1 /* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
2 * vim:expandtab:shiftwidth=8:tabstop=8:
4 * Copyright (c) 2002, 2003 Cluster File Systems, Inc.
6 * This file is part of Lustre, http://www.lustre.org.
8 * Lustre is free software; you can redistribute it and/or
9 * modify it under the terms of version 2 of the GNU General Public
10 * License as published by the Free Software Foundation.
12 * Lustre is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
15 * GNU General Public License for more details.
17 * You should have received a copy of the GNU General Public License
18 * along with Lustre; if not, write to the Free Software
19 * Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
23 #define DEBUG_SUBSYSTEM S_RPC
27 #include <liblustre.h>
30 #include <linux/obd_support.h>
31 #include <linux/obd_class.h>
32 #include <linux/lustre_lib.h>
33 #include <linux/lustre_ha.h>
34 #include <linux/lustre_import.h>
36 #include "ptlrpc_internal.h"
38 void ptlrpc_init_client(int req_portal, int rep_portal, char *name,
39 struct ptlrpc_client *cl)
41 cl->cli_request_portal = req_portal;
42 cl->cli_reply_portal = rep_portal;
46 struct ptlrpc_connection *ptlrpc_uuid_to_connection(struct obd_uuid *uuid)
48 struct ptlrpc_connection *c;
49 struct ptlrpc_peer peer;
52 err = ptlrpc_uuid_to_peer(uuid, &peer);
54 CERROR("cannot find peer %s!\n", uuid->uuid);
58 c = ptlrpc_get_connection(&peer, uuid);
60 memcpy(c->c_remote_uuid.uuid,
61 uuid->uuid, sizeof(c->c_remote_uuid.uuid));
64 CDEBUG(D_INFO, "%s -> %p\n", uuid->uuid, c);
69 void ptlrpc_readdress_connection(struct ptlrpc_connection *conn,
70 struct obd_uuid *uuid)
72 struct ptlrpc_peer peer;
75 err = ptlrpc_uuid_to_peer(uuid, &peer);
77 CERROR("cannot find peer %s!\n", uuid->uuid);
81 memcpy(&conn->c_peer, &peer, sizeof (peer));
85 static inline struct ptlrpc_bulk_desc *new_bulk(int npages, int type, int portal)
87 struct ptlrpc_bulk_desc *desc;
89 OBD_ALLOC(desc, offsetof (struct ptlrpc_bulk_desc, bd_iov[npages]));
93 spin_lock_init(&desc->bd_lock);
94 init_waitqueue_head(&desc->bd_waitq);
95 desc->bd_max_iov = npages;
96 desc->bd_iov_count = 0;
97 desc->bd_md_h = PTL_INVALID_HANDLE;
98 desc->bd_portal = portal;
104 struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_imp (struct ptlrpc_request *req,
105 int npages, int type, int portal)
107 struct obd_import *imp = req->rq_import;
108 struct ptlrpc_bulk_desc *desc;
110 LASSERT(type == BULK_PUT_SINK || type == BULK_GET_SOURCE);
111 desc = new_bulk(npages, type, portal);
115 desc->bd_import_generation = req->rq_import_generation;
116 desc->bd_import = class_import_get(imp);
119 desc->bd_cbid.cbid_fn = client_bulk_callback;
120 desc->bd_cbid.cbid_arg = desc;
122 /* This makes req own desc, and free it when she frees herself */
128 struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_exp (struct ptlrpc_request *req,
129 int npages, int type, int portal)
131 struct obd_export *exp = req->rq_export;
132 struct ptlrpc_bulk_desc *desc;
134 LASSERT(type == BULK_PUT_SOURCE || type == BULK_GET_SINK);
136 desc = new_bulk(npages, type, portal);
140 desc->bd_export = class_export_get(exp);
143 desc->bd_cbid.cbid_fn = server_bulk_callback;
144 desc->bd_cbid.cbid_arg = desc;
146 /* NB we don't assign rq_bulk here; server-side requests are
147 * re-used, and the handler frees the bulk desc explicitly. */
152 void ptlrpc_prep_bulk_page(struct ptlrpc_bulk_desc *desc,
153 struct page *page, int pageoffset, int len)
155 LASSERT(desc->bd_iov_count < desc->bd_max_iov);
156 LASSERT(page != NULL);
157 LASSERT(pageoffset >= 0);
159 LASSERT(pageoffset + len <= PAGE_SIZE);
163 ptlrpc_add_bulk_page(desc, page, pageoffset, len);
166 void ptlrpc_free_bulk(struct ptlrpc_bulk_desc *desc)
170 LASSERT(desc != NULL);
171 LASSERT(desc->bd_iov_count != LI_POISON); /* not freed already */
172 LASSERT(!desc->bd_network_rw); /* network hands off or */
173 LASSERT((desc->bd_export != NULL) ^ (desc->bd_import != NULL));
175 class_export_put(desc->bd_export);
177 class_import_put(desc->bd_import);
179 OBD_FREE(desc, offsetof(struct ptlrpc_bulk_desc,
180 bd_iov[desc->bd_max_iov]));
184 struct ptlrpc_request *ptlrpc_prep_req(struct obd_import *imp, int opcode,
185 int count, int *lengths, char **bufs)
187 struct ptlrpc_request *request;
191 LASSERT((unsigned long)imp > 0x1000);
193 OBD_ALLOC(request, sizeof(*request));
195 CERROR("request allocation out of memory\n");
199 rc = lustre_pack_request(request, count, lengths, bufs);
201 CERROR("cannot pack request %d\n", rc);
202 OBD_FREE(request, sizeof(*request));
206 if (imp->imp_server_timeout)
207 request->rq_timeout = obd_timeout / 2;
209 request->rq_timeout = obd_timeout;
210 request->rq_send_state = LUSTRE_IMP_FULL;
211 request->rq_type = PTL_RPC_MSG_REQUEST;
212 request->rq_import = class_import_get(imp);
214 request->rq_req_cbid.cbid_fn = request_out_callback;
215 request->rq_req_cbid.cbid_arg = request;
217 request->rq_reply_cbid.cbid_fn = reply_in_callback;
218 request->rq_reply_cbid.cbid_arg = request;
220 request->rq_phase = RQ_PHASE_NEW;
222 /* XXX FIXME bug 249 */
223 request->rq_request_portal = imp->imp_client->cli_request_portal;
224 request->rq_reply_portal = imp->imp_client->cli_reply_portal;
226 spin_lock_init(&request->rq_lock);
227 INIT_LIST_HEAD(&request->rq_list);
228 INIT_LIST_HEAD(&request->rq_replay_list);
229 INIT_LIST_HEAD(&request->rq_set_chain);
230 init_waitqueue_head(&request->rq_reply_waitq);
231 request->rq_xid = ptlrpc_next_xid();
232 atomic_set(&request->rq_refcount, 1);
234 request->rq_reqmsg->opc = opcode;
235 request->rq_reqmsg->flags = 0;
240 struct ptlrpc_request_set *ptlrpc_prep_set(void)
242 struct ptlrpc_request_set *set;
244 OBD_ALLOC(set, sizeof *set);
247 INIT_LIST_HEAD(&set->set_requests);
248 init_waitqueue_head(&set->set_waitq);
249 set->set_remaining = 0;
250 spin_lock_init(&set->set_new_req_lock);
251 INIT_LIST_HEAD(&set->set_new_requests);
256 /* Finish with this set; opposite of prep_set. */
257 void ptlrpc_set_destroy(struct ptlrpc_request_set *set)
259 struct list_head *tmp;
260 struct list_head *next;
265 /* Requests on the set should either all be completed, or all be new */
266 expected_phase = (set->set_remaining == 0) ?
