diff options
| author | Lars Ellenberg <lars.ellenberg@linbit.com> | 2012-03-23 14:42:19 +0100 |
|---|---|---|
| committer | Philipp Reisner <philipp.reisner@linbit.com> | 2012-05-09 15:16:50 +0200 |
| commit | d64957c9a9757642f59aa4a63dadf159b2694bab (patch) | |
| tree | a127150d74df0c9cb9ada387427db790fa489c5f /drivers/block | |
| parent | 41c4a0035b36d400b79cbd945390a76e909711a7 (diff) | |
| download | linux-d64957c9a9757642f59aa4a63dadf159b2694bab.tar.gz linux-d64957c9a9757642f59aa4a63dadf159b2694bab.tar.xz linux-d64957c9a9757642f59aa4a63dadf159b2694bab.zip | |
drbd: fix WRITE_ACKED_BY_PEER_AND_SIS to not set RQ_NET_DONE
Just because this request happened during a resync does
not mean it may pretend to have been barrier-acked.
Signed-off-by: Philipp Reisner <philipp.reisner@linbit.com>
Signed-off-by: Lars Ellenberg <lars.ellenberg@linbit.com>
Diffstat (limited to 'drivers/block')
| -rw-r--r-- | drivers/block/drbd/drbd_req.c | 15 |
1 files changed, 5 insertions, 10 deletions
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c index 2a246ac84d7f..cd7687fad9e6 100644 --- a/drivers/block/drbd/drbd_req.c +++ b/drivers/block/drbd/drbd_req.c @@ -627,8 +627,6 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what, _req_may_be_done(req, m); /* Allowed while state.susp */ break; - case write_acked_by_peer_and_sis: - req->rq_state |= RQ_NET_SIS; case conflict_discarded_by_peer: /* for discarded conflicting writes of multiple primaries, * there is no need to keep anything in the tl, potential @@ -639,18 +637,15 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what, (unsigned long long)req->sector, req->size); req->rq_state |= RQ_NET_DONE; /* fall through */ + case write_acked_by_peer_and_sis: case write_acked_by_peer: + if (what == write_acked_by_peer_and_sis) + req->rq_state |= RQ_NET_SIS; /* protocol C; successfully written on peer. - * Nothing to do here. + * Nothing more to do here. * We want to keep the tl in place for all protocols, to cater - * for volatile write-back caches on lower level devices. - * - * A barrier request is expected to have forced all prior - * requests onto stable storage, so completion of a barrier - * request could set NET_DONE right here, and not wait for the - * P_BARRIER_ACK, but that is an unnecessary optimization. */ + * for volatile write-back caches on lower level devices. */ - /* this makes it effectively the same as for: */ case recv_acked_by_peer: /* protocol B; pretends to be successfully written on peer. * see also notes above in handed_over_to_network about |