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>

1 files changed, 3 insertions, 0 deletions

diff --git a/drivers/block/drbd/drbd_req.h b/drivers/block/drbd/drbd_req.h
index 68f54050b7ca..492f81d3765a 100644
--- a/drivers/block/drbd/drbd_req.h
+++ b/drivers/block/drbd/drbd_req.h
@@ -268,6 +268,9 @@ extern void request_timer_fn(unsigned long data);
 extern void tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what);
 extern void _tl_restart(struct drbd_tconn *tconn, enum drbd_req_event what);
 
+/* this is in drbd_main.c */
+extern void drbd_restart_write(struct drbd_request *req);
+
 /* use this if you don't want to deal with calling complete_master_bio()
  * outside the spinlock, e.g. when walking some list on cleanup. */
 static inline int _req_mod(struct drbd_request *req, enum drbd_req_event what)