Patch "md/raid5: avoid races when changing cache size." has been added to the 4.1-stable tree

Patch "md/raid5: avoid races when changing cache size." has been added to the 4.1-stable tree

[gregkh]

This is a note to let you know that I've just added the patch titled

    md/raid5: avoid races when changing cache size.

to the 4.1-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     md-raid5-avoid-races-when-changing-cache-size.patch
and it can be found in the queue-4.1 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From 2d5b569b665ea6d0b15c52529ff06300de81a7ce Mon Sep 17 00:00:00 2001
From: NeilBrown <neilb@suse.com>
Date: Mon, 6 Jul 2015 12:49:23 +1000
Subject: md/raid5: avoid races when changing cache size.

From: NeilBrown <neilb@suse.com>

commit 2d5b569b665ea6d0b15c52529ff06300de81a7ce upstream.

Cache size can grow or shrink due to various pressures at
any time.  So when we resize the cache as part of a 'grow'
operation (i.e. change the size to allow more devices) we need
to blocks that automatic growing/shrinking.

So introduce a mutex.  auto grow/shrink uses mutex_trylock()
and just doesn't bother if there is a blockage.
Resizing the whole cache holds the mutex to ensure that
the correct number of new stripes is allocated.

This bug can result in some stripes not being freed when an
array is stopped.  This leads to the kmem_cache not being
freed and a subsequent array can try to use the same kmem_cache
and get confused.

Fixes: edbe83ab4c27 ("md/raid5: allow the stripe_cache to grow and shrink.")
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>


---
 drivers/md/raid5.c |   31 +++++++++++++++++++++++++------
 drivers/md/raid5.h |    3 ++-
 2 files changed, 27 insertions(+), 7 deletions(-)

--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -2151,6 +2151,9 @@ static int resize_stripes(struct r5conf
 	if (!sc)
 		return -ENOMEM;
 
+	/* Need to ensure auto-resizing doesn't interfere */
+	mutex_lock(&conf->cache_size_mutex);
+
 	for (i = conf->max_nr_stripes; i; i--) {
 		nsh = alloc_stripe(sc, GFP_KERNEL);
 		if (!nsh)
@@ -2167,6 +2170,7 @@ static int resize_stripes(struct r5conf
 			kmem_cache_free(sc, nsh);
 		}
 		kmem_cache_destroy(sc);
+		mutex_unlock(&conf->cache_size_mutex);
 		return -ENOMEM;
 	}
 	/* Step 2 - Must use GFP_NOIO now.
@@ -2213,6 +2217,7 @@ static int resize_stripes(struct r5conf
 	} else
 		err = -ENOMEM;
 
+	mutex_unlock(&conf->cache_size_mutex);
 	/* Step 4, return new stripes to service */
 	while(!list_empty(&newstripes)) {
 		nsh = list_entry(newstripes.next, struct stripe_head, lru);
@@ -5846,12 +5851,14 @@ static void raid5d(struct md_thread *thr
 	pr_debug("%d stripes handled\n", handled);
 
 	spin_unlock_irq(&conf->device_lock);
-	if (test_and_clear_bit(R5_ALLOC_MORE, &conf->cache_state)) {
+	if (test_and_clear_bit(R5_ALLOC_MORE, &conf->cache_state) &&
+	    mutex_trylock(&conf->cache_size_mutex)) {
 		grow_one_stripe(conf, __GFP_NOWARN);
 		/* Set flag even if allocation failed.  This helps
 		 * slow down allocation requests when mem is short
 		 */
 		set_bit(R5_DID_ALLOC, &conf->cache_state);
+		mutex_unlock(&conf->cache_size_mutex);
 	}
 
 	async_tx_issue_pending_all();
@@ -5883,18 +5890,22 @@ raid5_set_cache_size(struct mddev *mddev
 		return -EINVAL;
 
 	conf->min_nr_stripes = size;
+	mutex_lock(&conf->cache_size_mutex);
 	while (size < conf->max_nr_stripes &&
 	       drop_one_stripe(conf))
 		;
+	mutex_unlock(&conf->cache_size_mutex);
 
 
 	err = md_allow_write(mddev);
 	if (err)
 		return err;
 
+	mutex_lock(&conf->cache_size_mutex);
 	while (size > conf->max_nr_stripes)
 		if (!grow_one_stripe(conf, GFP_KERNEL))
 			break;
+	mutex_unlock(&conf->cache_size_mutex);
 
 	return 0;
 }
@@ -6360,11 +6371,18 @@ static unsigned long raid5_cache_scan(st
 				      struct shrink_control *sc)
 {
 	struct r5conf *conf = container_of(shrink, struct r5conf, shrinker);
-	int ret = 0;
-	while (ret < sc->nr_to_scan) {
-		if (drop_one_stripe(conf) == 0)
-			return SHRINK_STOP;
-		ret++;
+	unsigned long ret = SHRINK_STOP;
+
+	if (mutex_trylock(&conf->cache_size_mutex)) {
+		ret= 0;
+		while (ret < sc->nr_to_scan) {
+			if (drop_one_stripe(conf) == 0) {
+				ret = SHRINK_STOP;
+				break;
+			}
+			ret++;
+		}
+		mutex_unlock(&conf->cache_size_mutex);
 	}
 	return ret;
 }
@@ -6433,6 +6451,7 @@ static struct r5conf *setup_conf(struct
 		goto abort;
 	spin_lock_init(&conf->device_lock);
 	seqcount_init(&conf->gen_lock);
+	mutex_init(&conf->cache_size_mutex);
 	init_waitqueue_head(&conf->wait_for_stripe);
 	init_waitqueue_head(&conf->wait_for_overlap);
 	INIT_LIST_HEAD(&conf->handle_list);
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -482,7 +482,8 @@ struct r5conf {
 	 */
 	int			active_name;
 	char			cache_name[2][32];
-	struct kmem_cache		*slab_cache; /* for allocating stripes */
+	struct kmem_cache	*slab_cache; /* for allocating stripes */
+	struct mutex		cache_size_mutex; /* Protect changes to cache size */
 
 	int			seq_flush, seq_write;
 	int			quiesce;


Patches currently in stable-queue which might be from neilb@suse.com are

queue-4.1/md-raid5-avoid-races-when-changing-cache-size.patch
queue-4.1/md-flush-event_work-before-stopping-array.patch
queue-4.1/md-raid5-don-t-let-shrink_slab-shrink-too-far.patch
queue-4.1/nfsv4-don-t-set-setattr-for-o_rdonly-o_excl.patch
queue-4.1/md-raid10-always-set-reshape_safe-when-initializing-reshape_position.patch