mm: fix GFP_THISNODE callers and clarify

GFP_THISNODE is for callers that implement their own clever fallback to
remote nodes.  It restricts the allocation to the specified node and
does not invoke reclaim, assuming that the caller will take care of it
when the fallback fails, e.g.  through a subsequent allocation request
without GFP_THISNODE set.

However, many current GFP_THISNODE users only want the node exclusive
aspect of the flag, without actually implementing their own fallback or
triggering reclaim if necessary.  This results in things like page
migration failing prematurely even when there is easily reclaimable
memory available, unless kswapd happens to be running already or a
concurrent allocation attempt triggers the necessary reclaim.

Convert all callsites that don't implement their own fallback strategy
to __GFP_THISNODE.  This restricts the allocation a single node too, but
at the same time allows the allocator to enter the slowpath, wake
kswapd, and invoke direct reclaim if necessary, to make the allocation
happen when memory is full.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Jan Stancek <jstancek@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

diff --git a/arch/ia64/kernel/uncached.c b/arch/ia64/kernel/uncached.c
index a96bcf8..20e8a9b 100644 (file)
--- a/arch/ia64/kernel/uncached.c
+++ b/arch/ia64/kernel/uncached.c
@@ -98,7 +98,7 @@ static int uncached_add_chunk(struct uncached_pool *uc_pool, int nid)
        /* attempt to allocate a granule's worth of cached memory pages */
 
        page = alloc_pages_exact_node(nid,
-                               GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+                               GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE,
                                IA64_GRANULE_SHIFT-PAGE_SHIFT);
        if (!page) {
                mutex_unlock(&uc_pool->add_chunk_mutex);

diff --git a/arch/powerpc/platforms/cell/ras.c b/arch/powerpc/platforms/cell/ras.c
index 5ec1e47..e865d74 100644 (file)
--- a/arch/powerpc/platforms/cell/ras.c
+++ b/arch/powerpc/platforms/cell/ras.c
@@ -123,7 +123,8 @@ static int __init cbe_ptcal_enable_on_node(int nid, int order)
 
        area->nid = nid;
        area->order = order;
-       area->pages = alloc_pages_exact_node(area->nid, GFP_KERNEL|GFP_THISNODE,
+       area->pages = alloc_pages_exact_node(area->nid,
+                                               GFP_KERNEL|__GFP_THISNODE,
                                                area->order);
 
        if (!area->pages) {

diff --git a/drivers/misc/sgi-xp/xpc_uv.c b/drivers/misc/sgi-xp/xpc_uv.c
index b9e2000..95c8944 100644 (file)
--- a/drivers/misc/sgi-xp/xpc_uv.c
+++ b/drivers/misc/sgi-xp/xpc_uv.c
@@ -240,7 +240,7 @@ xpc_create_gru_mq_uv(unsigned int mq_size, int cpu, char *irq_name,
 
        nid = cpu_to_node(cpu);
        page = alloc_pages_exact_node(nid,
-                                     GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+                                     GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE,
                                      pg_order);
        if (page == NULL) {
                dev_err(xpc_part, "xpc_create_gru_mq_uv() failed to alloc %d "

diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 0437439..39b81dc 100644 (file)
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -123,6 +123,10 @@ struct vm_area_struct;
                         __GFP_NOMEMALLOC | __GFP_NORETRY | __GFP_NOWARN | \
                         __GFP_NO_KSWAPD)
 
+/*
+ * GFP_THISNODE does not perform any reclaim, you most likely want to
+ * use __GFP_THISNODE to allocate from a given node without fallback!
+ */
 #ifdef CONFIG_NUMA
 #define GFP_THISNODE   (__GFP_THISNODE | __GFP_NOWARN | __GFP_NORETRY)
 #else

diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 5f2052c..9b61b9b 100644 (file)
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -590,10 +590,10 @@ static inline bool zone_is_empty(struct zone *zone)
 
 /*
  * The NUMA zonelists are doubled because we need zonelists that restrict the
- * allocations to a single node for GFP_THISNODE.
+ * allocations to a single node for __GFP_THISNODE.
  *
  * [0] : Zonelist with fallback
- * [1] : No fallback (GFP_THISNODE)
+ * [1] : No fallback (__GFP_THISNODE)
  */
 #define MAX_ZONELISTS 2
 

diff --git a/include/linux/slab.h b/include/linux/slab.h
index 9260abd..b5b2df6 100644 (file)
--- a/include/linux/slab.h
+++ b/include/linux/slab.h
@@ -410,7 +410,7 @@ static __always_inline void *kmalloc_large(size_t size, gfp_t flags)
  *
  * %GFP_NOWAIT - Allocation will not sleep.
  *
- * %GFP_THISNODE - Allocate node-local memory only.
+ * %__GFP_THISNODE - Allocate node-local memory only.
  *
  * %GFP_DMA - Allocation suitable for DMA.
  *   Should only be used for kmalloc() caches. Otherwise, use a

diff --git a/kernel/profile.c b/kernel/profile.c
index 6631e1e..ebdd9c1 100644 (file)
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -549,14 +549,14 @@ static int create_hash_tables(void)
                struct page *page;
 
                page = alloc_pages_exact_node(node,
-                               GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+                               GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE,
                                0);
                if (!page)
                        goto out_cleanup;
                per_cpu(cpu_profile_hits, cpu)[1]
                                = (struct profile_hit *)page_address(page);
                page = alloc_pages_exact_node(node,
-                               GFP_KERNEL | __GFP_ZERO | GFP_THISNODE,
+                               GFP_KERNEL | __GFP_ZERO | __GFP_THISNODE,
                                0);
                if (!page)
                        goto out_cleanup;

diff --git a/mm/migrate.c b/mm/migrate.c
index 482a33d..b494fdb 100644 (file)
--- a/mm/migrate.c
+++ b/mm/migrate.c
@@ -1158,7 +1158,7 @@ static struct page *new_page_node(struct page *p, unsigned long private,
                                        pm->node);
        else
                return alloc_pages_exact_node(pm->node,
-                               GFP_HIGHUSER_MOVABLE | GFP_THISNODE, 0);
+                               GFP_HIGHUSER_MOVABLE | __GFP_THISNODE, 0);
 }
 
 /*
@@ -1544,9 +1544,9 @@ static struct page *alloc_misplaced_dst_page(struct page *page,
        struct page *newpage;
 
        newpage = alloc_pages_exact_node(nid,
-                                        (GFP_HIGHUSER_MOVABLE | GFP_THISNODE |
-                                         __GFP_NOMEMALLOC | __GFP_NORETRY |
-                                         __GFP_NOWARN) &
+                                        (GFP_HIGHUSER_MOVABLE |
+                                         __GFP_THISNODE | __GFP_NOMEMALLOC |
+                                         __GFP_NORETRY | __GFP_NOWARN) &
                                         ~GFP_IOFS, 0);
 
        return newpage;
@@ -1747,7 +1747,8 @@ int migrate_misplaced_transhuge_page(struct mm_struct *mm,
                goto out_dropref;
 
        new_page = alloc_pages_node(node,
-               (GFP_TRANSHUGE | GFP_THISNODE) & ~__GFP_WAIT, HPAGE_PMD_ORDER);
+               (GFP_TRANSHUGE | __GFP_THISNODE) & ~__GFP_WAIT,
+               HPAGE_PMD_ORDER);
        if (!new_page)
                goto out_fail;