dm: avoid destroying table in dm_any_congested
dm_any_congested() just checks for the DMF_BLOCK_IO and has no
code to make sure that suspend waits for dm_any_congested() to
complete. This patch adds such a check.
Without it, a race can occur with dm_table_put() attempting to
destroying the table in the wrong thread, the one running
dm_any_congested() which is meant to be quick and return
immediately.
Two examples of problems:
1. Sleeping functions called from congested code, the caller
of which holds a spin lock.
2. An ABBA deadlock between pdflush and multipathd. The two locks
in contention are inode lock and kernel lock.
Signed-off-by: Chandra Seetharaman <sekharan@us.ibm.com>
Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index dc25d8a..c99e4728 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -937,16 +937,24 @@
static int dm_any_congested(void *congested_data, int bdi_bits)
{
- int r;
- struct mapped_device *md = (struct mapped_device *) congested_data;
- struct dm_table *map = dm_get_table(md);
+ int r = bdi_bits;
+ struct mapped_device *md = congested_data;
+ struct dm_table *map;
- if (!map || test_bit(DMF_BLOCK_IO, &md->flags))
- r = bdi_bits;
- else
- r = dm_table_any_congested(map, bdi_bits);
+ atomic_inc(&md->pending);
- dm_table_put(map);
+ if (!test_bit(DMF_BLOCK_IO, &md->flags)) {
+ map = dm_get_table(md);
+ if (map) {
+ r = dm_table_any_congested(map, bdi_bits);
+ dm_table_put(map);
+ }
+ }
+
+ if (!atomic_dec_return(&md->pending))
+ /* nudge anyone waiting on suspend queue */
+ wake_up(&md->wait);
+
return r;
}