Merge with master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6.git

12 files changed, 202 insertions, 162 deletions

diff --git a/mm/filemap.c b/mm/filemap.c
index 93595c327bb..d5fdae2eb18 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -123,8 +123,7 @@ void remove_from_page_cache(struct page *page)
 {
 	struct address_space *mapping = page->mapping;
 
-	if (unlikely(!PageLocked(page)))
-		PAGE_BUG(page);
+	BUG_ON(!PageLocked(page));
 
 	write_lock_irq(&mapping->tree_lock);
 	__remove_from_page_cache(page);
@@ -139,7 +138,25 @@ static int sync_page(void *word)
 	page = container_of((page_flags_t *)word, struct page, flags);
 
 	/*
-	 * FIXME, fercrissake.  What is this barrier here for?
+	 * page_mapping() is being called without PG_locked held.
+	 * Some knowledge of the state and use of the page is used to
+	 * reduce the requirements down to a memory barrier.
+	 * The danger here is of a stale page_mapping() return value
+	 * indicating a struct address_space different from the one it's
+	 * associated with when it is associated with one.
+	 * After smp_mb(), it's either the correct page_mapping() for
+	 * the page, or an old page_mapping() and the page's own
+	 * page_mapping() has gone NULL.
+	 * The ->sync_page() address_space operation must tolerate
+	 * page_mapping() going NULL. By an amazing coincidence,
+	 * this comes about because none of the users of the page
+	 * in the ->sync_page() methods make essential use of the
+	 * page_mapping(), merely passing the page down to the backing
+	 * device's unplug functions when it's non-NULL, which in turn
+	 * ignore it for all cases but swap, where only page->private is
+	 * of interest. When page_mapping() does go NULL, the entire
+	 * call stack gracefully ignores the page and returns.
+	 * -- wli
 	 */
 	smp_mb();
 	mapping = page_mapping(page);
@@ -152,9 +169,10 @@ static int sync_page(void *word)
 /**
  * filemap_fdatawrite_range - start writeback against all of a mapping's
  * dirty pages that lie within the byte offsets <start, end>
- * @mapping: address space structure to write
- * @start: offset in bytes where the range starts
- * @end : offset in bytes where the range ends
+ * @mapping:	address space structure to write
+ * @start:	offset in bytes where the range starts
+ * @end:	offset in bytes where the range ends
+ * @sync_mode:	enable synchronous operation
  *
  * If sync_mode is WB_SYNC_ALL then this is a "data integrity" operation, as
  * opposed to a regular memory * cleansing writeback.  The difference between
@@ -518,8 +536,8 @@ EXPORT_SYMBOL(find_trylock_page);
 /**
  * find_lock_page - locate, pin and lock a pagecache page
  *
- * @mapping - the address_space to search
- * @offset - the page index
+ * @mapping: the address_space to search
+ * @offset: the page index
  *
  * Locates the desired pagecache page, locks it, increments its reference
  * count and returns its address.
@@ -558,9 +576,9 @@ EXPORT_SYMBOL(find_lock_page);
 /**
  * find_or_create_page - locate or add a pagecache page
  *
- * @mapping - the page's address_space
- * @index - the page's index into the mapping
- * @gfp_mask - page allocation mode
+ * @mapping: the page's address_space
+ * @index: the page's index into the mapping
+ * @gfp_mask: page allocation mode
  *
  * Locates a page in the pagecache.  If the page is not present, a new page
  * is allocated using @gfp_mask and is added to the pagecache and to the VM's
@@ -1949,7 +1967,7 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov,
 		buf = iov->iov_base + written;
 	else {
 		filemap_set_next_iovec(&cur_iov, &iov_base, written);
-		buf = iov->iov_base + iov_base;
+		buf = cur_iov->iov_base + iov_base;
 	}
 
 	do {
@@ -2007,9 +2025,11 @@ generic_file_buffered_write(struct kiocb *iocb, const struct iovec *iov,
 				count -= status;
 				pos += status;
 				buf += status;
-				if (unlikely(nr_segs > 1))
+				if (unlikely(nr_segs > 1)) {
 					filemap_set_next_iovec(&cur_iov,
 							&iov_base, status);
+					buf = cur_iov->iov_base + iov_base;
+				}
 			}
 		}
 		if (unlikely(copied != bytes))
diff --git a/mm/highmem.c b/mm/highmem.c
index d01276506b0..40091159946 100644
--- a/mm/highmem.c
+++ b/mm/highmem.c
@@ -325,6 +325,7 @@ static void bounce_end_io(struct bio *bio, mempool_t *pool, int err)
 			continue;
 
