memory-hotplug: add sysfs valid_zones attribute

Currently memory-hotplug has two limits: 1. If the memory block is in ZONE_NORMAL, you can change it to ZONE_MOVABLE, but this memory block must be adjacent to ZONE_MOVABLE. 2. If the memory block is in ZONE_MOVABLE, you can change it to ZONE_NORMAL, but this memory block must be adjacent to ZONE_NORMAL. With this patch, we can easy to know a memory block can be onlined to which zone, and don't need to know the above two limits. Updated the related Documentation. [akpm@linux-foundation.org: use conventional comment layout] [akpm@linux-foundation.org: fix build with CONFIG_MEMORY_HOTREMOVE=n] [akpm@linux-foundation.org: remove unused local zone_prev] Signed-off-by: Zhang Zhen <zhenzhang.zhang@huawei.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: David Rientjes <rientjes@google.com> Cc: Toshi Kani <toshi.kani@hp.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Wang Nan <wangnan0@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

memory-hotplug: add sysfs valid_zones attribute
Currently memory-hotplug has two limits: 1. If the memory block is in ZONE_NORMAL, you can change it to ZONE_MOVABLE, but this memory block must be adjacent to ZONE_MOVABLE. 2. If the memory block is in ZONE_MOVABLE, you can change it to ZONE_NORMAL, but this memory block must be adjacent to ZONE_NORMAL. With this patch, we can easy to know a memory block can be onlined to which zone, and don't need to know the above two limits. Updated the related Documentation. [akpm@linux-foundation.org: use conventional comment layout] [akpm@linux-foundation.org: fix build with CONFIG_MEMORY_HOTREMOVE=n] [akpm@linux-foundation.org: remove unused local zone_prev] Signed-off-by: Zhang Zhen <zhenzhang.zhang@huawei.com> Cc: Dave Hansen <dave.hansen@intel.com> Cc: David Rientjes <rientjes@google.com> Cc: Toshi Kani <toshi.kani@hp.com> Cc: Yasuaki Ishimatsu <isimatu.yasuaki@jp.fujitsu.com> Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com> Cc: Wang Nan <wangnan0@huawei.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Zhang Zhen · Linus Torvalds
1 parent cc71aba348
Showing 5 changed files with 62 additions and 2 deletions Side-by-side Diff
Documentation/ABI/testing/sysfs-devices-memory
Documentation/memory-hotplug.txt
drivers/base/memory.c
include/linux/memory_hotplug.h
mm/memory_hotplug.c
@@ -61,6 +61,14 @@
 		http://www.ibm.com/developerworks/wikis/display/LinuxP/powerpc-utils
  
  
+What:           /sys/devices/system/memory/memoryX/valid_zones
+Date:           July 2014
+Contact:	Zhang Zhen <zhenzhang.zhang@huawei.com>
+Description:
+		The file /sys/devices/system/memory/memoryX/valid_zones	is
+		read-only and is designed to show which zone this memory
+		block can be onlined to.
+
 What:		/sys/devices/system/memoryX/nodeY
 Date:		October 2009
 Contact:	Linux Memory Management list <linux-mm@kvack.org>
@@ -155,6 +155,7 @@
 /sys/devices/system/memory/memoryXXX/phys_device
 /sys/devices/system/memory/memoryXXX/state
 /sys/devices/system/memory/memoryXXX/removable
+/sys/devices/system/memory/memoryXXX/valid_zones
  
 'phys_index'      : read-only and contains memory block id, same as XXX.
 'state'           : read-write
@@ -170,6 +171,15 @@
                     block is removable and a value of 0 indicates that
                     it is not removable. A memory block is removable only if
                     every section in the block is removable.
+'valid_zones'     : read-only: designed to show which zones this memory block
+		    can be onlined to.
+		    The first column shows it's default zone.
+		    "memory6/valid_zones: Normal Movable" shows this memoryblock
+		    can be onlined to ZONE_NORMAL by default and to ZONE_MOVABLE
+		    by online_movable.
+		    "memory7/valid_zones: Movable Normal" shows this memoryblock
+		    can be onlined to ZONE_MOVABLE by default and to ZONE_NORMAL
+		    by online_kernel.
  
