drbd: Pass a peer device to the resync and online verify functions

Originally-from: Andreas Grünbacher <agruen@linbit.com>
Signed-off-by: Christoph Böhmwalder <christoph.boehmwalder@linbit.com>
Link: https://lore.kernel.org/r/20230330102744.2128122-2-christoph.boehmwalder@linbit.comSigned-off-by: Jens Axboe <axboe@kernel.dk>

drivers/block/drbd/drbd_actlog.c

@@ -735,8 +735,9 @@ static bool update_rs_extent(struct drbd_device *device,
 	return false;
 }
 
-void drbd_advance_rs_marks(struct drbd_device *device, unsigned long still_to_go)
+void drbd_advance_rs_marks(struct drbd_peer_device *peer_device, unsigned long still_to_go)
 {
+	struct drbd_device *device = peer_device->device;
 	unsigned long now = jiffies;
 	unsigned long last = device->rs_mark_time[device->rs_last_mark];
 	int next = (device->rs_last_mark + 1) % DRBD_SYNC_MARKS;
@@ -819,7 +820,7 @@ static int update_sync_bits(struct drbd_device *device,
 		if (mode == SET_IN_SYNC) {
 			unsigned long still_to_go = drbd_bm_total_weight(device);
 			bool rs_is_done = (still_to_go <= device->rs_failed);
-			drbd_advance_rs_marks(device, still_to_go);
+			drbd_advance_rs_marks(first_peer_device(device), still_to_go);
 			if (cleared || rs_is_done)
 				maybe_schedule_on_disk_bitmap_update(device, rs_is_done);
 		} else if (mode == RECORD_RS_FAILED)
@@ -843,10 +844,11 @@ static bool plausible_request_size(int size)
  * called by worker on C_SYNC_TARGET and receiver on SyncSource.
  *
  */
-int __drbd_change_sync(struct drbd_device *device, sector_t sector, int size,
+int __drbd_change_sync(struct drbd_peer_device *peer_device, sector_t sector, int size,
 		enum update_sync_bits_mode mode)
 {
 	/* Is called from worker and receiver context _only_ */
+	struct drbd_device *device = peer_device->device;
 	unsigned long sbnr, ebnr, lbnr;
 	unsigned long count = 0;
 	sector_t esector, nr_sectors;
@@ -1009,14 +1011,15 @@ int drbd_rs_begin_io(struct drbd_device *device, sector_t sector)
  * tries to set it to BME_LOCKED. Returns 0 upon success, and -EAGAIN
  * if there is still application IO going on in this area.
  */
-int drbd_try_rs_begin_io(struct drbd_device *device, sector_t sector)
+int drbd_try_rs_begin_io(struct drbd_peer_device *peer_device, sector_t sector)
 {
+	struct drbd_device *device = peer_device->device;
 	unsigned int enr = BM_SECT_TO_EXT(sector);
 	const unsigned int al_enr = enr*AL_EXT_PER_BM_SECT;
 	struct lc_element *e;
 	struct bm_extent *bm_ext;
 	int i;
-	bool throttle = drbd_rs_should_slow_down(device, sector, true);
+	bool throttle = drbd_rs_should_slow_down(peer_device, sector, true);
 
 	/* If we need to throttle, a half-locked (only marked BME_NO_WRITES,
 	 * not yet BME_LOCKED) extent needs to be kicked out explicitly if we

