md: warn about updating super block failure

Many personalities will handle IO error from daemon thread(like raid1d,
raid10d, raid5d), and sb will require to be clean before hanlding these
failed IO. However update sb can fail, for example array is broken by
IO failure, or user config sysfs api array_state.

This patch adds warning if updating sb failed first, in case this will
be related to IO hang.

Link: https://lore.kernel.org/linux-raid/20251117085557.770572-2-yukuai@fnnas.com
Signed-off-by: Yu Kuai <yukuai@fnnas.com>
Reviewed-by: Li Nan <linan122@huawei.com>

diff --git a/drivers/md/md.c b/drivers/md/md.c
index 7b5c5967568fba1310bbea57072cf2d183c5222c..345b1e623abaf44f02d39c29535a3f2c58e7838f 100644 (file)
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2788,6 +2788,7 @@ void md_update_sb(struct mddev *mddev, int force_change)
        if (!md_is_rdwr(mddev)) {
                if (force_change)
                        set_bit(MD_SB_CHANGE_DEVS, &mddev->sb_flags);
+               pr_err("%s: can't update sb for read-only array %s\n", __func__, mdname(mddev));
                return;
        }