267 RQ_PHASE_COMPLETE : RQ_PHASE_NEW;
268 list_for_each (tmp, &set->set_requests) {
269 struct ptlrpc_request *req =
270 list_entry(tmp, struct ptlrpc_request, rq_set_chain);
272 LASSERT(req->rq_phase == expected_phase);
276 LASSERT(set->set_remaining == 0 || set->set_remaining == n);
278 list_for_each_safe(tmp, next, &set->set_requests) {
279 struct ptlrpc_request *req =
280 list_entry(tmp, struct ptlrpc_request, rq_set_chain);
281 list_del_init(&req->rq_set_chain);
283 LASSERT(req->rq_phase == expected_phase);
285 if (req->rq_phase == RQ_PHASE_NEW) {
287 if (req->rq_interpret_reply != NULL) {
288 int (*interpreter)(struct ptlrpc_request *,
290 req->rq_interpret_reply;
292 /* higher level (i.e. LOV) failed;
293 * let the sub reqs clean up */
294 req->rq_status = -EBADR;
295 interpreter(req, &req->rq_async_args,
298 set->set_remaining--;
302 ptlrpc_req_finished (req);
305 LASSERT(set->set_remaining == 0);
307 OBD_FREE(set, sizeof(*set));
311 void ptlrpc_set_add_req(struct ptlrpc_request_set *set,
312 struct ptlrpc_request *req)
314 /* The set takes over the caller's request reference */
315 list_add_tail(&req->rq_set_chain, &set->set_requests);
317 set->set_remaining++;
318 atomic_inc(&req->rq_import->imp_inflight);
321 /* lock so many callers can add things, the context that owns the set
322 * is supposed to notice these and move them into the set proper. */
323 void ptlrpc_set_add_new_req(struct ptlrpc_request_set *set,
324 struct ptlrpc_request *req)
327 spin_lock_irqsave(&set->set_new_req_lock, flags);
328 /* The set takes over the caller's request reference */
329 list_add_tail(&req->rq_set_chain, &set->set_new_requests);
331 spin_unlock_irqrestore(&set->set_new_req_lock, flags);
335 * Based on the current state of the import, determine if the request
336 * can be sent, is an error, or should be delayed.
338 * Returns true if this request should be delayed. If false, and
339 * *status is set, then the request can not be sent and *status is the
340 * error code. If false and status is 0, then request can be sent.
342 * The imp->imp_lock must be held.
344 static int ptlrpc_import_delay_req(struct obd_import *imp,
345 struct ptlrpc_request *req, int *status)
350 LASSERT (status != NULL);
353 if (imp->imp_state == LUSTRE_IMP_NEW) {
354 DEBUG_REQ(D_ERROR, req, "Uninitialized import.");
358 else if (imp->imp_state == LUSTRE_IMP_CLOSED) {
359 DEBUG_REQ(D_ERROR, req, "IMP_CLOSED ");
362 /* allow CONNECT even if import is invalid */
363 else if (req->rq_send_state == LUSTRE_IMP_CONNECTING &&
364 imp->imp_state == LUSTRE_IMP_CONNECTING) {
368 * If the import has been invalidated (such as by an OST failure), the
369 * request must fail with -EIO.
371 else if (imp->imp_invalid) {
372 DEBUG_REQ(D_ERROR, req, "IMP_INVALID");
375 else if (req->rq_import_generation != imp->imp_generation) {
376 DEBUG_REQ(D_ERROR, req, "req wrong generation:");
379 else if (req->rq_send_state != imp->imp_state) {
380 if (imp->imp_obd->obd_no_recov || imp->imp_dlm_fake
382 *status = -EWOULDBLOCK;
390 static int ptlrpc_check_reply(struct ptlrpc_request *req)
396 /* serialise with network callback */
397 spin_lock_irqsave (&req->rq_lock, flags);
399 if (req->rq_replied) {
400 DEBUG_REQ(D_NET, req, "REPLIED:");
404 if (req->rq_net_err && !req->rq_timedout) {
405 spin_unlock_irqrestore (&req->rq_lock, flags);
406 rc = ptlrpc_expire_one_request(req);
407 spin_lock_irqsave (&req->rq_lock, flags);
412 DEBUG_REQ(D_ERROR, req, "ABORTED:");
416 if (req->rq_resend) {
417 DEBUG_REQ(D_ERROR, req, "RESEND:");
421 if (req->rq_restart) {
422 DEBUG_REQ(D_ERROR, req, "RESTART:");
427 spin_unlock_irqrestore (&req->rq_lock, flags);
428 DEBUG_REQ(D_NET, req, "rc = %d for", rc);
432 static int ptlrpc_check_status(struct ptlrpc_request *req)
437 err = req->rq_repmsg->status;
438 if (req->rq_repmsg->type == PTL_RPC_MSG_ERR) {
439 DEBUG_REQ(D_ERROR, req, "type == PTL_RPC_MSG_ERR, err == %d",
441 RETURN(err < 0 ? err : -EINVAL);
445 DEBUG_REQ(D_INFO, req, "status is %d", err);
446 } else if (err > 0) {
447 /* XXX: translate this error from net to host */
448 DEBUG_REQ(D_INFO, req, "status is %d", err);
454 static int after_reply(struct ptlrpc_request *req)
457 struct obd_import *imp = req->rq_import;
461 LASSERT(!req->rq_receiving_reply);
463 /* NB Until this point, the whole of the incoming message,
464 * including buflens, status etc is in the sender's byte order. */
467 /* Clear reply swab mask; this is a new reply in sender's byte order */
468 req->rq_rep_swab_mask = 0;
470 LASSERT (req->rq_nob_received <= req->rq_replen);
471 rc = lustre_unpack_msg(req->rq_repmsg, req->rq_nob_received);
473 CERROR("unpack_rep failed: %d\n", rc);
477 if (req->rq_repmsg->type != PTL_RPC_MSG_REPLY &&
478 req->rq_repmsg->type != PTL_RPC_MSG_ERR) {
479 CERROR("invalid packet type received (type=%u)\n",
480 req->rq_repmsg->type);
484 /* Store transno in reqmsg for replay. */
