/* -*- mode: c; c-basic-offset: 8; -*- * vim: noexpandtab sw=8 ts=8 sts=0: * * dlmmod.c * * standalone DLM module * * Copyright (C) 2004 Oracle. All rights reserved. * * This program is free software; you can redistribute it and/or * modify it under the terms of the GNU General Public * License as published by the Free Software Foundation; either * version 2 of the License, or (at your option) any later version. * * This program 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 this program; if not, write to the * Free Software Foundation, Inc., 59 Temple Place - Suite 330, * Boston, MA 021110-1307, USA. * */ #include #include #include #include #include #include #include #include #include #include #include #include #include #include #include "cluster/heartbeat.h" #include "cluster/nodemanager.h" #include "cluster/tcp.h" #include "dlmapi.h" #include "dlmcommon.h" #include "dlmdomain.h" #define MLOG_MASK_PREFIX (ML_DLM|ML_DLM_MASTER) #include "cluster/masklog.h" enum dlm_mle_type { DLM_MLE_BLOCK, DLM_MLE_MASTER, DLM_MLE_MIGRATION }; struct dlm_lock_name { u8 len; u8 name[DLM_LOCKID_NAME_MAX]; }; struct dlm_master_list_entry { struct list_head list; struct list_head hb_events; struct dlm_ctxt *dlm; spinlock_t spinlock; wait_queue_head_t wq; atomic_t woken; struct kref mle_refs; int inuse; unsigned long maybe_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long vote_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long response_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; unsigned long node_map[BITS_TO_LONGS(O2NM_MAX_NODES)]; u8 master; u8 new_master; enum dlm_mle_type type; struct o2hb_callback_func mle_hb_up; struct o2hb_callback_func mle_hb_down; union { struct dlm_lock_resource *res; struct dlm_lock_name name; } u; }; static void dlm_mle_node_down(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle, struct o2nm_node *node, int idx); static void dlm_mle_node_up(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle, struct o2nm_node *node, int idx); static void dlm_assert_master_worker(struct dlm_work_item *item, void *data); static int dlm_do_assert_master(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, void *nodemap, u32 flags); static inline int dlm_mle_equal(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle, const char *name, unsigned int namelen) { struct dlm_lock_resource *res; if (dlm != mle->dlm) return 0; if (mle->type == DLM_MLE_BLOCK || mle->type == DLM_MLE_MIGRATION) { if (namelen != mle->u.name.len || memcmp(name, mle->u.name.name, namelen)!=0) return 0; } else { res = mle->u.res; if (namelen != res->lockname.len || memcmp(res->lockname.name, name, namelen) != 0) return 0; } return 1; } #define dlm_print_nodemap(m) _dlm_print_nodemap(m,#m) static void _dlm_print_nodemap(unsigned long *map, const char *mapname) { int i; printk("%s=[ ", mapname); for (i=0; imaybe_map, *vote = mle->vote_map, *resp = mle->response_map, *node = mle->node_map; k = &mle->mle_refs; if (mle->type == DLM_MLE_BLOCK) type = "BLK"; else if (mle->type == DLM_MLE_MASTER) type = "MAS"; else type = "MIG"; refs = atomic_read(&k->refcount); master = mle->master; attached = (list_empty(&mle->hb_events) ? 'N' : 'Y'); if (mle->type != DLM_MLE_MASTER) { namelen = mle->u.name.len; name = mle->u.name.name; } else { namelen = mle->u.res->lockname.len; name = mle->u.res->lockname.name; } mlog(ML_NOTICE, "%.*s: %3s refs=%3d mas=%3u new=%3u evt=%c inuse=%d ", namelen, name, type, refs, master, mle->new_master, attached, mle->inuse); dlm_print_nodemap(maybe); printk(", "); dlm_print_nodemap(vote); printk(", "); dlm_print_nodemap(resp); printk(", "); dlm_print_nodemap(node); printk(", "); printk("\n"); } #if 0 /* Code here is included but defined out as it aids debugging */ static void dlm_dump_mles(struct dlm_ctxt *dlm) { struct dlm_master_list_entry *mle; struct list_head *iter; mlog(ML_NOTICE, "dumping all mles for domain %s:\n", dlm->name); spin_lock(&dlm->master_lock); list_for_each(iter, &dlm->master_list) { mle = list_entry(iter, struct dlm_master_list_entry, list); dlm_print_one_mle(mle); } spin_unlock(&dlm->master_lock); } int dlm_dump_all_mles(const char __user *data, unsigned int len) { struct list_head *iter; struct dlm_ctxt *dlm; spin_lock(&dlm_domain_lock); list_for_each(iter, &dlm_domains) { dlm = list_entry (iter, struct dlm_ctxt, list); mlog(ML_NOTICE, "found dlm: %p, name=%s\n", dlm, dlm->name); dlm_dump_mles(dlm); } spin_unlock(&dlm_domain_lock); return len; } EXPORT_SYMBOL_GPL(dlm_dump_all_mles); #endif /* 0 */ static struct kmem_cache *dlm_mle_cache = NULL; static void dlm_mle_release(struct kref *kref); static void dlm_init_mle(struct dlm_master_list_entry *mle, enum dlm_mle_type type, struct dlm_ctxt *dlm, struct dlm_lock_resource *res, const char *name, unsigned int namelen); static void dlm_put_mle(struct dlm_master_list_entry *mle); static void __dlm_put_mle(struct dlm_master_list_entry *mle); static int dlm_find_mle(struct dlm_ctxt *dlm, struct dlm_master_list_entry **mle, char *name, unsigned int namelen); static int dlm_do_master_request(struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, int to); static int dlm_wait_for_lock_mastery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, int *blocked); static int dlm_restart_lock_mastery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, int blocked); static int dlm_add_migration_mle(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, struct dlm_master_list_entry **oldmle, const char *name, unsigned int namelen, u8 new_master, u8 master); static u8 dlm_pick_migration_target(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); static int dlm_mark_lockres_migrating(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 target); static int dlm_pre_master_reco_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res); int dlm_is_host_down(int errno) { switch (errno) { case -EBADF: case -ECONNREFUSED: case -ENOTCONN: case -ECONNRESET: case -EPIPE: case -EHOSTDOWN: case -EHOSTUNREACH: case -ETIMEDOUT: case -ECONNABORTED: case -ENETDOWN: case -ENETUNREACH: case -ENETRESET: case -ESHUTDOWN: case -ENOPROTOOPT: case -EINVAL: /* if returned from our tcp code, this means there is no socket */ return 1; } return 0; } /* * MASTER LIST FUNCTIONS */ /* * regarding master list entries and heartbeat callbacks: * * in order to avoid sleeping and allocation that occurs in * heartbeat, master list entries are simply attached to the * dlm's established heartbeat callbacks. the mle is attached * when it is created, and since the dlm->spinlock is held at * that time, any heartbeat event will be properly discovered * by the mle. the mle needs to be detached from the * dlm->mle_hb_events list as soon as heartbeat events are no * longer useful to the mle, and before the mle is freed. * * as a general rule, heartbeat events are no longer needed by * the mle once an "answer" regarding the lock master has been * received. */ static inline void __dlm_mle_attach_hb_events(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle) { assert_spin_locked(&dlm->spinlock); list_add_tail(&mle->hb_events, &dlm->mle_hb_events); } static inline void __dlm_mle_detach_hb_events(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle) { if (!list_empty(&mle->hb_events)) list_del_init(&mle->hb_events); } static inline void dlm_mle_detach_hb_events(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle) { spin_lock(&dlm->spinlock); __dlm_mle_detach_hb_events(dlm, mle); spin_unlock(&dlm->spinlock); } static void dlm_get_mle_inuse(struct dlm_master_list_entry *mle) { struct dlm_ctxt *dlm; dlm = mle->dlm; assert_spin_locked(&dlm->spinlock); assert_spin_locked(&dlm->master_lock); mle->inuse++; kref_get(&mle->mle_refs); } static void dlm_put_mle_inuse(struct dlm_master_list_entry *mle) { struct dlm_ctxt *dlm; dlm = mle->dlm; spin_lock(&dlm->spinlock); spin_lock(&dlm->master_lock); mle->inuse--; __dlm_put_mle(mle); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); } /* remove from list and free */ static void __dlm_put_mle(struct dlm_master_list_entry *mle) { struct dlm_ctxt *dlm; dlm = mle->dlm; assert_spin_locked(&dlm->spinlock); assert_spin_locked(&dlm->master_lock); if (!atomic_read(&mle->mle_refs.refcount)) { /* this may or may not crash, but who cares. * it's a BUG. */ mlog(ML_ERROR, "bad mle: %p\n", mle); dlm_print_one_mle(mle); BUG(); } else kref_put(&mle->mle_refs, dlm_mle_release); } /* must not have any spinlocks coming in */ static void dlm_put_mle(struct dlm_master_list_entry *mle) { struct dlm_ctxt *dlm; dlm = mle->dlm; spin_lock(&dlm->spinlock); spin_lock(&dlm->master_lock); __dlm_put_mle(mle); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); } static inline void dlm_get_mle(struct dlm_master_list_entry *mle) { kref_get(&mle->mle_refs); } static void dlm_init_mle(struct dlm_master_list_entry *mle, enum dlm_mle_type type, struct dlm_ctxt *dlm, struct dlm_lock_resource *res, const char *name, unsigned int namelen) { assert_spin_locked(&dlm->spinlock); mle->dlm = dlm; mle->type = type; INIT_LIST_HEAD(&mle->list); INIT_LIST_HEAD(&mle->hb_events); memset(mle->maybe_map, 0, sizeof(mle->maybe_map)); spin_lock_init(&mle->spinlock); init_waitqueue_head(&mle->wq); atomic_set(&mle->woken, 