From: Changsheng Liu <liuchangcheng@xxxxxxxxxx>
After the user config CONFIG_MOVABLE_NODE,
When the memory is hot added, should_add_memory_movable() return 0
because all zones including ZONE_MOVABLE are empty,
so the memory that was hot added will be assigned to ZONE_NORMAL
and ZONE_NORMAL will be created firstly.
But we want the whole node to be added to ZONE_MOVABLE by default.
So we change should_add_memory_movable(): if the user config
CONFIG_MOVABLE_NODE and sysctl parameter hotadd_memory_as_movable is 1
and the ZONE_NORMAL is empty or the pfn of the hot-added memory
is after the end of the ZONE_NORMAL it will always return 1
and then the whole node will be added to ZONE_MOVABLE by default.
If we want the node to be assigned to ZONE_NORMAL,
we can do it as follows:
"echo online_kernel > /sys/devices/system/memory/memoryXXX/state"
By the patch, the behavious of kernel is changed by sysctl,
user can automatically create movable memory
by only the following udev rule:
SUBSYSTEM=="memory", ACTION=="add",
ATTR{state}=="offline", ATTR{state}="online"
Signed-off-by: Changsheng Liu <liuchangsheng@xxxxxxxxxx>
Signed-off-by: Xiaofeng Yan <yanxiaofeng@xxxxxxxxxx>
Tested-by: Dongdong Fan <fandd@xxxxxxxxxx>
Cc: Wang Nan <wangnan0@xxxxxxxxxx>
Cc: Dave Hansen <dave.hansen@xxxxxxxxx>
Cc: Yinghai Lu <yinghai@xxxxxxxxxx>
Cc: Tang Chen <tangchen@xxxxxxxxxxxxxx>
Cc: Yasuaki Ishimatsu <isimatu.yasuaki@xxxxxxxxxxxxxx>
Cc: Toshi Kani <toshi.kani@xxxxxx>
Cc: Xishi Qiu <qiuxishi@xxxxxxxxxx>
---
Documentation/memory-hotplug.txt | 5 ++++-
kernel/sysctl.c | 15 +++++++++++++++
mm/memory_hotplug.c | 24 ++++++++++++++++++++++++
3 files changed, 43 insertions(+), 1 deletions(-)
diff --git a/Documentation/memory-hotplug.txt b/Documentation/memory-hotplug.txt
index ce2cfcf..7ac7485 100644
--- a/Documentation/memory-hotplug.txt
+++ b/Documentation/memory-hotplug.txt
@@ -277,7 +277,7 @@ And if the memory block is in ZONE_MOVABLE, you can change it to ZONE_NORMAL:
After this, memory block XXX's state will be 'online' and the amount of
available memory will be increased.
-Currently, newly added memory is added as ZONE_NORMAL (for powerpc, ZONE_DMA).
+Currently, newly added memory is added as ZONE_NORMAL or ZONE_MOVABLE (for powerpc, ZONE_DMA).
This may be changed in future.
@@ -319,6 +319,9 @@ creates ZONE_MOVABLE as following.
Size of memory not for movable pages (not for offline) is TOTAL - ZZZZ.
Size of memory for movable pages (for offline) is ZZZZ.
+And a sysctl parameter for assigning the hot added memory to ZONE_MOVABLE is
+supported. If the value of "kernel/hotadd_memory_as_movable" is 1,the hot added
+memory will be assigned to ZONE_MOVABLE by default.
Note: Unfortunately, there is no information to show which memory block belongs
to ZONE_MOVABLE. This is TBD.
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 19b62b5..16b1501 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -166,6 +166,10 @@ extern int unaligned_dump_stack;
extern int no_unaligned_warning;
#endif
+#ifdef CONFIG_MOVABLE_NODE
+extern int hotadd_memory_as_movable;
+#endif
+
#ifdef CONFIG_PROC_SYSCTL
#define SYSCTL_WRITES_LEGACY -1
@@ -1139,6 +1143,17 @@ static struct ctl_table kern_table[] = {
.proc_handler = timer_migration_handler,
},
#endif
+/*If the value of "kernel/hotadd_memory_as_movable" is 1,the hot added
+ * memory will be assigned to ZONE_MOVABLE by default.*/
+#ifdef CONFIG_MOVABLE_NODE
+ {
+ .procname = "hotadd_memory_as_movable",
+ .data = &hotadd_memory_as_movable,
+ .maxlen = sizeof(int),
+ .mode = 0644,
+ .proc_handler = proc_dointvec,
+ },
+#endif
{ }
};
diff --git a/mm/memory_hotplug.c b/mm/memory_hotplug.c
index 26fbba7..eca5512 100644
--- a/mm/memory_hotplug.c
+++ b/mm/memory_hotplug.c
@@ -37,6 +37,11 @@
#include "internal.h"
+/*If the global variable value is 1,
+ * the hot added memory will be assigned to ZONE_MOVABLE by default
+ */
+int hotadd_memory_as_movable;
+
/*
* online_page_callback contains pointer to current page onlining function.
* Initially it is generic_online_page(). If it is required it could be
@@ -1190,6 +1195,9 @@ static int check_hotplug_memory_range(u64 start, u64 size)
/*
* If movable zone has already been setup, newly added memory should be check.
* If its address is higher than movable zone, it should be added as movable.
+ * And if system config CONFIG_MOVABLE_NODE and set the sysctl parameter
+ * "hotadd_memory_as_movable" and added memory does not overlap the zone
+ * before MOVABLE_ZONE,the memory will be added as movable.
* Without this check, movable zone may overlap with other zone.
*/
static int should_add_memory_movable(int nid, u64 start, u64 size)
@@ -1197,6 +1205,22 @@ static int should_add_memory_movable(int nid, u64 start, u64 size)
unsigned long start_pfn = start >> PAGE_SHIFT;
pg_data_t *pgdat = NODE_DATA(nid);
struct zone *movable_zone = pgdat->node_zones + ZONE_MOVABLE;
+ struct zone *pre_zone = pgdat->node_zones + (ZONE_MOVABLE - 1);
+ /*
+ * The system configs CONFIG_MOVABLE_NODE to assign a node
+ * which has only movable memory,so the hot-added memory should
+ * be assigned to ZONE_MOVABLE by default,
+ * but the function zone_for_memory() assign the hot-added memory
+ * to ZONE_NORMAL(x86_64) by default.Kernel does not allow to
+ * create ZONE_MOVABLE before ZONE_NORMAL,So if the value of
+ * sysctl parameter "hotadd_memory_as_movable" is 1
+ * and the ZONE_NORMAL is empty or the pfn of the hot-added memory
+ * is after the end of ZONE_NORMAL
+ * the hot-added memory will be assigned to ZONE_MOVABLE.
+ */
+ if (hotadd_memory_as_movable
+ && (zone_is_empty(pre_zone) || zone_end_pfn(pre_zone) <= start_pfn))
+ return 1;
if (zone_is_empty(movable_zone))
return 0;
--
1.7.1
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>