 		mempool_free(bvec->bv_page, pool);	
+		dec_page_state(nr_bounce);
 	}
 
 	bio_endio(bio_orig, bio_orig->bi_size, err);
@@ -405,6 +406,7 @@ static void __blk_queue_bounce(request_queue_t *q, struct bio **bio_orig,
 		to->bv_page = mempool_alloc(pool, q->bounce_gfp);
 		to->bv_len = from->bv_len;
 		to->bv_offset = from->bv_offset;
+		inc_page_state(nr_bounce);
 
 		if (rw == WRITE) {
 			char *vto, *vfrom;
diff --git a/mm/mempool.c b/mm/mempool.c
index b014ffeaa41..c9f3d462042 100644
--- a/mm/mempool.c
+++ b/mm/mempool.c
@@ -198,31 +198,22 @@ void * mempool_alloc(mempool_t *pool, unsigned int __nocast gfp_mask)
 	void *element;
 	unsigned long flags;
 	DEFINE_WAIT(wait);
-	int gfp_nowait = gfp_mask & ~(__GFP_WAIT | __GFP_IO);
+	int gfp_temp;
 
 	might_sleep_if(gfp_mask & __GFP_WAIT);
+
+	gfp_mask |= __GFP_NOMEMALLOC;	/* don't allocate emergency reserves */
+	gfp_mask |= __GFP_NORETRY;	/* don't loop in __alloc_pages */
+	gfp_mask |= __GFP_NOWARN;	/* failures are OK */
+
+	gfp_temp = gfp_mask & ~(__GFP_WAIT|__GFP_IO);
+
 repeat_alloc:
-	element = pool->alloc(gfp_nowait|__GFP_NOWARN, pool->pool_data);
+
+	element = pool->alloc(gfp_temp, pool->pool_data);
 	if (likely(element != NULL))
 		return element;
 
-	/*
-	 * If the pool is less than 50% full and we can perform effective
-	 * page reclaim then try harder to allocate an element.
-	 */
-	mb();
-	if ((gfp_mask & __GFP_FS) && (gfp_mask != gfp_nowait) &&
-				(pool->curr_nr <= pool->min_nr/2)) {
-		element = pool->alloc(gfp_mask, pool->pool_data);
-		if (likely(element != NULL))
-			return element;
-	}
-
-	/*
-	 * Kick the VM at this point.
-	 */
-	wakeup_bdflush(0);
-
 	spin_lock_irqsave(&pool->lock, flags);
 	if (likely(pool->curr_nr)) {
 		element = remove_element(pool);
@@ -235,8 +226,10 @@ repeat_alloc:
 	if (!(gfp_mask & __GFP_WAIT))
 		return NULL;
 
+	/* Now start performing page reclaim */
+	gfp_temp = gfp_mask;
 	prepare_to_wait(&pool->wait, &wait, TASK_UNINTERRUPTIBLE);
-	mb();
+	smp_mb();
 	if (!pool->curr_nr)
 		io_schedule();
 	finish_wait(&pool->wait, &wait);
@@ -257,7 +250,7 @@ void mempool_free(void *element, mempool_t *pool)
 {
 	unsigned long flags;
 
-	mb();
+	smp_mb();
 	if (pool->curr_nr < pool->min_nr) {
 		spin_lock_irqsave(&pool->lock, flags);
 		if (pool->curr_nr < pool->min_nr) {
diff --git a/mm/mmap.c b/mm/mmap.c
index 6ea204cc751..01f9793591f 100644
--- a/mm/mmap.c
+++ b/mm/mmap.c
@@ -937,9 +937,10 @@ unsigned long do_mmap_pgoff(struct file * file, unsigned long addr,
 	/* mlock MCL_FUTURE? */
 	if (vm_flags & VM_LOCKED) {
 		unsigned long locked, lock_limit;
-		locked = mm->locked_vm << PAGE_SHIFT;
+		locked = len >> PAGE_SHIFT;
+		locked += mm->locked_vm;
 		lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur;
-		locked += len;
+		lock_limit >>= PAGE_SHIFT;
 		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
 			return -EAGAIN;
 	}
@@ -1009,8 +1010,7 @@ munmap_back:
 	}
 