0); kref_init(&mle->mle_refs); memset(mle->response_map, 0, sizeof(mle->response_map)); mle->master = O2NM_MAX_NODES; mle->new_master = O2NM_MAX_NODES; mle->inuse = 0; if (mle->type == DLM_MLE_MASTER) { BUG_ON(!res); mle->u.res = res; } else if (mle->type == DLM_MLE_BLOCK) { BUG_ON(!name); memcpy(mle->u.name.name, name, namelen); mle->u.name.len = namelen; } else /* DLM_MLE_MIGRATION */ { BUG_ON(!name); memcpy(mle->u.name.name, name, namelen); mle->u.name.len = namelen; } /* copy off the node_map and register hb callbacks on our copy */ memcpy(mle->node_map, dlm->domain_map, sizeof(mle->node_map)); memcpy(mle->vote_map, dlm->domain_map, sizeof(mle->vote_map)); clear_bit(dlm->node_num, mle->vote_map); clear_bit(dlm->node_num, mle->node_map); /* attach the mle to the domain node up/down events */ __dlm_mle_attach_hb_events(dlm, mle); } /* returns 1 if found, 0 if not */ static int dlm_find_mle(struct dlm_ctxt *dlm, struct dlm_master_list_entry **mle, char *name, unsigned int namelen) { struct dlm_master_list_entry *tmpmle; struct list_head *iter; assert_spin_locked(&dlm->master_lock); list_for_each(iter, &dlm->master_list) { tmpmle = list_entry(iter, struct dlm_master_list_entry, list); if (!dlm_mle_equal(dlm, tmpmle, name, namelen)) continue; dlm_get_mle(tmpmle); *mle = tmpmle; return 1; } return 0; } void dlm_hb_event_notify_attached(struct dlm_ctxt *dlm, int idx, int node_up) { struct dlm_master_list_entry *mle; struct list_head *iter; assert_spin_locked(&dlm->spinlock); list_for_each(iter, &dlm->mle_hb_events) { mle = list_entry(iter, struct dlm_master_list_entry, hb_events); if (node_up) dlm_mle_node_up(dlm, mle, NULL, idx); else dlm_mle_node_down(dlm, mle, NULL, idx); } } static void dlm_mle_node_down(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle, struct o2nm_node *node, int idx) { spin_lock(&mle->spinlock); if (!test_bit(idx, mle->node_map)) mlog(0, "node %u already removed from nodemap!\n", idx); else clear_bit(idx, mle->node_map); spin_unlock(&mle->spinlock); } static void dlm_mle_node_up(struct dlm_ctxt *dlm, struct dlm_master_list_entry *mle, struct o2nm_node *node, int idx) { spin_lock(&mle->spinlock); if (test_bit(idx, mle->node_map)) mlog(0, "node %u already in node map!\n", idx); else set_bit(idx, mle->node_map); spin_unlock(&mle->spinlock); } int dlm_init_mle_cache(void) { dlm_mle_cache = kmem_cache_create("dlm_mle_cache", sizeof(struct dlm_master_list_entry), 0, SLAB_HWCACHE_ALIGN, NULL, NULL); if (dlm_mle_cache == NULL) return -ENOMEM; return 0; } void dlm_destroy_mle_cache(void) { if (dlm_mle_cache) kmem_cache_destroy(dlm_mle_cache); } static void dlm_mle_release(struct kref *kref) { struct dlm_master_list_entry *mle; struct dlm_ctxt *dlm; mlog_entry_void(); mle = container_of(kref, struct dlm_master_list_entry, mle_refs); dlm = mle->dlm; if (mle->type != DLM_MLE_MASTER) { mlog(0, "calling mle_release for %.*s, type %d\n", mle->u.name.len, mle->u.name.name, mle->type); } else { mlog(0, "calling mle_release for %.*s, type %d\n", mle->u.res->lockname.len, mle->u.res->lockname.name, mle->type); } assert_spin_locked(&dlm->spinlock); assert_spin_locked(&dlm->master_lock); /* remove from list if not already */ if (!list_empty(&mle->list)) list_del_init(&mle->list); /* detach the mle from the domain node up/down events */ __dlm_mle_detach_hb_events(dlm, mle); /* NOTE: kfree under spinlock here. * if this is bad, we can move this to a freelist. */ kmem_cache_free(dlm_mle_cache, mle); } /* * LOCK RESOURCE FUNCTIONS */ static void dlm_set_lockres_owner(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 owner) { assert_spin_locked(&res->spinlock); mlog_entry("%.*s, %u\n", res->lockname.len, res->lockname.name, owner); if (owner == dlm->node_num) atomic_inc(&dlm->local_resources); else if (owner == DLM_LOCK_RES_OWNER_UNKNOWN) atomic_inc(&dlm->unknown_resources); else atomic_inc(&dlm->remote_resources); res->owner = owner; } void dlm_change_lockres_owner(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 owner) { assert_spin_locked(&res->spinlock); if (owner == res->owner) return; if (res->owner == dlm->node_num) atomic_dec(&dlm->local_resources); else if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN) atomic_dec(&dlm->unknown_resources); else atomic_dec(&dlm->remote_resources); dlm_set_lockres_owner(dlm, res, owner); } static void dlm_lockres_release(struct kref *kref) { struct dlm_lock_resource *res; res = container_of(kref, struct dlm_lock_resource, refs); /* This should not happen -- all lockres' have a name * associated with them at init time. */ BUG_ON(!res->lockname.name); mlog(0, "destroying lockres %.*s\n", res->lockname.len, res->lockname.name); if (!hlist_unhashed(&res->hash_node) || !list_empty(&res->granted) || !list_empty(&res->converting) || !list_empty(&res->blocked) || !list_empty(&res->dirty) || !list_empty(&res->recovering) || !list_empty(&res->purge)) { mlog(ML_ERROR, "Going to BUG for resource %.*s." " We're on a list! [%c%c%c%c%c%c%c]\n", res->lockname.len, res->lockname.name, !hlist_unhashed(&res->hash_node) ? 'H' : ' ', !list_empty(&res->granted) ? 'G' : ' ', !list_empty(&res->converting) ? 'C' : ' ', !list_empty(&res->blocked) ? 'B' : ' ', !list_empty(&res->dirty) ? 'D' : ' ', !list_empty(&res->recovering) ? 'R' : ' ', !list_empty(&res->purge) ? 'P' : ' '); dlm_print_one_lock_resource(res); } /* By the time we're ready to blow this guy away, we shouldn't * be on any lists. */ BUG_ON(!hlist_unhashed(&res->hash_node)); BUG_ON(!list_empty(&res->granted)); BUG_ON(!list_empty(&res->converting)); BUG_ON(!list_empty(&res->blocked)); BUG_ON(!list_empty(&res->dirty)); BUG_ON(!list_empty(&res->recovering)); BUG_ON(!list_empty(&res->purge)); kfree(res->lockname.name); kfree(res); } void dlm_lockres_put(struct dlm_lock_resource *res) { kref_put(&res->refs, dlm_lockres_release); } static void dlm_init_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, const char *name, unsigned int namelen) { char *qname; /* If we memset here, we lose our reference to the kmalloc'd * res->lockname.name, so be sure to init every field * correctly! */ qname = (char *) res->lockname.name; memcpy(qname, name, namelen); res->lockname.len = namelen; res->lockname.hash = dlm_lockid_hash(name, namelen); init_waitqueue_head(&res->wq); spin_lock_init(&res->spinlock); INIT_HLIST_NODE(&res->hash_node); INIT_LIST_HEAD(&res->granted); INIT_LIST_HEAD(&res->converting); INIT_LIST_HEAD(&res->blocked); INIT_LIST_HEAD(&res->dirty); INIT_LIST_HEAD(&res->recovering); INIT_LIST_HEAD(&res->purge); atomic_set(&res->asts_reserved, 0); res->migration_pending = 0; res->inflight_locks = 0; kref_init(&res->refs); /* just for consistency */ spin_lock(&res->spinlock); dlm_set_lockres_owner(dlm, res, DLM_LOCK_RES_OWNER_UNKNOWN); spin_unlock(&res->spinlock); res->state = DLM_LOCK_RES_IN_PROGRESS; res->last_used = 0; memset(res->lvb, 0, DLM_LVB_LEN); memset(res->refmap, 0, sizeof(res->refmap)); } struct dlm_lock_resource *dlm_new_lockres(struct dlm_ctxt *dlm, const char *name, unsigned int namelen) { struct dlm_lock_resource *res; res = kmalloc(sizeof(struct dlm_lock_resource), GFP_NOFS); if (!res) return NULL; res->lockname.name = kmalloc(namelen, GFP_NOFS); if (!res->lockname.name) { kfree(res); return NULL; } dlm_init_lockres(dlm, res, name, namelen); return res; } void __dlm_lockres_grab_inflight_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, int new_lockres, const char *file, int line) { if (!new_lockres) assert_spin_locked(&res->spinlock); if (!test_bit(dlm->node_num, res->refmap)) { BUG_ON(res->inflight_locks != 0); dlm_lockres_set_refmap_bit(dlm->node_num, res); } res->inflight_locks++; mlog(0, "%s:%.*s: inflight++: now %u\n", dlm->name, res->lockname.len, res->lockname.name, res->inflight_locks); } void __dlm_lockres_drop_inflight_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, const char *file, int line) { assert_spin_locked(&res->spinlock); BUG_ON(res->inflight_locks == 0); res->inflight_locks--; mlog(0, "%s:%.*s: inflight--: now %u\n", dlm->name, res->lockname.len, res->lockname.name, res->inflight_locks); if (res->inflight_locks == 0) dlm_lockres_clear_refmap_bit(dlm->node_num, res); wake_up(&res->wq); } /* * lookup a lock resource by name. * may already exist in the hashtable. * lockid is null terminated * * if not, allocate enough for the lockres and for * the temporary structure used in doing the mastering. * * also, do a lookup in the dlm->master_list to see * if another node has begun mastering the same lock. * if so, there should be a block entry in there * for this name, and we should *not* attempt to master * the lock here. need to wait around for that node * to assert_master (or die). * */ struct dlm_lock_resource * dlm_get_lock_resource(struct dlm_ctxt *dlm, const char *lockid, int namelen, int flags) { struct dlm_lock_resource *tmpres=NULL, *res=NULL; struct dlm_master_list_entry *mle = NULL; struct dlm_master_list_entry *alloc_mle = NULL; int blocked = 0; int ret, nodenum; struct dlm_node_iter iter; unsigned int hash; int tries = 0; int bit, wait_on_recovery = 0; int drop_inflight_if_nonlocal = 0; BUG_ON(!lockid); hash = dlm_lockid_hash(lockid, namelen); mlog(0, "get lockres %s (len %d)\n", lockid, namelen); lookup: spin_lock(&dlm->spinlock); tmpres = __dlm_lookup_lockres_full(dlm, lockid, namelen, hash); if (tmpres) { int dropping_ref = 0; spin_lock(&tmpres->spinlock); if (tmpres->owner == dlm->node_num) { BUG_ON(tmpres->state & DLM_LOCK_RES_DROPPING_REF); dlm_lockres_grab_inflight_ref(dlm, tmpres); } else if (tmpres->state & DLM_LOCK_RES_DROPPING_REF) dropping_ref = 1; spin_unlock(&tmpres->spinlock); spin_unlock(&dlm->spinlock); /* wait until done messaging the master, drop our ref to allow * the lockres to be purged, start over. */ if (dropping_ref) { spin_lock(&tmpres->spinlock); __dlm_wait_on_lockres_flags(tmpres, DLM_LOCK_RES_DROPPING_REF); spin_unlock(&tmpres->spinlock); dlm_lockres_put(tmpres); tmpres = NULL; goto lookup; } mlog(0, "found in hash!