diff --git a/drivers/block/drbd/drbd_main.c b/drivers/block/drbd/drbd_main.c
index 11d8ff6016ac776093128082bc5520ca90d8ab46..157d1e4343c23a957c751778cd7781cca1ee7833 100644
--- a/drivers/block/drbd/drbd_main.c
+++ b/drivers/block/drbd/drbd_main.c
@@ -867,10 +867,9 @@ static union drbd_state sanitize_state(struct drbd_conf *mdev, union drbd_state
 	}
 
 	if (fp == FP_STONITH &&
-	    (ns.role == R_PRIMARY &&
-	     ns.conn < C_CONNECTED &&
-	     ns.pdsk > D_OUTDATED))
-			ns.susp = 1;
+	    (ns.role == R_PRIMARY && ns.conn < C_CONNECTED && ns.pdsk > D_OUTDATED) &&
+	    !(os.role == R_PRIMARY && os.conn < C_CONNECTED && os.pdsk > D_OUTDATED))
+		ns.susp = 1;
 
 	if (ns.aftr_isp || ns.peer_isp || ns.user_isp) {
 		if (ns.conn == C_SYNC_SOURCE)
diff --git a/drivers/block/drbd/drbd_nl.c b/drivers/block/drbd/drbd_nl.c
index 22538d9628f11dfeb57dced35cde611f97783bb0..e2a5875a07b1cbd181046913627b25ba409c5e9f 100644
--- a/drivers/block/drbd/drbd_nl.c
+++ b/drivers/block/drbd/drbd_nl.c
@@ -894,11 +894,6 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
 		min_md_device_sectors = MD_RESERVED_SECT * (nbc->dc.meta_dev_idx + 1);
 	}
 
-	if (drbd_get_capacity(nbc->md_bdev) > max_possible_sectors)
-		dev_warn(DEV, "truncating very big lower level device "
-		     "to currently maximum possible %llu sectors\n",
-		     (unsigned long long) max_possible_sectors);
-
 	if (drbd_get_capacity(nbc->md_bdev) < min_md_device_sectors) {
 		retcode = ERR_MD_DISK_TO_SMALL;
 		dev_warn(DEV, "refusing attach: md-device too small, "
@@ -917,6 +912,15 @@ static int drbd_nl_disk_conf(struct drbd_conf *mdev, struct drbd_nl_cfg_req *nlp
 
 	nbc->known_size = drbd_get_capacity(nbc->backing_bdev);
 
+	if (nbc->known_size > max_possible_sectors) {
+		dev_warn(DEV, "==> truncating very big lower level device "
+			"to currently maximum possible %llu sectors <==\n",
+			(unsigned long long) max_possible_sectors);
+		if (nbc->dc.meta_dev_idx >= 0)
+			dev_warn(DEV, "==>> using internal or flexible "
+				      "meta data may help <<==\n");
+	}
+
 	drbd_suspend_io(mdev);
 	/* also wait for the last barrier ack. */
 	wait_event(mdev->misc_wait, !atomic_read(&mdev->ap_pending_cnt));
diff --git a/drivers/block/drbd/drbd_receiver.c b/drivers/block/drbd/drbd_receiver.c
index 360baf60f574a0b67154bfc5c1eb3afe28316719..9bbc509443e5bfae8b2953f1815f91b365bae9b4 100644
--- a/drivers/block/drbd/drbd_receiver.c
+++ b/drivers/block/drbd/drbd_receiver.c
@@ -2099,7 +2099,7 @@ static int drbd_asb_recover_0p(struct drbd_conf *mdev) __must_hold(local)
 			break;
 		}
 		/* Else fall through to one of the other strategies... */
-		dev_warn(DEV, "Discard younger/older primary did not found a decision\n"
+		dev_warn(DEV, "Discard younger/older primary did not find a decision\n"
 		     "Using discard-least-changes instead\n");
 	case ASB_DISCARD_ZERO_CHG:
 		if (ch_peer == 0 && ch_self == 0) {
@@ -3619,10 +3619,6 @@ static void drbd_disconnect(struct drbd_conf *mdev)
 	set_bit(STOP_SYNC_TIMER, &mdev->flags);
 	resync_timer_fn((unsigned long)mdev);
 