drivers/block/drbd/drbd_int.h

@@ -1433,21 +1433,24 @@ void drbd_resync_after_changed(struct drbd_device *device);
 extern void drbd_start_resync(struct drbd_device *device, enum drbd_conns side);
 extern void resume_next_sg(struct drbd_device *device);
 extern void suspend_other_sg(struct drbd_device *device);
-extern int drbd_resync_finished(struct drbd_device *device);
+extern int drbd_resync_finished(struct drbd_peer_device *peer_device);
 /* maybe rather drbd_main.c ? */
 extern void *drbd_md_get_buffer(struct drbd_device *device, const char *intent);
 extern void drbd_md_put_buffer(struct drbd_device *device);
 extern int drbd_md_sync_page_io(struct drbd_device *device,
 		struct drbd_backing_dev *bdev, sector_t sector, enum req_op op);
-extern void drbd_ov_out_of_sync_found(struct drbd_device *, sector_t, int);
+extern void drbd_ov_out_of_sync_found(struct drbd_peer_device *peer_device,
+		sector_t sector, int size);
 extern void wait_until_done_or_force_detached(struct drbd_device *device,
 		struct drbd_backing_dev *bdev, unsigned int *done);
-extern void drbd_rs_controller_reset(struct drbd_device *device);
+extern void drbd_rs_controller_reset(struct drbd_peer_device *peer_device);
 
-static inline void ov_out_of_sync_print(struct drbd_device *device)
+static inline void ov_out_of_sync_print(struct drbd_peer_device *peer_device)
 {
+	struct drbd_device *device = peer_device->device;
+
 	if (device->ov_last_oos_size) {
-		drbd_err(device, "Out of sync: start=%llu, size=%lu (sectors)\n",
+		drbd_err(peer_device, "Out of sync: start=%llu, size=%lu (sectors)\n",
 		     (unsigned long long)device->ov_last_oos_start,
 		     (unsigned long)device->ov_last_oos_size);
 	}
@@ -1486,7 +1489,7 @@ extern int drbd_ack_receiver(struct drbd_thread *thi);
 extern void drbd_send_ping_wf(struct work_struct *ws);
 extern void drbd_send_acks_wf(struct work_struct *ws);
 extern bool drbd_rs_c_min_rate_throttle(struct drbd_device *device);
-extern bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector,
+extern bool drbd_rs_should_slow_down(struct drbd_peer_device *peer_device, sector_t sector,
 		bool throttle_if_app_is_waiting);
 extern int drbd_submit_peer_request(struct drbd_peer_request *peer_req);
 extern int drbd_free_peer_reqs(struct drbd_device *, struct list_head *);
@@ -1542,22 +1545,22 @@ extern void drbd_al_begin_io(struct drbd_device *device, struct drbd_interval *i
 extern void drbd_al_complete_io(struct drbd_device *device, struct drbd_interval *i);
 extern void drbd_rs_complete_io(struct drbd_device *device, sector_t sector);
 extern int drbd_rs_begin_io(struct drbd_device *device, sector_t sector);
-extern int drbd_try_rs_begin_io(struct drbd_device *device, sector_t sector);
+extern int drbd_try_rs_begin_io(struct drbd_peer_device *peer_device, sector_t sector);
 extern void drbd_rs_cancel_all(struct drbd_device *device);
 extern int drbd_rs_del_all(struct drbd_device *device);
-extern void drbd_rs_failed_io(struct drbd_device *device,
+extern void drbd_rs_failed_io(struct drbd_peer_device *peer_device,
 		sector_t sector, int size);
-extern void drbd_advance_rs_marks(struct drbd_device *device, unsigned long still_to_go);
+extern void drbd_advance_rs_marks(struct drbd_peer_device *peer_device, unsigned long still_to_go);
 
 enum update_sync_bits_mode { RECORD_RS_FAILED, SET_OUT_OF_SYNC, SET_IN_SYNC };
-extern int __drbd_change_sync(struct drbd_device *device, sector_t sector, int size,
+extern int __drbd_change_sync(struct drbd_peer_device *peer_device, sector_t sector, int size,
 		enum update_sync_bits_mode mode);
-#define drbd_set_in_sync(device, sector, size) \
-	__drbd_change_sync(device, sector, size, SET_IN_SYNC)
-#define drbd_set_out_of_sync(device, sector, size) \
-	__drbd_change_sync(device, sector, size, SET_OUT_OF_SYNC)
-#define drbd_rs_failed_io(device, sector, size) \
-	__drbd_change_sync(device, sector, size, RECORD_RS_FAILED)
+#define drbd_set_in_sync(peer_device, sector, size) \
+	__drbd_change_sync(peer_device, sector, size, SET_IN_SYNC)
+#define drbd_set_out_of_sync(peer_device, sector, size) \
+	__drbd_change_sync(peer_device, sector, size, SET_OUT_OF_SYNC)
+#define drbd_rs_failed_io(peer_device, sector, size) \
+	__drbd_change_sync(peer_device, sector, size, RECORD_RS_FAILED)
 extern void drbd_al_shrink(struct drbd_device *device);
 extern int drbd_al_initialize(struct drbd_device *, void *);
 