\n"); if (res) dlm_lockres_put(res); res = tmpres; goto leave; } if (!res) { spin_unlock(&dlm->spinlock); mlog(0, "allocating a new resource\n"); /* nothing found and we need to allocate one. */ alloc_mle = (struct dlm_master_list_entry *) kmem_cache_alloc(dlm_mle_cache, GFP_NOFS); if (!alloc_mle) goto leave; res = dlm_new_lockres(dlm, lockid, namelen); if (!res) goto leave; goto lookup; } mlog(0, "no lockres found, allocated our own: %p\n", res); if (flags & LKM_LOCAL) { /* caller knows it's safe to assume it's not mastered elsewhere * DONE! return right away */ spin_lock(&res->spinlock); dlm_change_lockres_owner(dlm, res, dlm->node_num); __dlm_insert_lockres(dlm, res); dlm_lockres_grab_inflight_ref(dlm, res); spin_unlock(&res->spinlock); spin_unlock(&dlm->spinlock); /* lockres still marked IN_PROGRESS */ goto wake_waiters; } /* check master list to see if another node has started mastering it */ spin_lock(&dlm->master_lock); /* if we found a block, wait for lock to be mastered by another node */ blocked = dlm_find_mle(dlm, &mle, (char *)lockid, namelen); if (blocked) { int mig; if (mle->type == DLM_MLE_MASTER) { mlog(ML_ERROR, "master entry for nonexistent lock!\n"); BUG(); } mig = (mle->type == DLM_MLE_MIGRATION); /* if there is a migration in progress, let the migration * finish before continuing. we can wait for the absence * of the MIGRATION mle: either the migrate finished or * one of the nodes died and the mle was cleaned up. * if there is a BLOCK here, but it already has a master * set, we are too late. the master does not have a ref * for us in the refmap. detach the mle and drop it. * either way, go back to the top and start over. */ if (mig || mle->master != O2NM_MAX_NODES) { BUG_ON(mig && mle->master == dlm->node_num); /* we arrived too late. the master does not * have a ref for us. retry. */ mlog(0, "%s:%.*s: late on %s\n", dlm->name, namelen, lockid, mig ? "MIGRATION" : "BLOCK"); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); /* master is known, detach */ if (!mig) dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle(mle); mle = NULL; /* this is lame, but we cant wait on either * the mle or lockres waitqueue here */ if (mig) msleep(100); goto lookup; } } else { /* go ahead and try to master lock on this node */ mle = alloc_mle; /* make sure this does not get freed below */ alloc_mle = NULL; dlm_init_mle(mle, DLM_MLE_MASTER, dlm, res, NULL, 0); set_bit(dlm->node_num, mle->maybe_map); list_add(&mle->list, &dlm->master_list); /* still holding the dlm spinlock, check the recovery map * to see if there are any nodes that still need to be * considered. these will not appear in the mle nodemap * but they might own this lockres. wait on them. */ bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) { mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to" "recover before lock mastery can begin\n", dlm->name, namelen, (char *)lockid, bit); wait_on_recovery = 1; } } /* at this point there is either a DLM_MLE_BLOCK or a * DLM_MLE_MASTER on the master list, so it's safe to add the * lockres to the hashtable. anyone who finds the lock will * still have to wait on the IN_PROGRESS. */ /* finally add the lockres to its hash bucket */ __dlm_insert_lockres(dlm, res); /* since this lockres is new it doesnt not require the spinlock */ dlm_lockres_grab_inflight_ref_new(dlm, res); /* if this node does not become the master make sure to drop * this inflight reference below */ drop_inflight_if_nonlocal = 1; /* get an extra ref on the mle in case this is a BLOCK * if so, the creator of the BLOCK may try to put the last * ref at this time in the assert master handler, so we * need an extra one to keep from a bad ptr deref. */ dlm_get_mle_inuse(mle); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); redo_request: while (wait_on_recovery) { /* any cluster changes that occurred after dropping the * dlm spinlock would be detectable be a change on the mle, * so we only need to clear out the recovery map once. */ if (dlm_is_recovery_lock(lockid, namelen)) { mlog(ML_NOTICE, "%s: recovery map is not empty, but " "must master $RECOVERY lock now\n", dlm->name); if (!dlm_pre_master_reco_lockres(dlm, res)) wait_on_recovery = 0; else { mlog(0, "%s: waiting 500ms for heartbeat state " "change\n", dlm->name); msleep(500); } continue; } dlm_kick_recovery_thread(dlm); msleep(1000); dlm_wait_for_recovery(dlm); spin_lock(&dlm->spinlock); bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); if (bit < O2NM_MAX_NODES) { mlog(ML_NOTICE, "%s:%.*s: at least one node (%d) to" "recover before lock mastery can begin\n", dlm->name, namelen, (char *)lockid, bit); wait_on_recovery = 1; } else wait_on_recovery = 0; spin_unlock(&dlm->spinlock); if (wait_on_recovery) dlm_wait_for_node_recovery(dlm, bit, 10000); } /* must wait for lock to be mastered elsewhere */ if (blocked) goto wait; ret = -EINVAL; dlm_node_iter_init(mle->vote_map, &iter); while ((nodenum = dlm_node_iter_next(&iter)) >= 0) { ret = dlm_do_master_request(res, mle, nodenum); if (ret < 0) mlog_errno(ret); if (mle->master != O2NM_MAX_NODES) { /* found a master ! */ if (mle->master <= nodenum) break; /* if our master request has not reached the master * yet, keep going until it does. this is how the * master will know that asserts are needed back to * the lower nodes. */ mlog(0, "%s:%.*s: requests only up to %u but master " "is %u, keep going\n", dlm->name, namelen, lockid, nodenum, mle->master); } } wait: /* keep going until the response map includes all nodes */ ret = dlm_wait_for_lock_mastery(dlm, res, mle, &blocked); if (ret < 0) { wait_on_recovery = 1; mlog(0, "%s:%.*s: node map changed, redo the " "master request now, blocked=%d\n", dlm->name, res->lockname.len, res->lockname.name, blocked); if (++tries > 20) { mlog(ML_ERROR, "%s:%.*s: spinning on " "dlm_wait_for_lock_mastery, blocked=%d\n", dlm->name, res->lockname.len, res->lockname.name, blocked); dlm_print_one_lock_resource(res); dlm_print_one_mle(mle); tries = 0; } goto redo_request; } mlog(0, "lockres mastered by %u\n", res->owner); /* make sure we never continue without this */ BUG_ON(res->owner == O2NM_MAX_NODES); /* master is known, detach if not already detached */ dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle(mle); /* put the extra ref */ dlm_put_mle_inuse(mle); wake_waiters: spin_lock(&res->spinlock); if (res->owner != dlm->node_num && drop_inflight_if_nonlocal) dlm_lockres_drop_inflight_ref(dlm, res); res->state &= ~DLM_LOCK_RES_IN_PROGRESS; spin_unlock(&res->spinlock); wake_up(&res->wq); leave: /* need to free the unused mle */ if (alloc_mle) kmem_cache_free(dlm_mle_cache, alloc_mle); return res; } #define DLM_MASTERY_TIMEOUT_MS 5000 static int dlm_wait_for_lock_mastery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, int *blocked) { u8 m; int ret, bit; int map_changed, voting_done; int assert, sleep; recheck: ret = 0; assert = 0; /* check if another node has already become the owner */ spin_lock(&res->spinlock); if (res->owner != DLM_LOCK_RES_OWNER_UNKNOWN) { mlog(0, "%s:%.*s: owner is suddenly %u\n", dlm->name, res->lockname.len, res->lockname.name, res->owner); spin_unlock(&res->spinlock); /* this will cause the master to re-assert across * the whole cluster, freeing up mles */ if (res->owner != dlm->node_num) { ret = dlm_do_master_request(res, mle, res->owner); if (ret < 0) { /* give recovery a chance to run */ mlog(ML_ERROR, "link to %u went down?: %d\n", res->owner, ret); msleep(500); goto recheck; } } ret = 0; goto leave; } spin_unlock(&res->spinlock); spin_lock(&mle->spinlock); m = mle->master; map_changed = (memcmp(mle->vote_map, mle->node_map, sizeof(mle->vote_map)) != 0); voting_done = (memcmp(mle->vote_map, mle->response_map, sizeof(mle->vote_map)) == 0); /* restart if we hit any errors */ if (map_changed) { int b; mlog(0, "%s: %.*s: node map changed, restarting\n", dlm->name, res->lockname.len, res->lockname.name); ret = dlm_restart_lock_mastery(dlm, res, mle, *blocked); b = (mle->type == DLM_MLE_BLOCK); if ((*blocked && !b) || (!*blocked && b)) { mlog(0, "%s:%.*s: status change: old=%d new=%d\n", dlm->name, res->lockname.len, res->lockname.name, *blocked, b); *blocked = b; } spin_unlock(&mle->spinlock); if (ret < 0) { mlog_errno(ret); goto leave; } mlog(0, "%s:%.*s: restart lock mastery succeeded, " "rechecking now\n", dlm->name, res->lockname.len, res->lockname.name); goto recheck; } else { if (!voting_done) { mlog(0, "map not changed and voting not done " "for %s:%.