/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*-
 * vim:expandtab:shiftwidth=8:tabstop=8:
 *
 *  Copyright (c) 2002, 2003 Cluster File Systems, Inc.
 *
 *   This file is part of Lustre, http://www.lustre.org.
 *
 *   Lustre is free software; you can redistribute it and/or
 *   modify it under the terms of version 2 of the GNU General Public
 *   License as published by the Free Software Foundation.
 *
 *   Lustre is distributed in the hope that it will be useful,
 *   but WITHOUT ANY WARRANTY; without even the implied warranty of
 *   MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *   GNU General Public License for more details.
 *
 *   You should have received a copy of the GNU General Public License
 *   along with Lustre; if not, write to the Free Software
 *   Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
 *
 */

#define DEBUG_SUBSYSTEM S_RPC
#ifndef __KERNEL__
#include <errno.h>
#include <signal.h>
#include <liblustre.h>
#endif

#include <linux/obd_support.h>
#include <linux/obd_class.h>
#include <linux/lustre_lib.h>
#include <linux/lustre_ha.h>
#include <linux/lustre_import.h>

#include "ptlrpc_internal.h"

void ptlrpc_init_client(int req_portal, int rep_portal, char *name,
                        struct ptlrpc_client *cl)
{
        cl->cli_request_portal = req_portal;
        cl->cli_reply_portal   = rep_portal;
        cl->cli_name           = name;
}

struct obd_uuid *ptlrpc_req_to_uuid(struct ptlrpc_request *req)
{
        return &req->rq_connection->c_remote_uuid;
}

struct ptlrpc_connection *ptlrpc_uuid_to_connection(struct obd_uuid *uuid)
{
        struct ptlrpc_connection *c;
        struct ptlrpc_peer peer;
        int err;

        err = ptlrpc_uuid_to_peer(uuid, &peer);
        if (err != 0) {
                CERROR("cannot find peer %s!\n", uuid->uuid);
                return NULL;
        }

        c = ptlrpc_get_connection(&peer, uuid);
        if (c) {
                memcpy(c->c_remote_uuid.uuid,
                       uuid->uuid, sizeof(c->c_remote_uuid.uuid));
                c->c_epoch++;
        }

        CDEBUG(D_INFO, "%s -> %p\n", uuid->uuid, c);

        return c;
}

void ptlrpc_readdress_connection(struct ptlrpc_connection *conn,
                                 struct obd_uuid *uuid)
{
        struct ptlrpc_peer peer;
        int err;

        err = ptlrpc_uuid_to_peer(uuid, &peer);
        if (err != 0) {
                CERROR("cannot find peer %s!\n", uuid->uuid);
                return;
        }

        memcpy(&conn->c_peer, &peer, sizeof (peer));
        return;
}

static inline struct ptlrpc_bulk_desc *new_bulk(void)
{
        struct ptlrpc_bulk_desc *desc;

        OBD_ALLOC(desc, sizeof(*desc));
        if (!desc)
                return NULL;

        spin_lock_init(&desc->bd_lock);
        init_waitqueue_head(&desc->bd_waitq);
        INIT_LIST_HEAD(&desc->bd_page_list);
        desc->bd_md_h = PTL_HANDLE_NONE;
        desc->bd_me_h = PTL_HANDLE_NONE;

        return desc;
}

struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_imp (struct ptlrpc_request *req,
                                               int type, int portal)
{
        struct obd_import *imp = req->rq_import;
        struct ptlrpc_bulk_desc *desc;

        LASSERT(type == BULK_PUT_SINK || type == BULK_GET_SOURCE);

        desc = new_bulk();
        if (desc == NULL)
                RETURN(NULL);

        desc->bd_import_generation = req->rq_import_generation;
        desc->bd_import = class_import_get(imp);
        desc->bd_req = req;
        desc->bd_type = type;
        desc->bd_portal = portal;

        /* This makes req own desc, and free it when she frees herself */
        req->rq_bulk = desc;

        return desc;
}

struct ptlrpc_bulk_desc *ptlrpc_prep_bulk_exp (struct ptlrpc_request *req,
                                               int type, int portal)
{
        struct obd_export *exp = req->rq_export;
        struct ptlrpc_bulk_desc *desc;

        LASSERT(type == BULK_PUT_SOURCE || type == BULK_GET_SINK);

        desc = new_bulk();
        if (desc == NULL)
                RETURN(NULL);

        desc->bd_export = class_export_get(exp);
        desc->bd_req = req;
        desc->bd_type = type;
        desc->bd_portal = portal;

        /* NB we don't assign rq_bulk here; server-side requests are
         * re-used, and the handler frees the bulk desc explicitly. */

        return desc;
}

int ptlrpc_prep_bulk_page(struct ptlrpc_bulk_desc *desc,
                          struct page *page, int pageoffset, int len)
{
        struct ptlrpc_bulk_page *bulk;

        OBD_ALLOC(bulk, sizeof(*bulk));
        if (bulk == NULL)
                return -ENOMEM;

        LASSERT(page != NULL);
        LASSERT(pageoffset >= 0);
        LASSERT(len > 0);
        LASSERT(pageoffset + len <= PAGE_SIZE);

        bulk->bp_page = page;
        bulk->bp_pageoffset = pageoffset;
        bulk->bp_buflen = len;

        bulk->bp_desc = desc;
        list_add_tail(&bulk->bp_link, &desc->bd_page_list);
        desc->bd_page_count++;
        return 0;
}

void ptlrpc_free_bulk(struct ptlrpc_bulk_desc *desc)
{
        struct list_head *tmp, *next;
        ENTRY;

        LASSERT(desc != NULL);
        LASSERT(desc->bd_page_count != 0x5a5a5a5a); /* not freed already */
        LASSERT(!desc->bd_network_rw);         /* network hands off or */

        list_for_each_safe(tmp, next, &desc->bd_page_list) {
                struct ptlrpc_bulk_page *bulk;
                bulk = list_entry(tmp, struct ptlrpc_bulk_page, bp_link);
                ptlrpc_free_bulk_page(bulk);
        }

        LASSERT(desc->bd_page_count == 0);
        LASSERT((desc->bd_export != NULL) ^ (desc->bd_import != NULL));

        if (desc->bd_export)
                class_export_put(desc->bd_export);
        else
                class_import_put(desc->bd_import);

        OBD_FREE(desc, sizeof(*desc));
        EXIT;
}

void ptlrpc_free_bulk_page(struct ptlrpc_bulk_page *bulk)
{
        LASSERT(bulk != NULL);

        list_del(&bulk->bp_link);
        bulk->bp_desc->bd_page_count--;
        OBD_FREE(bulk, sizeof(*bulk));
}

struct ptlrpc_request *ptlrpc_prep_req(struct obd_import *imp, int opcode,
                                       int count, int *lengths, char **bufs)
{
        struct ptlrpc_request *request;
        int rc;
        ENTRY;

