[SCSI] bsg: fix bsg_mutex hang with device removal

We don't need to hold bsg_mutex during bsg_complete_all_commands(). It
leads to a problem that we block bsg_unregister_queue during
bsg_complete_all_commands (untill all the outstanding commands
complete).

Thanks to Pete Wyckoff for finding the bug and testing the patch.

The detailed bug report is:

http://marc.info/?l=linux-scsi&m=121182137132145&w=2

Tested-by: Pete Wyckoff <pw@osc.edu>
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@HansenPartnership.com>

1 files changed, 6 insertions, 3 deletions

diff --git a/block/bsg.c b/block/bsg.c
index f0b7cd343216..7cdec32205d8 100644
--- a/block/bsg.c
+++ b/block/bsg.c
@@ -724,8 +724,13 @@ static int bsg_put_device(struct bsg_device *bd)
         mutex_lock(&bsg_mutex);
 
         do_free = atomic_dec_and_test(&bd->ref_count);
-        if (!do_free)
+        if (!do_free) {
+                mutex_unlock(&bsg_mutex);
                 goto out;
+        }
+
+        hlist_del(&bd->dev_list);
+        mutex_unlock(&bsg_mutex);
 
         dprintk("%s: tearing down\n", bd->name);
 
@@ -741,10 +746,8 @@ static int bsg_put_device(struct bsg_device *bd)
          */
         ret = bsg_complete_all_commands(bd);
 
-        hlist_del(&bd->dev_list);
         kfree(bd);
 out:
-        mutex_unlock(&bsg_mutex);
         kref_put(&q->bsg_dev.ref, bsg_kref_release_function);
         if (do_free)
                 blk_put_queue(q);