*s\n", dlm->name, res->lockname.len, res->lockname.name); } } if (m != O2NM_MAX_NODES) { /* another node has done an assert! * all done! */ sleep = 0; } else { sleep = 1; /* have all nodes responded? */ if (voting_done && !*blocked) { bit = find_next_bit(mle->maybe_map, O2NM_MAX_NODES, 0); if (dlm->node_num <= bit) { /* my node number is lowest. * now tell other nodes that I am * mastering this. */ mle->master = dlm->node_num; /* ref was grabbed in get_lock_resource * will be dropped in dlmlock_master */ assert = 1; sleep = 0; } /* if voting is done, but we have not received * an assert master yet, we must sleep */ } } spin_unlock(&mle->spinlock); /* sleep if we haven't finished voting yet */ if (sleep) { unsigned long timeo = msecs_to_jiffies(DLM_MASTERY_TIMEOUT_MS); /* if (atomic_read(&mle->mle_refs.refcount) < 2) mlog(ML_ERROR, "mle (%p) refs=%d, name=%.*s\n", mle, atomic_read(&mle->mle_refs.refcount), res->lockname.len, res->lockname.name); */ atomic_set(&mle->woken, 0); (void)wait_event_timeout(mle->wq, (atomic_read(&mle->woken) == 1), timeo); if (res->owner == O2NM_MAX_NODES) { mlog(0, "%s:%.*s: waiting again\n", dlm->name, res->lockname.len, res->lockname.name); goto recheck; } mlog(0, "done waiting, master is %u\n", res->owner); ret = 0; goto leave; } ret = 0; /* done */ if (assert) { m = dlm->node_num; mlog(0, "about to master %.*s here, this=%u\n", res->lockname.len, res->lockname.name, m); ret = dlm_do_assert_master(dlm, res, mle->vote_map, 0); if (ret) { /* This is a failure in the network path, * not in the response to the assert_master * (any nonzero response is a BUG on this node). * Most likely a socket just got disconnected * due to node death. */ mlog_errno(ret); } /* no longer need to restart lock mastery. * all living nodes have been contacted. */ ret = 0; } /* set the lockres owner */ spin_lock(&res->spinlock); /* mastery reference obtained either during * assert_master_handler or in get_lock_resource */ dlm_change_lockres_owner(dlm, res, m); spin_unlock(&res->spinlock); leave: return ret; } struct dlm_bitmap_diff_iter { int curnode; unsigned long *orig_bm; unsigned long *cur_bm; unsigned long diff_bm[BITS_TO_LONGS(O2NM_MAX_NODES)]; }; enum dlm_node_state_change { NODE_DOWN = -1, NODE_NO_CHANGE = 0, NODE_UP }; static void dlm_bitmap_diff_iter_init(struct dlm_bitmap_diff_iter *iter, unsigned long *orig_bm, unsigned long *cur_bm) { unsigned long p1, p2; int i; iter->curnode = -1; iter->orig_bm = orig_bm; iter->cur_bm = cur_bm; for (i = 0; i < BITS_TO_LONGS(O2NM_MAX_NODES); i++) { p1 = *(iter->orig_bm + i); p2 = *(iter->cur_bm + i); iter->diff_bm[i] = (p1 & ~p2) | (p2 & ~p1); } } static int dlm_bitmap_diff_iter_next(struct dlm_bitmap_diff_iter *iter, enum dlm_node_state_change *state) { int bit; if (iter->curnode >= O2NM_MAX_NODES) return -ENOENT; bit = find_next_bit(iter->diff_bm, O2NM_MAX_NODES, iter->curnode+1); if (bit >= O2NM_MAX_NODES) { iter->curnode = O2NM_MAX_NODES; return -ENOENT; } /* if it was there in the original then this node died */ if (test_bit(bit, iter->orig_bm)) *state = NODE_DOWN; else *state = NODE_UP; iter->curnode = bit; return bit; } static int dlm_restart_lock_mastery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, int blocked) { struct dlm_bitmap_diff_iter bdi; enum dlm_node_state_change sc; int node; int ret = 0; mlog(0, "something happened such that the " "master process may need to be restarted!\n"); assert_spin_locked(&mle->spinlock); dlm_bitmap_diff_iter_init(&bdi, mle->vote_map, mle->node_map); node = dlm_bitmap_diff_iter_next(&bdi, &sc); while (node >= 0) { if (sc == NODE_UP) { /* a node came up. clear any old vote from * the response map and set it in the vote map * then restart the mastery. */ mlog(ML_NOTICE, "node %d up while restarting\n", node); /* redo the master request, but only for the new node */ mlog(0, "sending request to new node\n"); clear_bit(node, mle->response_map); set_bit(node, mle->vote_map); } else { mlog(ML_ERROR, "node down! %d\n", node); if (blocked) { int lowest = find_next_bit(mle->maybe_map, O2NM_MAX_NODES, 0); /* act like it was never there */ clear_bit(node, mle->maybe_map); if (node == lowest) { mlog(0, "expected master %u died" " while this node was blocked " "waiting on it!\n", node); lowest = find_next_bit(mle->maybe_map, O2NM_MAX_NODES, lowest+1); if (lowest < O2NM_MAX_NODES) { mlog(0, "%s:%.*s:still " "blocked. waiting on %u " "now\n", dlm->name, res->lockname.len, res->lockname.name, lowest); } else { /* mle is an MLE_BLOCK, but * there is now nothing left to * block on. we need to return * all the way back out and try * again with an MLE_MASTER. * dlm_do_local_recovery_cleanup * has already run, so the mle * refcount is ok */ mlog(0, "%s:%.*s: no " "longer blocking. try to " "master this here\n", dlm->name, res->lockname.len, res->lockname.name); mle->type = DLM_MLE_MASTER; mle->u.res = res; } } } /* now blank out everything, as if we had never * contacted anyone */ memset(mle->maybe_map, 0, sizeof(mle->maybe_map)); memset(mle->response_map, 0, sizeof(mle->response_map)); /* reset the vote_map to the current node_map */ memcpy(mle->vote_map, mle->node_map, sizeof(mle->node_map)); /* put myself into the maybe map */ if (mle->type != DLM_MLE_BLOCK) set_bit(dlm->node_num, mle->maybe_map); } ret = -EAGAIN; node = dlm_bitmap_diff_iter_next(&bdi, &sc); } return ret; } /* * DLM_MASTER_REQUEST_MSG * * returns: 0 on success, * -errno on a network error * * on error, the caller should assume the target node is "dead" * */ static int dlm_do_master_request(struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, int to) { struct dlm_ctxt *dlm = mle->dlm; struct dlm_master_request request; int ret, response=0, resend; memset(&request, 0, sizeof(request)); request.node_idx = dlm->node_num; BUG_ON(mle->type == DLM_MLE_MIGRATION); if (mle->type != DLM_MLE_MASTER) { request.namelen = mle->u.name.len; memcpy(request.name, mle->u.name.name, request.namelen); } else { request.namelen = mle->u.res->lockname.len; memcpy(request.name, mle->u.res->lockname.name, request.namelen); } again: ret = o2net_send_message(DLM_MASTER_REQUEST_MSG, dlm->key, &request, sizeof(request), to, &response); if (ret < 0) { if (ret == -ESRCH) { /* should never happen */ mlog(ML_ERROR, "TCP stack not ready!\n"); BUG(); } else if (ret == -EINVAL) { mlog(ML_ERROR, "bad args passed to o2net!\n"); BUG(); } else if (ret == -ENOMEM) { mlog(ML_ERROR, "out of memory while trying to send " "network message! retrying\n"); /* this is totally crude */ msleep(50); goto again; } else if (!dlm_is_host_down(ret)) { /* not a network error. bad. */ mlog_errno(ret); mlog(ML_ERROR, "unhandled error!"); BUG(); } /* all other errors should be network errors, * and likely indicate node death */ mlog(ML_ERROR, "link to %d went down!\n", to); goto out; } ret = 0; resend = 0; spin_lock(&mle->spinlock); switch (response) { case DLM_MASTER_RESP_YES: set_bit(to, mle->response_map); mlog(0, "node %u is the master, response=YES\n", to); mlog(0, "%s:%.*s: master node %u now knows I have a " "reference\n", dlm->name, res->lockname.len, res->lockname.name, to); mle->master = to; break; case DLM_MASTER_RESP_NO: mlog(0, "node %u not master, response=NO\n", to); set_bit(to, mle->response_map); break; case DLM_MASTER_RESP_MAYBE: mlog(0, "node %u not master, response=MAYBE\n", to); set_bit(to, mle->response_map); set_bit(to, mle->maybe_map); break; case DLM_MASTER_RESP_ERROR: mlog(0, "node %u hit an error, resending\n", to); resend = 1; response = 0; break; default: mlog(ML_ERROR, "bad response! %u\n", response); BUG(); } spin_unlock(&mle->spinlock); if (resend) { /* this is also totally crude */ msleep(50); goto again; } out: return ret; } /* * locks that can be taken here: * dlm->spinlock * res->spinlock * mle->spinlock * dlm->master_list * * if possible, TRIM THIS DOWN!!! */ int dlm_master_request_handler(struct o2net_msg *msg, u32 len, void *data) { u8 response = DLM_MASTER_RESP_MAYBE; struct dlm_ctxt *dlm = data; struct dlm_lock_resource *res = NULL; struct dlm_master_request *request = (struct dlm_master_request *) msg->buf; struct dlm_master_list_entry *mle = NULL, *tmpmle = NULL; char *name; unsigned int namelen, hash; int found, ret; int set_maybe; int dispatch_assert = 0; if (!dlm_grab(dlm)) return DLM_MASTER_RESP_NO; if (!dlm_domain_fully_joined(dlm)) { response = DLM_MASTER_RESP_NO; goto send_response; } name = request->name; namelen = request->namelen; hash = dlm_lockid_hash(name, namelen); if (namelen > DLM_LOCKID_NAME_MAX) { response = DLM_IVBUFLEN; goto send_response; } way_up_top: spin_lock(&dlm->spinlock); res = __dlm_lookup_lockres(dlm, name, namelen, hash); if (res) { spin_unlock(&dlm->spinlock); /* take care of the easy cases up front */ spin_lock(&res->spinlock); if (res->state & (DLM_LOCK_RES_RECOVERING| DLM_LOCK_RES_MIGRATING)) { spin_unlock(&res->spinlock); mlog(0, "returning DLM_MASTER_RESP_ERROR since res is " "being recovered/migrated\n"); response = DLM_MASTER_RESP_ERROR; if (mle) kmem_cache_free(dlm_mle_cache, mle); goto send_response; } if (res->owner == dlm->node_num) { mlog(0, "%s:%.