 	/* Check against address space limit. */
-	if ((mm->total_vm << PAGE_SHIFT) + len
-	    > current->signal->rlim[RLIMIT_AS].rlim_cur)
+	if (!may_expand_vm(mm, len >> PAGE_SHIFT))
 		return -ENOMEM;
 
 	if (accountable && (!(flags & MAP_NORESERVE) ||
@@ -1421,7 +1421,7 @@ static int acct_stack_growth(struct vm_area_struct * vma, unsigned long size, un
 	struct rlimit *rlim = current->signal->rlim;
 
 	/* address space limit tests */
-	if (mm->total_vm + grow > rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT)
+	if (!may_expand_vm(mm, grow))
 		return -ENOMEM;
 
 	/* Stack limit test */
@@ -1823,9 +1823,10 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
 	 */
 	if (mm->def_flags & VM_LOCKED) {
 		unsigned long locked, lock_limit;
-		locked = mm->locked_vm << PAGE_SHIFT;
+		locked = len >> PAGE_SHIFT;
+		locked += mm->locked_vm;
 		lock_limit = current->signal->rlim[RLIMIT_MEMLOCK].rlim_cur;
-		locked += len;
+		lock_limit >>= PAGE_SHIFT;
 		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
 			return -EAGAIN;
 	}
@@ -1848,8 +1849,7 @@ unsigned long do_brk(unsigned long addr, unsigned long len)
 	}
 
 	/* Check against address space limits *after* clearing old maps... */
-	if ((mm->total_vm << PAGE_SHIFT) + len
-	    > current->signal->rlim[RLIMIT_AS].rlim_cur)
+	if (!may_expand_vm(mm, len >> PAGE_SHIFT))
 		return -ENOMEM;
 
 	if (mm->map_count > sysctl_max_map_count)
@@ -2019,3 +2019,19 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
 	}
 	return new_vma;
 }
+
+/*
+ * Return true if the calling process may expand its vm space by the passed
+ * number of pages
+ */
+int may_expand_vm(struct mm_struct *mm, unsigned long npages)
+{
+	unsigned long cur = mm->total_vm;	/* pages */
+	unsigned long lim;
+
+	lim = current->signal->rlim[RLIMIT_AS].rlim_cur >> PAGE_SHIFT;
+
+	if (cur + npages > lim)
+		return 0;
+	return 1;
+}
diff --git a/mm/mremap.c b/mm/mremap.c
index 0d1c1b9c7a0..0dd7ace94e5 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -347,10 +347,10 @@ unsigned long do_mremap(unsigned long addr,
 		if (locked > lock_limit && !capable(CAP_IPC_LOCK))
 			goto out;
 	}
-	ret = -ENOMEM;
-	if ((current->mm->total_vm << PAGE_SHIFT) + (new_len - old_len)
-	    > current->signal->rlim[RLIMIT_AS].rlim_cur)
+	if (!may_expand_vm(current->mm, (new_len - old_len) >> PAGE_SHIFT)) {
+		ret = -ENOMEM;
 		goto out;
+	}
 
 	if (vma->vm_flags & VM_ACCOUNT) {
 		charged = (new_len - old_len) >> PAGE_SHIFT;
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 6ddd6a29c73..613b99a5591 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -255,7 +255,7 @@ static void balance_dirty_pages(struct address_space *mapping)
 
 /**
  * balance_dirty_pages_ratelimited - balance dirty memory state
- * @mapping - address_space which was dirtied
+ * @mapping: address_space which was dirtied
  *
  * Processes which are dirtying memory should call in here once for each page
  * which was newly dirtied.  The function will periodically check the system's
@@ -562,8 +562,8 @@ int do_writepages(struct address_space *mapping, struct writeback_control *wbc)
 /**
  * write_one_page - write out a single page and optionally wait on I/O
  *
- * @page - the page to write
- * @wait - if true, wait on writeout
+ * @page: the page to write
+ * @wait: if true, wait on writeout
  *
  * The page must be locked by the caller and will be unlocked upon return.
  *
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index c73dbbc1cd8..fc1b1064c50 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -799,14 +799,18 @@ __alloc_pages(unsigned int __nocast gfp_mask, unsigned int order,
 	}
 
