summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorLee Schermerhorn <lee.schermerhorn@hp.com>2008-10-19 07:26:53 +0400
committerLinus Torvalds <torvalds@linux-foundation.org>2008-10-20 19:52:31 +0400
commitaf936a1606246a10c145feac3770f6287f483f02 (patch)
tree8b1ca7fabb5c749ffdecd654519889c6c2ed2fb6
parent64d6519dda3905dfb94d3f93c07c5f263f41813f (diff)
downloadlinux-af936a1606246a10c145feac3770f6287f483f02.tar.xz
vmscan: unevictable LRU scan sysctl
This patch adds a function to scan individual or all zones' unevictable lists and move any pages that have become evictable onto the respective zone's inactive list, where shrink_inactive_list() will deal with them. Adds sysctl to scan all nodes, and per node attributes to individual nodes' zones. Kosaki: If evictable page found in unevictable lru when write /proc/sys/vm/scan_unevictable_pages, print filename and file offset of these pages. [akpm@linux-foundation.org: fix one CONFIG_MMU=n build error] [kosaki.motohiro@jp.fujitsu.com: adapt vmscan-unevictable-lru-scan-sysctl.patch to new sysfs API] Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com> Signed-off-by: Rik van Riel <riel@redhat.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com> Signed-off-by: Hugh Dickins <hugh@veritas.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
-rw-r--r--drivers/base/node.c5
-rw-r--r--include/linux/rmap.h3
-rw-r--r--include/linux/swap.h15
-rw-r--r--kernel/sysctl.c10
-rw-r--r--mm/rmap.c4
-rw-r--r--mm/vmscan.c166
6 files changed, 201 insertions, 2 deletions
diff --git a/drivers/base/node.c b/drivers/base/node.c
index fb45d88a2446..f5207090885a 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -13,6 +13,7 @@
#include <linux/nodemask.h>
#include <linux/cpu.h>
#include <linux/device.h>
+#include <linux/swap.h>
static struct sysdev_class node_class = {
.name = "node",
@@ -191,6 +192,8 @@ int register_node(struct node *node, int num, struct node *parent)
sysdev_create_file(&node->sysdev, &attr_meminfo);
sysdev_create_file(&node->sysdev, &attr_numastat);
sysdev_create_file(&node->sysdev, &attr_distance);
+
+ scan_unevictable_register_node(node);
}
return error;
}
@@ -210,6 +213,8 @@ void unregister_node(struct node *node)
sysdev_remove_file(&node->sysdev, &attr_numastat);
sysdev_remove_file(&node->sysdev, &attr_distance);
+ scan_unevictable_unregister_node(node);
+
sysdev_unregister(&node->sysdev);
}
diff --git a/include/linux/rmap.h b/include/linux/rmap.h
index 955667e6a52d..1da48db8db09 100644
--- a/include/linux/rmap.h
+++ b/include/linux/rmap.h
@@ -75,6 +75,9 @@ void anon_vma_unlink(struct vm_area_struct *);
void anon_vma_link(struct vm_area_struct *);
void __anon_vma_link(struct vm_area_struct *);
+extern struct anon_vma *page_lock_anon_vma(struct page *page);
+extern void page_unlock_anon_vma(struct anon_vma *anon_vma);
+
/*
* rmap interfaces called when adding or removing pte of page
*/
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 07eda69412fb..a3af95b2cb6d 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -7,6 +7,7 @@
#include <linux/list.h>
#include <linux/memcontrol.h>
#include <linux/sched.h>
+#include <linux/node.h>
#include <asm/atomic.h>
#include <asm/page.h>
@@ -235,15 +236,29 @@ static inline int zone_reclaim(struct zone *z, gfp_t mask, unsigned int order)
#ifdef CONFIG_UNEVICTABLE_LRU
extern int page_evictable(struct page *page, struct vm_area_struct *vma);
extern void scan_mapping_unevictable_pages(struct address_space *);
+
+extern unsigned long scan_unevictable_pages;
+extern int scan_unevictable_handler(struct ctl_table *, int, struct file *,
+ void __user *, size_t *, loff_t *);
+extern int scan_unevictable_register_node(struct node *node);
+extern void scan_unevictable_unregister_node(struct node *node);
#else
static inline int page_evictable(struct page *page,
struct vm_area_struct *vma)
{
return 1;
}
