5 files changed, 153 insertions, 15 deletions

diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 41a59afc70fa..78b4bc64c006 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -23,6 +23,12 @@ void reset_vma_resv_huge_pages(struct vm_area_struct *vma);
 int hugetlb_sysctl_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
 int hugetlb_overcommit_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
 int hugetlb_treat_movable_handler(struct ctl_table *, int, void __user *, size_t *, loff_t *);
+
+#ifdef CONFIG_NUMA
+int hugetlb_mempolicy_sysctl_handler(struct ctl_table *, int,
+                                        void __user *, size_t *, loff_t *);
+#endif
+
 int copy_hugetlb_page_range(struct mm_struct *, struct mm_struct *, struct vm_area_struct *);
 int follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *,
                         struct page **, struct vm_area_struct **,
diff --git a/include/linux/mempolicy.h b/include/linux/mempolicy.h
index 085c903fe0f1..1cc966cd3e5f 100644
--- a/include/linux/mempolicy.h
+++ b/include/linux/mempolicy.h
@@ -201,6 +201,7 @@ extern void mpol_fix_fork_child_flag(struct task_struct *p);
 extern struct zonelist *huge_zonelist(struct vm_area_struct *vma,
                                 unsigned long addr, gfp_t gfp_flags,
                                 struct mempolicy **mpol, nodemask_t **nodemask);
+extern bool init_nodemask_of_mempolicy(nodemask_t *mask);
 extern unsigned slab_node(struct mempolicy *policy);
 
 extern enum zone_type policy_zone;
@@ -328,6 +329,8 @@ static inline struct zonelist *huge_zonelist(struct vm_area_struct *vma,
         return node_zonelist(0, gfp_flags);
 }
 
+static inline bool init_nodemask_of_mempolicy(nodemask_t *m) { return false; }
+
 static inline int do_migrate_pages(struct mm_struct *mm,
                         const nodemask_t *from_nodes,
                         const nodemask_t *to_nodes, int flags)
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 554ac4894f0f..60fc93131095 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -1051,7 +1051,7 @@ static struct ctl_table vm_table[] = {
                 .extra2         = &one_hundred,
         },
 #ifdef CONFIG_HUGETLB_PAGE
-         {
+        {
                 .procname       = "nr_hugepages",
                 .data           = NULL,
                 .maxlen         = sizeof(unsigned long),
@@ -1059,7 +1059,18 @@ static struct ctl_table vm_table[] = {
                 .proc_handler   = hugetlb_sysctl_handler,
                 .extra1         = (void *)&hugetlb_zero,
                 .extra2         = (void *)&hugetlb_infinity,
-         },
+        },
+#ifdef CONFIG_NUMA
+        {
+                .procname       = "nr_hugepages_mempolicy",
+                .data           = NULL,
+                .maxlen         = sizeof(unsigned long),
+                .mode           = 0644,
+                .proc_handler   = &hugetlb_mempolicy_sysctl_handler,
+                .extra1         = (void *)&hugetlb_zero,
+                .extra2         = (void *)&hugetlb_infinity,
+        },
+#endif
          {
                 .procname       = "hugetlb_shm_group",
                 .data           = &sysctl_hugetlb_shm_group,
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 324d1abae876..1125d818ea06 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1330,29 +1330,71 @@ static struct hstate *kobj_to_hstate(struct kobject *kobj)
         return NULL;
 }
 
-static ssize_t nr_hugepages_show(struct kobject *kobj,
+static ssize_t nr_hugepages_show_common(struct kobject *kobj,
                                         struct kobj_attribute *attr, char *buf)
 {
         struct hstate *h = kobj_to_hstate(kobj);
         return sprintf(buf, "%lu\n", h->nr_huge_pages);
 }
-static ssize_t nr_hugepages_store(struct kobject *kobj,
-                struct kobj_attribute *attr, const char *buf, size_t count)
+static ssize_t nr_hugepages_store_common(bool obey_mempolicy,
+                        struct kobject *kobj, struct kobj_attribute *attr,
+                        const char *buf, size_t len)
 {
         int err;
-        unsigned long input;
+        unsigned long count;
         struct hstate *h = kobj_to_hstate(kobj);
+        NODEMASK_ALLOC(nodemask_t, nodes_allowed);
 
-        err = strict_strtoul(buf, 10, &input);
+        err = strict_strtoul(buf, 10, &count);
         if (err)
                 return 0;
 
-        h->max_huge_pages = set_max_huge_pages(h, input, &node_online_map);
+        if (!(obey_mempolicy && init_nodemask_of_mempolicy(nodes_allowed))) {
+                NODEMASK_FREE(nodes_allowed);
+                nodes_allowed = &node_online_map;
+        }
+        h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
 
-        return count;
+        if (nodes_allowed != &node_online_map)
+                NODEMASK_FREE(nodes_allowed);
+
+        return len;
+}
+
+static ssize_t nr_hugepages_show(struct kobject *kobj,
+                                       struct kobj_attribute *attr, char *buf)
+{
+        return nr_hugepages_show_common(kobj, attr, buf);
+}
+
+static ssize_t nr_hugepages_store(struct kobject *kobj,
+               struct kobj_attribute *attr, const char *buf, size_t len)
+{
+        return nr_hugepages_store_common(false, kobj, attr, buf, len);
 }
 HSTATE_ATTR(nr_hugepages);
 
