Merge tag 'md/4.0-fixes' of git://neil.brown.name/md

Pull md fixes from Neil Brown:
 "Three md fixes:

   - fix a read-balance problem that was reported 2 years ago, but that
     I never noticed the report :-(

   - fix for rare RAID6 problem causing incorrect bitmap updates when
     two devices fail.

   - add __ATTR_PREALLOC annotation now that it is possible"

* tag 'md/4.0-fixes' of git://neil.brown.name/md:
  md: mark some attributes as pre-alloc
  raid5: check faulty flag for array status during recovery.
  md/raid1: fix read balance when a drive is write-mostly.

3 files changed, 20 insertions, 12 deletions

diff --git a/drivers/md/md.c b/drivers/md/md.c
index c8d2bac4e28b..cadf9cc02b25 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2555,7 +2555,7 @@ state_store(struct md_rdev *rdev, const char *buf, size_t len)
 	return err ? err : len;
 }
 static struct rdev_sysfs_entry rdev_state =
-__ATTR(state, S_IRUGO|S_IWUSR, state_show, state_store);
+__ATTR_PREALLOC(state, S_IRUGO|S_IWUSR, state_show, state_store);
 
 static ssize_t
 errors_show(struct md_rdev *rdev, char *page)
@@ -3638,7 +3638,8 @@ resync_start_store(struct mddev *mddev, const char *buf, size_t len)
 	return err ?: len;
 }
 static struct md_sysfs_entry md_resync_start =
-__ATTR(resync_start, S_IRUGO|S_IWUSR, resync_start_show, resync_start_store);
+__ATTR_PREALLOC(resync_start, S_IRUGO|S_IWUSR,
+		resync_start_show, resync_start_store);
 
 /*
  * The array state can be:
@@ -3851,7 +3852,7 @@ array_state_store(struct mddev *mddev, const char *buf, size_t len)
 	return err ?: len;
 }
 static struct md_sysfs_entry md_array_state =
-__ATTR(array_state, S_IRUGO|S_IWUSR, array_state_show, array_state_store);
+__ATTR_PREALLOC(array_state, S_IRUGO|S_IWUSR, array_state_show, array_state_store);
 
 static ssize_t
 max_corrected_read_errors_show(struct mddev *mddev, char *page) {
@@ -4101,7 +4102,7 @@ out_unlock:
 }
 
 static struct md_sysfs_entry md_metadata =
-__ATTR(metadata_version, S_IRUGO|S_IWUSR, metadata_show, metadata_store);
+__ATTR_PREALLOC(metadata_version, S_IRUGO|S_IWUSR, metadata_show, metadata_store);
 
 static ssize_t
 action_show(struct mddev *mddev, char *page)
@@ -4189,7 +4190,7 @@ action_store(struct mddev *mddev, const char *page, size_t len)
 }
 
 static struct md_sysfs_entry md_scan_mode =
-__ATTR(sync_action, S_IRUGO|S_IWUSR, action_show, action_store);
+__ATTR_PREALLOC(sync_action, S_IRUGO|S_IWUSR, action_show, action_store);
 
 static ssize_t
 last_sync_action_show(struct mddev *mddev, char *page)
@@ -4335,7 +4336,8 @@ sync_completed_show(struct mddev *mddev, char *page)
 	return sprintf(page, "%llu / %llu\n", resync, max_sectors);
 }
 
-static struct md_sysfs_entry md_sync_completed = __ATTR_RO(sync_completed);
+static struct md_sysfs_entry md_sync_completed =
+	__ATTR_PREALLOC(sync_completed, S_IRUGO, sync_completed_show, NULL);
 
 static ssize_t
 min_sync_show(struct mddev *mddev, char *page)
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index 4153da5d4011..d34e238afa54 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -560,7 +560,7 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
 		if (test_bit(WriteMostly, &rdev->flags)) {
 			/* Don't balance among write-mostly, just
 			 * use the first as a last resort */
-			if (best_disk < 0) {
+			if (best_dist_disk < 0) {
 				if (is_badblock(rdev, this_sector, sectors,
 						&first_bad, &bad_sectors)) {
 					if (first_bad < this_sector)
@@ -569,7 +569,8 @@ static int read_balance(struct r1conf *conf, struct r1bio *r1_bio, int *max_sect
 					best_good_sectors = first_bad - this_sector;
 				} else
 					best_good_sectors = sectors;
-				best_disk = disk;
+				best_dist_disk = disk;
+				best_pending_disk = disk;
 			}
 			continue;
 		}
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index e75d48c0421a..cd2f96b2c572 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -5121,12 +5121,17 @@ static inline sector_t sync_request(struct mddev *mddev, sector_t sector_nr, int
 		schedule_timeout_uninterruptible(1);
 	}
 	/* Need to check if array will still be degraded after recovery/resync
-	 * We don't need to check the 'failed' flag as when that gets set,
-	 * recovery aborts.
+	 * Note in case of > 1 drive failures it's possible we're rebuilding
+	 * one drive while leaving another faulty drive in array.
 	 */
-	for (i = 0; i < conf->raid_disks; i++)
-		if (conf->disks[i].rdev == NULL)
+	rcu_read_lock();
+	for (i = 0; i < conf->raid_disks; i++) {
+		struct md_rdev *rdev = ACCESS_ONCE(conf->disks[i].rdev);
+
+		if (rdev == NULL || test_bit(Faulty, &rdev->flags))
 			still_degraded = 1;
+	}
+	rcu_read_unlock();
 
 	bitmap_start_sync(mddev->bitmap, sector_nr, &sync_blocks, still_degraded);