*s: setting bit %u in refmap\n", dlm->name, namelen, name, request->node_idx); dlm_lockres_set_refmap_bit(request->node_idx, res); spin_unlock(&res->spinlock); response = DLM_MASTER_RESP_YES; if (mle) kmem_cache_free(dlm_mle_cache, mle); /* this node is the owner. * there is some extra work that needs to * happen now. the requesting node has * caused all nodes up to this one to * create mles. this node now needs to * go back and clean those up. */ dispatch_assert = 1; goto send_response; } else if (res->owner != DLM_LOCK_RES_OWNER_UNKNOWN) { spin_unlock(&res->spinlock); // mlog(0, "node %u is the master\n", res->owner); response = DLM_MASTER_RESP_NO; if (mle) kmem_cache_free(dlm_mle_cache, mle); goto send_response; } /* ok, there is no owner. either this node is * being blocked, or it is actively trying to * master this lock. */ if (!(res->state & DLM_LOCK_RES_IN_PROGRESS)) { mlog(ML_ERROR, "lock with no owner should be " "in-progress!\n"); BUG(); } // mlog(0, "lockres is in progress...\n"); spin_lock(&dlm->master_lock); found = dlm_find_mle(dlm, &tmpmle, name, namelen); if (!found) { mlog(ML_ERROR, "no mle found for this lock!\n"); BUG(); } set_maybe = 1; spin_lock(&tmpmle->spinlock); if (tmpmle->type == DLM_MLE_BLOCK) { // mlog(0, "this node is waiting for " // "lockres to be mastered\n"); response = DLM_MASTER_RESP_NO; } else if (tmpmle->type == DLM_MLE_MIGRATION) { mlog(0, "node %u is master, but trying to migrate to " "node %u.\n", tmpmle->master, tmpmle->new_master); if (tmpmle->master == dlm->node_num) { mlog(ML_ERROR, "no owner on lockres, but this " "node is trying to migrate it to %u?!\n", tmpmle->new_master); BUG(); } else { /* the real master can respond on its own */ response = DLM_MASTER_RESP_NO; } } else if (tmpmle->master != DLM_LOCK_RES_OWNER_UNKNOWN) { set_maybe = 0; if (tmpmle->master == dlm->node_num) { response = DLM_MASTER_RESP_YES; /* this node will be the owner. * go back and clean the mles on any * other nodes */ dispatch_assert = 1; dlm_lockres_set_refmap_bit(request->node_idx, res); mlog(0, "%s:%.*s: setting bit %u in refmap\n", dlm->name, namelen, name, request->node_idx); } else response = DLM_MASTER_RESP_NO; } else { // mlog(0, "this node is attempting to " // "master lockres\n"); response = DLM_MASTER_RESP_MAYBE; } if (set_maybe) set_bit(request->node_idx, tmpmle->maybe_map); spin_unlock(&tmpmle->spinlock); spin_unlock(&dlm->master_lock); spin_unlock(&res->spinlock); /* keep the mle attached to heartbeat events */ dlm_put_mle(tmpmle); if (mle) kmem_cache_free(dlm_mle_cache, mle); goto send_response; } /* * lockres doesn't exist on this node * if there is an MLE_BLOCK, return NO * if there is an MLE_MASTER, return MAYBE * otherwise, add an MLE_BLOCK, return NO */ spin_lock(&dlm->master_lock); found = dlm_find_mle(dlm, &tmpmle, name, namelen); if (!found) { /* this lockid has never been seen on this node yet */ // mlog(0, "no mle found\n"); if (!mle) { spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); mle = (struct dlm_master_list_entry *) kmem_cache_alloc(dlm_mle_cache, GFP_NOFS); if (!mle) { response = DLM_MASTER_RESP_ERROR; mlog_errno(-ENOMEM); goto send_response; } goto way_up_top; } // mlog(0, "this is second time thru, already allocated, " // "add the block.\n"); dlm_init_mle(mle, DLM_MLE_BLOCK, dlm, NULL, name, namelen); set_bit(request->node_idx, mle->maybe_map); list_add(&mle->list, &dlm->master_list); response = DLM_MASTER_RESP_NO; } else { // mlog(0, "mle was found\n"); set_maybe = 1; spin_lock(&tmpmle->spinlock); if (tmpmle->master == dlm->node_num) { mlog(ML_ERROR, "no lockres, but an mle with this node as master!\n"); BUG(); } if (tmpmle->type == DLM_MLE_BLOCK) response = DLM_MASTER_RESP_NO; else if (tmpmle->type == DLM_MLE_MIGRATION) { mlog(0, "migration mle was found (%u->%u)\n", tmpmle->master, tmpmle->new_master); /* real master can respond on its own */ response = DLM_MASTER_RESP_NO; } else response = DLM_MASTER_RESP_MAYBE; if (set_maybe) set_bit(request->node_idx, tmpmle->maybe_map); spin_unlock(&tmpmle->spinlock); } spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); if (found) { /* keep the mle attached to heartbeat events */ dlm_put_mle(tmpmle); } send_response: if (dispatch_assert) { if (response != DLM_MASTER_RESP_YES) mlog(ML_ERROR, "invalid response %d\n", response); if (!res) { mlog(ML_ERROR, "bad lockres while trying to assert!\n"); BUG(); } mlog(0, "%u is the owner of %.*s, cleaning everyone else\n", dlm->node_num, res->lockname.len, res->lockname.name); ret = dlm_dispatch_assert_master(dlm, res, 0, request->node_idx, DLM_ASSERT_MASTER_MLE_CLEANUP); if (ret < 0) { mlog(ML_ERROR, "failed to dispatch assert master work\n"); response = DLM_MASTER_RESP_ERROR; } } dlm_put(dlm); return response; } /* * DLM_ASSERT_MASTER_MSG */ /* * NOTE: this can be used for debugging * can periodically run all locks owned by this node * and re-assert across the cluster... */ int dlm_do_assert_master(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, void *nodemap, u32 flags) { struct dlm_assert_master assert; int to, tmpret; struct dlm_node_iter iter; int ret = 0; int reassert; const char *lockname = res->lockname.name; unsigned int namelen = res->lockname.len; BUG_ON(namelen > O2NM_MAX_NAME_LEN); again: reassert = 0; /* note that if this nodemap is empty, it returns 0 */ dlm_node_iter_init(nodemap, &iter); while ((to = dlm_node_iter_next(&iter)) >= 0) { int r = 0; struct dlm_master_list_entry *mle = NULL; mlog(0, "sending assert master to %d (%.*s)\n", to, namelen, lockname); memset(&assert, 0, sizeof(assert)); assert.node_idx = dlm->node_num; assert.namelen = namelen; memcpy(assert.name, lockname, namelen); assert.flags = cpu_to_be32(flags); tmpret = o2net_send_message(DLM_ASSERT_MASTER_MSG, dlm->key, &assert, sizeof(assert), to, &r); if (tmpret < 0) { mlog(0, "assert_master returned %d!\n", tmpret); if (!dlm_is_host_down(tmpret)) { mlog(ML_ERROR, "unhandled error=%d!\n", tmpret); BUG(); } /* a node died. finish out the rest of the nodes. */ mlog(0, "link to %d went down!\n", to); /* any nonzero status return will do */ ret = tmpret; r = 0; } else if (r < 0) { /* ok, something horribly messed. kill thyself. */ mlog(ML_ERROR,"during assert master of %.*s to %u, " "got %d.\n", namelen, lockname, to, r); spin_lock(&dlm->spinlock); spin_lock(&dlm->master_lock); if (dlm_find_mle(dlm, &mle, (char *)lockname, namelen)) { dlm_print_one_mle(mle); __dlm_put_mle(mle); } spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); BUG(); } if (r & DLM_ASSERT_RESPONSE_REASSERT && !(r & DLM_ASSERT_RESPONSE_MASTERY_REF)) { mlog(ML_ERROR, "%.*s: very strange, " "master MLE but no lockres on %u\n", namelen, lockname, to); } if (r & DLM_ASSERT_RESPONSE_REASSERT) { mlog(0, "%.*s: node %u create mles on other " "nodes and requests a re-assert\n", namelen, lockname, to); reassert = 1; } if (r & DLM_ASSERT_RESPONSE_MASTERY_REF) { mlog(0, "%.*s: node %u has a reference to this " "lockres, set the bit in the refmap\n", namelen, lockname, to); spin_lock(&res->spinlock); dlm_lockres_set_refmap_bit(to, res); spin_unlock(&res->spinlock); } } if (reassert) goto again; return ret; } /* * locks that can be taken here: * dlm->spinlock * res->spinlock * mle->spinlock * dlm->master_list * * if possible, TRIM THIS DOWN!!! */ int dlm_assert_master_handler(struct o2net_msg *msg, u32 len, void *data) { struct dlm_ctxt *dlm = data; struct dlm_master_list_entry *mle = NULL; struct dlm_assert_master *assert = (struct dlm_assert_master *)msg->buf; struct dlm_lock_resource *res = NULL; char *name; unsigned int namelen, hash; u32 flags; int master_request = 0, have_lockres_ref = 0; int ret = 0; if (!dlm_grab(dlm)) return 0; name = assert->name; namelen = assert->namelen; hash = dlm_lockid_hash(name, namelen); flags = be32_to_cpu(assert->flags); if (namelen > DLM_LOCKID_NAME_MAX) { mlog(ML_ERROR, "Invalid name length!"); goto done; } spin_lock(&dlm->spinlock); if (flags) mlog(0, "assert_master with flags: %u\n", flags); /* find the MLE */ spin_lock(&dlm->master_lock); if (!dlm_find_mle(dlm, &mle, name, namelen)) { /* not an error, could be master just re-asserting */ mlog(0, "just got an assert_master from %u, but no " "MLE for it! (%.*s)\n", assert->node_idx, namelen, name); } else { int bit = find_next_bit (mle->maybe_map, O2NM_MAX_NODES, 0); if (bit >= O2NM_MAX_NODES) { /* not necessarily an error, though less likely. * could be master just re-asserting. */ mlog(0, "no bits set in the maybe_map, but %u " "is asserting! (%.*s)\n", assert->node_idx, namelen, name); } else if (bit != assert->node_idx) { if (flags & DLM_ASSERT_MASTER_MLE_CLEANUP) { mlog(0, "master %u was found, %u should " "back off\n", assert->node_idx, bit); } else { /* with the fix for bug 569, a higher node * number winning the mastery will respond * YES to mastery requests, but this node * had no way of knowing. let it pass. */ mlog(0, "%u is the lowest node, " "%u is asserting. (%.*s) %u must " "have begun after %u won.\n", bit, assert->node_idx, namelen, name, bit, assert->node_idx); } } if (mle->type == DLM_MLE_MIGRATION) { if (flags & DLM_ASSERT_MASTER_MLE_CLEANUP) { mlog(0, "%s:%.*s: got cleanup assert" " from %u for migration\n", dlm->name, namelen, name, assert->node_idx); } else if (!(flags & DLM_ASSERT_MASTER_FINISH_MIGRATION)) { mlog(0, "%s:%.