mm: fix-up zone present pages

I think zone->present_pages indicates pages that buddy system can management, it should be: zone->present_pages = spanned pages - absent pages - bootmem pages, but is now: zone->present_pages = spanned pages - absent pages - memmap pages. spanned pages: total size, including holes. absent pages: holes. bootmem pages: pages used in system boot, managed by bootmem allocator. memmap pages: pages used by page structs. This may cause zone->present_pages less than it should be. For example, numa node 1 has ZONE_NORMAL and ZONE_MOVABLE, it's memmap and other bootmem will be allocated from ZONE_MOVABLE, so ZONE_NORMAL's present_pages should be spanned pages - absent pages, but now it also minus memmap pages(free_area_init_core), which are actually allocated from ZONE_MOVABLE. When offlining all memory of a zone, this will cause zone->present_pages less than 0, because present_pages is unsigned long type, it is actually a very large integer, it indirectly caused zone->watermark[WMARK_MIN] becomes a large integer(setup_per_zone_wmarks()), than cause totalreserve_pages become a large integer(calculate_totalreserve_pages()), and finally cause memory allocating failure when fork process(__vm_enough_memory()). [root@localhost ~]# dmesg -bash: fork: Cannot allocate memory I think the bug described in http://marc.info/?l=linux-mm&m=134502182714186&w=2 is also caused by wrong zone present pages. This patch intends to fix-up zone->present_pages when memory are freed to buddy system on x86_64 and IA64 platforms. Signed-off-by: Jianguo Wu <wujianguo@huawei.com> Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Reported-by: Petr Tesarik <ptesarik@suse.cz> Tested-by: Petr Tesarik <ptesarik@suse.cz> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Minchan Kim <minchan.kim@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

mm: fix-up zone present pages
I think zone->present_pages indicates pages that buddy system can management, it should be: zone->present_pages = spanned pages - absent pages - bootmem pages, but is now: zone->present_pages = spanned pages - absent pages - memmap pages. spanned pages: total size, including holes. absent pages: holes. bootmem pages: pages used in system boot, managed by bootmem allocator. memmap pages: pages used by page structs. This may cause zone->present_pages less than it should be. For example, numa node 1 has ZONE_NORMAL and ZONE_MOVABLE, it's memmap and other bootmem will be allocated from ZONE_MOVABLE, so ZONE_NORMAL's present_pages should be spanned pages - absent pages, but now it also minus memmap pages(free_area_init_core), which are actually allocated from ZONE_MOVABLE. When offlining all memory of a zone, this will cause zone->present_pages less than 0, because present_pages is unsigned long type, it is actually a very large integer, it indirectly caused zone->watermark[WMARK_MIN] becomes a large integer(setup_per_zone_wmarks()), than cause totalreserve_pages become a large integer(calculate_totalreserve_pages()), and finally cause memory allocating failure when fork process(__vm_enough_memory()). [root@localhost ~]# dmesg -bash: fork: Cannot allocate memory I think the bug described in http://marc.info/?l=linux-mm&m=134502182714186&w=2 is also caused by wrong zone present pages. This patch intends to fix-up zone->present_pages when memory are freed to buddy system on x86_64 and IA64 platforms. Signed-off-by: Jianguo Wu <wujianguo@huawei.com> Signed-off-by: Jiang Liu <jiang.liu@huawei.com> Reported-by: Petr Tesarik <ptesarik@suse.cz> Tested-by: Petr Tesarik <ptesarik@suse.cz> Cc: "Luck, Tony" <tony.luck@intel.com> Cc: Mel Gorman <mel@csn.ul.ie> Cc: Yinghai Lu <yinghai@kernel.org> Cc: Minchan Kim <minchan.kim@gmail.com> Cc: Johannes Weiner <hannes@cmpxchg.org> Cc: David Rientjes <rientjes@google.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Jianguo Wu · Linus Torvalds
1 parent 05106e6a54
Showing 6 changed files with 58 additions and 1 deletions Side-by-side Diff
arch/ia64/mm/init.c
include/linux/mm.h
mm/bootmem.c
mm/memory_hotplug.c
mm/nobootmem.c
mm/page_alloc.c
@@ -637,6 +637,7 @@
  
 	high_memory = __va(max_low_pfn * PAGE_SIZE);
  
+	reset_zone_present_pages();
 	for_each_online_pgdat(pgdat)
 		if (pgdat->bdata->node_bootmem_map)
 			totalram_pages += free_all_bootmem_node(pgdat);
@@ -1684,6 +1684,10 @@
 static inline bool page_is_guard(struct page *page) { return false; }
 #endif /* CONFIG_DEBUG_PAGEALLOC */
  
