicy;
 
-	if ((gfp & __GFP_WAIT) && !in_interrupt())
-		cpuset_update_task_memory_state();
 	if (!pol || in_interrupt() || (gfp & __GFP_THISNODE))
 		pol = &default_policy;
 
@@ -1854,6 +1894,8 @@ static int shared_policy_replace(struct shared_policy *sp, unsigned long start,
  */
 void mpol_shared_policy_init(struct shared_policy *sp, struct mempolicy *mpol)
 {
+	int ret;
+
 	sp->root = RB_ROOT;		/* empty tree == default mempolicy */
 	spin_lock_init(&sp->lock);
 
@@ -1863,9 +1905,19 @@ void mpol_shared_policy_init(struct shared_policy *sp, struct mempolicy *mpol)
 
 		/* contextualize the tmpfs mount point mempolicy */
 		new = mpol_new(mpol->mode, mpol->flags, &mpol->w.user_nodemask);
+		if (IS_ERR(new)) {
 			mpol_put(mpol);	/* drop our ref on sb mpol */
-		if (IS_ERR(new))
 			return;		/* no valid nodemask intersection */
+		}
+
+		task_lock(current);
+		ret = mpol_set_nodemask(new, &mpol->w.user_nodemask);
+		task_unlock(current);
+		mpol_put(mpol);	/* drop our ref on sb mpol */
+		if (ret) {
+			mpol_put(new);
+			return;
+		}
 
 		/* Create pseudo-vma that contains just the policy */
 		memset(&pvma, 0, sizeof(struct vm_area_struct));
@@ -2086,8 +2138,19 @@ int mpol_parse_str(char *str, struct mempolicy **mpol, int no_context)
 	new = mpol_new(mode, mode_flags, &nodes);
 	if (IS_ERR(new))
 		err = 1;
-	else if (no_context)
-		new->w.user_nodemask = nodes;	/* save for contextualization */
+	else {
+		int ret;
+
+		task_lock(current);
+		ret = mpol_set_nodemask(new, &nodes);
+		task_unlock(current);
+		if (ret)
+			err = 1;
+		else if (no_context) {
+			/* save for contextualization */
+			new->w.user_nodemask = nodes;
+		}
+	}
 
 out:
 	/* Restore string for error message */

mm/page_alloc.c

@@ -1569,10 +1569,7 @@ __alloc_pages_internal(gfp_t gfp_mask, unsigned int order,
 
 	/* We now go into synchronous reclaim */
 	cpuset_memory_pressure_bump();
-	/*
-	 * The task's cpuset might have expanded its set of allowable nodes
-	 */
-	cpuset_update_task_memory_state();
+
 	p->flags |= PF_MEMALLOC;
 
 	lockdep_set_current_reclaim_state(gfp_mask);