 	/* This allocation should allow future memory freeing. */
-	if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE))) && !in_interrupt()) {
-		/* go through the zonelist yet again, ignoring mins */
-		for (i = 0; (z = zones[i]) != NULL; i++) {
-			if (!cpuset_zone_allowed(z))
-				continue;
-			page = buffered_rmqueue(z, order, gfp_mask);
-			if (page)
-				goto got_pg;
+
+	if (((p->flags & PF_MEMALLOC) || unlikely(test_thread_flag(TIF_MEMDIE)))
+			&& !in_interrupt()) {
+		if (!(gfp_mask & __GFP_NOMEMALLOC)) {
+			/* go through the zonelist yet again, ignoring mins */
+			for (i = 0; (z = zones[i]) != NULL; i++) {
+				if (!cpuset_zone_allowed(z))
+					continue;
+				page = buffered_rmqueue(z, order, gfp_mask);
+				if (page)
+					goto got_pg;
+			}
 		}
 		goto nopage;
 	}
@@ -1351,8 +1355,7 @@ static int __init build_zonelists_node(pg_data_t *pgdat, struct zonelist *zoneli
 #define MAX_NODE_LOAD (num_online_nodes())
 static int __initdata node_load[MAX_NUMNODES];
 /**
- * find_next_best_node - find the next node that should appear in a given
- *    node's fallback list
+ * find_next_best_node - find the next node that should appear in a given node's fallback list
  * @node: node whose fallback list we're appending
  * @used_node_mask: nodemask_t of already used nodes
  *
@@ -1671,6 +1674,18 @@ static void __init free_area_init_core(struct pglist_data *pgdat,
 		if (batch < 1)
 			batch = 1;
 
+		/*
+		 * Clamp the batch to a 2^n - 1 value. Having a power
+		 * of 2 value was found to be more likely to have
+		 * suboptimal cache aliasing properties in some cases.
+		 *
+		 * For example if 2 tasks are alternately allocating
+		 * batches of pages, one task can end up with a lot
+		 * of pages of one half of the possible page colors
+		 * and the other with pages of the other colors.
+		 */
+		batch = (1 << fls(batch + batch/2)) - 1;
+
 		for (cpu = 0; cpu < NR_CPUS; cpu++) {
 			struct per_cpu_pages *pcp;
 
@@ -1881,6 +1896,7 @@ static char *vmstat_text[] = {
 	"allocstall",
 
 	"pgrotated",
+	"nr_bounce",
 };
 
 static void *vmstat_start(struct seq_file *m, loff_t *pos)
diff --git a/mm/rmap.c b/mm/rmap.c
index 884d6d1928b..378de234c12 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -243,6 +243,42 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
 }
 
 /*
+ * Check that @page is mapped at @address into @mm.
+ *
+ * On success returns with mapped pte and locked mm->page_table_lock.
+ */
+static pte_t *page_check_address(struct page *page, struct mm_struct *mm,
+					unsigned long address)
+{
+	pgd_t *pgd;
+	pud_t *pud;
+	pmd_t *pmd;
+	pte_t *pte;
+
+	/*
+	 * We need the page_table_lock to protect us from page faults,
+	 * munmap, fork, etc...
+	 */
+	spin_lock(&mm->page_table_lock);
+	pgd = pgd_offset(mm, address);
+	if (likely(pgd_present(*pgd))) {
+		pud = pud_offset(pgd, address);
+		if (likely(pud_present(*pud))) {
+			pmd = pmd_offset(pud, address);
+			if (likely(pmd_present(*pmd))) {
+				pte = pte_offset_map(pmd, address);
+				if (likely(pte_present(*pte) &&
+					   page_to_pfn(page) == pte_pfn(*pte)))
+					return pte;
+				pte_unmap(pte);
+			}
+		}
+	}
+	spin_unlock(&mm->page_table_lock);
+	return ERR_PTR(-ENOENT);
+}
+
+/*
  * Subfunctions of page_referenced: page_referenced_one called
  * repeatedly from either page_referenced_anon or page_referenced_file.
  */
@@ -251,9 +287,6 @@ static int page_referenced_one(struct page *page,
 {
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long address;
-	pgd_t *pgd;
-	pud_t *pud;
-	pmd_t *pmd;
 	pte_t *pte;
 	int referenced = 0;
 
