aboutsummaryrefslogtreecommitdiff
path: root/drivers/block/drbd/drbd_receiver.c
diff options
context:
space:
mode:
authorLars Ellenberg <lars.ellenberg@linbit.com>2011-11-24 10:36:25 +0100
committerPhilipp Reisner <philipp.reisner@linbit.com>2012-11-08 16:58:21 +0100
commit2312f0b3c5ab794fbac9e9bebe90c784c9d449c5 (patch)
tree0b4b74af502169d61e026855593b828fa202665a /drivers/block/drbd/drbd_receiver.c
parentf9916d61a40e7ad43c2a20444894f85c45512f91 (diff)
downloadvexpress-lsk-2312f0b3c5ab794fbac9e9bebe90c784c9d449c5.tar.gz
drbd: fix potential deadlock during "restart" of conflicting writes
w_restart_write(), run from worker context, calls __drbd_make_request() and further drbd_al_begin_io(, delegate=true), which then potentially deadlocks. The previous patch moved a BUG_ON to expose such call paths, which would now be triggered. Also, if we call __drbd_make_request() from resource worker context, like w_restart_write() did, and that should block for whatever reason (!drbd_state_is_stable(), resource suspended, ...), we potentially deadlock the whole resource, as the worker is needed for state changes and other things. Create a dedicated retry workqueue for this instead. Also make sure that inc_ap_bio()/dec_ap_bio() are properly paired, even if do_retry() needs to retry itself, in case __drbd_make_request() returns != 0. Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com> Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Diffstat (limited to 'drivers/block/drbd/drbd_receiver.c')
-rw-r--r--drivers/block/drbd/drbd_receiver.c32
1 files changed, 3 insertions, 29 deletions
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 8a7f61ba74a..b159ad15abe 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -1748,30 +1748,6 @@ static int receive_RSDataReply(struct drbd_tconn *tconn, struct packet_info *pi)
return err;
}
-static int w_restart_write(struct drbd_work *w, int cancel)
-{
- struct drbd_request *req = container_of(w, struct drbd_request, w);
- struct drbd_conf *mdev = w->mdev;
- struct bio *bio;
- unsigned long start_time;
- unsigned long flags;
-
- spin_lock_irqsave(&mdev->tconn->req_lock, flags);
- if (!expect(req->rq_state & RQ_POSTPONED)) {
- spin_unlock_irqrestore(&mdev->tconn->req_lock, flags);
- return -EIO;
- }
- bio = req->master_bio;
- start_time = req->start_time;
- /* Postponed requests will not have their master_bio completed! */
- __req_mod(req, DISCARD_WRITE, NULL);
- spin_unlock_irqrestore(&mdev->tconn->req_lock, flags);
-
- while (__drbd_make_request(mdev, bio, start_time))
- /* retry */ ;
- return 0;
-}
-
static void restart_conflicting_writes(struct drbd_conf *mdev,
sector_t sector, int size)
{
@@ -1785,11 +1761,9 @@ static void restart_conflicting_writes(struct drbd_conf *mdev,
if (req->rq_state & RQ_LOCAL_PENDING ||
!(req->rq_state & RQ_POSTPONED))
continue;
- if (expect(list_empty(&req->w.list))) {
- req->w.mdev = mdev;
- req->w.cb = w_restart_write;
- drbd_queue_work(&mdev->tconn->data.work, &req->w);
- }
+ /* as it is RQ_POSTPONED, this will cause it to
+ * be queued on the retry workqueue. */
+ __req_mod(req, DISCARD_WRITE, NULL);
}
}