        LASSERT((unsigned long)imp > 0x1000);

        OBD_ALLOC(request, sizeof(*request));
        if (!request) {
                CERROR("request allocation out of memory\n");
                RETURN(NULL);
        }

        rc = lustre_pack_msg(count, lengths, bufs,
                             &request->rq_reqlen, &request->rq_reqmsg);
        if (rc) {
                CERROR("cannot pack request %d\n", rc);
                OBD_FREE(request, sizeof(*request));
                RETURN(NULL);
        }

        request->rq_timeout = obd_timeout;
        request->rq_level = LUSTRE_CONN_FULL;
        request->rq_type = PTL_RPC_MSG_REQUEST;
        request->rq_import = class_import_get(imp);
        request->rq_phase = RQ_PHASE_NEW;

        /* XXX FIXME bug 249 */
        request->rq_request_portal = imp->imp_client->cli_request_portal;
        request->rq_reply_portal = imp->imp_client->cli_reply_portal;

        request->rq_connection = ptlrpc_connection_addref(imp->imp_connection);

        spin_lock_init(&request->rq_lock);
        INIT_LIST_HEAD(&request->rq_list);
        init_waitqueue_head(&request->rq_wait_for_rep);
        request->rq_xid = ptlrpc_next_xid();
        atomic_set(&request->rq_refcount, 1);

        request->rq_reqmsg->opc = opcode;
        request->rq_reqmsg->flags = 0;

        RETURN(request);
}

struct ptlrpc_request_set *ptlrpc_prep_set(void)
{
        struct ptlrpc_request_set *set;

        OBD_ALLOC(set, sizeof *set);
        if (!set)
                RETURN(NULL);
        INIT_LIST_HEAD(&set->set_requests);
        init_waitqueue_head(&set->set_waitq);
        set->set_remaining = 0;

        RETURN(set);
}

/* Finish with this set; opposite of prep_set. */
void ptlrpc_set_destroy(struct ptlrpc_request_set *set)
{
        struct list_head *tmp;
        struct list_head *next;
        int               expected_phase;
        int               n = 0;
        ENTRY;

        /* Requests on the set should either all be completed, or all be new */
        expected_phase = (set->set_remaining == 0) ?
                         RQ_PHASE_COMPLETE : RQ_PHASE_NEW;
        list_for_each (tmp, &set->set_requests) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_set_chain);

                LASSERT(req->rq_phase == expected_phase);
                n++;
        }

        LASSERT(set->set_remaining == 0 || set->set_remaining == n);

        list_for_each_safe(tmp, next, &set->set_requests) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_set_chain);
                list_del_init(&req->rq_set_chain);

                LASSERT(req->rq_phase == expected_phase);

                if (req->rq_phase == RQ_PHASE_NEW) {

                        if (req->rq_interpret_reply != NULL) {
                                int (*interpreter)(struct ptlrpc_request *,
                                                   void *, int) =
                                        req->rq_interpret_reply;

                                /* higher level (i.e. LOV) failed;
                                 * let the sub reqs clean up */
                                req->rq_status = -EBADR;
                                interpreter(req, &req->rq_async_args,
                                            req->rq_status);
                        }
                        set->set_remaining--;
                }

                req->rq_set = NULL;
                ptlrpc_req_finished (req);
        }

        LASSERT(set->set_remaining == 0);

        OBD_FREE(set, sizeof(*set));
        EXIT;
}

void ptlrpc_set_add_req(struct ptlrpc_request_set *set,
                        struct ptlrpc_request *req)
{
        /* The set takes over the caller's request reference */
        list_add_tail(&req->rq_set_chain, &set->set_requests);
        req->rq_set = set;
        set->set_remaining++;
}

static int ptlrpc_check_reply(struct ptlrpc_request *req)
{
        unsigned long flags;
        int rc = 0;
        ENTRY;

        /* serialise with network callback */
        spin_lock_irqsave (&req->rq_lock, flags);

        if (req->rq_replied) {
                DEBUG_REQ(D_NET, req, "REPLIED:");
                GOTO(out, rc = 1);
        }

        if (req->rq_err) {
                DEBUG_REQ(D_ERROR, req, "ABORTED:");
                GOTO(out, rc = 1);
        }

        if (req->rq_resend) {
                DEBUG_REQ(D_ERROR, req, "RESEND:");
                GOTO(out, rc = 1);
        }

        if (req->rq_restart) {
                DEBUG_REQ(D_ERROR, req, "RESTART:");
                GOTO(out, rc = 1);
        }
        EXIT;
 out:
        spin_unlock_irqrestore (&req->rq_lock, flags);
        DEBUG_REQ(D_NET, req, "rc = %d for", rc);
        return rc;
}

static int ptlrpc_check_status(struct ptlrpc_request *req)
{
        int err;
        ENTRY;

        err = req->rq_repmsg->status;
        if (req->rq_repmsg->type == PTL_RPC_MSG_ERR) {
                DEBUG_REQ(D_ERROR, req, "type == PTL_RPC_MSG_ERR (%d)", err);
                if (err >= 0)
                        CERROR("Error Reply has >= zero status\n");
                RETURN(err < 0 ? err : -EINVAL);
        }

        if (err < 0) {
                DEBUG_REQ(D_INFO, req, "status is %d", err);
        } else if (err > 0) {
                /* XXX: translate this error from net to host */
                DEBUG_REQ(D_INFO, req, "status is %d", err);
        }

        RETURN(err);
}

static int after_reply(struct ptlrpc_request *req, int *restartp)
{
        unsigned long flags;
        struct obd_import *imp = req->rq_import;
        int rc;
        ENTRY;

        LASSERT(!req->rq_receiving_reply);
        LASSERT(req->rq_replied);

        if (restartp != NULL)
                *restartp = 0;

        /* NB Until this point, the whole of the incoming message,
         * including buflens, status etc is in the sender's byte order. */

#if SWAB_PARANOIA
        /* Clear reply swab mask; this is a new reply in sender's byte order */
        req->rq_rep_swab_mask = 0;
#endif
        rc = lustre_unpack_msg(req->rq_repmsg, req->rq_replen);
        if (rc) {
                CERROR("unpack_rep failed: %d\n", rc);
                RETURN(-EPROTO);
        }

        if (req->rq_repmsg->type != PTL_RPC_MSG_REPLY &&
            req->rq_repmsg->type != PTL_RPC_MSG_ERR) {
                CERROR("invalid packet type received (type=%u)\n",
                       req->rq_repmsg->type);
                RETURN(-EPROTO);
        }

        /* Store transno in reqmsg for replay. */
        req->rq_reqmsg->transno = req->rq_transno = req->rq_repmsg->transno;

        rc = ptlrpc_check_status(req);

