[PATCH v4] writeback: safer lock nesting

[Greg Thelen]

lock_page_memcg()/unlock_page_memcg() use spin_lock_irqsave/restore() if
the page's memcg is undergoing move accounting, which occurs when a
process leaves its memcg for a new one that has
memory.move_charge_at_immigrate set.

unlocked_inode_to_wb_begin,end() use spin_lock_irq/spin_unlock_irq() if the
given inode is switching writeback domains.  Switches occur when enough
writes are issued from a new domain.

This existing pattern is thus suspicious:
    lock_page_memcg(page);
    unlocked_inode_to_wb_begin(inode, &locked);
    ...
    unlocked_inode_to_wb_end(inode, locked);
    unlock_page_memcg(page);

If both inode switch and process memcg migration are both in-flight then
unlocked_inode_to_wb_end() will unconditionally enable interrupts while
still holding the lock_page_memcg() irq spinlock.  This suggests the
possibility of deadlock if an interrupt occurs before
unlock_page_memcg().

    truncate
    __cancel_dirty_page
    lock_page_memcg
    unlocked_inode_to_wb_begin
    unlocked_inode_to_wb_end
    <interrupts mistakenly enabled>
                                    <interrupt>
                                    end_page_writeback
                                    test_clear_page_writeback
                                    lock_page_memcg
                                    <deadlock>
    unlock_page_memcg

Due to configuration limitations this deadlock is not currently possible
because we don't mix cgroup writeback (a cgroupv2 feature) and
memory.move_charge_at_immigrate (a cgroupv1 feature).

If the kernel is hacked to always claim inode switching and memcg
moving_account, then this script triggers lockup in less than a minute:
  cd /mnt/cgroup/memory
  mkdir a b
  echo 1 > a/memory.move_charge_at_immigrate
  echo 1 > b/memory.move_charge_at_immigrate
  (
    echo $BASHPID > a/cgroup.procs
    while true; do
      dd if=/dev/zero of=/mnt/big bs=1M count=256
    done
  ) &
  while true; do
    sync
  done &
  sleep 1h &
  SLEEP=$!
  while true; do
    echo $SLEEP > a/cgroup.procs
    echo $SLEEP > b/cgroup.procs
  done

The deadlock does not seem possible, so it's debatable if there's
any reason to modify the kernel.  I suggest we should to prevent future
surprises.  And Wang Long said "this deadlock occurs three times in our
environment", so there's more reason to apply this, even to stable.
Stable 4.4 has minor conflicts applying this patch.
For a clean 4.4 patch see "[PATCH for-4.4] writeback: safer lock nesting"
https://lkml.org/lkml/2018/4/11/146

Fixes: 682aa8e1a6a1 ("writeback: implement unlocked_inode_to_wb transaction and use it for stat updates")
Cc: stable@xxxxxxxxxxxxxxx # v4.2+
Reported-by: Wang Long <wanglong19@xxxxxxxxxxx>
Signed-off-by: Greg Thelen <gthelen@xxxxxxxxxx>
Acked-by: Michal Hocko <mhocko@xxxxxxxx>
Acked-by: Wang Long <wanglong19@xxxxxxxxxxx>
---

Changelog since v3:
- initialize wb_lock_cookie wiht {} rather than {0}.
- comment grammar fix
- commit log footer cleanup (-Change-Id, +Fixes, +Acks, +stable), though patch
  does not cleanly apply to 4.4.  I'll post a 4.4-stable specific patch.

Changelog since v2:
- explicitly initialize wb_lock_cookie to silence compiler warnings.