485 req->rq_reqmsg->transno = req->rq_transno = req->rq_repmsg->transno;
487 rc = ptlrpc_check_status(req);
489 /* Either we've been evicted, or the server has failed for
490 * some reason. Try to reconnect, and if that fails, punt to the
492 if (rc == -ENOTCONN) {
493 if (req->rq_send_state != LUSTRE_IMP_FULL ||
494 imp->imp_obd->obd_no_recov || imp->imp_dlm_fake) {
498 ptlrpc_request_handle_notconn(req);
503 if (req->rq_import->imp_replayable) {
504 spin_lock_irqsave(&imp->imp_lock, flags);
505 if (req->rq_replay || req->rq_transno != 0)
506 ptlrpc_retain_replayable_request(req, imp);
507 else if (req->rq_commit_cb != NULL) {
508 spin_unlock_irqrestore(&imp->imp_lock, flags);
509 req->rq_commit_cb(req);
510 spin_lock_irqsave(&imp->imp_lock, flags);
513 if (req->rq_transno > imp->imp_max_transno)
514 imp->imp_max_transno = req->rq_transno;
516 /* Replay-enabled imports return commit-status information. */
517 if (req->rq_repmsg->last_committed)
518 imp->imp_peer_committed_transno =
519 req->rq_repmsg->last_committed;
520 ptlrpc_free_committed(imp);
521 spin_unlock_irqrestore(&imp->imp_lock, flags);
527 static int ptlrpc_send_new_req(struct ptlrpc_request *req)
529 char str[PTL_NALFMT_SIZE];
530 struct obd_import *imp;
535 LASSERT(req->rq_phase == RQ_PHASE_NEW);
536 req->rq_phase = RQ_PHASE_RPC;
538 imp = req->rq_import;
539 spin_lock_irqsave(&imp->imp_lock, flags);
541 req->rq_import_generation = imp->imp_generation;
543 if (ptlrpc_import_delay_req(imp, req, &rc)) {
544 spin_lock (&req->rq_lock);
546 spin_unlock (&req->rq_lock);
548 DEBUG_REQ(D_HA, req, "req from PID %d waiting for recovery: "
550 req->rq_reqmsg->status,
551 ptlrpc_import_state_name(req->rq_send_state),
552 ptlrpc_import_state_name(imp->imp_state));
553 LASSERT(list_empty (&req->rq_list));
555 list_add_tail(&req->rq_list, &imp->imp_delayed_list);
556 spin_unlock_irqrestore(&imp->imp_lock, flags);
561 spin_unlock_irqrestore(&imp->imp_lock, flags);
563 req->rq_phase = RQ_PHASE_INTERPRET;
567 /* XXX this is the same as ptlrpc_queue_wait */
568 LASSERT(list_empty(&req->rq_list));
569 list_add_tail(&req->rq_list, &imp->imp_sending_list);
570 spin_unlock_irqrestore(&imp->imp_lock, flags);
572 req->rq_reqmsg->status = current->pid;
573 CDEBUG(D_RPCTRACE, "Sending RPC pname:cluuid:pid:xid:ni:nid:opc"
574 " %s:%s:%d:"LPU64":%s:%s:%d\n", current->comm,
575 imp->imp_obd->obd_uuid.uuid, req->rq_reqmsg->status,
577 imp->imp_connection->c_peer.peer_ni->pni_name,
578 ptlrpc_peernid2str(&imp->imp_connection->c_peer, str),
579 req->rq_reqmsg->opc);
581 rc = ptl_send_rpc(req);
583 DEBUG_REQ(D_HA, req, "send failed (%d); expect timeout", rc);
590 int ptlrpc_check_set(struct ptlrpc_request_set *set)
592 char str[PTL_NALFMT_SIZE];
594 struct list_head *tmp;
595 int force_timer_recalc = 0;
598 if (set->set_remaining == 0)
601 list_for_each(tmp, &set->set_requests) {
602 struct ptlrpc_request *req =
603 list_entry(tmp, struct ptlrpc_request, rq_set_chain);
604 struct obd_import *imp = req->rq_import;
607 if (req->rq_phase == RQ_PHASE_NEW &&
608 ptlrpc_send_new_req(req)) {
609 force_timer_recalc = 1;
612 if (!(req->rq_phase == RQ_PHASE_RPC ||
613 req->rq_phase == RQ_PHASE_BULK ||
614 req->rq_phase == RQ_PHASE_INTERPRET ||
615 req->rq_phase == RQ_PHASE_COMPLETE)) {
616 DEBUG_REQ(D_ERROR, req, "bad phase %x", req->rq_phase);
620 if (req->rq_phase == RQ_PHASE_COMPLETE)
623 if (req->rq_phase == RQ_PHASE_INTERPRET)
624 GOTO(interpret, req->rq_status);
626 if (req->rq_net_err && !req->rq_timedout)
627 ptlrpc_expire_one_request(req);
630 ptlrpc_unregister_reply(req);
631 if (req->rq_status == 0)
632 req->rq_status = -EIO;
633 req->rq_phase = RQ_PHASE_INTERPRET;
635 spin_lock_irqsave(&imp->imp_lock, flags);
636 list_del_init(&req->rq_list);
637 spin_unlock_irqrestore(&imp->imp_lock, flags);
639 GOTO(interpret, req->rq_status);
642 /* ptlrpc_queue_wait->l_wait_event guarantees that rq_intr
643 * will only be set after rq_timedout, but the oig waiting
644 * path sets rq_intr irrespective of whether ptlrpcd has
645 * seen a timeout. our policy is to only interpret
646 * interrupted rpcs after they have timed out */
647 if (req->rq_intr && (req->rq_timedout || req->rq_waiting)) {
648 /* NB could be on delayed list */
649 ptlrpc_unregister_reply(req);
650 req->rq_status = -EINTR;
651 req->rq_phase = RQ_PHASE_INTERPRET;
653 spin_lock_irqsave(&imp->imp_lock, flags);
654 list_del_init(&req->rq_list);
655 spin_unlock_irqrestore(&imp->imp_lock, flags);
657 GOTO(interpret, req->rq_status);
660 if (req->rq_phase == RQ_PHASE_RPC) {
661 if (req->rq_waiting || req->rq_resend) {
664 ptlrpc_unregister_reply(req);
666 spin_lock_irqsave(&imp->imp_lock, flags);
668 if (ptlrpc_import_delay_req(imp, req, &status)){
669 spin_unlock_irqrestore(&imp->imp_lock,
674 list_del_init(&req->rq_list);
676 req->rq_status = status;
677 req->rq_phase = RQ_PHASE_INTERPRET;
678 spin_unlock_irqrestore(&imp->imp_lock,
