7 files changed, 88 insertions, 15 deletions

diff --git a/mm/filemap.c b/mm/filemap.c
index 768687f1d46b..5d6e4c2000dc 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -1030,8 +1030,8 @@ __generic_file_aio_read(struct kiocb *iocb, const struct iovec *iov,
                         desc.error = 0;
                         do_generic_file_read(filp,ppos,&desc,file_read_actor);
                         retval += desc.written;
-                        if (!retval) {
-                                retval = desc.error;
+                        if (desc.error) {
+                                retval = retval ?: desc.error;
                                 break;
                         }
                 }
diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 2076b1542b8a..5abc57c2b8bd 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -457,6 +457,7 @@ long do_get_mempolicy(int *policy, nodemask_t *nmask,
         struct vm_area_struct *vma = NULL;
         struct mempolicy *pol = current->mempolicy;
 
+        cpuset_update_current_mems_allowed();
         if (flags & ~(unsigned long)(MPOL_F_NODE|MPOL_F_ADDR))
                 return -EINVAL;
         if (flags & MPOL_F_ADDR) {
@@ -1206,3 +1207,66 @@ void numa_default_policy(void)
 {
         do_set_mempolicy(MPOL_DEFAULT, NULL);
 }
+
+/* Migrate a policy to a different set of nodes */
+static void rebind_policy(struct mempolicy *pol, const nodemask_t *old,
+                                                        const nodemask_t *new)
+{
+        nodemask_t tmp;
+
+        if (!pol)
+                return;
+
+        switch (pol->policy) {
+        case MPOL_DEFAULT:
+                break;
+        case MPOL_INTERLEAVE:
+                nodes_remap(tmp, pol->v.nodes, *old, *new);
+                pol->v.nodes = tmp;
+                current->il_next = node_remap(current->il_next, *old, *new);
+                break;
+        case MPOL_PREFERRED:
+                pol->v.preferred_node = node_remap(pol->v.preferred_node,
+                                                                *old, *new);
+                break;
+        case MPOL_BIND: {
+                nodemask_t nodes;
+                struct zone **z;
+                struct zonelist *zonelist;
+
+                nodes_clear(nodes);
+                for (z = pol->v.zonelist->zones; *z; z++)
+                        node_set((*z)->zone_pgdat->node_id, nodes);
+                nodes_remap(tmp, nodes, *old, *new);
+                nodes = tmp;
+
+                zonelist = bind_zonelist(&nodes);
+
+                /* If no mem, then zonelist is NULL and we keep old zonelist.
+                 * If that old zonelist has no remaining mems_allowed nodes,
+                 * then zonelist_policy() will "FALL THROUGH" to MPOL_DEFAULT.
+                 */
+
+                if (zonelist) {
+                        /* Good - got mem - substitute new zonelist */
+                        kfree(pol->v.zonelist);
+                        pol->v.zonelist = zonelist;
+                }
+                break;
+        }
+        default:
+                BUG();
+                break;
+        }
+}
+
+/*
+ * Someone moved this task to different nodes.  Fixup mempolicies.
+ *
+ * TODO - fixup current->mm->vma and shmfs/tmpfs/hugetlbfs policies as well,
+ * once we have a cpuset mechanism to mark which cpuset subtree is migrating.
+ */
+void numa_policy_rebind(const nodemask_t *old, const nodemask_t *new)
+{
+        rebind_policy(current->mempolicy, old, new);
+}
diff --git a/mm/mmap.c b/mm/mmap.c
index 5ecc2cf3e1d7..320dda1778c3 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -1840,7 +1840,7 @@ asmlinkage long sys_munmap(unsigned long addr, size_t len)
 
 static inline void verify_mm_writelocked(struct mm_struct *mm)
 {
-#ifdef CONFIG_DEBUG_KERNEL
+#ifdef CONFIG_DEBUG_VM
         if (unlikely(down_read_trylock(&mm->mmap_sem))) {
                 WARN_ON(1);
                 up_read(&mm->mmap_sem);
diff --git a/mm/pdflush.c b/mm/pdflush.c
index d6781951267e..52822c98c489 100644
--- a/mm/pdflush.c
+++ b/mm/pdflush.c
@@ -20,6 +20,7 @@
 #include <linux/fs.h>           // Needed by writeback.h
 #include <linux/writeback.h>    // Prototypes pdflush_operation()
 #include <linux/kthread.h>
+#include <linux/cpuset.h>
 