        /* Either we've been evicted, or the server has failed for
         * some reason. Try to reconnect, and if that fails, punt to the
         * upcall. */
        if (rc == -ENOTCONN) {
                if (req->rq_level < LUSTRE_CONN_FULL || req->rq_no_recov ||
                    imp->imp_obd->obd_no_recov || imp->imp_dlm_fake) {
                        RETURN(-ENOTCONN);
                }

                ptlrpc_request_handle_notconn(req);

                if (req->rq_err)
                        RETURN(-EIO);

                if (req->rq_no_resend)
                        RETURN(rc); /* -ENOTCONN */

                if (req->rq_resend) {
                        if (restartp == NULL)
                                LBUG(); /* async resend not supported yet */
                        spin_lock_irqsave (&req->rq_lock, flags);
                        req->rq_resend = 0;
                        spin_unlock_irqrestore (&req->rq_lock, flags);
                        *restartp = 1;
                        lustre_msg_add_flags(req->rq_reqmsg, MSG_RESENT);
                        DEBUG_REQ(D_HA, req, "resending: ");
                        RETURN(0);
                }

                CERROR("request should be err or resend: %p\n", req);
                LBUG();
        }

        if (req->rq_import->imp_replayable) {
                spin_lock_irqsave(&imp->imp_lock, flags);
                if ((req->rq_replay || req->rq_transno != 0) && rc >= 0)
                        ptlrpc_retain_replayable_request(req, imp);

                if (req->rq_transno > imp->imp_max_transno)
                        imp->imp_max_transno = req->rq_transno;

                /* Replay-enabled imports return commit-status information. */
                if (req->rq_repmsg->last_committed)
                        imp->imp_peer_committed_transno =
                                req->rq_repmsg->last_committed;
                ptlrpc_free_committed(imp);
                spin_unlock_irqrestore(&imp->imp_lock, flags);
        }

        RETURN(rc);
}

int ptlrpc_check_set(struct ptlrpc_request_set *set)
{
        unsigned long flags;
        struct list_head *tmp;
        ENTRY;

        if (set->set_remaining == 0)
                RETURN(1);

        list_for_each(tmp, &set->set_requests) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_set_chain);
                struct obd_import *imp = req->rq_import;
                int rc = 0;

                if (!(req->rq_phase == RQ_PHASE_RPC ||
                      req->rq_phase == RQ_PHASE_BULK ||
                      req->rq_phase == RQ_PHASE_INTERPRET ||
                      req->rq_phase == RQ_PHASE_COMPLETE)) {
                        DEBUG_REQ(D_ERROR, req, "bad phase %x", req->rq_phase);
                        LBUG();
                }

                if (req->rq_phase == RQ_PHASE_COMPLETE)
                        continue;

                if (req->rq_phase == RQ_PHASE_INTERPRET)
                        GOTO(interpret, req->rq_status);

                if (req->rq_err) {
                        ptlrpc_unregister_reply(req);
                        if (req->rq_status == 0)
                                req->rq_status = -EIO;
                        req->rq_phase = RQ_PHASE_INTERPRET;

                        spin_lock_irqsave(&imp->imp_lock, flags);
                        list_del_init(&req->rq_list);
                        spin_unlock_irqrestore(&imp->imp_lock, flags);

                        GOTO(interpret, req->rq_status);
                }

                if (req->rq_intr) {
                        /* NB could be on delayed list */
                        ptlrpc_unregister_reply(req);
                        req->rq_status = -EINTR;
                        req->rq_phase = RQ_PHASE_INTERPRET;

                        spin_lock_irqsave(&imp->imp_lock, flags);
                        list_del_init(&req->rq_list);
                        spin_unlock_irqrestore(&imp->imp_lock, flags);

                        GOTO(interpret, req->rq_status);
                }

                if (req->rq_phase == RQ_PHASE_RPC) {
                        int do_restart = 0;
                        if (req->rq_waiting || req->rq_resend) {
                                spin_lock_irqsave(&imp->imp_lock, flags);

                                if (req->rq_level > imp->imp_level) {
                                        spin_unlock_irqrestore(&imp->imp_lock,
                                                               flags);
                                        continue;
                                }

                                list_del(&req->rq_list);
                                list_add_tail(&req->rq_list,
                                              &imp->imp_sending_list);

                                if (req->rq_import_generation <
                                    imp->imp_generation) {
                                        req->rq_status = -EIO;
                                        req->rq_phase = RQ_PHASE_INTERPRET;
                                        spin_unlock_irqrestore(&imp->imp_lock,
                                                               flags);
                                        GOTO(interpret, req->rq_status);
                                }
                                spin_unlock_irqrestore(&imp->imp_lock, flags);

                                req->rq_waiting = 0;
                                if (req->rq_resend) {
                                        lustre_msg_add_flags(req->rq_reqmsg,
                                                             MSG_RESENT);
                                        spin_lock_irqsave(&req->rq_lock, flags);
                                        req->rq_resend = 0;
                                        spin_unlock_irqrestore(&req->rq_lock,
                                                               flags);

                                        ptlrpc_unregister_reply(req);
                                        if (req->rq_bulk)
                                                ptlrpc_unregister_bulk(req);
                                }

                                rc = ptl_send_rpc(req);
                                if (rc) {
                                        req->rq_status = rc;
                                        req->rq_phase = RQ_PHASE_INTERPRET;
                                        GOTO(interpret, req->rq_status);
                                }

                        }

                        /* Ensure the network callback returned */
                        spin_lock_irqsave (&req->rq_lock, flags);
                        if (!req->rq_replied) {
                                spin_unlock_irqrestore (&req->rq_lock, flags);
                                continue;
                        }
                        spin_unlock_irqrestore (&req->rq_lock, flags);

                        spin_lock_irqsave(&imp->imp_lock, flags);
                        list_del_init(&req->rq_list);
                        spin_unlock_irqrestore(&imp->imp_lock, flags);

                        req->rq_status = after_reply(req, &do_restart);
                        if (do_restart) {
                                spin_lock_irqsave (&req->rq_lock, flags);
                                req->rq_resend = 1; /* ugh */
                                spin_unlock_irqrestore (&req->rq_lock, flags);
                                continue;
                        }

                        /* If there is no bulk associated with this request,
                         * then we're done and should let the interpreter
                         * process the reply.  Similarly if the RPC returned
                         * an error, and therefore the bulk will never arrive.
                         */
                        if (req->rq_bulk == NULL || req->rq_status != 0) {
                                req->rq_phase = RQ_PHASE_INTERPRET;
                                GOTO(interpret, req->rq_status);
                        }

                        req->rq_phase = RQ_PHASE_BULK;
                }

                LASSERT(req->rq_phase == RQ_PHASE_BULK);
                if (!ptlrpc_bulk_complete (req->rq_bulk))
                        continue;

                req->rq_phase = RQ_PHASE_INTERPRET;

        interpret:
                LASSERT(req->rq_phase == RQ_PHASE_INTERPRET);
                LASSERT(!req->rq_receiving_reply);

                ptlrpc_unregister_reply(req);
                if (req->rq_bulk != NULL)
                        ptlrpc_unregister_bulk (req);

                if (req->rq_interpret_reply != NULL) {
                        int (*interpreter)(struct ptlrpc_request *,void *,int) =
                                req->rq_interpret_reply;
                        req->rq_status = interpreter(req, &req->rq_async_args,
                                                     req->rq_status);
                }