@@ -1945,15 +1948,16 @@ static inline int __dec_ap_pending(struct drbd_device *device)
  * C_SYNC_SOURCE sends P_RS_DATA_REPLY   (and expects P_WRITE_ACK with ID_SYNCER)
  *					   (or P_NEG_ACK with ID_SYNCER)
  */
-static inline void inc_rs_pending(struct drbd_device *device)
+static inline void inc_rs_pending(struct drbd_peer_device *peer_device)
 {
-	atomic_inc(&device->rs_pending_cnt);
+	atomic_inc(&peer_device->device->rs_pending_cnt);
 }
 
-#define dec_rs_pending(device) ((void)expect((device), __dec_rs_pending(device) >= 0))
-static inline int __dec_rs_pending(struct drbd_device *device)
+#define dec_rs_pending(peer_device) \
+	((void)expect((peer_device), __dec_rs_pending(peer_device) >= 0))
+static inline int __dec_rs_pending(struct drbd_peer_device *peer_device)
 {
-	return atomic_dec_return(&device->rs_pending_cnt);
+	return atomic_dec_return(&peer_device->device->rs_pending_cnt);
 }
 
 /* counts how many answers we still need to send to the peer.

drivers/block/drbd/drbd_receiver.c

@@ -2044,11 +2044,11 @@ static int e_end_resync_block(struct drbd_work *w, int unused)
 	D_ASSERT(device, drbd_interval_empty(&peer_req->i));
 
 	if (likely((peer_req->flags & EE_WAS_ERROR) == 0)) {
-		drbd_set_in_sync(device, sector, peer_req->i.size);
+		drbd_set_in_sync(peer_device, sector, peer_req->i.size);
 		err = drbd_send_ack(peer_device, P_RS_WRITE_ACK, peer_req);
 	} else {
 		/* Record failure to sync */
-		drbd_rs_failed_io(device, sector, peer_req->i.size);
+		drbd_rs_failed_io(peer_device, sector, peer_req->i.size);
 
 		err  = drbd_send_ack(peer_device, P_NEG_ACK, peer_req);
 	}
@@ -2067,7 +2067,7 @@ static int recv_resync_read(struct drbd_peer_device *peer_device, sector_t secto
 	if (!peer_req)
 		goto fail;
 
-	dec_rs_pending(device);
+	dec_rs_pending(peer_device);
 
 	inc_unacked(device);
 	/* corresponding dec_unacked() in e_end_resync_block()
@@ -2220,7 +2220,7 @@ static int e_end_block(struct drbd_work *w, int cancel)
 				P_RS_WRITE_ACK : P_WRITE_ACK;
 			err = drbd_send_ack(peer_device, pcmd, peer_req);
 			if (pcmd == P_RS_WRITE_ACK)
-				drbd_set_in_sync(device, sector, peer_req->i.size);
+				drbd_set_in_sync(peer_device, sector, peer_req->i.size);
 		} else {
 			err = drbd_send_ack(peer_device, P_NEG_ACK, peer_req);
 			/* we expect it to be marked out of sync anyways...
@@ -2691,7 +2691,7 @@ static int receive_Data(struct drbd_connection *connection, struct packet_info *
 