@@ -263,39 +296,18 @@ static int page_referenced_one(struct page *page,
 	if (address == -EFAULT)
 		goto out;
 
-	spin_lock(&mm->page_table_lock);
-
-	pgd = pgd_offset(mm, address);
-	if (!pgd_present(*pgd))
-		goto out_unlock;
-
-	pud = pud_offset(pgd, address);
-	if (!pud_present(*pud))
-		goto out_unlock;
-
-	pmd = pmd_offset(pud, address);
-	if (!pmd_present(*pmd))
-		goto out_unlock;
-
-	pte = pte_offset_map(pmd, address);
-	if (!pte_present(*pte))
-		goto out_unmap;
-
-	if (page_to_pfn(page) != pte_pfn(*pte))
-		goto out_unmap;
-
-	if (ptep_clear_flush_young(vma, address, pte))
-		referenced++;
-
-	if (mm != current->mm && !ignore_token && has_swap_token(mm))
-		referenced++;
+	pte = page_check_address(page, mm, address);
+	if (!IS_ERR(pte)) {
+		if (ptep_clear_flush_young(vma, address, pte))
+			referenced++;
 
-	(*mapcount)--;
+		if (mm != current->mm && !ignore_token && has_swap_token(mm))
+			referenced++;
 
-out_unmap:
-	pte_unmap(pte);
-out_unlock:
-	spin_unlock(&mm->page_table_lock);
+		(*mapcount)--;
+		pte_unmap(pte);
+		spin_unlock(&mm->page_table_lock);
+	}
 out:
 	return referenced;
 }
@@ -502,9 +514,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long address;
-	pgd_t *pgd;
-	pud_t *pud;
-	pmd_t *pmd;
 	pte_t *pte;
 	pte_t pteval;
 	int ret = SWAP_AGAIN;
@@ -515,30 +524,9 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma)
 	if (address == -EFAULT)
 		goto out;
 
-	/*
-	 * We need the page_table_lock to protect us from page faults,
-	 * munmap, fork, etc...
-	 */
-	spin_lock(&mm->page_table_lock);
-
-	pgd = pgd_offset(mm, address);
-	if (!pgd_present(*pgd))
-		goto out_unlock;
-
-	pud = pud_offset(pgd, address);
-	if (!pud_present(*pud))
-		goto out_unlock;
-
-	pmd = pmd_offset(pud, address);
-	if (!pmd_present(*pmd))
-		goto out_unlock;
-
-	pte = pte_offset_map(pmd, address);
-	if (!pte_present(*pte))
-		goto out_unmap;
-
-	if (page_to_pfn(page) != pte_pfn(*pte))
-		goto out_unmap;
+	pte = page_check_address(page, mm, address);
+	if (IS_ERR(pte))
+		goto out;
 
 	/*
 	 * If the page is mlock()d, we cannot swap it out.
@@ -604,7 +592,6 @@ static int try_to_unmap_one(struct page *page, struct vm_area_struct *vma)
 
 out_unmap:
 	pte_unmap(pte);
-out_unlock:
 	spin_unlock(&mm->page_table_lock);
 out:
 	return ret;
@@ -708,7 +695,6 @@ static void try_to_unmap_cluster(unsigned long cursor,
 	}
 
 	pte_unmap(pte);
-
 out_unlock:
 	spin_unlock(&mm->page_table_lock);
 }
@@ -860,3 +846,4 @@ int try_to_unmap(struct page *page)
 		ret = SWAP_SUCCESS;
 	return ret;
 }
+
diff --git a/mm/slab.c b/mm/slab.c
index ec660d85ddd..84074264115 100644
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -583,7 +583,7 @@ static inline struct array_cache *ac_data(kmem_cache_t *cachep)
 	return cachep->array[smp_processor_id()];
 }
 
