The patch titled
Memoryless nodes: Fix GFP_THISNODE behavior
has been removed from the -mm tree. Its filename was
memoryless-nodes-fix-gfp_thisnode-behavior.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
Subject: Memoryless nodes: Fix GFP_THISNODE behavior
From: Christoph Lameter <clameter@xxxxxxx>
GFP_THISNODE checks that the zone selected is within the pgdat (node) of the
first zone of a nodelist. That only works if the node has memory. A
memoryless node will have its first node on another pgdat (node).
GFP_THISNODE currently will return simply memory on the first pgdat. Thus it
is returning memory on other nodes. GFP_THISNODE should fail if there is no
local memory on a node.
Add a new set of zonelists for each node that only contain the nodes that
belong to the zones itself so that no fallback is possible.
Then modify gfp_type to pickup the right zone based on the presence of
__GFP_THISNODE.
Drop the existing GFP_THISNODE checks from the page_allocators hot path.
Signed-off-by: Christoph Lameter <clameter@xxxxxxx>
Acked-by: Nishanth Aravamudan <nacc@xxxxxxxxxx>
Tested-by: Lee Schermerhorn <lee.schermerhorn@xxxxxx>
Acked-by: Bob Picco <bob.picco@xxxxxx>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx>
Cc: Mel Gorman <mel@xxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
include/linux/gfp.h | 17 ++++++++++++-----
include/linux/mmzone.h | 14 +++++++++++++-
mm/page_alloc.c | 28 +++++++++++++++++++++++-----
3 files changed, 48 insertions(+), 11 deletions(-)
diff -puN include/linux/gfp.h~memoryless-nodes-fix-gfp_thisnode-behavior include/linux/gfp.h
--- a/include/linux/gfp.h~memoryless-nodes-fix-gfp_thisnode-behavior
+++ a/include/linux/gfp.h
@@ -98,22 +98,29 @@ struct vm_area_struct;
static inline enum zone_type gfp_zone(gfp_t flags)
{
+ int base = 0;
+
+#ifdef CONFIG_NUMA
+ if (flags & __GFP_THISNODE)
+ base = MAX_NR_ZONES;
+#endif
+
#ifdef CONFIG_ZONE_DMA
if (flags & __GFP_DMA)
- return ZONE_DMA;
+ return base + ZONE_DMA;
#endif
#ifdef CONFIG_ZONE_DMA32
if (flags & __GFP_DMA32)
- return ZONE_DMA32;
+ return base + ZONE_DMA32;
#endif
if ((flags & (__GFP_HIGHMEM | __GFP_MOVABLE)) ==
(__GFP_HIGHMEM | __GFP_MOVABLE))
- return ZONE_MOVABLE;
+ return base + ZONE_MOVABLE;
#ifdef CONFIG_HIGHMEM
if (flags & __GFP_HIGHMEM)
- return ZONE_HIGHMEM;
+ return base + ZONE_HIGHMEM;
#endif
- return ZONE_NORMAL;
+ return base + ZONE_NORMAL;
}
/*
diff -puN include/linux/mmzone.h~memoryless-nodes-fix-gfp_thisnode-behavior include/linux/mmzone.h
--- a/include/linux/mmzone.h~memoryless-nodes-fix-gfp_thisnode-behavior
+++ a/include/linux/mmzone.h
@@ -324,6 +324,17 @@ struct zone {
#define MAX_ZONES_PER_ZONELIST (MAX_NUMNODES * MAX_NR_ZONES)
#ifdef CONFIG_NUMA
+
+/*
+ * The NUMA zonelists are doubled becausse we need zonelists that restrict the
+ * allocations to a single node for GFP_THISNODE.
+ *
+ * [0 .. MAX_NR_ZONES -1] : Zonelists with fallback
+ * [MAZ_NR_ZONES ... MAZ_ZONELISTS -1] : No fallback (GFP_THISNODE)
+ */
+#define MAX_ZONELISTS (2 * MAX_NR_ZONES)
+
+
/*
* We cache key information from each zonelist for smaller cache
* footprint when scanning for free pages in get_page_from_freelist().
@@ -389,6 +400,7 @@ struct zonelist_cache {
unsigned long last_full_zap; /* when last zap'd (jiffies) */
};
#else
+#define MAX_ZONELISTS MAX_NR_ZONES
struct zonelist_cache;
#endif
@@ -455,7 +467,7 @@ extern struct page *mem_map;
struct bootmem_data;
typedef struct pglist_data {
struct zone node_zones[MAX_NR_ZONES];
- struct zonelist node_zonelists[MAX_NR_ZONES];
+ struct zonelist node_zonelists[MAX_ZONELISTS];
int nr_zones;
#ifdef CONFIG_FLAT_NODE_MEM_MAP
struct page *node_mem_map;
diff -puN mm/page_alloc.c~memoryless-nodes-fix-gfp_thisnode-behavior mm/page_alloc.c
--- a/mm/page_alloc.c~memoryless-nodes-fix-gfp_thisnode-behavior
+++ a/mm/page_alloc.c
@@ -1191,9 +1191,6 @@ zonelist_scan:
!zlc_zone_worth_trying(zonelist, z, allowednodes))
continue;
zone = *z;
- if (unlikely(NUMA_BUILD && (gfp_mask & __GFP_THISNODE) &&
- zone->zone_pgdat != zonelist->zones[0]->zone_pgdat))
- break;
if ((alloc_flags & ALLOC_CPUSET) &&
!cpuset_zone_allowed_softwall(zone, gfp_mask))
goto try_next_zone;
@@ -1262,7 +1259,10 @@ restart:
z = zonelist->zones; /* the list of zones suitable for gfp_mask */
if (unlikely(*z == NULL)) {
- /* Should this ever happen?? */
+ /*
+ * Happens if we have an empty zonelist as a result of
+ * GFP_THISNODE being used on a memoryless node
+ */
return NULL;
}
@@ -1858,6 +1858,22 @@ static void build_zonelists_in_node_orde
}
/*
+ * Build gfp_thisnode zonelists
+ */
+static void build_thisnode_zonelists(pg_data_t *pgdat)
+{
+ enum zone_type i;
+ int j;
+ struct zonelist *zonelist;
+
+ for (i = 0; i < MAX_NR_ZONES; i++) {
+ zonelist = pgdat->node_zonelists + MAX_NR_ZONES + i;
+ j = build_zonelists_node(pgdat, zonelist, 0, i);
+ zonelist->zones[j] = NULL;
+ }
+}
+
+/*
* Build zonelists ordered by zone and nodes within zones.
