From 3e45fb10a9a818a765ce0cc8139ea0c78cb42e1e Mon Sep 17 00:00:00 2001
From: Tim Chen <tim.c.chen@linux.intel.com>
Date: Fri, 25 Aug 2017 09:13:55 -0700
Subject: [PATCH] sched/wait: Introduce wakeup boomark in wake_up_page_bit

References: bnc#971975 VM performance -- page waitqueue
Patch-mainline: v4.14-rc1
Git-commit: 11a19c7b099f96d00a8dec52bfbb8475e89b6745

Now that we have added breaks in the wait queue scan and allow bookmark
on scan position, we put this logic in the wake_up_page_bit function.

We can have very long page wait list in large system where multiple
pages share the same wait list. We break the wake up walk here to allow
other cpus a chance to access the list, and not to disable the interrupts
when traversing the list for too long.  This reduces the interrupt and
rescheduling latency, and excessive page wait queue lock hold time.

[ v2: Remove bookmark_wake_function ]

Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Mel Gorman <mgorman@suse.de>
---
 include/linux/wait.h |  2 ++
 kernel/sched/wait.c  |  7 +++++++
 mm/filemap.c         | 22 +++++++++++++++++++++-
 3 files changed, 30 insertions(+), 1 deletion(-)

diff --git a/include/linux/wait.h b/include/linux/wait.h
index 80034e8176ac..ed7ba30883e4 100644
--- a/include/linux/wait.h
+++ b/include/linux/wait.h
@@ -185,6 +185,8 @@ __remove_wait_queue(struct wait_queue_head *wq_head, struct wait_queue_entry *wq
 
 void __wake_up(struct wait_queue_head *wq_head, unsigned int mode, int nr, void *key);
 void __wake_up_locked_key(struct wait_queue_head *wq_head, unsigned int mode, void *key);
+void __wake_up_locked_key_bookmark(struct wait_queue_head *wq_head,
+		unsigned int mode, void *key, wait_queue_entry_t *bookmark);
 void __wake_up_sync_key(struct wait_queue_head *wq_head, unsigned int mode, int nr, void *key);
 void __wake_up_locked(struct wait_queue_head *wq_head, unsigned int mode, int nr);
 void __wake_up_sync(struct wait_queue_head *wq_head, unsigned int mode, int nr);
diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c
index 70701ef50465..98feab7933c7 100644
--- a/kernel/sched/wait.c
+++ b/kernel/sched/wait.c
@@ -165,6 +165,13 @@ void __wake_up_locked_key(struct wait_queue_head *wq_head, unsigned int mode, vo
 }
 EXPORT_SYMBOL_GPL(__wake_up_locked_key);
 
+void __wake_up_locked_key_bookmark(struct wait_queue_head *wq_head,
+		unsigned int mode, void *key, wait_queue_entry_t *bookmark)
+{
+	__wake_up_common(wq_head, mode, 1, 0, key, bookmark);
+}
+EXPORT_SYMBOL_GPL(__wake_up_locked_key_bookmark);
+
 /**
  * __wake_up_sync_key - wake up threads blocked on a waitqueue.
  * @wq_head: the waitqueue
diff --git a/mm/filemap.c b/mm/filemap.c
index 78159c12bdc2..da5b72522aae 100644
--- a/mm/filemap.c
+++ b/mm/filemap.c
@@ -829,13 +829,33 @@ static void wake_up_page_bit(struct page *page, int bit_nr)
 	wait_queue_head_t *q = page_waitqueue(page);
 	struct wait_page_key key;
 	unsigned long flags;
+	wait_queue_entry_t bookmark;
 
 	key.page = page;
 	key.bit_nr = bit_nr;
 	key.page_match = 0;
 
+	bookmark.flags = 0;
+	bookmark.private = NULL;
+	bookmark.func = NULL;
+	INIT_LIST_HEAD(&bookmark.entry);
+
 	spin_lock_irqsave(&q->lock, flags);
-	__wake_up_locked_key(q, TASK_NORMAL, &key);
+	__wake_up_locked_key_bookmark(q, TASK_NORMAL, &key, &bookmark);
+
+	while (bookmark.flags & WQ_FLAG_BOOKMARK) {
+		/*
+		 * Take a breather from holding the lock,
+		 * allow pages that finish wake up asynchronously
+		 * to acquire the lock and remove themselves
+		 * from wait queue
+		 */
+		spin_unlock_irqrestore(&q->lock, flags);
+		cpu_relax();
+		spin_lock_irqsave(&q->lock, flags);
+		__wake_up_locked_key_bookmark(q, TASK_NORMAL, &key, &bookmark);
+	}
+
 	/*
 	 * It is possible for other pages to have collided on the waitqueue
 	 * hash, so in that case check for a page match. That prevents a long-