*s: got unrelated assert" " from %u for migration, ignoring\n", dlm->name, namelen, name, assert->node_idx); __dlm_put_mle(mle); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); goto done; } } } spin_unlock(&dlm->master_lock); /* ok everything checks out with the MLE * now check to see if there is a lockres */ res = __dlm_lookup_lockres(dlm, name, namelen, hash); if (res) { spin_lock(&res->spinlock); if (res->state & DLM_LOCK_RES_RECOVERING) { mlog(ML_ERROR, "%u asserting but %.*s is " "RECOVERING!\n", assert->node_idx, namelen, name); goto kill; } if (!mle) { if (res->owner != DLM_LOCK_RES_OWNER_UNKNOWN && res->owner != assert->node_idx) { mlog(ML_ERROR, "assert_master from " "%u, but current owner is " "%u! (%.*s)\n", assert->node_idx, res->owner, namelen, name); goto kill; } } else if (mle->type != DLM_MLE_MIGRATION) { if (res->owner != DLM_LOCK_RES_OWNER_UNKNOWN) { /* owner is just re-asserting */ if (res->owner == assert->node_idx) { mlog(0, "owner %u re-asserting on " "lock %.*s\n", assert->node_idx, namelen, name); goto ok; } mlog(ML_ERROR, "got assert_master from " "node %u, but %u is the owner! " "(%.*s)\n", assert->node_idx, res->owner, namelen, name); goto kill; } if (!(res->state & DLM_LOCK_RES_IN_PROGRESS)) { mlog(ML_ERROR, "got assert from %u, but lock " "with no owner should be " "in-progress! (%.*s)\n", assert->node_idx, namelen, name); goto kill; } } else /* mle->type == DLM_MLE_MIGRATION */ { /* should only be getting an assert from new master */ if (assert->node_idx != mle->new_master) { mlog(ML_ERROR, "got assert from %u, but " "new master is %u, and old master " "was %u (%.*s)\n", assert->node_idx, mle->new_master, mle->master, namelen, name); goto kill; } } ok: spin_unlock(&res->spinlock); } spin_unlock(&dlm->spinlock); // mlog(0, "woo! got an assert_master from node %u!\n", // assert->node_idx); if (mle) { int extra_ref = 0; int nn = -1; int rr, err = 0; spin_lock(&mle->spinlock); if (mle->type == DLM_MLE_BLOCK || mle->type == DLM_MLE_MIGRATION) extra_ref = 1; else { /* MASTER mle: if any bits set in the response map * then the calling node needs to re-assert to clear * up nodes that this node contacted */ while ((nn = find_next_bit (mle->response_map, O2NM_MAX_NODES, nn+1)) < O2NM_MAX_NODES) { if (nn != dlm->node_num && nn != assert->node_idx) master_request = 1; } } mle->master = assert->node_idx; atomic_set(&mle->woken, 1); wake_up(&mle->wq); spin_unlock(&mle->spinlock); if (res) { spin_lock(&res->spinlock); if (mle->type == DLM_MLE_MIGRATION) { mlog(0, "finishing off migration of lockres %.*s, " "from %u to %u\n", res->lockname.len, res->lockname.name, dlm->node_num, mle->new_master); res->state &= ~DLM_LOCK_RES_MIGRATING; dlm_change_lockres_owner(dlm, res, mle->new_master); BUG_ON(res->state & DLM_LOCK_RES_DIRTY); } else { dlm_change_lockres_owner(dlm, res, mle->master); } spin_unlock(&res->spinlock); have_lockres_ref = 1; } /* master is known, detach if not already detached. * ensures that only one assert_master call will happen * on this mle. */ spin_lock(&dlm->spinlock); spin_lock(&dlm->master_lock); rr = atomic_read(&mle->mle_refs.refcount); if (mle->inuse > 0) { if (extra_ref && rr < 3) err = 1; else if (!extra_ref && rr < 2) err = 1; } else { if (extra_ref && rr < 2) err = 1; else if (!extra_ref && rr < 1) err = 1; } if (err) { mlog(ML_ERROR, "%s:%.*s: got assert master from %u " "that will mess up this node, refs=%d, extra=%d, " "inuse=%d\n", dlm->name, namelen, name, assert->node_idx, rr, extra_ref, mle->inuse); dlm_print_one_mle(mle); } list_del_init(&mle->list); __dlm_mle_detach_hb_events(dlm, mle); __dlm_put_mle(mle); if (extra_ref) { /* the assert master message now balances the extra * ref given by the master / migration request message. * if this is the last put, it will be removed * from the list. */ __dlm_put_mle(mle); } spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); } else if (res) { if (res->owner != assert->node_idx) { mlog(0, "assert_master from %u, but current " "owner is %u (%.*s), no mle\n", assert->node_idx, res->owner, namelen, name); } } done: ret = 0; if (res) dlm_lockres_put(res); dlm_put(dlm); if (master_request) { mlog(0, "need to tell master to reassert\n"); /* positive. negative would shoot down the node. */ ret |= DLM_ASSERT_RESPONSE_REASSERT; if (!have_lockres_ref) { mlog(ML_ERROR, "strange, got assert from %u, MASTER " "mle present here for %s:%.*s, but no lockres!\n", assert->node_idx, dlm->name, namelen, name); } } if (have_lockres_ref) { /* let the master know we have a reference to the lockres */ ret |= DLM_ASSERT_RESPONSE_MASTERY_REF; mlog(0, "%s:%.*s: got assert from %u, need a ref\n", dlm->name, namelen, name, assert->node_idx); } return ret; kill: /* kill the caller! */ mlog(ML_ERROR, "Bad message received from another node. Dumping state " "and killing the other node now! This node is OK and can continue.\n"); __dlm_print_one_lock_resource(res); spin_unlock(&res->spinlock); spin_unlock(&dlm->spinlock); dlm_lockres_put(res); dlm_put(dlm); return -EINVAL; } int dlm_dispatch_assert_master(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, int ignore_higher, u8 request_from, u32 flags) { struct dlm_work_item *item; item = kzalloc(sizeof(*item), GFP_NOFS); if (!item) return -ENOMEM; /* queue up work for dlm_assert_master_worker */ dlm_grab(dlm); /* get an extra ref for the work item */ dlm_init_work_item(dlm, item, dlm_assert_master_worker, NULL); item->u.am.lockres = res; /* already have a ref */ /* can optionally ignore node numbers higher than this node */ item->u.am.ignore_higher = ignore_higher; item->u.am.request_from = request_from; item->u.am.flags = flags; if (ignore_higher) mlog(0, "IGNORE HIGHER: %.*s\n", res->lockname.len, res->lockname.name); spin_lock(&dlm->work_lock); list_add_tail(&item->list, &dlm->work_list); spin_unlock(&dlm->work_lock); queue_work(dlm->dlm_worker, &dlm->dispatched_work); return 0; } static void dlm_assert_master_worker(struct dlm_work_item *item, void *data) { struct dlm_ctxt *dlm = data; int ret = 0; struct dlm_lock_resource *res; unsigned long nodemap[BITS_TO_LONGS(O2NM_MAX_NODES)]; int ignore_higher; int bit; u8 request_from; u32 flags; dlm = item->dlm; res = item->u.am.lockres; ignore_higher = item->u.am.ignore_higher; request_from = item->u.am.request_from; flags = item->u.am.flags; spin_lock(&dlm->spinlock); memcpy(nodemap, dlm->domain_map, sizeof(nodemap)); spin_unlock(&dlm->spinlock); clear_bit(dlm->node_num, nodemap); if (ignore_higher) { /* if is this just to clear up mles for nodes below * this node, do not send the message to the original * caller or any node number higher than this */ clear_bit(request_from, nodemap); bit = dlm->node_num; while (1) { bit = find_next_bit(nodemap, O2NM_MAX_NODES, bit+1); if (bit >= O2NM_MAX_NODES) break; clear_bit(bit, nodemap); } } /* * If we're migrating this lock to someone else, we are no * longer allowed to assert out own mastery. OTOH, we need to * prevent migration from starting while we're still asserting * our dominance. The reserved ast delays migration. */ spin_lock(&res->spinlock); if (res->state & DLM_LOCK_RES_MIGRATING) { mlog(0, "Someone asked us to assert mastery, but we're " "in the middle of migration. Skipping assert, " "the new master will handle that.\n"); spin_unlock(&res->spinlock); goto put; } else __dlm_lockres_reserve_ast(res); spin_unlock(&res->spinlock); /* this call now finishes out the nodemap * even if one or more nodes die */ mlog(0, "worker about to master %.*s here, this=%u\n", res->lockname.len, res->lockname.name, dlm->node_num); ret = dlm_do_assert_master(dlm, res, nodemap, flags); if (ret < 0) { /* no need to restart, we are done */ if (!dlm_is_host_down(ret)) mlog_errno(ret); } /* Ok, we've asserted ourselves. Let's let migration start. */ dlm_lockres_release_ast(dlm, res); put: dlm_lockres_put(res); mlog(0, "finished with dlm_assert_master_worker\n"); } /* SPECIAL CASE for the $RECOVERY lock used by the recovery thread. * We cannot wait for node recovery to complete to begin mastering this * lockres because this lockres is used to kick off recovery! ;-) * So, do a pre-check on all living nodes to see if any of those nodes * think that $RECOVERY is currently mastered by a dead node. If so, * we wait a short time to allow that node to get notified by its own * heartbeat stack, then check again. All $RECOVERY lock resources * mastered by dead nodes are purged when the hearbeat callback is * fired, so we can know for sure that it is safe to continue once * the node returns a live node or no node. */ static int dlm_pre_master_reco_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { struct dlm_node_iter iter; int nodenum; int ret = 0; u8 master = DLM_LOCK_RES_OWNER_UNKNOWN; spin_lock(&dlm->spinlock); dlm_node_iter_init(dlm->domain_map, &iter); spin_unlock(&dlm->spinlock); while ((nodenum = dlm_node_iter_next(&iter)) >= 0) { /* do not send to self */ if (nodenum == dlm->node_num) continue; ret = dlm_do_master_requery(dlm, res, nodenum, &master); if (ret < 0) { mlog_errno(ret); if (!dlm_is_host_down(ret)) BUG(); /* host is down, so answer for that node would be * DLM_LOCK_RES_OWNER_UNKNOWN. continue. */ ret = 0; } if (master != DLM_LOCK_RES_OWNER_UNKNOWN) { /* check to see if this master is in the recovery map */ spin_lock(&dlm->spinlock); if (test_bit(master, dlm->recovery_map)) { mlog(ML_NOTICE, "%s: node %u has not seen " "node %u go down yet, and thinks the " "dead node is mastering the recovery " "lock. must wait.