 NOTE:
   These directories/files appear after physical memory hotplug phase.
@@ -408,7 +418,6 @@
   - allowing memory hot-add to ZONE_MOVABLE. maybe we need some switch like
     sysctl or new control file.
   - showing memory block and physical device relationship.
-  - showing memory block is under ZONE_MOVABLE or not
   - test and make it better memory offlining.
   - support HugeTLB page migration and offlining.
   - memmap removing at memory offline.
@@ -373,6 +373,45 @@
 	return sprintf(buf, "%d\n", mem->phys_device);
 }
  
+#ifdef CONFIG_MEMORY_HOTREMOVE
+static ssize_t show_valid_zones(struct device *dev,
+				struct device_attribute *attr, char *buf)
+{
+	struct memory_block *mem = to_memory_block(dev);
+	unsigned long start_pfn, end_pfn;
+	unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block;
+	struct page *first_page;
+	struct zone *zone;
+
+	start_pfn = section_nr_to_pfn(mem->start_section_nr);
+	end_pfn = start_pfn + nr_pages;
+	first_page = pfn_to_page(start_pfn);
+
+	/* The block contains more than one zone can not be offlined. */
+	if (!test_pages_in_a_zone(start_pfn, end_pfn))
+		return sprintf(buf, "none\n");
+
+	zone = page_zone(first_page);
+
+	if (zone_idx(zone) == ZONE_MOVABLE - 1) {
+		/*The mem block is the last memoryblock of this zone.*/
+		if (end_pfn == zone_end_pfn(zone))
+			return sprintf(buf, "%s %s\n",
+					zone->name, (zone + 1)->name);
+	}
+
+	if (zone_idx(zone) == ZONE_MOVABLE) {
+		/*The mem block is the first memoryblock of ZONE_MOVABLE.*/
+		if (start_pfn == zone->zone_start_pfn)
+			return sprintf(buf, "%s %s\n",
+					zone->name, (zone - 1)->name);
+	}
+
+	return sprintf(buf, "%s\n", zone->name);
+}
+static DEVICE_ATTR(valid_zones, 0444, show_valid_zones, NULL);
+#endif
+
 static DEVICE_ATTR(phys_index, 0444, show_mem_start_phys_index, NULL);
 static DEVICE_ATTR(state, 0644, show_mem_state, store_mem_state);
 static DEVICE_ATTR(phys_device, 0444, show_phys_device, NULL);
@@ -523,6 +562,9 @@
 	&dev_attr_state.attr,
 	&dev_attr_phys_device.attr,
 	&dev_attr_removable.attr,
+#ifdef CONFIG_MEMORY_HOTREMOVE
+	&dev_attr_valid_zones.attr,
+#endif
 	NULL
 };
  
@@ -84,6 +84,7 @@
 extern int add_one_highpage(struct page *page, int pfn, int bad_ppro);
 /* VM interface that may be used by firmware interface */
 extern int online_pages(unsigned long, unsigned long, int);
+extern int test_pages_in_a_zone(unsigned long, unsigned long);
 extern void __offline_isolated_pages(unsigned long, unsigned long);
  
