memcg: fix deadlock between lock_page_cgroup and mapping tree_lock
mapping->tree_lock can be acquired from interrupt context. Then,
following dead lock can occur.
Assume "A" as a page.
CPU0:
lock_page_cgroup(A)
interrupted
-> take mapping->tree_lock.
CPU1:
take mapping->tree_lock
-> lock_page_cgroup(A)
This patch tries to fix above deadlock by moving memcg's hook to out of
mapping->tree_lock. charge/uncharge of pagecache/swapcache is protected
by page lock, not tree_lock.
After this patch, lock_page_cgroup() is not called under mapping->tree_lock.
Signed-off-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: Balbir Singh <balbir@in.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/mm/filemap.c b/mm/filemap.c
index 379ff0b..1b60f30 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -121,7 +121,6 @@
mapping->nrpages--;
__dec_zone_page_state(page, NR_FILE_PAGES);
BUG_ON(page_mapped(page));
- mem_cgroup_uncharge_cache_page(page);
/*
* Some filesystems seem to re-dirty the page even after
@@ -145,6 +144,7 @@
spin_lock_irq(&mapping->tree_lock);
__remove_from_page_cache(page);
spin_unlock_irq(&mapping->tree_lock);
+ mem_cgroup_uncharge_cache_page(page);
}
static int sync_page(void *word)
@@ -476,13 +476,13 @@
if (likely(!error)) {
mapping->nrpages++;
__inc_zone_page_state(page, NR_FILE_PAGES);
+ spin_unlock_irq(&mapping->tree_lock);
} else {
page->mapping = NULL;
+ spin_unlock_irq(&mapping->tree_lock);
mem_cgroup_uncharge_cache_page(page);
page_cache_release(page);
}
-
- spin_unlock_irq(&mapping->tree_lock);
radix_tree_preload_end();
} else
mem_cgroup_uncharge_cache_page(page);
diff --git a/mm/memcontrol.c b/mm/memcontrol.c
index 01c2d8f1..4a747a2 100644
--- a/mm/memcontrol.c
+++ b/mm/memcontrol.c
@@ -1488,8 +1488,9 @@
__mem_cgroup_uncharge_common(page, MEM_CGROUP_CHARGE_TYPE_CACHE);
}
+#ifdef CONFIG_SWAP
/*
- * called from __delete_from_swap_cache() and drop "page" account.
+ * called after __delete_from_swap_cache() and drop "page" account.
* memcg information is recorded to swap_cgroup of "ent"
*/
void mem_cgroup_uncharge_swapcache(struct page *page, swp_entry_t ent)
@@ -1506,6 +1507,7 @@
if (memcg)
css_put(&memcg->css);
}
+#endif
#ifdef CONFIG_CGROUP_MEM_RES_CTLR_SWAP
/*
diff --git a/mm/swap_state.c b/mm/swap_state.c
index 3ecea98..1416e7e 100644
--- a/mm/swap_state.c
+++ b/mm/swap_state.c
@@ -109,8 +109,6 @@
*/
void __delete_from_swap_cache(struct page *page)
{
- swp_entry_t ent = {.val = page_private(page)};
-
VM_BUG_ON(!PageLocked(page));
VM_BUG_ON(!PageSwapCache(page));
VM_BUG_ON(PageWriteback(page));
@@ -121,7 +119,6 @@
total_swapcache_pages--;
__dec_zone_page_state(page, NR_FILE_PAGES);
INC_CACHE_INFO(del_total);
- mem_cgroup_uncharge_swapcache(page, ent);
}
/**
@@ -191,6 +188,7 @@
__delete_from_swap_cache(page);
spin_unlock_irq(&swapper_space.tree_lock);
+ mem_cgroup_uncharge_swapcache(page, entry);
swap_free(entry);
page_cache_release(page);
}
diff --git a/mm/truncate.c b/mm/truncate.c
index 55206fa..12e1579 100644
--- a/mm/truncate.c
+++ b/mm/truncate.c
@@ -359,6 +359,7 @@
BUG_ON(page_has_private(page));
__remove_from_page_cache(page);
spin_unlock_irq(&mapping->tree_lock);
+ mem_cgroup_uncharge_cache_page(page);
page_cache_release(page); /* pagecache ref */
return 1;
failed:
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 5fa3eda..d254306 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -470,10 +470,12 @@
swp_entry_t swap = { .val = page_private(page) };
__delete_from_swap_cache(page);
spin_unlock_irq(&mapping->tree_lock);
+ mem_cgroup_uncharge_swapcache(page, swap);
swap_free(swap);
} else {
__remove_from_page_cache(page);
spin_unlock_irq(&mapping->tree_lock);
+ mem_cgroup_uncharge_cache_page(page);
}
return 1;