Changelog since v1:
- add wb_lock_cookie to record lock context.

 fs/fs-writeback.c                |  7 ++++---
 include/linux/backing-dev-defs.h |  5 +++++
 include/linux/backing-dev.h      | 31 +++++++++++++++++--------------
 mm/page-writeback.c              | 18 +++++++++---------
 4 files changed, 35 insertions(+), 26 deletions(-)

diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c
index 1280f915079b..b1178acfcb08 100644
--- a/fs/fs-writeback.c
+++ b/fs/fs-writeback.c
@@ -745,11 +745,12 @@ int inode_congested(struct inode *inode, int cong_bits)
 	 */
 	if (inode && inode_to_wb_is_valid(inode)) {
 		struct bdi_writeback *wb;
-		bool locked, congested;
+		struct wb_lock_cookie lock_cookie = {};
+		bool congested;
 
-		wb = unlocked_inode_to_wb_begin(inode, &locked);
+		wb = unlocked_inode_to_wb_begin(inode, &lock_cookie);
 		congested = wb_congested(wb, cong_bits);
-		unlocked_inode_to_wb_end(inode, locked);
+		unlocked_inode_to_wb_end(inode, &lock_cookie);
 		return congested;
 	}
 
diff --git a/include/linux/backing-dev-defs.h b/include/linux/backing-dev-defs.h
index bfe86b54f6c1..0bd432a4d7bd 100644
--- a/include/linux/backing-dev-defs.h
+++ b/include/linux/backing-dev-defs.h
@@ -223,6 +223,11 @@ static inline void set_bdi_congested(struct backing_dev_info *bdi, int sync)
 	set_wb_congested(bdi->wb.congested, sync);
 }
 
+struct wb_lock_cookie {
+	bool locked;
+	unsigned long flags;
+};
+
 #ifdef CONFIG_CGROUP_WRITEBACK
 
 /**
diff --git a/include/linux/backing-dev.h b/include/linux/backing-dev.h
index 3e4ce54d84ab..96f4a3ddfb81 100644
--- a/include/linux/backing-dev.h
+++ b/include/linux/backing-dev.h
@@ -346,7 +346,7 @@ static inline struct bdi_writeback *inode_to_wb(const struct inode *inode)
 /**
  * unlocked_inode_to_wb_begin - begin unlocked inode wb access transaction
  * @inode: target inode
- * @lockedp: temp bool output param, to be passed to the end function
+ * @cookie: output param, to be passed to the end function
  *
  * The caller wants to access the wb associated with @inode but isn't
  * holding inode->i_lock, mapping->tree_lock or wb->list_lock.  This
@@ -354,12 +354,12 @@ static inline struct bdi_writeback *inode_to_wb(const struct inode *inode)
  * association doesn't change until the transaction is finished with
  * unlocked_inode_to_wb_end().
  *
- * The caller must call unlocked_inode_to_wb_end() with *@lockdep
- * afterwards and can't sleep during transaction.  IRQ may or may not be
- * disabled on return.
+ * The caller must call unlocked_inode_to_wb_end() with *@cookie afterwards and
+ * can't sleep during the transaction.  IRQs may or may not be disabled on
+ * return.
  */
 static inline struct bdi_writeback *
-unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp)
+unlocked_inode_to_wb_begin(struct inode *inode, struct wb_lock_cookie *cookie)
 {
 	rcu_read_lock();
 
@@ -367,10 +367,10 @@ unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp)
 	 * Paired with store_release in inode_switch_wb_work_fn() and
 	 * ensures that we see the new wb if we see cleared I_WB_SWITCH.
 	 */
-	*lockedp = smp_load_acquire(&inode->i_state) & I_WB_SWITCH;
+	cookie->locked = smp_load_acquire(&inode->i_state) & I_WB_SWITCH;
 
-	if (unlikely(*lockedp))
-		spin_lock_irq(&inode->i_mapping->tree_lock);
+	if (unlikely(cookie->locked))
+		spin_lock_irqsave(&inode->i_mapping->tree_lock, cookie->flags);
 