                CDEBUG(D_RPCTRACE, "Completed RPC pname:cluuid:pid:xid:ni:nid:"
                       "opc %s:%s:%d:"LPU64":%s:"LPX64":%d\n", current->comm,
                       imp->imp_obd->obd_uuid.uuid, req->rq_reqmsg->status,
                       req->rq_xid,
                       imp->imp_connection->c_peer.peer_ni->pni_name,
                       imp->imp_connection->c_peer.peer_nid,
                       req->rq_reqmsg->opc);

                req->rq_phase = RQ_PHASE_COMPLETE;
                set->set_remaining--;
        }

        RETURN(set->set_remaining == 0);
}

int ptlrpc_expire_one_request(struct ptlrpc_request *req)
{
        unsigned long      flags;
        struct obd_import *imp = req->rq_import;
        ENTRY;

        DEBUG_REQ(D_ERROR, req, "timeout");

        spin_lock_irqsave (&req->rq_lock, flags);
        req->rq_timedout = 1;
        spin_unlock_irqrestore (&req->rq_lock, flags);

        ptlrpc_unregister_reply (req);

        if (imp == NULL) {
                DEBUG_REQ(D_HA, req, "NULL import: already cleaned up?");
                RETURN(1);
        }

        /* The DLM server doesn't want recovery run on its imports. */
        if (imp->imp_dlm_fake)
                RETURN(1);

        /* If this request is for recovery or other primordial tasks,
         * don't go back to sleep, and don't start recovery again.. */
        if (req->rq_level < LUSTRE_CONN_FULL || req->rq_no_recov ||
            imp->imp_obd->obd_no_recov)
                RETURN(1);

        ptlrpc_fail_import(imp, req->rq_import_generation);

        RETURN(0);
}

static int expired_set(void *data)
{
        struct ptlrpc_request_set *set = data;
        struct list_head          *tmp;
        time_t                     now = LTIME_S (CURRENT_TIME);
        ENTRY;

        LASSERT(set != NULL);

        /* A timeout expired; see which reqs it applies to... */
        list_for_each (tmp, &set->set_requests) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_set_chain);

                /* request in-flight? */
                if (!((req->rq_phase == RQ_PHASE_RPC && !req->rq_waiting) ||
                      (req->rq_phase == RQ_PHASE_BULK)))
                        continue;

                if (req->rq_timedout ||           /* already dealt with */
                    req->rq_sent + req->rq_timeout > now) /* not expired */
                        continue;

                /* deal with this guy */
                ptlrpc_expire_one_request (req);
        }

        /* When waiting for a whole set, we always to break out of the
         * sleep so we can recalculate the timeout, or enable interrupts
         * iff everyone's timed out.
         */
        RETURN(1);
}

static void interrupted_set(void *data)
{
        struct ptlrpc_request_set *set = data;
        struct list_head *tmp;
        unsigned long flags;

        LASSERT(set != NULL);
        CERROR("INTERRUPTED SET %p\n", set);

        list_for_each(tmp, &set->set_requests) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_set_chain);

                if (req->rq_phase != RQ_PHASE_RPC)
                        continue;

                spin_lock_irqsave (&req->rq_lock, flags);
                req->rq_intr = 1;
                spin_unlock_irqrestore (&req->rq_lock, flags);
        }
}

int ptlrpc_set_wait(struct ptlrpc_request_set *set)
{
        struct list_head      *tmp;
        struct obd_import     *imp;
        struct ptlrpc_request *req;
        struct l_wait_info     lwi;
        unsigned long          flags;
        int                    rc;
        time_t                 now;
        time_t                 deadline;
        int                    timeout;
        ENTRY;

        SIGNAL_MASK_ASSERT(); /* XXX BUG 1511 */
        LASSERT(!list_empty(&set->set_requests));
        list_for_each(tmp, &set->set_requests) {
                req = list_entry(tmp, struct ptlrpc_request, rq_set_chain);

                LASSERT(req->rq_level == LUSTRE_CONN_FULL);
                LASSERT(req->rq_phase == RQ_PHASE_NEW);
                req->rq_phase = RQ_PHASE_RPC;

                imp = req->rq_import;
                spin_lock_irqsave(&imp->imp_lock, flags);

                if (imp->imp_invalid) {
                        spin_unlock_irqrestore(&imp->imp_lock, flags);
                        req->rq_status = -EIO;
                        req->rq_phase = RQ_PHASE_INTERPRET;
                        continue;
                }

                req->rq_import_generation = imp->imp_generation;

                if (req->rq_level > imp->imp_level) {
                        if (req->rq_no_recov || imp->imp_obd->obd_no_recov ||
                            imp->imp_dlm_fake) {
                                spin_unlock_irqrestore(&imp->imp_lock, flags);
                                req->rq_status = -EWOULDBLOCK;
                                req->rq_phase = RQ_PHASE_INTERPRET;
                                continue;
                        }

                        spin_lock (&req->rq_lock);
                        req->rq_waiting = 1;
                        spin_unlock (&req->rq_lock);
                        LASSERT(list_empty (&req->rq_list));
                        // list_del(&req->rq_list);
                        list_add_tail(&req->rq_list, &imp->imp_delayed_list);
                        spin_unlock_irqrestore(&imp->imp_lock, flags);
                        continue;
                }

                /* XXX this is the same as ptlrpc_queue_wait */
                LASSERT(list_empty(&req->rq_list));
                list_add_tail(&req->rq_list, &imp->imp_sending_list);
                spin_unlock_irqrestore(&imp->imp_lock, flags);

                req->rq_reqmsg->status = current->pid;
                CDEBUG(D_RPCTRACE, "Sending RPC pname:cluuid:pid:xid:ni:nid:opc"
                       " %s:%s:%d:"LPU64":%s:"LPX64":%d\n", current->comm,
                       imp->imp_obd->obd_uuid.uuid, req->rq_reqmsg->status,
                       req->rq_xid,
                       imp->imp_connection->c_peer.peer_ni->pni_name,
                       imp->imp_connection->c_peer.peer_nid,
                       req->rq_reqmsg->opc);

                rc = ptl_send_rpc(req);
                if (rc) {
                        req->rq_status = rc;
                        req->rq_phase = RQ_PHASE_INTERPRET;
                }
        }

        do {
                now = LTIME_S (CURRENT_TIME);
                timeout = 0;
                list_for_each (tmp, &set->set_requests) {
                        req = list_entry(tmp, struct ptlrpc_request,
                                         rq_set_chain);

                        /* request in-flight? */
                        if (!((req->rq_phase == RQ_PHASE_RPC &&
                               !req->rq_waiting) ||
                              (req->rq_phase == RQ_PHASE_BULK)))
                                continue;