 	if (device->state.pdsk < D_INCONSISTENT) {
 		/* In case we have the only disk of the cluster, */
-		drbd_set_out_of_sync(device, peer_req->i.sector, peer_req->i.size);
+		drbd_set_out_of_sync(peer_device, peer_req->i.sector, peer_req->i.size);
 		peer_req->flags &= ~EE_MAY_SET_IN_SYNC;
 		drbd_al_begin_io(device, &peer_req->i);
 		peer_req->flags |= EE_CALL_AL_COMPLETE_IO;
@@ -2730,9 +2730,10 @@ static int receive_Data(struct drbd_connection *connection, struct packet_info *
  * The current sync rate used here uses only the most recent two step marks,
  * to have a short time average so we can react faster.
  */
-bool drbd_rs_should_slow_down(struct drbd_device *device, sector_t sector,
+bool drbd_rs_should_slow_down(struct drbd_peer_device *peer_device, sector_t sector,
 		bool throttle_if_app_is_waiting)
 {
+	struct drbd_device *device = peer_device->device;
 	struct lc_element *tmp;
 	bool throttle = drbd_rs_c_min_rate_throttle(device);
 
@@ -2844,7 +2845,7 @@ static int receive_DataRequest(struct drbd_connection *connection, struct packet
 			break;
 		case P_OV_REPLY:
 			verb = 0;
-			dec_rs_pending(device);
+			dec_rs_pending(peer_device);
 			drbd_send_ack_ex(peer_device, P_OV_RESULT, sector, size, ID_IN_SYNC);
 			break;
 		default:
@@ -2915,7 +2916,7 @@ static int receive_DataRequest(struct drbd_connection *connection, struct packet
 			/* track progress, we may need to throttle */
 			atomic_add(size >> 9, &device->rs_sect_in);
 			peer_req->w.cb = w_e_end_ov_reply;
-			dec_rs_pending(device);
+			dec_rs_pending(peer_device);
 			/* drbd_rs_begin_io done when we sent this request,
 			 * but accounting still needs to be done. */
 			goto submit_for_resync;
@@ -2978,7 +2979,7 @@ static int receive_DataRequest(struct drbd_connection *connection, struct packet
 
 	update_receiver_timing_details(connection, drbd_rs_should_slow_down);
 	if (device->state.peer != R_PRIMARY
-	&& drbd_rs_should_slow_down(device, sector, false))
+	&& drbd_rs_should_slow_down(peer_device, sector, false))
 		schedule_timeout_uninterruptible(HZ/10);
 	update_receiver_timing_details(connection, drbd_rs_begin_io);
 	if (drbd_rs_begin_io(device, sector))
@@ -4450,7 +4451,7 @@ static int receive_state(struct drbd_connection *connection, struct packet_info
 		else if (os.conn >= C_SYNC_SOURCE &&
 			 peer_state.conn == C_CONNECTED) {
 			if (drbd_bm_total_weight(device) <= device->rs_failed)
-				drbd_resync_finished(device);
+				drbd_resync_finished(peer_device);
 			return 0;
 		}
 	}
@@ -4458,8 +4459,8 @@ static int receive_state(struct drbd_connection *connection, struct packet_info
 	/* explicit verify finished notification, stop sector reached. */
 	if (os.conn == C_VERIFY_T && os.disk == D_UP_TO_DATE &&
 	    peer_state.conn == C_CONNECTED && real_peer_disk == D_UP_TO_DATE) {
-		ov_out_of_sync_print(device);
-		drbd_resync_finished(device);
+		ov_out_of_sync_print(peer_device);
+		drbd_resync_finished(peer_device);
 		return 0;
 	}
 
@@ -4937,7 +4938,7 @@ static int receive_out_of_sync(struct drbd_connection *connection, struct packet
 				drbd_conn_str(device->state.conn));
 	}
 
-	drbd_set_out_of_sync(device, be64_to_cpu(p->sector), be32_to_cpu(p->blksize));
+	drbd_set_out_of_sync(peer_device, be64_to_cpu(p->sector), be32_to_cpu(p->blksize));
 
 	return 0;
 }
@@ -4958,7 +4959,7 @@ static int receive_rs_deallocated(struct drbd_connection *connection, struct pac
 	sector = be64_to_cpu(p->sector);
 	size = be32_to_cpu(p->blksize);
 