* This results in conserving DMA zone[s] until all Normal memory is
* exhausted, but results in overflowing to remote node while memory
@@ -1961,7 +1977,7 @@ static void build_zonelists(pg_data_t *p
int order = current_zonelist_order;
/* initialize zonelists */
- for (i = 0; i < MAX_NR_ZONES; i++) {
+ for (i = 0; i < MAX_ZONELISTS; i++) {
zonelist = pgdat->node_zonelists + i;
zonelist->zones[0] = NULL;
}
@@ -2006,6 +2022,8 @@ static void build_zonelists(pg_data_t *p
/* calculate node order -- i.e., DMA last! */
build_zonelists_in_zone_order(pgdat, j);
}
+
+ build_thisnode_zonelists(pgdat);
}
/* Construct the zonelist performance cache - see further mmzone.h */
_
Patches currently in -mm which might be from clameter@xxxxxxx are
origin.patch
pa-risc-use-page-allocator-instead-of-slab-allocator.patch
dma-use-dev_to_node-to-get-node-for-device-in-dma_alloc_pages.patch
x86-fix-cpu_to_node-references.patch
x86-convert-x86_cpu_to_apicid-to-be-a-per-cpu-variable.patch
x86-convert-cpu_llc_id-to-be-a-per-cpu-variable.patch
x86-acpi-use-cpu_physical_id.patch
x86-convert-cpuinfo_x86-array-to-a-per_cpu-array.patch
slub-simplify-irq-off-handling.patch
slab-api-remove-useless-ctor-parameter-and-reorder-parameters.patch
slab-api-remove-useless-ctor-parameter-and-reorder-parameters-fix.patch
slab-api-remove-useless-ctor-parameter-and-reorder-parameters-fix-2.patch
slab-api-remove-useless-ctor-parameter-and-reorder-parameters-vs-unionfs.patch
oom-move-prototypes-to-appropriate-header-file.patch
oom-move-constraints-to-enum.patch
oom-change-all_unreclaimable-zone-member-to-flags.patch
oom-change-all_unreclaimable-zone-member-to-flags-fix.patch
oom-add-per-zone-locking.patch
oom-serialize-out-of-memory-calls.patch
oom-add-oom_kill_allocating_task-sysctl.patch
oom-suppress-extraneous-stack-and-memory-dump.patch
oom-compare-cpuset-mems_allowed-instead-of-exclusive.patch
oom-do-not-take-callback_mutex.patch
oom-do-not-take-callback_mutex-fix.patch
oom-prevent-including-schedh-in-header-file.patch
oom-add-header-file-to-kbuild-as-unifdef.patch
oom-convert-zone_scan_lock-from-mutex-to-spinlock.patch
mm-test-and-set-zone-reclaim-lock-before-starting.patch
mm-test-and-set-zone-reclaim-lock-before-starting-cleanup.patch
avoid-negative-and-full-width-shifts-in-radix-treec.patch
cpu-hotplug-slab-cleanup-cpuup_callback.patch
cpu-hotplug-slab-fix-memory-leak-in-cpu-hotplug-error-path.patch
intel-iommu-dmar-detection-and-parsing-logic.patch
intel-iommu-pci-generic-helper-function.patch
intel-iommu-clflush_cache_range-now-takes-size-param.patch
intel-iommu-iova-allocation-and-management-routines.patch
intel-iommu-intel-iommu-driver.patch
intel-iommu-avoid-memory-allocation-failures-in-dma-map-api-calls.patch
intel-iommu-intel-iommu-cmdline-option-forcedac.patch
intel-iommu-dmar-fault-handling-support.patch
intel-iommu-iommu-gfx-workaround.patch
intel-iommu-iommu-floppy-workaround.patch
revoke-core-code.patch
slab-api-remove-useless-ctor-parameter-and-reorder-parameters-vs-revoke.patch
documentation-vm-slabinfoc-clean-up-this-code.patch
cpuset-zero-malloc-revert-the-old-cpuset-fix.patch
memcontrol-move-oom-task-exclusion-to-tasklist.patch
memcontrol-move-oom-task-exclusion-to-tasklist-fix.patch
oom-add-sysctl-to-enable-task-memory-dump.patch
hotplug-cpu-migrate-a-task-within-its-cpuset.patch
hotplug-cpu-migrate-a-task-within-its-cpuset-fix.patch
hotplug-cpu-migrate-a-task-within-its-cpuset-doc.patch
bit_spin_lock-use-lock-bitops.patch
ext3-support-large-blocksize-up-to-pagesize.patch
slab-api-remove-useless-ctor-parameter-and-reorder-parameters-vs-reiser4.patch
page-owner-tracking-leak-detector.patch
-
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