-static inline kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags)
+static inline kmem_cache_t *__find_general_cachep(size_t size, int gfpflags)
 {
 	struct cache_sizes *csizep = malloc_sizes;
 
@@ -607,6 +607,12 @@ static inline kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags)
 	return csizep->cs_cachep;
 }
 
+kmem_cache_t *kmem_find_general_cachep(size_t size, int gfpflags)
+{
+	return __find_general_cachep(size, gfpflags);
+}
+EXPORT_SYMBOL(kmem_find_general_cachep);
+
 /* Cal the num objs, wastage, and bytes left over for a given slab size. */
 static void cache_estimate(unsigned long gfporder, size_t size, size_t align,
 		 int flags, size_t *left_over, unsigned int *num)
@@ -672,14 +678,11 @@ static struct array_cache *alloc_arraycache(int cpu, int entries,
 	int memsize = sizeof(void*)*entries+sizeof(struct array_cache);
 	struct array_cache *nc = NULL;
 
-	if (cpu != -1) {
-		kmem_cache_t *cachep;
-		cachep = kmem_find_general_cachep(memsize, GFP_KERNEL);
-		if (cachep)
-			nc = kmem_cache_alloc_node(cachep, cpu_to_node(cpu));
-	}
-	if (!nc)
+	if (cpu == -1)
 		nc = kmalloc(memsize, GFP_KERNEL);
+	else
+		nc = kmalloc_node(memsize, GFP_KERNEL, cpu_to_node(cpu));
+
 	if (nc) {
 		nc->avail = 0;
 		nc->limit = entries;
@@ -1663,7 +1666,7 @@ int kmem_cache_destroy(kmem_cache_t * cachep)
 	}
 
 	if (unlikely(cachep->flags & SLAB_DESTROY_BY_RCU))
-		synchronize_kernel();
+		synchronize_rcu();
 
 	/* no cpu_online check required here since we clear the percpu
 	 * array on cpu offline and set this to NULL.
@@ -2361,7 +2364,7 @@ out:
  * and can sleep. And it will allocate memory on the given node, which
  * can improve the performance for cpu bound structures.
  */
-void *kmem_cache_alloc_node(kmem_cache_t *cachep, int nodeid)
+void *kmem_cache_alloc_node(kmem_cache_t *cachep, int flags, int nodeid)
 {
 	int loop;
 	void *objp;
@@ -2393,7 +2396,7 @@ void *kmem_cache_alloc_node(kmem_cache_t *cachep, int nodeid)
 		spin_unlock_irq(&cachep->spinlock);
 
 		local_irq_disable();
-		if (!cache_grow(cachep, GFP_KERNEL, nodeid)) {
+		if (!cache_grow(cachep, flags, nodeid)) {
 			local_irq_enable();
 			return NULL;
 		}
@@ -2435,6 +2438,16 @@ got_slabp:
 }
 EXPORT_SYMBOL(kmem_cache_alloc_node);
 
+void *kmalloc_node(size_t size, int flags, int node)
+{
+	kmem_cache_t *cachep;
+
+	cachep = kmem_find_general_cachep(size, flags);
+	if (unlikely(cachep == NULL))
+		return NULL;
+	return kmem_cache_alloc_node(cachep, flags, node);
+}
+EXPORT_SYMBOL(kmalloc_node);
 #endif
 
 /**
@@ -2462,7 +2475,12 @@ void *__kmalloc(size_t size, unsigned int __nocast flags)
 {
 	kmem_cache_t *cachep;
 
-	cachep = kmem_find_general_cachep(size, flags);
+	/* If you want to save a few bytes .text space: replace
+	 * __ with kmem_.
+	 * Then kmalloc uses the uninlined functions instead of the inline
+	 * functions.
+	 */
+	cachep = __find_general_cachep(size, flags);
 	if (unlikely(cachep == NULL))
 		return NULL;
 	return __cache_alloc(cachep, flags);
@@ -2489,9 +2507,8 @@ void *__alloc_percpu(size_t size, size_t align)
 	for (i = 0; i < NR_CPUS; i++) {
 		if (!cpu_possible(i))
 			continue;
-		pdata->ptrs[i] = kmem_cache_alloc_node(
-				kmem_find_general_cachep(size, GFP_KERNEL),
-				cpu_to_node(i));
+		pdata->ptrs[i] = kmalloc_node(size, GFP_KERNEL,
+						cpu_to_node(i));
 