                        if (req->rq_timedout)   /* already timed out */
                                continue;

                        deadline = req->rq_sent + req->rq_timeout;
                        if (deadline <= now)    /* actually expired already */
                                timeout = 1;    /* ASAP */
                        else if (timeout == 0 || timeout > deadline - now)
                                timeout = deadline - now;
                }

                /* wait until all complete, interrupted, or an in-flight
                 * req times out */
                CDEBUG(D_HA, "set %p going to sleep for %d seconds\n",
                       set, timeout);
                lwi = LWI_TIMEOUT_INTR((timeout ? timeout : 1) * HZ,
                                       expired_set, interrupted_set, set);
                rc = l_wait_event(set->set_waitq, ptlrpc_check_set(set), &lwi);

                LASSERT(rc == 0 || rc == -EINTR || rc == -ETIMEDOUT);

                /* -EINTR => all requests have been flagged rq_intr so next
                 * check completes.
                 * -ETIMEOUTD => someone timed out.  When all reqs have
                 * timed out, signals are enabled allowing completion with
                 * EINTR.
                 * I don't really care if we go once more round the loop in
                 * the error cases -eeb. */
        } while (rc != 0);

        LASSERT(set->set_remaining == 0);

        rc = 0;
        list_for_each(tmp, &set->set_requests) {
                req = list_entry(tmp, struct ptlrpc_request, rq_set_chain);

                LASSERT(req->rq_phase == RQ_PHASE_COMPLETE);
                if (req->rq_status != 0)
                        rc = req->rq_status;
        }

        if (set->set_interpret != NULL) {
                int (*interpreter)(struct ptlrpc_request_set *set,void *,int) =
                        set->set_interpret;
                rc = interpreter (set, &set->set_args, rc);
        }

        RETURN(rc);
}

static void __ptlrpc_free_req(struct ptlrpc_request *request, int locked)
{
        ENTRY;
        if (request == NULL) {
                EXIT;
                return;
        }

        LASSERT(!request->rq_receiving_reply);

        /* We must take it off the imp_replay_list first.  Otherwise, we'll set
         * request->rq_reqmsg to NULL while osc_close is dereferencing it. */
        if (request->rq_import != NULL) {
                unsigned long flags = 0;
                if (!locked)
                        spin_lock_irqsave(&request->rq_import->imp_lock, flags);
                list_del_init(&request->rq_list);
                if (!locked)
                        spin_unlock_irqrestore(&request->rq_import->imp_lock,
                                               flags);
        }

        if (atomic_read(&request->rq_refcount) != 0) {
                DEBUG_REQ(D_ERROR, request,
                          "freeing request with nonzero refcount");
                LBUG();
        }

        if (request->rq_repmsg != NULL) {
                OBD_FREE(request->rq_repmsg, request->rq_replen);
                request->rq_repmsg = NULL;
        }
        if (request->rq_reqmsg != NULL) {
                OBD_FREE(request->rq_reqmsg, request->rq_reqlen);
                request->rq_reqmsg = NULL;
        }
        if (request->rq_export != NULL) {
                class_export_put(request->rq_export);
                request->rq_export = NULL;
        }
        if (request->rq_import != NULL) {
                class_import_put(request->rq_import);
                request->rq_import = NULL;
        }
        if (request->rq_bulk != NULL)
                ptlrpc_free_bulk(request->rq_bulk);

        ptlrpc_put_connection(request->rq_connection);
        OBD_FREE(request, sizeof(*request));
        EXIT;
}

void ptlrpc_free_req(struct ptlrpc_request *request)
{
        __ptlrpc_free_req(request, 0);
}

static int __ptlrpc_req_finished(struct ptlrpc_request *request, int locked)
{
        ENTRY;
        if (request == NULL)
                RETURN(1);

        if (request == (void *)(long)(0x5a5a5a5a5a5a5a5a) ||
            request->rq_obd == (void *)(long)(0x5a5a5a5a5a5a5a5a)) {
                CERROR("dereferencing freed request (bug 575)\n");
                LBUG();
                RETURN(1);
        }

        DEBUG_REQ(D_INFO, request, "refcount now %u",
                  atomic_read(&request->rq_refcount) - 1);

        if (atomic_dec_and_test(&request->rq_refcount)) {
                __ptlrpc_free_req(request, locked);
                RETURN(1);
        }

        RETURN(0);
}

void ptlrpc_req_finished(struct ptlrpc_request *request)
{
        __ptlrpc_req_finished(request, 0);
}

static void ptlrpc_cleanup_request_buf(struct ptlrpc_request *request)
{
        OBD_FREE(request->rq_reqmsg, request->rq_reqlen);
        request->rq_reqmsg = NULL;
        request->rq_reqlen = 0;
}

/* Disengage the client's reply buffer from the network
 * NB does _NOT_ unregister any client-side bulk.
 * IDEMPOTENT, but _not_ safe against concurrent callers.
 * The request owner (i.e. the thread doing the I/O) must call...
 */
void ptlrpc_unregister_reply (struct ptlrpc_request *request)
{
        unsigned long flags;
        int           rc;
        ENTRY;

        LASSERT(!in_interrupt ());             /* might sleep */

        spin_lock_irqsave (&request->rq_lock, flags);
        if (!request->rq_receiving_reply) {     /* not waiting for a reply */
                spin_unlock_irqrestore (&request->rq_lock, flags);
                EXIT;
                /* NB reply buffer not freed here */
                return;
        }

        LASSERT(!request->rq_replied);         /* callback hasn't completed */
        spin_unlock_irqrestore (&request->rq_lock, flags);

        rc = PtlMDUnlink (request->rq_reply_md_h);
        switch (rc) {
        default:
                LBUG ();

        case PTL_OK:                            /* unlinked before completion */
                LASSERT(request->rq_receiving_reply);
                LASSERT(!request->rq_replied);
                spin_lock_irqsave (&request->rq_lock, flags);
                request->rq_receiving_reply = 0;
                spin_unlock_irqrestore (&request->rq_lock, flags);
                OBD_FREE(request->rq_repmsg, request->rq_replen);
                request->rq_repmsg = NULL;
                EXIT;
                return;

        case PTL_MD_INUSE:                      /* callback in progress */
                for (;;) {
                        /* Network access will complete in finite time but
                         * the timeout lets us CERROR for visibility */
                        struct l_wait_info lwi = LWI_TIMEOUT(10*HZ, NULL, NULL);

                        rc = l_wait_event (request->rq_wait_for_rep,
                                           request->rq_replied, &lwi);
                        LASSERT(rc == 0 || rc == -ETIMEDOUT);
                        if (rc == 0) {
                                spin_lock_irqsave (&request->rq_lock, flags);
                                /* Ensure the callback has completed scheduling
                                 * me and taken its hands off the request */
                                spin_unlock_irqrestore(&request->rq_lock,flags);
                                break;
                        }