 typedef void (*online_page_callback_t)(struct page *page);
@@ -1307,7 +1307,7 @@
 /*
  * Confirm all pages in a range [start, end) is belongs to the same zone.
  */
-static int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn)
+int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn)
 {
 	unsigned long pfn;
 	struct zone *zone = NULL;
...	...	@@ -61,6 +61,14 @@
61	61	http://www.ibm.com/developerworks/wikis/display/LinuxP/powerpc-utils
62	62
63	63
	64	+What: /sys/devices/system/memory/memoryX/valid_zones
	65	+Date: July 2014
	66	+Contact: Zhang Zhen <zhenzhang.zhang@huawei.com>
	67	+Description:
	68	+ The file /sys/devices/system/memory/memoryX/valid_zones is
	69	+ read-only and is designed to show which zone this memory
	70	+ block can be onlined to.
	71	+
64	72	What: /sys/devices/system/memoryX/nodeY
65	73	Date: October 2009
66	74	Contact: Linux Memory Management list <linux-mm@kvack.org>
...	...	@@ -155,6 +155,7 @@
155	155	/sys/devices/system/memory/memoryXXX/phys_device
156	156	/sys/devices/system/memory/memoryXXX/state
157	157	/sys/devices/system/memory/memoryXXX/removable
	158	+/sys/devices/system/memory/memoryXXX/valid_zones
158	159
159	160	'phys_index' : read-only and contains memory block id, same as XXX.
160	161	'state' : read-write
...	...	@@ -170,6 +171,15 @@
170	171	block is removable and a value of 0 indicates that
171	172	it is not removable. A memory block is removable only if
172	173	every section in the block is removable.
	174	+'valid_zones' : read-only: designed to show which zones this memory block
	175	+ can be onlined to.
	176	+ The first column shows it's default zone.
	177	+ "memory6/valid_zones: Normal Movable" shows this memoryblock
	178	+ can be onlined to ZONE_NORMAL by default and to ZONE_MOVABLE
	179	+ by online_movable.
	180	+ "memory7/valid_zones: Movable Normal" shows this memoryblock
	181	+ can be onlined to ZONE_MOVABLE by default and to ZONE_NORMAL
	182	+ by online_kernel.
173	183
174	184	NOTE:
175	185	These directories/files appear after physical memory hotplug phase.
...	...	@@ -408,7 +418,6 @@
408	418	- allowing memory hot-add to ZONE_MOVABLE. maybe we need some switch like
409	419	sysctl or new control file.
410	420	- showing memory block and physical device relationship.
411		- - showing memory block is under ZONE_MOVABLE or not
412	421	- test and make it better memory offlining.
413	422	- support HugeTLB page migration and offlining.
414	423	- memmap removing at memory offline.
...	...	@@ -373,6 +373,45 @@
373	373	return sprintf(buf, "%d\n", mem->phys_device);
374	374	}
375	375
	376	+#ifdef CONFIG_MEMORY_HOTREMOVE
	377	+static ssize_t show_valid_zones(struct device *dev,
	378	+ struct device_attribute attr, char buf)
	379	+{
	380	+ struct memory_block *mem = to_memory_block(dev);
	381	+ unsigned long start_pfn, end_pfn;
	382	+ unsigned long nr_pages = PAGES_PER_SECTION * sections_per_block;
	383	+ struct page *first_page;
	384	+ struct zone *zone;
	385	+
	386	+ start_pfn = section_nr_to_pfn(mem->start_section_nr);
	387	+ end_pfn = start_pfn + nr_pages;
	388	+ first_page = pfn_to_page(start_pfn);
	389	+
	390	+ /* The block contains more than one zone can not be offlined. */
	391	+ if (!test_pages_in_a_zone(start_pfn, end_pfn))
	392	+ return sprintf(buf, "none\n");
	393	+
	394	+ zone = page_zone(first_page);
	395	+
	396	+ if (zone_idx(zone) == ZONE_MOVABLE - 1) {
	397	+ /The mem block is the last memoryblock of this zone./
	398	+ if (end_pfn == zone_end_pfn(zone))
	399	+ return sprintf(buf, "%s %s\n",
	400	+ zone->name, (zone + 1)->name);
	401	+ }
	402	+
	403	+ if (zone_idx(zone) == ZONE_MOVABLE) {
	404	+ /The mem block is the first memoryblock of ZONE_MOVABLE./
	405	+ if (start_pfn == zone->zone_start_pfn)
	406	+ return sprintf(buf, "%s %s\n",
	407	+ zone->name, (zone - 1)->name);
	408	+ }
	409	+
	410	+ return sprintf(buf, "%s\n", zone->name);
	411	+}
	412	+static DEVICE_ATTR(valid_zones, 0444, show_valid_zones, NULL);
	413	+#endif
	414	+
376	415	static DEVICE_ATTR(phys_index, 0444, show_mem_start_phys_index, NULL);
377	416	static DEVICE_ATTR(state, 0644, show_mem_state, store_mem_state);
378	417	static DEVICE_ATTR(phys_device, 0444, show_phys_device, NULL);
...	...	@@ -523,6 +562,9 @@
523	562	&dev_attr_state.attr,
524	563	&dev_attr_phys_device.attr,
525	564	&dev_attr_removable.attr,
	565	+#ifdef CONFIG_MEMORY_HOTREMOVE
	566	+ &dev_attr_valid_zones.attr,
	567	+#endif
526	568	NULL
527	569	};
528	570
...	...	@@ -84,6 +84,7 @@
84	84	extern int add_one_highpage(struct page *page, int pfn, int bad_ppro);
85	85	/* VM interface that may be used by firmware interface */
86	86	extern int online_pages(unsigned long, unsigned long, int);
	87	+extern int test_pages_in_a_zone(unsigned long, unsigned long);
87	88	extern void __offline_isolated_pages(unsigned long, unsigned long);
88	89
89	90	typedef void (online_page_callback_t)(struct page page);
...	...	@@ -1307,7 +1307,7 @@
1307	1307	/*
1308	1308	* Confirm all pages in a range [start, end) is belongs to the same zone.
1309	1309	*/
1310		-static int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn)
	1310	+int test_pages_in_a_zone(unsigned long start_pfn, unsigned long end_pfn)
1311	1311	{
1312	1312	unsigned long pfn;
1313	1313	struct zone *zone = NULL;