680 GOTO(interpret, req->rq_status);
682 if (req->rq_no_resend) {
683 req->rq_status = -ENOTCONN;
684 req->rq_phase = RQ_PHASE_INTERPRET;
685 spin_unlock_irqrestore(&imp->imp_lock,
687 GOTO(interpret, req->rq_status);
689 list_add_tail(&req->rq_list,
690 &imp->imp_sending_list);
692 spin_unlock_irqrestore(&imp->imp_lock, flags);
695 if (req->rq_resend) {
696 lustre_msg_add_flags(req->rq_reqmsg,
699 __u64 old_xid = req->rq_xid;
701 ptlrpc_unregister_bulk (req);
703 /* ensure previous bulk fails */
704 req->rq_xid = ptlrpc_next_xid();
705 CDEBUG(D_HA, "resend bulk "
708 old_xid, req->rq_xid);
712 rc = ptl_send_rpc(req);
714 DEBUG_REQ(D_HA, req, "send failed (%d)",
716 force_timer_recalc = 1;
719 /* need to reset the timeout */
720 force_timer_recalc = 1;
723 /* Still waiting for a reply? */
724 if (ptlrpc_client_receiving_reply(req))
727 /* Did we actually receive a reply? */
728 if (!ptlrpc_client_replied(req))
731 spin_lock_irqsave(&imp->imp_lock, flags);
732 list_del_init(&req->rq_list);
733 spin_unlock_irqrestore(&imp->imp_lock, flags);
735 req->rq_status = after_reply(req);
736 if (req->rq_resend) {
737 /* Add this req to the delayed list so
738 it can be errored if the import is
739 evicted after recovery. */
740 spin_lock_irqsave (&req->rq_lock, flags);
741 list_add_tail(&req->rq_list,
742 &imp->imp_delayed_list);
743 spin_unlock_irqrestore(&req->rq_lock, flags);
747 /* If there is no bulk associated with this request,
748 * then we're done and should let the interpreter
749 * process the reply. Similarly if the RPC returned
750 * an error, and therefore the bulk will never arrive.
752 if (req->rq_bulk == NULL || req->rq_status != 0) {
753 req->rq_phase = RQ_PHASE_INTERPRET;
754 GOTO(interpret, req->rq_status);
757 req->rq_phase = RQ_PHASE_BULK;
760 LASSERT(req->rq_phase == RQ_PHASE_BULK);
761 if (ptlrpc_bulk_active(req->rq_bulk))
764 if (!req->rq_bulk->bd_success) {
765 /* The RPC reply arrived OK, but the bulk screwed
766 * up! Dead wierd since the server told us the RPC
767 * was good after getting the REPLY for her GET or
768 * the ACK for her PUT. */
769 DEBUG_REQ(D_ERROR, req, "bulk transfer failed");
773 req->rq_phase = RQ_PHASE_INTERPRET;
776 LASSERT(req->rq_phase == RQ_PHASE_INTERPRET);
777 LASSERT(!req->rq_receiving_reply);
779 ptlrpc_unregister_reply(req);
780 if (req->rq_bulk != NULL)
781 ptlrpc_unregister_bulk (req);
783 req->rq_phase = RQ_PHASE_COMPLETE;
785 if (req->rq_interpret_reply != NULL) {
786 int (*interpreter)(struct ptlrpc_request *,void *,int) =
787 req->rq_interpret_reply;
788 req->rq_status = interpreter(req, &req->rq_async_args,
792 CDEBUG(D_RPCTRACE, "Completed RPC pname:cluuid:pid:xid:ni:nid:"
793 "opc %s:%s:%d:"LPU64":%s:%s:%d\n", current->comm,
794 imp->imp_obd->obd_uuid.uuid, req->rq_reqmsg->status,
796 imp->imp_connection->c_peer.peer_ni->pni_name,
797 ptlrpc_peernid2str(&imp->imp_connection->c_peer, str),
798 req->rq_reqmsg->opc);
800 set->set_remaining--;
802 atomic_dec(&imp->imp_inflight);
803 wake_up(&imp->imp_recovery_waitq);
806 /* If we hit an error, we want to recover promptly. */
807 RETURN(set->set_remaining == 0 || force_timer_recalc);
810 int ptlrpc_expire_one_request(struct ptlrpc_request *req)
813 struct obd_import *imp = req->rq_import;
816 DEBUG_REQ(D_ERROR, req, "timeout");
818 spin_lock_irqsave (&req->rq_lock, flags);
819 req->rq_timedout = 1;
820 spin_unlock_irqrestore (&req->rq_lock, flags);
822 ptlrpc_unregister_reply (req);
824 if (req->rq_bulk != NULL)
825 ptlrpc_unregister_bulk (req);
828 DEBUG_REQ(D_HA, req, "NULL import: already cleaned up?");
832 /* The DLM server doesn't want recovery run on its imports. */
833 if (imp->imp_dlm_fake)
836 /* If this request is for recovery or other primordial tasks,
837 * then error it out here. */
838 if (req->rq_send_state != LUSTRE_IMP_FULL ||
839 imp->imp_obd->obd_no_recov) {
840 spin_lock_irqsave (&req->rq_lock, flags);
841 req->rq_status = -ETIMEDOUT;
843 spin_unlock_irqrestore (&req->rq_lock, flags);
847 ptlrpc_fail_import(imp, req->rq_import_generation);
852 int ptlrpc_expired_set(void *data)
854 struct ptlrpc_request_set *set = data;
855 struct list_head *tmp;
856 time_t now = LTIME_S (CURRENT_TIME);
859 LASSERT(set != NULL);
861 /* A timeout expired; see which reqs it applies to... */
862 list_for_each (tmp, &set->set_requests) {
863 struct ptlrpc_request *req =
864 list_entry(tmp, struct ptlrpc_request, rq_set_chain);
866 /* request in-flight? */
867 if (!((req->rq_phase == RQ_PHASE_RPC && !req->rq_waiting
868 && !req->rq_resend) ||
869 (req->rq_phase == RQ_PHASE_BULK)))
872 if (req->rq_timedout || /* already dealt with */
873 req->rq_sent + req->rq_timeout > now) /* not expired */
876 /* deal with this guy */
877 ptlrpc_expire_one_request (req);
880 /* When waiting for a whole set, we always to break out of the
881 * sleep so we can recalculate the timeout, or enable interrupts
882 * iff everyone's timed out.