+extern void reset_zone_present_pages(void);
+extern void fixup_zone_present_pages(int nid, unsigned long start_pfn,
+				unsigned long end_pfn);
+
 #endif /* __KERNEL__ */
 #endif /* _LINUX_MM_H */
@@ -198,6 +198,8 @@
 			int order = ilog2(BITS_PER_LONG);
  
 			__free_pages_bootmem(pfn_to_page(start), order);
+			fixup_zone_present_pages(page_to_nid(pfn_to_page(start)),
+					start, start + BITS_PER_LONG);
 			count += BITS_PER_LONG;
 			start += BITS_PER_LONG;
 		} else {
@@ -208,6 +210,9 @@
 				if (vec & 1) {
 					page = pfn_to_page(start + off);
 					__free_pages_bootmem(page, 0);
+					fixup_zone_present_pages(
+						page_to_nid(page),
+						start + off, start + off + 1);
 					count++;
 				}
 				vec >>= 1;
  
@@ -221,8 +226,11 @@
 	pages = bdata->node_low_pfn - bdata->node_min_pfn;
 	pages = bootmem_bootmap_pages(pages);
 	count += pages;
-	while (pages--)
+	while (pages--) {
+		fixup_zone_present_pages(page_to_nid(page),
+				page_to_pfn(page), page_to_pfn(page) + 1);
 		__free_pages_bootmem(page++, 0);
+	}
  
 	bdebug("nid=%td released=%lx\n", bdata - bootmem_node_data, count);
  
@@ -106,6 +106,7 @@
 void __ref put_page_bootmem(struct page *page)
 {
 	unsigned long type;
+	struct zone *zone;
  
 	type = (unsigned long) page->lru.next;
 	BUG_ON(type < MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE ||
@@ -116,6 +117,12 @@
 		set_page_private(page, 0);
 		INIT_LIST_HEAD(&page->lru);
 		__free_pages_bootmem(page, 0);
+
+		zone = page_zone(page);
+		zone_span_writelock(zone);
+		zone->present_pages++;
+		zone_span_writeunlock(zone);
+		totalram_pages++;
 	}
  
 }
@@ -116,6 +116,8 @@
 		return 0;
  
 	__free_pages_memory(start_pfn, end_pfn);
+	fixup_zone_present_pages(pfn_to_nid(start >> PAGE_SHIFT),
+			start_pfn, end_pfn);
  
 	return end_pfn - start_pfn;
 }
@@ -126,6 +128,7 @@
 	phys_addr_t start, end, size;
 	u64 i;
  
+	reset_zone_present_pages();
 	for_each_free_mem_range(i, MAX_NUMNODES, &start, &end, NULL)
 		count += __free_memory_core(start, end);
  