-	/* so we can be sure that all remote or resync reads
-	 * made it at least to net_ee */
-	wait_event(mdev->misc_wait, !atomic_read(&mdev->local_cnt));
-
 	/* wait for all w_e_end_data_req, w_e_end_rsdata_req, w_send_barrier,
 	 * w_make_resync_request etc. which may still be on the worker queue
 	 * to be "canceled" */
diff --git a/drivers/block/drbd/drbd_req.c b/drivers/block/drbd/drbd_req.c
index 3678d3d66c6c8a056c8b9a4a94c074e0c271647f..d09aac4a84ec41a289a256761f9e570f97f7b8bc 100644
--- a/drivers/block/drbd/drbd_req.c
+++ b/drivers/block/drbd/drbd_req.c
@@ -505,7 +505,7 @@ void __req_mod(struct drbd_request *req, enum drbd_req_event what,
 		 * corresponding hlist_del is in _req_may_be_done() */
 		hlist_add_head(&req->colision, ar_hash_slot(mdev, req->sector));
 
-		set_bit(UNPLUG_REMOTE, &mdev->flags); /* why? */
+		set_bit(UNPLUG_REMOTE, &mdev->flags);
 
 		D_ASSERT(req->rq_state & RQ_NET_PENDING);
 		req->rq_state |= RQ_NET_QUEUED;
@@ -536,6 +536,11 @@ void __req_mod(struct drbd_request *req, enum drbd_req_event what,
 		 *
 		 * Add req to the (now) current epoch (barrier). */
 
+		/* otherwise we may lose an unplug, which may cause some remote
+		 * io-scheduler timeout to expire, increasing maximum latency,
+		 * hurting performance. */
+		set_bit(UNPLUG_REMOTE, &mdev->flags);
+
 		/* see drbd_make_request_common,
 		 * just after it grabs the req_lock */
 		D_ASSERT(test_bit(CREATE_BARRIER, &mdev->flags) == 0);
diff --git a/include/linux/drbd.h b/include/linux/drbd.h
index 233db5c18b86f69a8144346eb805252b0268540b..18942ad115d902e99860071830c5a07f06ec09bb 100644
--- a/include/linux/drbd.h
+++ b/include/linux/drbd.h
@@ -53,7 +53,7 @@
 
 
 extern const char *drbd_buildtag(void);
-#define REL_VERSION "8.3.3rc2"
+#define REL_VERSION "8.3.5"
 #define API_VERSION 88
 #define PRO_VERSION_MIN 86
 #define PRO_VERSION_MAX 91
diff --git a/include/linux/drbd_limits.h b/include/linux/drbd_limits.h
index 9d067ce469607a2dfd225eee770ffe561ae066ed..51f47a586ad87a79bbf57bde32776c746ca4b67a 100644
--- a/include/linux/drbd_limits.h
+++ b/include/linux/drbd_limits.h
@@ -70,11 +70,11 @@
   /* I don't think that a tcp send buffer of more than 10M is usefull */
 #define DRBD_SNDBUF_SIZE_MIN  0
 #define DRBD_SNDBUF_SIZE_MAX  (10<<20)
-#define DRBD_SNDBUF_SIZE_DEF  (2*65535)
+#define DRBD_SNDBUF_SIZE_DEF  0
 
 #define DRBD_RCVBUF_SIZE_MIN  0
 #define DRBD_RCVBUF_SIZE_MAX  (10<<20)
-#define DRBD_RCVBUF_SIZE_DEF  (2*65535)
+#define DRBD_RCVBUF_SIZE_DEF  0
 
   /* @4k PageSize -> 128kB - 512MB */
 #define DRBD_MAX_BUFFERS_MIN  32