887 void ptlrpc_mark_interrupted(struct ptlrpc_request *req)
890 spin_lock_irqsave(&req->rq_lock, flags);
892 spin_unlock_irqrestore(&req->rq_lock, flags);
895 void ptlrpc_interrupted_set(void *data)
897 struct ptlrpc_request_set *set = data;
898 struct list_head *tmp;
900 LASSERT(set != NULL);
901 CERROR("INTERRUPTED SET %p\n", set);
903 list_for_each(tmp, &set->set_requests) {
904 struct ptlrpc_request *req =
905 list_entry(tmp, struct ptlrpc_request, rq_set_chain);
907 if (req->rq_phase != RQ_PHASE_RPC)
910 ptlrpc_mark_interrupted(req);
914 int ptlrpc_set_next_timeout(struct ptlrpc_request_set *set)
916 struct list_head *tmp;
917 time_t now = LTIME_S(CURRENT_TIME);
920 struct ptlrpc_request *req;
923 SIGNAL_MASK_ASSERT(); /* XXX BUG 1511 */
925 list_for_each(tmp, &set->set_requests) {
926 req = list_entry(tmp, struct ptlrpc_request, rq_set_chain);
928 /* request in-flight? */
929 if (!((req->rq_phase == RQ_PHASE_RPC && !req->rq_waiting) ||
930 (req->rq_phase == RQ_PHASE_BULK)))
933 if (req->rq_timedout) /* already timed out */
936 deadline = req->rq_sent + req->rq_timeout;
937 if (deadline <= now) /* actually expired already */
938 timeout = 1; /* ASAP */
939 else if (timeout == 0 || timeout > deadline - now)
940 timeout = deadline - now;
946 int ptlrpc_set_wait(struct ptlrpc_request_set *set)
948 struct list_head *tmp;
949 struct ptlrpc_request *req;
950 struct l_wait_info lwi;
954 LASSERT(!list_empty(&set->set_requests));
955 list_for_each(tmp, &set->set_requests) {
956 req = list_entry(tmp, struct ptlrpc_request, rq_set_chain);
957 if (req->rq_phase == RQ_PHASE_NEW)
958 (void)ptlrpc_send_new_req(req);
962 timeout = ptlrpc_set_next_timeout(set);
964 /* wait until all complete, interrupted, or an in-flight
966 CDEBUG(D_HA, "set %p going to sleep for %d seconds\n",
968 lwi = LWI_TIMEOUT_INTR((timeout ? timeout : 1) * HZ,
970 ptlrpc_interrupted_set, set);
971 rc = l_wait_event(set->set_waitq, ptlrpc_check_set(set), &lwi);
973 LASSERT(rc == 0 || rc == -EINTR || rc == -ETIMEDOUT);
975 /* -EINTR => all requests have been flagged rq_intr so next
977 * -ETIMEOUTD => someone timed out. When all reqs have
978 * timed out, signals are enabled allowing completion with
980 * I don't really care if we go once more round the loop in
981 * the error cases -eeb. */
984 LASSERT(set->set_remaining == 0);
987 list_for_each(tmp, &set->set_requests) {
988 req = list_entry(tmp, struct ptlrpc_request, rq_set_chain);
990 LASSERT(req->rq_phase == RQ_PHASE_COMPLETE);
991 if (req->rq_status != 0)
995 if (set->set_interpret != NULL) {
996 int (*interpreter)(struct ptlrpc_request_set *set,void *,int) =
998 rc = interpreter (set, &set->set_args, rc);
1004 static void __ptlrpc_free_req(struct ptlrpc_request *request, int locked)
1007 if (request == NULL) {
1012 LASSERTF(!request->rq_receiving_reply, "req %p\n", request);
1013 LASSERTF(request->rq_rqbd == NULL, "req %p\n",request);/* client-side */
1014 LASSERTF(list_empty(&request->rq_list), "req %p\n", request);
1015 LASSERTF(list_empty(&request->rq_set_chain), "req %p\n", request);
1017 /* We must take it off the imp_replay_list first. Otherwise, we'll set
1018 * request->rq_reqmsg to NULL while osc_close is dereferencing it. */
1019 if (request->rq_import != NULL) {
1020 unsigned long flags = 0;
1022 spin_lock_irqsave(&request->rq_import->imp_lock, flags);
1023 list_del_init(&request->rq_replay_list);
1025 spin_unlock_irqrestore(&request->rq_import->imp_lock,
1028 LASSERTF(list_empty(&request->rq_replay_list), "req %p\n", request);
1030 if (atomic_read(&request->rq_refcount) != 0) {
1031 DEBUG_REQ(D_ERROR, request,
1032 "freeing request with nonzero refcount");
1036 if (request->rq_repmsg != NULL) {
1037 OBD_FREE(request->rq_repmsg, request->rq_replen);
1038 request->rq_repmsg = NULL;
1040 if (request->rq_reqmsg != NULL) {
1041 OBD_FREE(request->rq_reqmsg, request->rq_reqlen);
1042 request->rq_reqmsg = NULL;
1044 if (request->rq_export != NULL) {
1045 class_export_put(request->rq_export);
1046 request->rq_export = NULL;
1048 if (request->rq_import != NULL) {
1049 class_import_put(request->rq_import);
1050 request->rq_import = NULL;
1052 if (request->rq_bulk != NULL)
1053 ptlrpc_free_bulk(request->rq_bulk);
1055 OBD_FREE(request, sizeof(*request));
1059 void ptlrpc_free_req(struct ptlrpc_request *request)
1061 __ptlrpc_free_req(request, 0);
1064 static int __ptlrpc_req_finished(struct ptlrpc_request *request, int locked);
1065 void ptlrpc_req_finished_with_imp_lock(struct ptlrpc_request *request)
1067 LASSERT_SPIN_LOCKED(&request->rq_import->imp_lock);
1068 (void)__ptlrpc_req_finished(request, 1);
1071 static int __ptlrpc_req_finished(struct ptlrpc_request *request, int locked)
1074 if (request == NULL)
1077 if (request == LP_POISON ||
1078 request->rq_reqmsg == LP_POISON) {
1079 CERROR("dereferencing freed request (bug 575)\n");
1084 DEBUG_REQ(D_INFO, request, "refcount now %u",
1085 atomic_read(&request->rq_refcount) - 1);
1087 if (atomic_dec_and_test(&request->rq_refcount)) {
1088 __ptlrpc_free_req(request, locked);
1095 void ptlrpc_req_finished(struct ptlrpc_request *request)
1097 __ptlrpc_req_finished(request, 0);
1100 /* Disengage the client's reply buffer from the network
1101 * NB does _NOT_ unregister any client-side bulk.