                        CERROR ("Unexpectedly long timeout: req %p\n", request);
                }
                /* fall through */

        case PTL_INV_MD:                        /* callback completed */
                LASSERT(!request->rq_receiving_reply);
                LASSERT(request->rq_replied);
                EXIT;
                return;
        }
        /* Not Reached */
}

/* caller must hold imp->imp_lock */
void ptlrpc_free_committed(struct obd_import *imp)
{
        struct list_head *tmp, *saved;
        struct ptlrpc_request *req;
        struct ptlrpc_request *last_req = NULL; /* temporary fire escape */
        ENTRY;

        LASSERT(imp != NULL);

#ifdef CONFIG_SMP
        LASSERT(spin_is_locked(&imp->imp_lock));
#endif

        CDEBUG(D_HA, "%s: committing for last_committed "LPU64"\n",
               imp->imp_obd->obd_name, imp->imp_peer_committed_transno);

        list_for_each_safe(tmp, saved, &imp->imp_replay_list) {
                req = list_entry(tmp, struct ptlrpc_request, rq_list);

                /* XXX ok to remove when 1357 resolved - rread 05/29/03  */
                LASSERT(req != last_req);
                last_req = req;

                if (req->rq_import_generation < imp->imp_generation) {
                        DEBUG_REQ(D_HA, req, "freeing request with old gen");
                        GOTO(free_req, 0);
                }

                if (req->rq_replay) {
                        DEBUG_REQ(D_HA, req, "keeping (FL_REPLAY)");
                        continue;
                }

                /* not yet committed */
                if (req->rq_transno > imp->imp_peer_committed_transno) {
                        DEBUG_REQ(D_HA, req, "stopping search");
                        break;
                }

                DEBUG_REQ(D_HA, req, "committing (last_committed "LPU64")",
                          imp->imp_peer_committed_transno);
free_req:
                list_del_init(&req->rq_list);
                __ptlrpc_req_finished(req, 1);
        }

        EXIT;
        return;
}

void ptlrpc_cleanup_client(struct obd_import *imp)
{
        ENTRY;
        EXIT;
        return;
}

void ptlrpc_resend_req(struct ptlrpc_request *req)
{
        unsigned long flags;

        DEBUG_REQ(D_HA, req, "resending");
        req->rq_reqmsg->handle.cookie = 0;
        ptlrpc_put_connection(req->rq_connection);
        req->rq_connection =
                ptlrpc_connection_addref(req->rq_import->imp_connection);
        req->rq_status = -EAGAIN;

        spin_lock_irqsave (&req->rq_lock, flags);
        req->rq_resend = 1;
        req->rq_timedout = 0;
        if (req->rq_set != NULL)
                wake_up (&req->rq_set->set_waitq);
        else
                wake_up(&req->rq_wait_for_rep);
        spin_unlock_irqrestore (&req->rq_lock, flags);
}

/* XXX: this function and rq_status are currently unused */
void ptlrpc_restart_req(struct ptlrpc_request *req)
{
        unsigned long flags;

        DEBUG_REQ(D_HA, req, "restarting (possibly-)completed request");
        req->rq_status = -ERESTARTSYS;

        spin_lock_irqsave (&req->rq_lock, flags);
        req->rq_restart = 1;
        req->rq_timedout = 0;
        if (req->rq_set != NULL)
                wake_up (&req->rq_set->set_waitq);
        else
                wake_up(&req->rq_wait_for_rep);
        spin_unlock_irqrestore (&req->rq_lock, flags);
}

static int expired_request(void *data)
{
        struct ptlrpc_request *req = data;
        ENTRY;

        RETURN(ptlrpc_expire_one_request(req));
}

static void interrupted_request(void *data)
{
        unsigned long flags;

        struct ptlrpc_request *req = data;
        DEBUG_REQ(D_HA, req, "request interrupted");
        spin_lock_irqsave (&req->rq_lock, flags);
        req->rq_intr = 1;
        spin_unlock_irqrestore (&req->rq_lock, flags);
}

struct ptlrpc_request *ptlrpc_request_addref(struct ptlrpc_request *req)
{
        ENTRY;
        atomic_inc(&req->rq_refcount);
        RETURN(req);
}

void ptlrpc_retain_replayable_request(struct ptlrpc_request *req,
                                      struct obd_import *imp)
{
        struct list_head *tmp;

#ifdef CONFIG_SMP
        LASSERT(spin_is_locked(&imp->imp_lock));
#endif

        LASSERT(imp->imp_replayable);
        /* Balanced in ptlrpc_free_committed, usually. */
        ptlrpc_request_addref(req);
        list_for_each_prev(tmp, &imp->imp_replay_list) {
                struct ptlrpc_request *iter =
                        list_entry(tmp, struct ptlrpc_request, rq_list);

                /* We may have duplicate transnos if we create and then
                 * open a file, or for closes retained if to match creating
                 * opens, so use req->rq_xid as a secondary key.
                 * (See bugs 684, 685, and 428.)
                 * XXX no longer needed, but all opens need transnos!
                 */
                if (iter->rq_transno > req->rq_transno)
                        continue;

                if (iter->rq_transno == req->rq_transno) {
                        LASSERT(iter->rq_xid != req->rq_xid);
                        if (iter->rq_xid > req->rq_xid)
                                continue;
                }

                list_add(&req->rq_list, &iter->rq_list);
                return;
        }

        list_add_tail(&req->rq_list, &imp->imp_replay_list);
}

int ptlrpc_queue_wait(struct ptlrpc_request *req)
{
        int rc = 0;
        int brc;
        struct l_wait_info lwi;
        struct obd_import *imp = req->rq_import;
        struct obd_device *obd = imp->imp_obd;
        unsigned long flags;
        int do_restart = 0;
        int timeout = 0;
        ENTRY;

        LASSERT(req->rq_set == NULL);
        LASSERT(!req->rq_receiving_reply);

        /* for distributed debugging */
        req->rq_reqmsg->status = current->pid;
        LASSERT(imp->imp_obd != NULL);
        CDEBUG(D_RPCTRACE, "Sending RPC pname:cluuid:pid:xid:ni:nid:opc "
               "%s:%s:%d:"LPU64":%s:"LPX64":%d\n", current->comm,
               imp->imp_obd->obd_uuid.uuid,
               req->rq_reqmsg->status, req->rq_xid,
               imp->imp_connection->c_peer.peer_ni->pni_name,
               imp->imp_connection->c_peer.peer_nid,
               req->rq_reqmsg->opc);