-	dec_rs_pending(device);
+	dec_rs_pending(peer_device);
 
 	if (get_ldev(device)) {
 		struct drbd_peer_request *peer_req;
@@ -5650,12 +5651,12 @@ static int got_IsInSync(struct drbd_connection *connection, struct packet_info *
 
 	if (get_ldev(device)) {
 		drbd_rs_complete_io(device, sector);
-		drbd_set_in_sync(device, sector, blksize);
+		drbd_set_in_sync(peer_device, sector, blksize);
 		/* rs_same_csums is supposed to count in units of BM_BLOCK_SIZE */
 		device->rs_same_csum += (blksize >> BM_BLOCK_SHIFT);
 		put_ldev(device);
 	}
-	dec_rs_pending(device);
+	dec_rs_pending(peer_device);
 	atomic_add(blksize >> 9, &device->rs_sect_in);
 
 	return 0;
@@ -5701,8 +5702,8 @@ static int got_BlockAck(struct drbd_connection *connection, struct packet_info *
 	update_peer_seq(peer_device, be32_to_cpu(p->seq_num));
 
 	if (p->block_id == ID_SYNCER) {
-		drbd_set_in_sync(device, sector, blksize);
-		dec_rs_pending(device);
+		drbd_set_in_sync(peer_device, sector, blksize);
+		dec_rs_pending(peer_device);
 		return 0;
 	}
 	switch (pi->cmd) {
@@ -5747,8 +5748,8 @@ static int got_NegAck(struct drbd_connection *connection, struct packet_info *pi
 	update_peer_seq(peer_device, be32_to_cpu(p->seq_num));
 
 	if (p->block_id == ID_SYNCER) {
-		dec_rs_pending(device);
-		drbd_rs_failed_io(device, sector, size);
+		dec_rs_pending(peer_device);
+		drbd_rs_failed_io(peer_device, sector, size);
 		return 0;
 	}
 
@@ -5761,7 +5762,7 @@ static int got_NegAck(struct drbd_connection *connection, struct packet_info *pi
 		   request is no longer in the collision hash. */
 		/* In Protocol B we might already have got a P_RECV_ACK
 		   but then get a P_NEG_ACK afterwards. */
-		drbd_set_out_of_sync(device, sector, size);
+		drbd_set_out_of_sync(peer_device, sector, size);
 	}
 	return 0;
 }
@@ -5806,13 +5807,13 @@ static int got_NegRSDReply(struct drbd_connection *connection, struct packet_inf
 
 	update_peer_seq(peer_device, be32_to_cpu(p->seq_num));
 
-	dec_rs_pending(device);
+	dec_rs_pending(peer_device);
 
 	if (get_ldev_if_state(device, D_FAILED)) {
 		drbd_rs_complete_io(device, sector);
 		switch (pi->cmd) {
 		case P_NEG_RS_DREPLY:
-			drbd_rs_failed_io(device, sector, size);
+			drbd_rs_failed_io(peer_device, sector, size);
 			break;
 		case P_RS_CANCEL:
 			break;
@@ -5869,21 +5870,21 @@ static int got_OVResult(struct drbd_connection *connection, struct packet_info *
 	update_peer_seq(peer_device, be32_to_cpu(p->seq_num));
 
 	if (be64_to_cpu(p->block_id) == ID_OUT_OF_SYNC)
-		drbd_ov_out_of_sync_found(device, sector, size);
+		drbd_ov_out_of_sync_found(peer_device, sector, size);
 	else
-		ov_out_of_sync_print(device);
+		ov_out_of_sync_print(peer_device);
 
 	if (!get_ldev(device))
 		return 0;
 
 	drbd_rs_complete_io(device, sector);
-	dec_rs_pending(device);
+	dec_rs_pending(peer_device);
 
 	--device->ov_left;
 