1102 * IDEMPOTENT, but _not_ safe against concurrent callers.
1103 * The request owner (i.e. the thread doing the I/O) must call...
1105 void ptlrpc_unregister_reply (struct ptlrpc_request *request)
1108 wait_queue_head_t *wq;
1109 struct l_wait_info lwi;
1111 LASSERT(!in_interrupt ()); /* might sleep */
1113 if (!ptlrpc_client_receiving_reply(request))
1116 PtlMDUnlink (request->rq_reply_md_h);
1118 /* We have to l_wait_event() whatever the result, to give liblustre
1119 * a chance to run reply_in_callback() */
1121 if (request->rq_set == NULL)
1122 wq = &request->rq_set->set_waitq;
1124 wq = &request->rq_reply_waitq;
1127 /* Network access will complete in finite time but the HUGE
1128 * timeout lets us CWARN for visibility of sluggish NALs */
1129 lwi = LWI_TIMEOUT(300 * HZ, NULL, NULL);
1130 rc = l_wait_event (*wq, !ptlrpc_client_receiving_reply(request), &lwi);
1134 LASSERT (rc == -ETIMEDOUT);
1135 DEBUG_REQ(D_WARNING, request, "Unexpectedly long timeout");
1139 /* caller must hold imp->imp_lock */
1140 void ptlrpc_free_committed(struct obd_import *imp)
1142 struct list_head *tmp, *saved;
1143 struct ptlrpc_request *req;
1144 struct ptlrpc_request *last_req = NULL; /* temporary fire escape */
1147 LASSERT(imp != NULL);
1149 LASSERT_SPIN_LOCKED(&imp->imp_lock);
1151 CDEBUG(D_HA, "%s: committing for last_committed "LPU64"\n",
1152 imp->imp_obd->obd_name, imp->imp_peer_committed_transno);
1154 list_for_each_safe(tmp, saved, &imp->imp_replay_list) {
1155 req = list_entry(tmp, struct ptlrpc_request, rq_replay_list);
1157 /* XXX ok to remove when 1357 resolved - rread 05/29/03 */
1158 LASSERT(req != last_req);
1161 if (req->rq_import_generation < imp->imp_generation) {
1162 DEBUG_REQ(D_HA, req, "freeing request with old gen");
1166 if (req->rq_replay) {
1167 DEBUG_REQ(D_HA, req, "keeping (FL_REPLAY)");
1171 /* not yet committed */
1172 if (req->rq_transno > imp->imp_peer_committed_transno) {
1173 DEBUG_REQ(D_HA, req, "stopping search");
1177 DEBUG_REQ(D_HA, req, "committing (last_committed "LPU64")",
1178 imp->imp_peer_committed_transno);
1180 if (req->rq_commit_cb != NULL)
1181 req->rq_commit_cb(req);
1182 list_del_init(&req->rq_replay_list);
1183 __ptlrpc_req_finished(req, 1);
1190 void ptlrpc_cleanup_client(struct obd_import *imp)
1197 void ptlrpc_resend_req(struct ptlrpc_request *req)
1199 unsigned long flags;
1201 DEBUG_REQ(D_HA, req, "going to resend");
1202 req->rq_reqmsg->handle.cookie = 0;
1203 req->rq_status = -EAGAIN;
1205 spin_lock_irqsave (&req->rq_lock, flags);
1207 req->rq_net_err = 0;
1208 req->rq_timedout = 0;
1210 __u64 old_xid = req->rq_xid;
1212 /* ensure previous bulk fails */
1213 req->rq_xid = ptlrpc_next_xid();
1214 CDEBUG(D_HA, "resend bulk old x"LPU64" new x"LPU64"\n",
1215 old_xid, req->rq_xid);
1217 ptlrpc_wake_client_req(req);
1218 spin_unlock_irqrestore (&req->rq_lock, flags);
1221 /* XXX: this function and rq_status are currently unused */
1222 void ptlrpc_restart_req(struct ptlrpc_request *req)
1224 unsigned long flags;
1226 DEBUG_REQ(D_HA, req, "restarting (possibly-)completed request");
1227 req->rq_status = -ERESTARTSYS;
1229 spin_lock_irqsave (&req->rq_lock, flags);
1230 req->rq_restart = 1;
1231 req->rq_timedout = 0;
1232 ptlrpc_wake_client_req(req);
1233 spin_unlock_irqrestore (&req->rq_lock, flags);
1236 static int expired_request(void *data)
1238 struct ptlrpc_request *req = data;
1241 RETURN(ptlrpc_expire_one_request(req));
1244 static void interrupted_request(void *data)
1246 unsigned long flags;
1248 struct ptlrpc_request *req = data;
1249 DEBUG_REQ(D_HA, req, "request interrupted");
1250 spin_lock_irqsave (&req->rq_lock, flags);
1252 spin_unlock_irqrestore (&req->rq_lock, flags);
1255 struct ptlrpc_request *ptlrpc_request_addref(struct ptlrpc_request *req)
1258 atomic_inc(&req->rq_refcount);
1262 void ptlrpc_retain_replayable_request(struct ptlrpc_request *req,
1263 struct obd_import *imp)
1265 struct list_head *tmp;
1267 LASSERT_SPIN_LOCKED(&imp->imp_lock);
1269 /* clear this for new requests that were resent as well
1270 as resent replayed requests. */
1271 lustre_msg_clear_flags(req->rq_reqmsg,
1274 /* don't re-add requests that have been replayed */
1275 if (!list_empty(&req->rq_replay_list))
1278 lustre_msg_add_flags(req->rq_reqmsg,
1281 LASSERT(imp->imp_replayable);
1282 /* Balanced in ptlrpc_free_committed, usually. */
1283 ptlrpc_request_addref(req);
1284 list_for_each_prev(tmp, &imp->imp_replay_list) {
1285 struct ptlrpc_request *iter =
1286 list_entry(tmp, struct ptlrpc_request, rq_replay_list);
1288 /* We may have duplicate transnos if we create and then
1289 * open a file, or for closes retained if to match creating
1290 * opens, so use req->rq_xid as a secondary key.
1291 * (See bugs 684, 685, and 428.)
1292 * XXX no longer needed, but all opens need transnos!