        /* Mark phase here for a little debug help */
        req->rq_phase = RQ_PHASE_RPC;

        spin_lock_irqsave(&imp->imp_lock, flags);
        req->rq_import_generation = imp->imp_generation;
restart:
        /*
         * If the import has been invalidated (such as by an OST failure), the
         * request must fail with -EIO.  Recovery requests are allowed to go
         * through, though, so that they have a chance to revalidate the
         * import.
         */
        if (req->rq_import->imp_invalid && req->rq_level == LUSTRE_CONN_FULL) {
                DEBUG_REQ(D_ERROR, req, "IMP_INVALID:");
                spin_unlock_irqrestore(&imp->imp_lock, flags);
                GOTO(out, rc = -EIO);
        }

        if (req->rq_import_generation < imp->imp_generation) {
                DEBUG_REQ(D_ERROR, req, "req old gen:");
                spin_unlock_irqrestore(&imp->imp_lock, flags);
                GOTO(out, rc = -EIO);
        }

        if (req->rq_level > imp->imp_level) {
                list_del(&req->rq_list);
                if (req->rq_no_recov || obd->obd_no_recov ||
                    imp->imp_dlm_fake) {
                        spin_unlock_irqrestore(&imp->imp_lock, flags);
                        GOTO(out, rc = -EWOULDBLOCK);
                }

                list_add_tail(&req->rq_list, &imp->imp_delayed_list);
                spin_unlock_irqrestore(&imp->imp_lock, flags);

                DEBUG_REQ(D_HA, req, "\"%s\" waiting for recovery: (%d > %d)",
                          current->comm, req->rq_level, imp->imp_level);
                lwi = LWI_INTR(NULL, NULL);
                rc = l_wait_event(req->rq_wait_for_rep,
                                  (req->rq_level <= imp->imp_level ||
                                   req->rq_err),
                                  &lwi);
                DEBUG_REQ(D_HA, req, "\"%s\" awake: (%d > %d or %d == 1)",
                          current->comm, imp->imp_level, req->rq_level,
                          req->rq_err);

                spin_lock_irqsave(&imp->imp_lock, flags);
                list_del_init(&req->rq_list);

                if (req->rq_err ||
                    req->rq_import_generation < imp->imp_generation)
                        rc = -EIO;

                if (rc) {
                        spin_unlock_irqrestore(&imp->imp_lock, flags);
                        GOTO(out, rc);
                }

                DEBUG_REQ(D_HA, req, "resumed");
        }

        /* XXX this is the same as ptlrpc_set_wait */
        LASSERT(list_empty(&req->rq_list));
        list_add_tail(&req->rq_list, &imp->imp_sending_list);
        spin_unlock_irqrestore(&imp->imp_lock, flags);

        rc = ptl_send_rpc(req);
        if (rc) {
                /* The DLM's fake imports want to avoid all forms of
                 * recovery. */
                if (imp->imp_dlm_fake) {
                        spin_lock_irqsave(&imp->imp_lock, flags);
                        list_del_init(&req->rq_list);
                        spin_unlock_irqrestore(&imp->imp_lock, flags);
                        GOTO(out, rc);
                }

                DEBUG_REQ(D_ERROR, req, "send failed (%d); recovering", rc);

                ptlrpc_fail_import(imp, req->rq_import_generation);

                /* If we've been told to not wait, we're done. */
                if (req->rq_level < LUSTRE_CONN_FULL || req->rq_no_recov ||
                    obd->obd_no_recov) {
                        spin_lock_irqsave(&imp->imp_lock, flags);
                        list_del_init(&req->rq_list);
                        spin_unlock_irqrestore(&imp->imp_lock, flags);
                        GOTO(out, rc);
                }

                /* If we errored, allow the user to interrupt immediately */
                timeout = 1;
        } else {
                timeout = req->rq_timeout * HZ;
                DEBUG_REQ(D_NET, req, "-- sleeping");
        }
#ifdef __KERNEL__
        lwi = LWI_TIMEOUT_INTR(timeout, expired_request, interrupted_request,
                               req);
        l_wait_event(req->rq_wait_for_rep, ptlrpc_check_reply(req), &lwi);
#else
        {
                extern int reply_in_callback(ptl_event_t *ev);
                ptl_event_t reply_ev;
                PtlEQWait(req->rq_connection->c_peer.peer_ni->pni_reply_in_eq_h,
                          &reply_ev);
                reply_in_callback(&reply_ev);

                LASSERT(reply_ev.mem_desc.user_ptr == (void *)req);
                // ptlrpc_check_reply(req);
                // not required now it only tests
        }
#endif

        DEBUG_REQ(D_NET, req, "-- done sleeping");

        CDEBUG(D_RPCTRACE, "Completed RPC pname:cluuid:pid:xid:ni:nid:opc "
               "%s:%s:%d:"LPU64":%s:"LPX64":%d\n", current->comm,
               imp->imp_obd->obd_uuid.uuid,
               req->rq_reqmsg->status, req->rq_xid,
               imp->imp_connection->c_peer.peer_ni->pni_name,
               imp->imp_connection->c_peer.peer_nid,
               req->rq_reqmsg->opc);

        spin_lock_irqsave(&imp->imp_lock, flags);
        list_del_init(&req->rq_list);
        spin_unlock_irqrestore(&imp->imp_lock, flags);

        /* If the reply was received normally, this just grabs the spinlock
         * (ensuring the reply callback has returned), sees that
         * req->rq_receiving_reply is clear and returns. */
        ptlrpc_unregister_reply (req);

        if (req->rq_err)
                GOTO(out, rc = -EIO);

        /* Resend if we need to, unless we were interrupted. */
        if (req->rq_resend && !req->rq_intr) {
                /* ...unless we were specifically told otherwise. */
                if (req->rq_no_resend) {
                        spin_lock_irqsave (&req->rq_lock, flags);
                        req->rq_no_resend = 0;
                        spin_unlock_irqrestore (&req->rq_lock, flags);
                        GOTO(out, rc = -ETIMEDOUT);
                }
                spin_lock_irqsave (&req->rq_lock, flags);
                req->rq_resend = 0;
                spin_unlock_irqrestore (&req->rq_lock, flags);
                lustre_msg_add_flags(req->rq_reqmsg, MSG_RESENT);

                if (req->rq_bulk != NULL)
                        ptlrpc_unregister_bulk (req);

                DEBUG_REQ(D_HA, req, "resending: ");
                spin_lock_irqsave(&imp->imp_lock, flags);
                goto restart;
        }

        if (req->rq_intr) {
                /* Should only be interrupted if we timed out. */
                if (!req->rq_timedout)
                        DEBUG_REQ(D_ERROR, req,
                                  "rq_intr set but rq_timedout not");
                GOTO(out, rc = -EINTR);
        }

        if (req->rq_timedout) {                 /* non-recoverable timeout */
                GOTO(out, rc = -ETIMEDOUT);
        }

        if (!req->rq_replied) {
                /* How can this be? -eeb */
                DEBUG_REQ(D_ERROR, req, "!rq_replied: ");
                LBUG();
                GOTO(out, rc = req->rq_status);
        }

        rc = after_reply (req, &do_restart);
        /* NB may return +ve success rc */
        if (do_restart) {
                if (req->rq_bulk != NULL)
                        ptlrpc_unregister_bulk (req);
                DEBUG_REQ(D_HA, req, "resending: ");
                spin_lock_irqsave(&imp->imp_lock, flags);
                goto restart;
        }

 out:
        if (req->rq_bulk != NULL) {
                if (rc >= 0) {                  /* success so far */
                        lwi = LWI_TIMEOUT(timeout, NULL, NULL);
                        brc = l_wait_event(req->rq_wait_for_rep,
                                           ptlrpc_bulk_complete(req->rq_bulk),
                                           &lwi);
                        if (brc != 0) {
                                LASSERT(brc == -ETIMEDOUT);
                                CERROR ("Timed out waiting for bulk\n");
                                rc = brc;
                        }
                }
                if (rc < 0) {
                        /* MDS blocks for put ACKs before replying */
                        /* OSC sets rq_no_resend for the time being */
                        LASSERT(req->rq_no_resend);
                        ptlrpc_unregister_bulk (req);
                }
        }