 
 /*
@@ -170,12 +171,24 @@ static int __pdflush(struct pdflush_work *my_work)
 static int pdflush(void *dummy)
 {
         struct pdflush_work my_work;
+        cpumask_t cpus_allowed;
 
         /*
          * pdflush can spend a lot of time doing encryption via dm-crypt.  We
          * don't want to do that at keventd's priority.
          */
         set_user_nice(current, 0);
+
+        /*
+         * Some configs put our parent kthread in a limited cpuset,
+         * which kthread() overrides, forcing cpus_allowed == CPU_MASK_ALL.
+         * Our needs are more modest - cut back to our cpusets cpus_allowed.
+         * This is needed as pdflush's are dynamically created and destroyed.
+         * The boottime pdflush's are easily placed w/o these 2 lines.
+         */
+        cpus_allowed = cpuset_cpus_allowed(current);
+        set_cpus_allowed(current, cpus_allowed);
+
         return __pdflush(&my_work);
 }
 
diff --git a/mm/swap.c b/mm/swap.c
index b89512877ec2..154ae13d8b7e 100644
--- a/mm/swap.c
+++ b/mm/swap.c
@@ -259,6 +259,8 @@ void __pagevec_release(struct pagevec *pvec)
         pagevec_reinit(pvec);
 }
 
+EXPORT_SYMBOL(__pagevec_release);
+
 /*
  * pagevec_release() for pages which are known to not be on the LRU
  *
@@ -270,7 +272,6 @@ void __pagevec_release_nonlru(struct pagevec *pvec)
         struct pagevec pages_to_free;
 
         pagevec_init(&pages_to_free, pvec->cold);
-        pages_to_free.cold = pvec->cold;
         for (i = 0; i < pagevec_count(pvec); i++) {
                 struct page *page = pvec->pages[i];
 
@@ -388,6 +389,7 @@ unsigned pagevec_lookup_tag(struct pagevec *pvec, struct address_space *mapping,
         return pagevec_count(pvec);
 }
 
+EXPORT_SYMBOL(pagevec_lookup_tag);
 
 #ifdef CONFIG_SMP
 /*
diff --git a/mm/tiny-shmem.c b/mm/tiny-shmem.c
index c13a2161bca2..b58abcf44ed6 100644
--- a/mm/tiny-shmem.c
+++ b/mm/tiny-shmem.c
@@ -31,11 +31,14 @@ static struct vfsmount *shm_mnt;
 
 static int __init init_tmpfs(void)
 {
-        register_filesystem(&tmpfs_fs_type);
+        BUG_ON(register_filesystem(&tmpfs_fs_type) != 0);
+
 #ifdef CONFIG_TMPFS
         devfs_mk_dir("shm");
 #endif
         shm_mnt = kern_mount(&tmpfs_fs_type);
+        BUG_ON(IS_ERR(shm_mnt));
+
         return 0;
 }
 module_init(init_tmpfs)
diff --git a/mm/truncate.c b/mm/truncate.c
index 60c8764bfac2..29c18f68dc35 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -13,18 +13,9 @@
 #include <linux/pagemap.h>
 #include <linux/pagevec.h>
 #include <linux/buffer_head.h>  /* grr. try_to_release_page,
-                                   block_invalidatepage */
+                                   do_invalidatepage */
 
 
-static int do_invalidatepage(struct page *page, unsigned long offset)
-{
-        int (*invalidatepage)(struct page *, unsigned long);
-        invalidatepage = page->mapping->a_ops->invalidatepage;
-        if (invalidatepage == NULL)
-                invalidatepage = block_invalidatepage;
-        return (*invalidatepage)(page, offset);
-}
-
 static inline void truncate_partial_page(struct page *page, unsigned partial)
 {
         memclear_highpage_flush(page, partial, PAGE_CACHE_SIZE-partial);