 		if (!pdata->ptrs[i])
 			goto unwind_oom;
diff --git a/mm/swap_state.c b/mm/swap_state.c
index a063a902ed0..4f251775ef9 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -143,7 +143,6 @@ void __delete_from_swap_cache(struct page *page)
 int add_to_swap(struct page * page)
 {
 	swp_entry_t entry;
-	int pf_flags;
 	int err;
 
 	if (!PageLocked(page))
@@ -154,29 +153,19 @@ int add_to_swap(struct page * page)
 		if (!entry.val)
 			return 0;
 
-		/* Radix-tree node allocations are performing
-		 * GFP_ATOMIC allocations under PF_MEMALLOC.  
-		 * They can completely exhaust the page allocator.  
-		 *
-		 * So PF_MEMALLOC is dropped here.  This causes the slab 
-		 * allocations to fail earlier, so radix-tree nodes will 
-		 * then be allocated from the mempool reserves.
+		/*
+		 * Radix-tree node allocations from PF_MEMALLOC contexts could
+		 * completely exhaust the page allocator. __GFP_NOMEMALLOC
+		 * stops emergency reserves from being allocated.
 		 *
-		 * We're still using __GFP_HIGH for radix-tree node
-		 * allocations, so some of the emergency pools are available,
-		 * just not all of them.
+		 * TODO: this could cause a theoretical memory reclaim
+		 * deadlock in the swap out path.
 		 */
-
-		pf_flags = current->flags;
-		current->flags &= ~PF_MEMALLOC;
-
 		/*
 		 * Add it to the swap cache and mark it dirty
 		 */
-		err = __add_to_swap_cache(page, entry, GFP_ATOMIC|__GFP_NOWARN);
-
-		if (pf_flags & PF_MEMALLOC)
-			current->flags |= PF_MEMALLOC;
+		err = __add_to_swap_cache(page, entry,
+				GFP_ATOMIC|__GFP_NOMEMALLOC|__GFP_NOWARN);
 
 		switch (err) {
 		case 0:				/* Success */
diff --git a/mm/truncate.c b/mm/truncate.c
index c9a63f0b69a..60c8764bfac 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -242,7 +242,7 @@ EXPORT_SYMBOL(invalidate_inode_pages);
 
 /**
  * invalidate_inode_pages2_range - remove range of pages from an address_space
- * @mapping - the address_space
+ * @mapping: the address_space
  * @start: the page offset 'from' which to invalidate
  * @end: the page offset 'to' which to invalidate (inclusive)
  *
@@ -322,7 +322,7 @@ EXPORT_SYMBOL_GPL(invalidate_inode_pages2_range);
 
 /**
  * invalidate_inode_pages2 - remove all pages from an address_space
- * @mapping - the address_space
+ * @mapping: the address_space
  *
  * Any pages which are found to be mapped into pagetables are unmapped prior to
  * invalidation.
diff --git a/mm/vmalloc.c b/mm/vmalloc.c
index c6182f6f130..2bd83e5c2bb 100644
--- a/mm/vmalloc.c
+++ b/mm/vmalloc.c
@@ -475,6 +475,10 @@ void *vmalloc(unsigned long size)
 
 EXPORT_SYMBOL(vmalloc);
 
+#ifndef PAGE_KERNEL_EXEC
+# define PAGE_KERNEL_EXEC PAGE_KERNEL
+#endif
+
 /**
  *	vmalloc_exec  -  allocate virtually contiguous, executable memory
  *
@@ -488,10 +492,6 @@ EXPORT_SYMBOL(vmalloc);
  *	use __vmalloc() instead.
  */
 
-#ifndef PAGE_KERNEL_EXEC
-# define PAGE_KERNEL_EXEC PAGE_KERNEL
-#endif
-
 void *vmalloc_exec(unsigned long size)
 {
 	return __vmalloc(size, GFP_KERNEL | __GFP_HIGHMEM, PAGE_KERNEL_EXEC);