 	/*
 	 * Protected by either !I_WB_SWITCH + rcu_read_lock() or tree_lock.
@@ -382,12 +382,14 @@ unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp)
 /**
  * unlocked_inode_to_wb_end - end inode wb access transaction
  * @inode: target inode
- * @locked: *@lockedp from unlocked_inode_to_wb_begin()
+ * @cookie: @cookie from unlocked_inode_to_wb_begin()
  */
-static inline void unlocked_inode_to_wb_end(struct inode *inode, bool locked)
+static inline void unlocked_inode_to_wb_end(struct inode *inode,
+					    struct wb_lock_cookie *cookie)
 {
-	if (unlikely(locked))
-		spin_unlock_irq(&inode->i_mapping->tree_lock);
+	if (unlikely(cookie->locked))
+		spin_unlock_irqrestore(&inode->i_mapping->tree_lock,
+				       cookie->flags);
 
 	rcu_read_unlock();
 }
@@ -434,12 +436,13 @@ static inline struct bdi_writeback *inode_to_wb(struct inode *inode)
 }
 
 static inline struct bdi_writeback *
-unlocked_inode_to_wb_begin(struct inode *inode, bool *lockedp)
+unlocked_inode_to_wb_begin(struct inode *inode, struct wb_lock_cookie *cookie)
 {
 	return inode_to_wb(inode);
 }
 
-static inline void unlocked_inode_to_wb_end(struct inode *inode, bool locked)
+static inline void unlocked_inode_to_wb_end(struct inode *inode,
+					    struct wb_lock_cookie *cookie)
 {
 }
 
diff --git a/mm/page-writeback.c b/mm/page-writeback.c
index 586f31261c83..8369572e1f7d 100644
--- a/mm/page-writeback.c
+++ b/mm/page-writeback.c
@@ -2501,13 +2501,13 @@ void account_page_redirty(struct page *page)
 	if (mapping && mapping_cap_account_dirty(mapping)) {
 		struct inode *inode = mapping->host;
 		struct bdi_writeback *wb;
-		bool locked;
+		struct wb_lock_cookie cookie = {};
 
-		wb = unlocked_inode_to_wb_begin(inode, &locked);
+		wb = unlocked_inode_to_wb_begin(inode, &cookie);
 		current->nr_dirtied--;
 		dec_node_page_state(page, NR_DIRTIED);
 		dec_wb_stat(wb, WB_DIRTIED);
-		unlocked_inode_to_wb_end(inode, locked);
+		unlocked_inode_to_wb_end(inode, &cookie);
 	}
 }
 EXPORT_SYMBOL(account_page_redirty);
@@ -2613,15 +2613,15 @@ void __cancel_dirty_page(struct page *page)
 	if (mapping_cap_account_dirty(mapping)) {
 		struct inode *inode = mapping->host;
 		struct bdi_writeback *wb;
-		bool locked;
+		struct wb_lock_cookie cookie = {};
 
 		lock_page_memcg(page);
-		wb = unlocked_inode_to_wb_begin(inode, &locked);
+		wb = unlocked_inode_to_wb_begin(inode, &cookie);
 
 		if (TestClearPageDirty(page))
 			account_page_cleaned(page, mapping, wb);
 
-		unlocked_inode_to_wb_end(inode, locked);
+		unlocked_inode_to_wb_end(inode, &cookie);
 		unlock_page_memcg(page);
 	} else {
 		ClearPageDirty(page);
@@ -2653,7 +2653,7 @@ int clear_page_dirty_for_io(struct page *page)
 	if (mapping && mapping_cap_account_dirty(mapping)) {
 		struct inode *inode = mapping->host;
 		struct bdi_writeback *wb;
-		bool locked;
+		struct wb_lock_cookie cookie = {};
 
 		/*
 		 * Yes, Virginia, this is indeed insane.
@@ -2690,14 +2690,14 @@ int clear_page_dirty_for_io(struct page *page)
 		 * always locked coming in here, so we get the desired
 		 * exclusion.
 		 */
-		wb = unlocked_inode_to_wb_begin(inode, &locked);
+		wb = unlocked_inode_to_wb_begin(inode, &cookie);
 		if (TestClearPageDirty(page)) {
 			dec_lruvec_page_state(page, NR_FILE_DIRTY);
 			dec_zone_page_state(page, NR_ZONE_WRITE_PENDING);
 			dec_wb_stat(wb, WB_RECLAIMABLE);
 			ret = 1;
 		}
-		unlocked_inode_to_wb_end(inode, locked);
+		unlocked_inode_to_wb_end(inode, &cookie);
 		return ret;
 	}
 	return TestClearPageDirty(page);
-- 
2.17.0.484.g0c8726318c-goog