\n", dlm->name, nodenum, master); ret = -EAGAIN; } spin_unlock(&dlm->spinlock); mlog(0, "%s: reco lock master is %u\n", dlm->name, master); break; } } return ret; } /* * DLM_DEREF_LOCKRES_MSG */ int dlm_drop_lockres_ref(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { struct dlm_deref_lockres deref; int ret = 0, r; const char *lockname; unsigned int namelen; lockname = res->lockname.name; namelen = res->lockname.len; BUG_ON(namelen > O2NM_MAX_NAME_LEN); mlog(0, "%s:%.*s: sending deref to %d\n", dlm->name, namelen, lockname, res->owner); memset(&deref, 0, sizeof(deref)); deref.node_idx = dlm->node_num; deref.namelen = namelen; memcpy(deref.name, lockname, namelen); ret = o2net_send_message(DLM_DEREF_LOCKRES_MSG, dlm->key, &deref, sizeof(deref), res->owner, &r); if (ret < 0) mlog_errno(ret); else if (r < 0) { /* BAD. other node says I did not have a ref. */ mlog(ML_ERROR,"while dropping ref on %s:%.*s " "(master=%u) got %d.\n", dlm->name, namelen, lockname, res->owner, r); dlm_print_one_lock_resource(res); BUG(); } return ret; } int dlm_deref_lockres_handler(struct o2net_msg *msg, u32 len, void *data) { struct dlm_ctxt *dlm = data; struct dlm_deref_lockres *deref = (struct dlm_deref_lockres *)msg->buf; struct dlm_lock_resource *res = NULL; char *name; unsigned int namelen; int ret = -EINVAL; u8 node; unsigned int hash; if (!dlm_grab(dlm)) return 0; name = deref->name; namelen = deref->namelen; node = deref->node_idx; if (namelen > DLM_LOCKID_NAME_MAX) { mlog(ML_ERROR, "Invalid name length!"); goto done; } if (deref->node_idx >= O2NM_MAX_NODES) { mlog(ML_ERROR, "Invalid node number: %u\n", node); goto done; } hash = dlm_lockid_hash(name, namelen); spin_lock(&dlm->spinlock); res = __dlm_lookup_lockres_full(dlm, name, namelen, hash); if (!res) { spin_unlock(&dlm->spinlock); mlog(ML_ERROR, "%s:%.*s: bad lockres name\n", dlm->name, namelen, name); goto done; } spin_unlock(&dlm->spinlock); spin_lock(&res->spinlock); BUG_ON(res->state & DLM_LOCK_RES_DROPPING_REF); if (test_bit(node, res->refmap)) { ret = 0; dlm_lockres_clear_refmap_bit(node, res); } else { mlog(ML_ERROR, "%s:%.*s: node %u trying to drop ref " "but it is already dropped!\n", dlm->name, namelen, name, node); __dlm_print_one_lock_resource(res); } spin_unlock(&res->spinlock); if (!ret) dlm_lockres_calc_usage(dlm, res); done: if (res) dlm_lockres_put(res); dlm_put(dlm); return ret; } /* * DLM_MIGRATE_LOCKRES */ static int dlm_migrate_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 target) { struct dlm_master_list_entry *mle = NULL; struct dlm_master_list_entry *oldmle = NULL; struct dlm_migratable_lockres *mres = NULL; int ret = -EINVAL; const char *name; unsigned int namelen; int mle_added = 0; struct list_head *queue, *iter; int i; struct dlm_lock *lock; int empty = 1; if (!dlm_grab(dlm)) return -EINVAL; name = res->lockname.name; namelen = res->lockname.len; mlog(0, "migrating %.*s to %u\n", namelen, name, target); /* * ensure this lockres is a proper candidate for migration */ spin_lock(&res->spinlock); if (res->owner == DLM_LOCK_RES_OWNER_UNKNOWN) { mlog(0, "cannot migrate lockres with unknown owner!\n"); spin_unlock(&res->spinlock); goto leave; } if (res->owner != dlm->node_num) { mlog(0, "cannot migrate lockres this node doesn't own!\n"); spin_unlock(&res->spinlock); goto leave; } mlog(0, "checking queues...\n"); queue = &res->granted; for (i=0; i<3; i++) { list_for_each(iter, queue) { lock = list_entry (iter, struct dlm_lock, list); empty = 0; if (lock->ml.node == dlm->node_num) { mlog(0, "found a lock owned by this node " "still on the %s queue! will not " "migrate this lockres\n", i==0 ? "granted" : (i==1 ? "converting" : "blocked")); spin_unlock(&res->spinlock); ret = -ENOTEMPTY; goto leave; } } queue++; } mlog(0, "all locks on this lockres are nonlocal. continuing\n"); spin_unlock(&res->spinlock); /* no work to do */ if (empty) { mlog(0, "no locks were found on this lockres! done!\n"); ret = 0; goto leave; } /* * preallocate up front * if this fails, abort */ ret = -ENOMEM; mres = (struct dlm_migratable_lockres *) __get_free_page(GFP_NOFS); if (!mres) { mlog_errno(ret); goto leave; } mle = (struct dlm_master_list_entry *) kmem_cache_alloc(dlm_mle_cache, GFP_NOFS); if (!mle) { mlog_errno(ret); goto leave; } ret = 0; /* * find a node to migrate the lockres to */ mlog(0, "picking a migration node\n"); spin_lock(&dlm->spinlock); /* pick a new node */ if (!test_bit(target, dlm->domain_map) || target >= O2NM_MAX_NODES) { target = dlm_pick_migration_target(dlm, res); } mlog(0, "node %u chosen for migration\n", target); if (target >= O2NM_MAX_NODES || !test_bit(target, dlm->domain_map)) { /* target chosen is not alive */ ret = -EINVAL; } if (ret) { spin_unlock(&dlm->spinlock); goto fail; } mlog(0, "continuing with target = %u\n", target); /* * clear any existing master requests and * add the migration mle to the list */ spin_lock(&dlm->master_lock); ret = dlm_add_migration_mle(dlm, res, mle, &oldmle, name, namelen, target, dlm->node_num); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); if (ret == -EEXIST) { mlog(0, "another process is already migrating it\n"); goto fail; } mle_added = 1; /* * set the MIGRATING flag and flush asts * if we fail after this we need to re-dirty the lockres */ if (dlm_mark_lockres_migrating(dlm, res, target) < 0) { mlog(ML_ERROR, "tried to migrate %.*s to %u, but " "the target went down.\n", res->lockname.len, res->lockname.name, target); spin_lock(&res->spinlock); res->state &= ~DLM_LOCK_RES_MIGRATING; spin_unlock(&res->spinlock); ret = -EINVAL; } fail: if (oldmle) { /* master is known, detach if not already detached */ dlm_mle_detach_hb_events(dlm, oldmle); dlm_put_mle(oldmle); } if (ret < 0) { if (mle_added) { dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle(mle); } else if (mle) { kmem_cache_free(dlm_mle_cache, mle); } goto leave; } /* * at this point, we have a migration target, an mle * in the master list, and the MIGRATING flag set on * the lockres */ /* now that remote nodes are spinning on the MIGRATING flag, * ensure that all assert_master work is flushed. */ flush_workqueue(dlm->dlm_worker); /* get an extra reference on the mle. * otherwise the assert_master from the new * master will destroy this. * also, make sure that all callers of dlm_get_mle * take both dlm->spinlock and dlm->master_lock */ spin_lock(&dlm->spinlock); spin_lock(&dlm->master_lock); dlm_get_mle_inuse(mle); spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); /* notify new node and send all lock state */ /* call send_one_lockres with migration flag. * this serves as notice to the target node that a * migration is starting. */ ret = dlm_send_one_lockres(dlm, res, mres, target, DLM_MRES_MIGRATION); if (ret < 0) { mlog(0, "migration to node %u failed with %d\n", target, ret); /* migration failed, detach and clean up mle */ dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle(mle); dlm_put_mle_inuse(mle); spin_lock(&res->spinlock); res->state &= ~DLM_LOCK_RES_MIGRATING; spin_unlock(&res->spinlock); goto leave; } /* at this point, the target sends a message to all nodes, * (using dlm_do_migrate_request). this node is skipped since * we had to put an mle in the list to begin the process. this * node now waits for target to do an assert master. this node * will be the last one notified, ensuring that the migration * is complete everywhere. if the target dies while this is * going on, some nodes could potentially see the target as the * master, so it is important that my recovery finds the migration * mle and sets the master to UNKNONWN. */ /* wait for new node to assert master */ while (1) { ret = wait_event_interruptible_timeout(mle->wq, (atomic_read(&mle->woken) == 1), msecs_to_jiffies(5000)); if (ret >= 0) { if (atomic_read(&mle->woken) == 1 || res->owner == target) break; mlog(0, "%s:%.*s: timed out during migration\n", dlm->name, res->lockname.len, res->lockname.name); /* avoid hang during shutdown when migrating lockres * to a node which also goes down */ if (dlm_is_node_dead(dlm, target)) { mlog(0, "%s:%.*s: expected migration " "target %u is no longer up, restarting\n", dlm->name, res->lockname.len, res->lockname.name, target); ret = -EINVAL; /* migration failed, detach and clean up mle */ dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle(mle); dlm_put_mle_inuse(mle); spin_lock(&res->spinlock); res->state &= ~DLM_LOCK_RES_MIGRATING; spin_unlock(&res->spinlock); goto leave; } } else mlog(0, "%s:%.