+
static inline void scan_mapping_unevictable_pages(struct address_space *mapping)
{
}
+
+static inline int scan_unevictable_register_node(struct node *node)
+{
+ return 0;
+}
+
+static inline void scan_unevictable_unregister_node(struct node *node) { }
#endif
extern int kswapd_run(int nid);
diff --git a/kernel/sysctl.c b/kernel/sysctl.c
index 617d41e4d6a0..b3cc73931d1f 100644
--- a/kernel/sysctl.c
+++ b/kernel/sysctl.c
@@ -833,6 +833,16 @@ static struct ctl_table kern_table[] = {
.proc_handler = &proc_dointvec,
},
#endif
+#ifdef CONFIG_UNEVICTABLE_LRU
+ {
+ .ctl_name = CTL_UNNUMBERED,
+ .procname = "scan_unevictable_pages",
+ .data = &scan_unevictable_pages,
+ .maxlen = sizeof(scan_unevictable_pages),
+ .mode = 0644,
+ .proc_handler = &scan_unevictable_handler,
+ },
+#endif
/*
* NOTE: do not add new entries to this table unless you have read
* Documentation/sysctl/ctl_unnumbered.txt
diff --git a/mm/rmap.c b/mm/rmap.c
index 7e60df99018e..7e90bebbeb6c 100644
--- a/mm/rmap.c
+++ b/mm/rmap.c
@@ -181,7 +181,7 @@ void __init anon_vma_init(void)
* Getting a lock on a stable anon_vma from a page off the LRU is
* tricky: page_lock_anon_vma rely on RCU to guard against the races.
*/
-static struct anon_vma *page_lock_anon_vma(struct page *page)
+struct anon_vma *page_lock_anon_vma(struct page *page)
{
struct anon_vma *anon_vma;
unsigned long anon_mapping;
@@ -201,7 +201,7 @@ out:
return NULL;
}
-static void page_unlock_anon_vma(struct anon_vma *anon_vma)
+void page_unlock_anon_vma(struct anon_vma *anon_vma)
{
spin_unlock(&anon_vma->lock);
rcu_read_unlock();
diff --git a/mm/vmscan.c b/mm/vmscan.c
index e5aaaad159ef..ca64e3e0c518 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -39,6 +39,7 @@
#include <linux/freezer.h>
#include <linux/memcontrol.h>
#include <linux/delayacct.h>
+#include <linux/sysctl.h>
#include <asm/tlbflush.h>
#include <asm/div64.h>
@@ -2363,6 +2364,39 @@ int page_evictable(struct page *page, struct vm_area_struct *vma)
return 1;
}
+static void show_page_path(struct page *page)
+{
+ char buf[256];
+ if (page_is_file_cache(page)) {
+ struct address_space *mapping = page->mapping;
+ struct dentry *dentry;
+ pgoff_t pgoff = page->index << (PAGE_CACHE_SHIFT - PAGE_SHIFT);
+
+ spin_lock(&mapping->i_mmap_lock);
+ dentry = d_find_alias(mapping->host);
+ printk(KERN_INFO "rescued: %s %lu\n",
+ dentry_path(dentry, buf, 256), pgoff);
+ spin_unlock(&mapping->i_mmap_lock);
+ } else {
+#if defined(CONFIG_MM_OWNER) && defined(CONFIG_MMU)
+ struct anon_vma *anon_vma;
+ struct vm_area_struct *vma;
+
+ anon_vma = page_lock_anon_vma(page);
+ if (!anon_vma)
+ return;
+
+ list_for_each_entry(vma, &anon_vma->head, anon_vma_node) {
+ printk(KERN_INFO "rescued: anon %s\n",
+ vma->vm_mm->owner->comm);
+ break;
+ }
+ page_unlock_anon_vma(anon_vma);
+#endif
+ }
+}
+
+
/**
* check_move_unevictable_page - check page for evictability and move to appropriate zone lru list
* @page: page to check evictability and move to appropriate lru list
@@ -2382,6 +2416,9 @@ retry:
ClearPageUnevictable(page);
if (page_evictable(page, NULL)) {
enum lru_list l = LRU_INACTIVE_ANON + page_is_file_cache(page);
+
+ show_page_path(page);
+
__dec_zone_state(zone, NR_UNEVICTABLE);
list_move(&page->lru, &zone->lru[l].list);
__inc_zone_state(zone, NR_INACTIVE_ANON + l);
@@ -2451,4 +2488,133 @@ void scan_mapping_unevictable_pages(struct address_space *mapping)
}
}
+
+/**
+ * scan_zone_unevictable_pages - check unevictable list for evictable pages
+ * @zone - zone of which to scan the unevictable list
+ *
+ * Scan @zone's unevictable LRU lists to check for pages that have become
+ * evictable. Move those that have to @zone's inactive list where they
+ * become candidates for reclaim, unless shrink_inactive_zone() decides
+ * to reactivate them. Pages that are still unevictable are rotated
+ * back onto @zone's unevictable list.
