From: NeilBrown <neilb@suse.de>
Date: Thu, 14 Nov 2013 04:16:15 +0000 (+1100)
Subject: md/raid5.c: add proper locking to error path of raid5_start_reshape.
X-Git-Tag: v3.13-rc1~31^2~16
X-Git-Url: https://git.karo-electronics.de/?a=commitdiff_plain;h=ba8805b97320416e7c5bb8f55d2bd06d5c319e7d;p=karo-tx-linux.git

md/raid5.c: add proper locking to error path of raid5_start_reshape.

If raid5_start_reshape errors out, we need to reset all the fields
that were updated (not just some), and need to use the seq_counter
to ensure make_request() doesn't use an inconsitent state.

Signed-off-by: NeilBrown <neilb@suse.de>
---

diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index f013d3662665..4bbcb7e26d12 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -6384,12 +6384,18 @@ static int raid5_start_reshape(struct mddev *mddev)
 	if (!mddev->sync_thread) {
 		mddev->recovery = 0;
 		spin_lock_irq(&conf->device_lock);
+		write_seqcount_begin(&conf->gen_lock);
 		mddev->raid_disks = conf->raid_disks = conf->previous_raid_disks;
+		mddev->new_chunk_sectors =
+			conf->chunk_sectors = conf->prev_chunk_sectors;
+		mddev->new_layout = conf->algorithm = conf->prev_algo;
 		rdev_for_each(rdev, mddev)
 			rdev->new_data_offset = rdev->data_offset;
 		smp_wmb();
+		conf->generation --;
 		conf->reshape_progress = MaxSector;
 		mddev->reshape_position = MaxSector;
+		write_seqcount_end(&conf->gen_lock);
 		spin_unlock_irq(&conf->device_lock);
 		return -EAGAIN;
 	}