 	/* let's advance progress step marks only for every other megabyte */
 	if ((device->ov_left & 0x200) == 0x200)
-		drbd_advance_rs_marks(device, device->ov_left);
+		drbd_advance_rs_marks(peer_device, device->ov_left);
 
 	if (device->ov_left == 0) {
 		dw = kmalloc(sizeof(*dw), GFP_NOIO);
@@ -5893,8 +5894,8 @@ static int got_OVResult(struct drbd_connection *connection, struct packet_info *
 			drbd_queue_work(&peer_device->connection->sender_work, &dw->w);
 		} else {
 			drbd_err(device, "kmalloc(dw) failed.");
-			ov_out_of_sync_print(device);
-			drbd_resync_finished(device);
+			ov_out_of_sync_print(peer_device);
+			drbd_resync_finished(peer_device);
 		}
 	}
 	put_ldev(device);

drivers/block/drbd/drbd_req.c

@@ -122,12 +122,13 @@ void drbd_req_destroy(struct kref *kref)
 		 * before it even was submitted or sent.
 		 * In that case we do not want to touch the bitmap at all.
 		 */
+		struct drbd_peer_device *peer_device = first_peer_device(device);
 		if ((s & (RQ_POSTPONED|RQ_LOCAL_MASK|RQ_NET_MASK)) != RQ_POSTPONED) {
 			if (!(s & RQ_NET_OK) || !(s & RQ_LOCAL_OK))
-				drbd_set_out_of_sync(device, req->i.sector, req->i.size);
+				drbd_set_out_of_sync(peer_device, req->i.sector, req->i.size);
 
 			if ((s & RQ_NET_OK) && (s & RQ_LOCAL_OK) && (s & RQ_NET_SIS))
-				drbd_set_in_sync(device, req->i.sector, req->i.size);
+				drbd_set_in_sync(peer_device, req->i.sector, req->i.size);
 		}
 
 		/* one might be tempted to move the drbd_al_complete_io
@@ -620,7 +621,7 @@ int __req_mod(struct drbd_request *req, enum drbd_req_event what,
 		break;
 
 	case READ_COMPLETED_WITH_ERROR:
-		drbd_set_out_of_sync(device, req->i.sector, req->i.size);
+		drbd_set_out_of_sync(peer_device, req->i.sector, req->i.size);
 		drbd_report_io_error(device, req);
 		__drbd_chk_io_error(device, DRBD_READ_ERROR);
 		fallthrough;
@@ -1131,7 +1132,7 @@ static int drbd_process_write_request(struct drbd_request *req)
 	if (remote) {
 		_req_mod(req, TO_BE_SENT, peer_device);
 		_req_mod(req, QUEUE_FOR_NET_WRITE, peer_device);
-	} else if (drbd_set_out_of_sync(device, req->i.sector, req->i.size))
+	} else if (drbd_set_out_of_sync(peer_device, req->i.sector, req->i.size))
 		_req_mod(req, QUEUE_FOR_SEND_OOS, peer_device);
 
 	return remote;

drivers/block/drbd/drbd_state.c

@@ -1222,9 +1222,11 @@ void drbd_resume_al(struct drbd_device *device)
 }
 
 /* helper for _drbd_set_state */
-static void set_ov_position(struct drbd_device *device, enum drbd_conns cs)
+static void set_ov_position(struct drbd_peer_device *peer_device, enum drbd_conns cs)
 {
-	if (first_peer_device(device)->connection->agreed_pro_version < 90)
+	struct drbd_device *device = peer_device->device;
+
+	if (peer_device->connection->agreed_pro_version < 90)
 		device->ov_start_sector = 0;
 	device->rs_total = drbd_bm_bits(device);
 	device->ov_position = 0;
@@ -1387,7 +1389,7 @@ _drbd_set_state(struct drbd_device *device, union drbd_state ns,
 		unsigned long now = jiffies;
 		int i;
 
-		set_ov_position(device, ns.conn);
+		set_ov_position(peer_device, ns.conn);
 		device->rs_start = now;
 		device->rs_last_sect_ev = 0;
 		device->ov_last_oos_size = 0;
@@ -1398,7 +1400,7 @@ _drbd_set_state(struct drbd_device *device, union drbd_state ns,
 			device->rs_mark_time[i] = now;
 		}
 
-		drbd_rs_controller_reset(device);
+		drbd_rs_controller_reset(peer_device);
 
 		if (ns.conn == C_VERIFY_S) {
 			drbd_info(device, "Starting Online Verify from sector %llu\n",