1294 if (iter->rq_transno > req->rq_transno)
1297 if (iter->rq_transno == req->rq_transno) {
1298 LASSERT(iter->rq_xid != req->rq_xid);
1299 if (iter->rq_xid > req->rq_xid)
1303 list_add(&req->rq_replay_list, &iter->rq_replay_list);
1307 list_add_tail(&req->rq_replay_list, &imp->imp_replay_list);
1310 int ptlrpc_queue_wait(struct ptlrpc_request *req)
1312 char str[PTL_NALFMT_SIZE];
1315 struct l_wait_info lwi;
1316 struct obd_import *imp = req->rq_import;
1317 unsigned long flags;
1321 LASSERT(req->rq_set == NULL);
1322 LASSERT(!req->rq_receiving_reply);
1323 atomic_inc(&imp->imp_inflight);
1325 /* for distributed debugging */
1326 req->rq_reqmsg->status = current->pid;
1327 LASSERT(imp->imp_obd != NULL);
1328 CDEBUG(D_RPCTRACE, "Sending RPC pname:cluuid:pid:xid:ni:nid:opc "
1329 "%s:%s:%d:"LPU64":%s:%s:%d\n", current->comm,
1330 imp->imp_obd->obd_uuid.uuid,
1331 req->rq_reqmsg->status, req->rq_xid,
1332 imp->imp_connection->c_peer.peer_ni->pni_name,
1333 ptlrpc_peernid2str(&imp->imp_connection->c_peer, str),
1334 req->rq_reqmsg->opc);
1336 /* Mark phase here for a little debug help */
1337 req->rq_phase = RQ_PHASE_RPC;
1339 spin_lock_irqsave(&imp->imp_lock, flags);
1340 req->rq_import_generation = imp->imp_generation;
1342 if (ptlrpc_import_delay_req(imp, req, &rc)) {
1343 list_del(&req->rq_list);
1345 list_add_tail(&req->rq_list, &imp->imp_delayed_list);
1346 spin_unlock_irqrestore(&imp->imp_lock, flags);
1348 DEBUG_REQ(D_HA, req, "\"%s\" waiting for recovery: (%s != %s)",
1350 ptlrpc_import_state_name(req->rq_send_state),
1351 ptlrpc_import_state_name(imp->imp_state));
1352 lwi = LWI_INTR(interrupted_request, req);
1353 rc = l_wait_event(req->rq_reply_waitq,
1354 (req->rq_send_state == imp->imp_state ||
1357 DEBUG_REQ(D_HA, req, "\"%s\" awake: (%s == %s or %d == 1)",
1359 ptlrpc_import_state_name(imp->imp_state),
1360 ptlrpc_import_state_name(req->rq_send_state),
1363 spin_lock_irqsave(&imp->imp_lock, flags);
1364 list_del_init(&req->rq_list);
1369 else if (req->rq_intr) {
1372 else if (req->rq_no_resend) {
1373 spin_unlock_irqrestore(&imp->imp_lock, flags);
1374 GOTO(out, rc = -ETIMEDOUT);
1382 list_del_init(&req->rq_list);
1383 spin_unlock_irqrestore(&imp->imp_lock, flags);
1384 req->rq_status = rc; // XXX this ok?
1388 if (req->rq_resend) {
1389 lustre_msg_add_flags(req->rq_reqmsg, MSG_RESENT);
1391 if (req->rq_bulk != NULL)
1392 ptlrpc_unregister_bulk (req);
1394 DEBUG_REQ(D_HA, req, "resending: ");
1397 /* XXX this is the same as ptlrpc_set_wait */
1398 LASSERT(list_empty(&req->rq_list));
1399 list_add_tail(&req->rq_list, &imp->imp_sending_list);
1400 spin_unlock_irqrestore(&imp->imp_lock, flags);
1402 rc = ptl_send_rpc(req);
1404 DEBUG_REQ(D_HA, req, "send failed (%d); recovering", rc);
1407 timeout = MAX(req->rq_timeout * HZ, 1);
1408 DEBUG_REQ(D_NET, req, "-- sleeping");
1410 lwi = LWI_TIMEOUT_INTR(timeout, expired_request, interrupted_request,
1412 l_wait_event(req->rq_reply_waitq, ptlrpc_check_reply(req), &lwi);
1413 DEBUG_REQ(D_NET, req, "-- done sleeping");
1415 CDEBUG(D_RPCTRACE, "Completed RPC pname:cluuid:pid:xid:ni:nid:opc "
1416 "%s:%s:%d:"LPU64":%s:%s:%d\n", current->comm,
1417 imp->imp_obd->obd_uuid.uuid,
1418 req->rq_reqmsg->status, req->rq_xid,
1419 imp->imp_connection->c_peer.peer_ni->pni_name,
1420 ptlrpc_peernid2str(&imp->imp_connection->c_peer, str),
1421 req->rq_reqmsg->opc);
1423 spin_lock_irqsave(&imp->imp_lock, flags);
1424 list_del_init(&req->rq_list);
1425 spin_unlock_irqrestore(&imp->imp_lock, flags);
1427 /* If the reply was received normally, this just grabs the spinlock
1428 * (ensuring the reply callback has returned), sees that
1429 * req->rq_receiving_reply is clear and returns. */
1430 ptlrpc_unregister_reply (req);
1433 GOTO(out, rc = -EIO);
1435 /* Resend if we need to, unless we were interrupted. */
1436 if (req->rq_resend && !req->rq_intr) {
1437 /* ...unless we were specifically told otherwise. */
1438 if (req->rq_no_resend)
1439 GOTO(out, rc = -ETIMEDOUT);
1440 spin_lock_irqsave(&imp->imp_lock, flags);
1445 /* Should only be interrupted if we timed out. */
1446 if (!req->rq_timedout)
1447 DEBUG_REQ(D_ERROR, req,
1448 "rq_intr set but rq_timedout not");
1449 GOTO(out, rc = -EINTR);
1452 if (req->rq_timedout) { /* non-recoverable timeout */
1453 GOTO(out, rc = -ETIMEDOUT);
1456 if (!req->rq_replied) {
1457 /* How can this be? -eeb */
1458 DEBUG_REQ(D_ERROR, req, "!rq_replied: ");
1460 GOTO(out, rc = req->rq_status);
1463 rc = after_reply (req);
1464 /* NB may return +ve success rc */
1465 if (req->rq_resend) {
1466 spin_lock_irqsave(&imp->imp_lock, flags);
1471 if (req->rq_bulk != NULL) {
1473 /* success so far. Note that anything going wrong
1474 * with bulk now, is EXTREMELY strange, since the
1475 * server must have believed that the bulk
1476 * tranferred OK before she replied with success to
1478 lwi = LWI_TIMEOUT(timeout, NULL, NULL);
1479 brc = l_wait_event(req->rq_reply_waitq,
