btrfs: get rid of endio_repair_workers
authorOmar Sandoval <osandov@fb.com>
Thu, 16 Apr 2020 21:46:24 +0000 (14:46 -0700)
committerDavid Sterba <dsterba@suse.com>
Mon, 25 May 2020 09:25:27 +0000 (11:25 +0200)
commit5c047a699aa9433ad92136343a9306d985134c24
tree6b50a96455fab869499f944b0f7d4cc4eecf013f
parentfd9d6670edba68cedd87b3db04cbdcdfc492cc0a
btrfs: get rid of endio_repair_workers

This was originally added in commit 8b110e393c5a ("Btrfs: implement
repair function when direct read fails") to avoid a deadlock. In that
commit, the direct I/O read endio executes on the endio_workers
workqueue, submits a repair bio, and waits for it to complete. The
repair bio endio must execute on a different workqueue, otherwise it
could block on the endio_workers workqueue becoming available, which
won't happen because the original endio is blocked on the repair bio.

As of the previous commit, the original endio doesn't wait for the
repair bio, so this separate workqueue is unnecessary.

Reviewed-by: Josef Bacik <josef@toxicpanda.com>
Signed-off-by: Omar Sandoval <osandov@fb.com>
Signed-off-by: David Sterba <dsterba@suse.com>
fs/btrfs/ctree.h
fs/btrfs/disk-io.c
fs/btrfs/disk-io.h
fs/btrfs/inode.c