@@ -6087,4 +6087,38 @@
 	dump_page_flags(page->flags);
 	mem_cgroup_print_bad_page(page);
 }
+
+/* reset zone->present_pages */
+void reset_zone_present_pages(void)
+{
+	struct zone *z;
+	int i, nid;
+
+	for_each_node_state(nid, N_HIGH_MEMORY) {
+		for (i = 0; i < MAX_NR_ZONES; i++) {
+			z = NODE_DATA(nid)->node_zones + i;
+			z->present_pages = 0;
+		}
+	}
+}
+
+/* calculate zone's present pages in buddy system */
+void fixup_zone_present_pages(int nid, unsigned long start_pfn,
+				unsigned long end_pfn)
+{
+	struct zone *z;
+	unsigned long zone_start_pfn, zone_end_pfn;
+	int i;
+
+	for (i = 0; i < MAX_NR_ZONES; i++) {
+		z = NODE_DATA(nid)->node_zones + i;
+		zone_start_pfn = z->zone_start_pfn;
+		zone_end_pfn = zone_start_pfn + z->spanned_pages;
+
+		/* if the two regions intersect */
+		if (!(zone_start_pfn >= end_pfn	|| zone_end_pfn <= start_pfn))
+			z->present_pages += min(end_pfn, zone_end_pfn) -
+					    max(start_pfn, zone_start_pfn);
+	}
+}
...	...	@@ -637,6 +637,7 @@
637	637
638	638	high_memory = __va(max_low_pfn * PAGE_SIZE);
639	639
	640	+ reset_zone_present_pages();
640	641	for_each_online_pgdat(pgdat)
641	642	if (pgdat->bdata->node_bootmem_map)
642	643	totalram_pages += free_all_bootmem_node(pgdat);
...	...	@@ -1684,6 +1684,10 @@
1684	1684	static inline bool page_is_guard(struct page *page) { return false; }
1685	1685	#endif /* CONFIG_DEBUG_PAGEALLOC */
1686	1686
	1687	+extern void reset_zone_present_pages(void);
	1688	+extern void fixup_zone_present_pages(int nid, unsigned long start_pfn,
	1689	+ unsigned long end_pfn);
	1690	+
1687	1691	#endif /* __KERNEL__ */
1688	1692	#endif /* _LINUX_MM_H */
...	...	@@ -198,6 +198,8 @@
198	198	int order = ilog2(BITS_PER_LONG);
199	199
200	200	__free_pages_bootmem(pfn_to_page(start), order);
	201	+ fixup_zone_present_pages(page_to_nid(pfn_to_page(start)),
	202	+ start, start + BITS_PER_LONG);
201	203	count += BITS_PER_LONG;
202	204	start += BITS_PER_LONG;
203	205	} else {
...	...	@@ -208,6 +210,9 @@
208	210	if (vec & 1) {
209	211	page = pfn_to_page(start + off);
210	212	__free_pages_bootmem(page, 0);
	213	+ fixup_zone_present_pages(
	214	+ page_to_nid(page),
	215	+ start + off, start + off + 1);
211	216	count++;
212	217	}
213	218	vec >>= 1;
214	219
...	...	@@ -221,8 +226,11 @@
221	226	pages = bdata->node_low_pfn - bdata->node_min_pfn;
222	227	pages = bootmem_bootmap_pages(pages);
223	228	count += pages;
224		- while (pages--)
	229	+ while (pages--) {
	230	+ fixup_zone_present_pages(page_to_nid(page),
	231	+ page_to_pfn(page), page_to_pfn(page) + 1);
225	232	__free_pages_bootmem(page++, 0);
	233	+ }
226	234
227	235	bdebug("nid=%td released=%lx\n", bdata - bootmem_node_data, count);
228	236
...	...	@@ -106,6 +106,7 @@
106	106	void __ref put_page_bootmem(struct page *page)
107	107	{
108	108	unsigned long type;
	109	+ struct zone *zone;
109	110
110	111	type = (unsigned long) page->lru.next;
111	112	BUG_ON(type < MEMORY_HOTPLUG_MIN_BOOTMEM_TYPE \|\|
...	...	@@ -116,6 +117,12 @@
116	117	set_page_private(page, 0);
117	118	INIT_LIST_HEAD(&page->lru);
118	119	__free_pages_bootmem(page, 0);
	120	+
	121	+ zone = page_zone(page);
	122	+ zone_span_writelock(zone);
	123	+ zone->present_pages++;
	124	+ zone_span_writeunlock(zone);
	125	+ totalram_pages++;
119	126	}
120	127
121	128	}
...	...	@@ -116,6 +116,8 @@
116	116	return 0;
117	117
118	118	__free_pages_memory(start_pfn, end_pfn);
	119	+ fixup_zone_present_pages(pfn_to_nid(start >> PAGE_SHIFT),
	120	+ start_pfn, end_pfn);
119	121
120	122	return end_pfn - start_pfn;
121	123	}
...	...	@@ -126,6 +128,7 @@
126	128	phys_addr_t start, end, size;
127	129	u64 i;
128	130
	131	+ reset_zone_present_pages();
129	132	for_each_free_mem_range(i, MAX_NUMNODES, &start, &end, NULL)
130	133	count += __free_memory_core(start, end);
131	134
...	...	@@ -6087,4 +6087,38 @@
6087	6087	dump_page_flags(page->flags);
6088	6088	mem_cgroup_print_bad_page(page);
6089	6089	}
	6090	+
	6091	+/* reset zone->present_pages */
	6092	+void reset_zone_present_pages(void)
	6093	+{
	6094	+ struct zone *z;
	6095	+ int i, nid;
	6096	+
	6097	+ for_each_node_state(nid, N_HIGH_MEMORY) {
	6098	+ for (i = 0; i < MAX_NR_ZONES; i++) {
	6099	+ z = NODE_DATA(nid)->node_zones + i;
	6100	+ z->present_pages = 0;
	6101	+ }
	6102	+ }
	6103	+}
	6104	+
	6105	+/* calculate zone's present pages in buddy system */
	6106	+void fixup_zone_present_pages(int nid, unsigned long start_pfn,
	6107	+ unsigned long end_pfn)
	6108	+{
	6109	+ struct zone *z;
	6110	+ unsigned long zone_start_pfn, zone_end_pfn;
	6111	+ int i;
	6112	+
	6113	+ for (i = 0; i < MAX_NR_ZONES; i++) {
	6114	+ z = NODE_DATA(nid)->node_zones + i;
	6115	+ zone_start_pfn = z->zone_start_pfn;
	6116	+ zone_end_pfn = zone_start_pfn + z->spanned_pages;
	6117	+
	6118	+ /* if the two regions intersect */
	6119	+ if (!(zone_start_pfn >= end_pfn \|\| zone_end_pfn <= start_pfn))
	6120	+ z->present_pages += min(end_pfn, zone_end_pfn) -
	6121	+ max(start_pfn, zone_start_pfn);
	6122	+ }
	6123	+}