        LASSERT(!req->rq_receiving_reply);
        req->rq_phase = RQ_PHASE_INTERPRET;
        RETURN(rc);
}

int ptlrpc_replay_req(struct ptlrpc_request *req)
{
        int rc = 0, old_level, old_status = 0;
        // struct ptlrpc_client *cli = req->rq_import->imp_client;
        struct l_wait_info lwi;
        ENTRY;

        /* I don't touch rq_phase here, so the debug log can show what
         * state it was left in */

        /* Not handling automatic bulk replay yet (or ever?) */
        LASSERT(req->rq_bulk == NULL);

        DEBUG_REQ(D_NET, req, "about to replay");

        /* Update request's state, since we might have a new connection. */
        ptlrpc_put_connection(req->rq_connection);
        req->rq_connection =
                ptlrpc_connection_addref(req->rq_import->imp_connection);

        /* temporarily set request to RECOVD level (reset at out:) */
        old_level = req->rq_level;
        if (req->rq_replied)
                old_status = req->rq_repmsg->status;
        req->rq_level = LUSTRE_CONN_RECOVER;
        rc = ptl_send_rpc(req);
        if (rc) {
                CERROR("error %d, opcode %d\n", rc, req->rq_reqmsg->opc);
                ptlrpc_cleanup_request_buf(req);
                // up(&cli->cli_rpc_sem);
                GOTO(out, rc = -rc);
        }

        CDEBUG(D_OTHER, "-- sleeping\n");
        lwi = LWI_INTR(NULL, NULL); /* XXX needs timeout, nested recovery */
        l_wait_event(req->rq_wait_for_rep, ptlrpc_check_reply(req), &lwi);
        CDEBUG(D_OTHER, "-- done\n");

        // up(&cli->cli_rpc_sem);

        /* If the reply was received normally, this just grabs the spinlock
         * (ensuring the reply callback has returned), sees that
         * req->rq_receiving_reply is clear and returns. */
        ptlrpc_unregister_reply (req);

        if (!req->rq_replied) {
                CERROR("Unknown reason for wakeup\n");
                /* XXX Phil - I end up here when I kill obdctl */
                /* ...that's because signals aren't all masked in
                 * l_wait_event() -eeb */
                GOTO(out, rc = -EINTR);
        }

#if SWAB_PARANOIA
        /* Clear reply swab mask; this is a new reply in sender's byte order */
        req->rq_rep_swab_mask = 0;
#endif
        rc = lustre_unpack_msg(req->rq_repmsg, req->rq_replen);
        if (rc) {
                CERROR("unpack_rep failed: %d\n", rc);
                GOTO(out, rc = -EPROTO);
        }
#if 0
        /* FIXME: Enable when BlueArc makes new release */
        if (req->rq_repmsg->type != PTL_RPC_MSG_REPLY &&
            req->rq_repmsg->type != PTL_RPC_MSG_ERR) {
                CERROR("invalid packet type received (type=%u)\n",
                       req->rq_repmsg->type);
                GOTO(out, rc = -EPROTO);
        }
#endif

        /* The transno had better not change over replay. */
        LASSERT(req->rq_reqmsg->transno == req->rq_repmsg->transno);

        CDEBUG(D_NET, "got rep "LPD64"\n", req->rq_xid);

        /* let the callback do fixups, possibly including in the request */
        if (req->rq_replay_cb)
                req->rq_replay_cb(req);

        if (req->rq_replied && req->rq_repmsg->status != old_status) {
                DEBUG_REQ(D_HA, req, "status %d, old was %d",
                          req->rq_repmsg->status, old_status);
        }

 out:
        req->rq_level = old_level;
        RETURN(rc);
}

void ptlrpc_abort_inflight(struct obd_import *imp)
{
        unsigned long flags;
        struct list_head *tmp, *n;
        ENTRY;

        /* Make sure that no new requests get processed for this import.
         * ptlrpc_{queue,set}_wait must (and does) hold imp_lock while testing
         * this flag and then putting requests on sending_list or delayed_list.
         */
        spin_lock_irqsave(&imp->imp_lock, flags);

        /* XXX locking?  Maybe we should remove each request with the list
         * locked?  Also, how do we know if the requests on the list are
         * being freed at this time?
         */
        list_for_each_safe(tmp, n, &imp->imp_sending_list) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_list);

                DEBUG_REQ(D_HA, req, "inflight");

                spin_lock (&req->rq_lock);
                if (req->rq_import_generation < imp->imp_generation) {
                        req->rq_err = 1;
                        if (req->rq_set != NULL)
                                wake_up(&req->rq_set->set_waitq);
                        else
                                wake_up(&req->rq_wait_for_rep);
                }
                spin_unlock (&req->rq_lock);
        }

        list_for_each_safe(tmp, n, &imp->imp_delayed_list) {
                struct ptlrpc_request *req =
                        list_entry(tmp, struct ptlrpc_request, rq_list);

                DEBUG_REQ(D_HA, req, "aborting waiting req");

                spin_lock (&req->rq_lock);
                if (req->rq_import_generation < imp->imp_generation) {
                        req->rq_err = 1;
                        if (req->rq_set != NULL)
                                wake_up(&req->rq_set->set_waitq);
                        else
                                wake_up(&req->rq_wait_for_rep);
                }
                spin_unlock (&req->rq_lock);
        }

        /* Last chance to free reqs left on the replay list, but we
         * will still leak reqs that haven't comitted.  */
        if (imp->imp_replayable)
                ptlrpc_free_committed(imp);

        spin_unlock_irqrestore(&imp->imp_lock, flags);

        EXIT;
}

static __u64 ptlrpc_last_xid = 0;
static spinlock_t ptlrpc_last_xid_lock = SPIN_LOCK_UNLOCKED;

__u64 ptlrpc_next_xid(void)
{
        __u64 tmp;
        spin_lock(&ptlrpc_last_xid_lock);
        tmp = ++ptlrpc_last_xid;
        spin_unlock(&ptlrpc_last_xid_lock);
        return tmp;
}