+ */
+#define SCAN_UNEVICTABLE_BATCH_SIZE 16UL /* arbitrary lock hold batch size */
+void scan_zone_unevictable_pages(struct zone *zone)
+{
+ struct list_head *l_unevictable = &zone->lru[LRU_UNEVICTABLE].list;
+ unsigned long scan;
+ unsigned long nr_to_scan = zone_page_state(zone, NR_UNEVICTABLE);
+
+ while (nr_to_scan > 0) {
+ unsigned long batch_size = min(nr_to_scan,
+ SCAN_UNEVICTABLE_BATCH_SIZE);
+
+ spin_lock_irq(&zone->lru_lock);
+ for (scan = 0; scan < batch_size; scan++) {
+ struct page *page = lru_to_page(l_unevictable);
+
+ if (!trylock_page(page))
+ continue;
+
+ prefetchw_prev_lru_page(page, l_unevictable, flags);
+
+ if (likely(PageLRU(page) && PageUnevictable(page)))
+ check_move_unevictable_page(page, zone);
+
+ unlock_page(page);
+ }
+ spin_unlock_irq(&zone->lru_lock);
+
+ nr_to_scan -= batch_size;
+ }
+}
+
+
+/**
+ * scan_all_zones_unevictable_pages - scan all unevictable lists for evictable pages
+ *
+ * A really big hammer: scan all zones' unevictable LRU lists to check for
+ * pages that have become evictable. Move those back to the zones'
+ * inactive list where they become candidates for reclaim.
+ * This occurs when, e.g., we have unswappable pages on the unevictable lists,
+ * and we add swap to the system. As such, it runs in the context of a task
+ * that has possibly/probably made some previously unevictable pages
+ * evictable.
+ */
+void scan_all_zones_unevictable_pages(void)
+{
+ struct zone *zone;
+
+ for_each_zone(zone) {
+ scan_zone_unevictable_pages(zone);
+ }
+}
+
+/*
+ * scan_unevictable_pages [vm] sysctl handler. On demand re-scan of
+ * all nodes' unevictable lists for evictable pages
+ */
+unsigned long scan_unevictable_pages;
+
+int scan_unevictable_handler(struct ctl_table *table, int write,
+ struct file *file, void __user *buffer,
+ size_t *length, loff_t *ppos)
+{
+ proc_doulongvec_minmax(table, write, file, buffer, length, ppos);
+
+ if (write && *(unsigned long *)table->data)
+ scan_all_zones_unevictable_pages();
+
+ scan_unevictable_pages = 0;
+ return 0;
+}
+
+/*
+ * per node 'scan_unevictable_pages' attribute. On demand re-scan of
+ * a specified node's per zone unevictable lists for evictable pages.
+ */
+
+static ssize_t read_scan_unevictable_node(struct sys_device *dev,
+ struct sysdev_attribute *attr,
+ char *buf)
+{
+ return sprintf(buf, "0\n"); /* always zero; should fit... */
+}
+
+static ssize_t write_scan_unevictable_node(struct sys_device *dev,
+ struct sysdev_attribute *attr,
+ const char *buf, size_t count)
+{
+ struct zone *node_zones = NODE_DATA(dev->id)->node_zones;
+ struct zone *zone;
+ unsigned long res;
+ unsigned long req = strict_strtoul(buf, 10, &res);
+
+ if (!req)
+ return 1; /* zero is no-op */
+
+ for (zone = node_zones; zone - node_zones < MAX_NR_ZONES; ++zone) {
+ if (!populated_zone(zone))
+ continue;
+ scan_zone_unevictable_pages(zone);
+ }
+ return 1;
+}
+
+
+static SYSDEV_ATTR(scan_unevictable_pages, S_IRUGO | S_IWUSR,
+ read_scan_unevictable_node,
+ write_scan_unevictable_node);
+
+int scan_unevictable_register_node(struct node *node)
+{
+ return sysdev_create_file(&node->sysdev, &attr_scan_unevictable_pages);
+}
+
+void scan_unevictable_unregister_node(struct node *node)
+{
+ sysdev_remove_file(&node->sysdev, &attr_scan_unevictable_pages);
+}
+
#endif