1480 !ptlrpc_bulk_active(req->rq_bulk),
1482 LASSERT(brc == 0 || brc == -ETIMEDOUT);
1484 LASSERT(brc == -ETIMEDOUT);
1485 DEBUG_REQ(D_ERROR, req, "bulk timed out");
1487 } else if (!req->rq_bulk->bd_success) {
1488 DEBUG_REQ(D_ERROR, req, "bulk transfer failed");
1493 ptlrpc_unregister_bulk (req);
1496 LASSERT(!req->rq_receiving_reply);
1497 req->rq_phase = RQ_PHASE_INTERPRET;
1499 atomic_dec(&imp->imp_inflight);
1500 wake_up(&imp->imp_recovery_waitq);
1504 struct ptlrpc_replay_async_args {
1506 int praa_old_status;
1509 static int ptlrpc_replay_interpret(struct ptlrpc_request *req,
1510 void * data, int rc)
1512 struct ptlrpc_replay_async_args *aa = data;
1513 struct obd_import *imp = req->rq_import;
1514 unsigned long flags;
1516 atomic_dec(&imp->imp_replay_inflight);
1518 if (!req->rq_replied) {
1519 CERROR("request replay timed out, restarting recovery\n");
1520 GOTO(out, rc = -ETIMEDOUT);
1524 /* Clear reply swab mask; this is a new reply in sender's byte order */
1525 req->rq_rep_swab_mask = 0;
1527 LASSERT (req->rq_nob_received <= req->rq_replen);
1528 rc = lustre_unpack_msg(req->rq_repmsg, req->rq_nob_received);
1530 CERROR("unpack_rep failed: %d\n", rc);
1531 GOTO(out, rc = -EPROTO);
1534 if (req->rq_repmsg->type == PTL_RPC_MSG_ERR &&
1535 req->rq_repmsg->status == -ENOTCONN)
1536 GOTO(out, rc = req->rq_repmsg->status);
1538 /* The transno had better not change over replay. */
1539 LASSERT(req->rq_reqmsg->transno == req->rq_repmsg->transno);
1541 DEBUG_REQ(D_HA, req, "got rep");
1543 /* let the callback do fixups, possibly including in the request */
1544 if (req->rq_replay_cb)
1545 req->rq_replay_cb(req);
1547 if (req->rq_replied && req->rq_repmsg->status != aa->praa_old_status) {
1548 DEBUG_REQ(D_ERROR, req, "status %d, old was %d",
1549 req->rq_repmsg->status, aa->praa_old_status);
1551 /* Put it back for re-replay. */
1552 req->rq_repmsg->status = aa->praa_old_status;
1555 spin_lock_irqsave(&imp->imp_lock, flags);
1556 imp->imp_last_replay_transno = req->rq_transno;
1557 spin_unlock_irqrestore(&imp->imp_lock, flags);
1559 /* continue with recovery */
1560 rc = ptlrpc_import_recovery_state_machine(imp);
1562 req->rq_send_state = aa->praa_old_state;
1565 /* this replay failed, so restart recovery */
1566 ptlrpc_connect_import(imp, NULL);
1572 int ptlrpc_replay_req(struct ptlrpc_request *req)
1574 struct ptlrpc_replay_async_args *aa;
1577 LASSERT(req->rq_import->imp_state == LUSTRE_IMP_REPLAY);
1579 /* Not handling automatic bulk replay yet (or ever?) */
1580 LASSERT(req->rq_bulk == NULL);
1582 DEBUG_REQ(D_HA, req, "REPLAY");
1584 LASSERT (sizeof (*aa) <= sizeof (req->rq_async_args));
1585 aa = (struct ptlrpc_replay_async_args *)&req->rq_async_args;
1586 memset(aa, 0, sizeof *aa);
1588 /* Prepare request to be resent with ptlrpcd */
1589 aa->praa_old_state = req->rq_send_state;
1590 req->rq_send_state = LUSTRE_IMP_REPLAY;
1591 req->rq_phase = RQ_PHASE_NEW;
1592 aa->praa_old_status = req->rq_repmsg->status;
1595 req->rq_interpret_reply = ptlrpc_replay_interpret;
1596 atomic_inc(&req->rq_import->imp_replay_inflight);
1597 ptlrpc_request_addref(req); /* ptlrpcd needs a ref */
1599 ptlrpcd_add_req(req);
1603 void ptlrpc_abort_inflight(struct obd_import *imp)
1605 unsigned long flags;
1606 struct list_head *tmp, *n;
1609 /* Make sure that no new requests get processed for this import.
1610 * ptlrpc_{queue,set}_wait must (and does) hold imp_lock while testing
1611 * this flag and then putting requests on sending_list or delayed_list.
1613 spin_lock_irqsave(&imp->imp_lock, flags);
1615 /* XXX locking? Maybe we should remove each request with the list
1616 * locked? Also, how do we know if the requests on the list are
1617 * being freed at this time?
1619 list_for_each_safe(tmp, n, &imp->imp_sending_list) {
1620 struct ptlrpc_request *req =
1621 list_entry(tmp, struct ptlrpc_request, rq_list);
1623 DEBUG_REQ(D_HA, req, "inflight");
1625 spin_lock (&req->rq_lock);
1626 if (req->rq_import_generation < imp->imp_generation) {
1628 ptlrpc_wake_client_req(req);
1630 spin_unlock (&req->rq_lock);
1633 list_for_each_safe(tmp, n, &imp->imp_delayed_list) {
1634 struct ptlrpc_request *req =
1635 list_entry(tmp, struct ptlrpc_request, rq_list);
1637 DEBUG_REQ(D_HA, req, "aborting waiting req");
1639 spin_lock (&req->rq_lock);
1640 if (req->rq_import_generation < imp->imp_generation) {
1642 ptlrpc_wake_client_req(req);
1644 spin_unlock (&req->rq_lock);
1647 /* Last chance to free reqs left on the replay list, but we
1648 * will still leak reqs that haven't comitted. */
1649 if (imp->imp_replayable)
1650 ptlrpc_free_committed(imp);
1652 spin_unlock_irqrestore(&imp->imp_lock, flags);
1657 static __u64 ptlrpc_last_xid = 0;
1658 static spinlock_t ptlrpc_last_xid_lock = SPIN_LOCK_UNLOCKED;
1660 __u64 ptlrpc_next_xid(void)
1663 spin_lock(&ptlrpc_last_xid_lock);
1664 tmp = ++ptlrpc_last_xid;
1665 spin_unlock(&ptlrpc_last_xid_lock);