+#ifdef CONFIG_NUMA
+
+/*
+ * hstate attribute for optionally mempolicy-based constraint on persistent
+ * huge page alloc/free.
+ */
+static ssize_t nr_hugepages_mempolicy_show(struct kobject *kobj,
+                                       struct kobj_attribute *attr, char *buf)
+{
+        return nr_hugepages_show_common(kobj, attr, buf);
+}
+
+static ssize_t nr_hugepages_mempolicy_store(struct kobject *kobj,
+               struct kobj_attribute *attr, const char *buf, size_t len)
+{
+        return nr_hugepages_store_common(true, kobj, attr, buf, len);
+}
+HSTATE_ATTR(nr_hugepages_mempolicy);
+#endif
+
+
 static ssize_t nr_overcommit_hugepages_show(struct kobject *kobj,
                                         struct kobj_attribute *attr, char *buf)
 {
@@ -1408,6 +1450,9 @@ static struct attribute *hstate_attrs[] = {
         &free_hugepages_attr.attr,
         &resv_hugepages_attr.attr,
         &surplus_hugepages_attr.attr,
+#ifdef CONFIG_NUMA
+        &nr_hugepages_mempolicy_attr.attr,
+#endif
         NULL,
 };
 
@@ -1574,9 +1619,9 @@ static unsigned int cpuset_mems_nr(unsigned int *array)
 }
 
 #ifdef CONFIG_SYSCTL
-int hugetlb_sysctl_handler(struct ctl_table *table, int write,
-                           void __user *buffer,
-                           size_t *length, loff_t *ppos)
+static int hugetlb_sysctl_handler_common(bool obey_mempolicy,
+                         struct ctl_table *table, int write,
+                         void __user *buffer, size_t *length, loff_t *ppos)
 {
         struct hstate *h = &default_hstate;
         unsigned long tmp;
@@ -1588,13 +1633,39 @@ int hugetlb_sysctl_handler(struct ctl_table *table, int write,
         table->maxlen = sizeof(unsigned long);
         proc_doulongvec_minmax(table, write, buffer, length, ppos);
 
-        if (write)
-                h->max_huge_pages = set_max_huge_pages(h, tmp,
-                                                        &node_online_map);
+        if (write) {
+                NODEMASK_ALLOC(nodemask_t, nodes_allowed);
+                if (!(obey_mempolicy &&
+                               init_nodemask_of_mempolicy(nodes_allowed))) {
+                        NODEMASK_FREE(nodes_allowed);
+                        nodes_allowed = &node_states[N_HIGH_MEMORY];
+                }
+                h->max_huge_pages = set_max_huge_pages(h, tmp, nodes_allowed);
+
+                if (nodes_allowed != &node_states[N_HIGH_MEMORY])
+                        NODEMASK_FREE(nodes_allowed);
+        }
 
         return 0;
 }
 
+int hugetlb_sysctl_handler(struct ctl_table *table, int write,
+                          void __user *buffer, size_t *length, loff_t *ppos)
+{
+
+        return hugetlb_sysctl_handler_common(false, table, write,
+                                                        buffer, length, ppos);
+}
+
+#ifdef CONFIG_NUMA
+int hugetlb_mempolicy_sysctl_handler(struct ctl_table *table, int write,
+                          void __user *buffer, size_t *length, loff_t *ppos)
+{
+        return hugetlb_sysctl_handler_common(true, table, write,
+                                                        buffer, length, ppos);
+}
+#endif /* CONFIG_NUMA */
+
 int hugetlb_treat_movable_handler(struct ctl_table *table, int write,
                         void __user *buffer,
                         size_t *length, loff_t *ppos)
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 0f89eabbaf3e..f11fdad06204 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -1568,6 +1568,53 @@ struct zonelist *huge_zonelist(struct vm_area_struct *vma, unsigned long addr,
         }
         return zl;
 }
+
+/*
+ * init_nodemask_of_mempolicy
+ *
+ * If the current task's mempolicy is "default" [NULL], return 'false'
+ * to indicate default policy.  Otherwise, extract the policy nodemask
+ * for 'bind' or 'interleave' policy into the argument nodemask, or
+ * initialize the argument nodemask to contain the single node for
+ * 'preferred' or 'local' policy and return 'true' to indicate presence
+ * of non-default mempolicy.
+ *
+ * We don't bother with reference counting the mempolicy [mpol_get/put]
+ * because the current task is examining it's own mempolicy and a task's
+ * mempolicy is only ever changed by the task itself.
+ *
+ * N.B., it is the caller's responsibility to free a returned nodemask.
+ */
+bool init_nodemask_of_mempolicy(nodemask_t *mask)
+{
+        struct mempolicy *mempolicy;
+        int nid;
+
+        if (!(mask && current->mempolicy))
+                return false;
+
+        mempolicy = current->mempolicy;
+        switch (mempolicy->mode) {
+        case MPOL_PREFERRED:
+                if (mempolicy->flags & MPOL_F_LOCAL)
+                        nid = numa_node_id();
+                else
+                        nid = mempolicy->v.preferred_node;
+                init_nodemask_of_node(mask, nid);
+                break;
+
+        case MPOL_BIND:
+                /* Fall through */
+        case MPOL_INTERLEAVE:
+                *mask =  mempolicy->v.nodes;
+                break;
+
+        default:
+                BUG();
+        }
+
+        return true;
+}
 #endif
 
 /* Allocate a page in interleaved policy.