*s: caught signal during migration\n", dlm->name, res->lockname.len, res->lockname.name); } /* all done, set the owner, clear the flag */ spin_lock(&res->spinlock); dlm_set_lockres_owner(dlm, res, target); res->state &= ~DLM_LOCK_RES_MIGRATING; dlm_remove_nonlocal_locks(dlm, res); spin_unlock(&res->spinlock); wake_up(&res->wq); /* master is known, detach if not already detached */ dlm_mle_detach_hb_events(dlm, mle); dlm_put_mle_inuse(mle); ret = 0; dlm_lockres_calc_usage(dlm, res); leave: /* re-dirty the lockres if we failed */ if (ret < 0) dlm_kick_thread(dlm, res); /* TODO: cleanup */ if (mres) free_page((unsigned long)mres); dlm_put(dlm); mlog(0, "returning %d\n", ret); return ret; } #define DLM_MIGRATION_RETRY_MS 100 /* Should be called only after beginning the domain leave process. * There should not be any remaining locks on nonlocal lock resources, * and there should be no local locks left on locally mastered resources. * * Called with the dlm spinlock held, may drop it to do migration, but * will re-acquire before exit. * * Returns: 1 if dlm->spinlock was dropped/retaken, 0 if never dropped */ int dlm_empty_lockres(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { int ret; int lock_dropped = 0; if (res->owner != dlm->node_num) { if (!__dlm_lockres_unused(res)) { mlog(ML_ERROR, "%s:%.*s: this node is not master, " "trying to free this but locks remain\n", dlm->name, res->lockname.len, res->lockname.name); } goto leave; } /* Wheee! Migrate lockres here! Will sleep so drop spinlock. */ spin_unlock(&dlm->spinlock); lock_dropped = 1; while (1) { ret = dlm_migrate_lockres(dlm, res, O2NM_MAX_NODES); if (ret >= 0) break; if (ret == -ENOTEMPTY) { mlog(ML_ERROR, "lockres %.*s still has local locks!\n", res->lockname.len, res->lockname.name); BUG(); } mlog(0, "lockres %.*s: migrate failed, " "retrying\n", res->lockname.len, res->lockname.name); msleep(DLM_MIGRATION_RETRY_MS); } spin_lock(&dlm->spinlock); leave: return lock_dropped; } int dlm_lock_basts_flushed(struct dlm_ctxt *dlm, struct dlm_lock *lock) { int ret; spin_lock(&dlm->ast_lock); spin_lock(&lock->spinlock); ret = (list_empty(&lock->bast_list) && !lock->bast_pending); spin_unlock(&lock->spinlock); spin_unlock(&dlm->ast_lock); return ret; } static int dlm_migration_can_proceed(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 mig_target) { int can_proceed; spin_lock(&res->spinlock); can_proceed = !!(res->state & DLM_LOCK_RES_MIGRATING); spin_unlock(&res->spinlock); /* target has died, so make the caller break out of the * wait_event, but caller must recheck the domain_map */ spin_lock(&dlm->spinlock); if (!test_bit(mig_target, dlm->domain_map)) can_proceed = 1; spin_unlock(&dlm->spinlock); return can_proceed; } static int dlm_lockres_is_dirty(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { int ret; spin_lock(&res->spinlock); ret = !!(res->state & DLM_LOCK_RES_DIRTY); spin_unlock(&res->spinlock); return ret; } static int dlm_mark_lockres_migrating(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 target) { int ret = 0; mlog(0, "dlm_mark_lockres_migrating: %.*s, from %u to %u\n", res->lockname.len, res->lockname.name, dlm->node_num, target); /* need to set MIGRATING flag on lockres. this is done by * ensuring that all asts have been flushed for this lockres. */ spin_lock(&res->spinlock); BUG_ON(res->migration_pending); res->migration_pending = 1; /* strategy is to reserve an extra ast then release * it below, letting the release do all of the work */ __dlm_lockres_reserve_ast(res); spin_unlock(&res->spinlock); /* now flush all the pending asts */ dlm_kick_thread(dlm, res); /* before waiting on DIRTY, block processes which may * try to dirty the lockres before MIGRATING is set */ spin_lock(&res->spinlock); BUG_ON(res->state & DLM_LOCK_RES_BLOCK_DIRTY); res->state |= DLM_LOCK_RES_BLOCK_DIRTY; spin_unlock(&res->spinlock); /* now wait on any pending asts and the DIRTY state */ wait_event(dlm->ast_wq, !dlm_lockres_is_dirty(dlm, res)); dlm_lockres_release_ast(dlm, res); mlog(0, "about to wait on migration_wq, dirty=%s\n", res->state & DLM_LOCK_RES_DIRTY ? "yes" : "no"); /* if the extra ref we just put was the final one, this * will pass thru immediately. otherwise, we need to wait * for the last ast to finish. */ again: ret = wait_event_interruptible_timeout(dlm->migration_wq, dlm_migration_can_proceed(dlm, res, target), msecs_to_jiffies(1000)); if (ret < 0) { mlog(0, "woken again: migrating? %s, dead? %s\n", res->state & DLM_LOCK_RES_MIGRATING ? "yes":"no", test_bit(target, dlm->domain_map) ? "no":"yes"); } else { mlog(0, "all is well: migrating? %s, dead? %s\n", res->state & DLM_LOCK_RES_MIGRATING ? "yes":"no", test_bit(target, dlm->domain_map) ? "no":"yes"); } if (!dlm_migration_can_proceed(dlm, res, target)) { mlog(0, "trying again...\n"); goto again; } /* now that we are sure the MIGRATING state is there, drop * the unneded state which blocked threads trying to DIRTY */ spin_lock(&res->spinlock); BUG_ON(!(res->state & DLM_LOCK_RES_BLOCK_DIRTY)); BUG_ON(!(res->state & DLM_LOCK_RES_MIGRATING)); res->state &= ~DLM_LOCK_RES_BLOCK_DIRTY; spin_unlock(&res->spinlock); /* did the target go down or die? */ spin_lock(&dlm->spinlock); if (!test_bit(target, dlm->domain_map)) { mlog(ML_ERROR, "aha. migration target %u just went down\n", target); ret = -EHOSTDOWN; } spin_unlock(&dlm->spinlock); /* * at this point: * * o the DLM_LOCK_RES_MIGRATING flag is set * o there are no pending asts on this lockres * o all processes trying to reserve an ast on this * lockres must wait for the MIGRATING flag to clear */ return ret; } /* last step in the migration process. * original master calls this to free all of the dlm_lock * structures that used to be for other nodes. */ static void dlm_remove_nonlocal_locks(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { struct list_head *iter, *iter2; struct list_head *queue = &res->granted; int i, bit; struct dlm_lock *lock; assert_spin_locked(&res->spinlock); BUG_ON(res->owner == dlm->node_num); for (i=0; i<3; i++) { list_for_each_safe(iter, iter2, queue) { lock = list_entry (iter, struct dlm_lock, list); if (lock->ml.node != dlm->node_num) { mlog(0, "putting lock for node %u\n", lock->ml.node); /* be extra careful */ BUG_ON(!list_empty(&lock->ast_list)); BUG_ON(!list_empty(&lock->bast_list)); BUG_ON(lock->ast_pending); BUG_ON(lock->bast_pending); dlm_lockres_clear_refmap_bit(lock->ml.node, res); list_del_init(&lock->list); dlm_lock_put(lock); } } queue++; } bit = 0; while (1) { bit = find_next_bit(res->refmap, O2NM_MAX_NODES, bit); if (bit >= O2NM_MAX_NODES) break; /* do not clear the local node reference, if there is a * process holding this, let it drop the ref itself */ if (bit != dlm->node_num) { mlog(0, "%s:%.*s: node %u had a ref to this " "migrating lockres, clearing\n", dlm->name, res->lockname.len, res->lockname.name, bit); dlm_lockres_clear_refmap_bit(bit, res); } bit++; } } /* for now this is not too intelligent. we will * need stats to make this do the right thing. * this just finds the first lock on one of the * queues and uses that node as the target. */ static u8 dlm_pick_migration_target(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { int i; struct list_head *queue = &res->granted; struct list_head *iter; struct dlm_lock *lock; int nodenum; assert_spin_locked(&dlm->spinlock); spin_lock(&res->spinlock); for (i=0; i<3; i++) { list_for_each(iter, queue) { /* up to the caller to make sure this node * is alive */ lock = list_entry (iter, struct dlm_lock, list); if (lock->ml.node != dlm->node_num) { spin_unlock(&res->spinlock); return lock->ml.node; } } queue++; } spin_unlock(&res->spinlock); mlog(0, "have not found a suitable target yet! checking domain map\n"); /* ok now we're getting desperate. pick anyone alive. */ nodenum = -1; while (1) { nodenum = find_next_bit(dlm->domain_map, O2NM_MAX_NODES, nodenum+1); mlog(0, "found %d in domain map\n", nodenum); if (nodenum >= O2NM_MAX_NODES) break; if (nodenum != dlm->node_num) { mlog(0, "picking %d\n", nodenum); return nodenum; } } mlog(0, "giving up. no master to migrate to\n"); return DLM_LOCK_RES_OWNER_UNKNOWN; } /* this is called by the new master once all lockres * data has been received */ static int dlm_do_migrate_request(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 master, u8 new_master, struct dlm_node_iter *iter) { struct dlm_migrate_request migrate; int ret, status = 0; int nodenum; memset(&migrate, 0, sizeof(migrate)); migrate.namelen = res->lockname.len; memcpy(migrate.name, res->lockname.name, migrate.namelen); migrate.new_master = new_master; migrate.master = master; ret = 0; /* send message to all nodes, except the master and myself */ while ((nodenum = dlm_node_iter_next(iter)) >= 0) { if (nodenum == master || nodenum == new_master) continue; ret = o2net_send_message(DLM_MIGRATE_REQUEST_MSG, dlm->key, &migrate, sizeof(migrate), nodenum, &status); if (ret < 0) mlog_errno(ret); else if (status < 0) { mlog(0, "migrate request (node %u) returned %d!\n", nodenum, status); ret = status; } else if (status == DLM_MIGRATE_RESPONSE_MASTERY_REF) { /* during the migration request we short-circuited * the mastery of the lockres. make sure we have * a mastery ref for nodenum */ mlog(0, "%s:%.*s: need ref for node %u\n", dlm->name, res->lockname.len, res->lockname.name, nodenum); spin_lock(&res->spinlock); dlm_lockres_set_refmap_bit(nodenum, res); spin_unlock(&res->spinlock); } } if (ret < 0) mlog_errno(ret); mlog(0, "returning ret=%d\n", ret); return ret; } /* if there is an existing mle for this lockres, we now know who the master is. * (the one who sent us *this* message) we can clear it up right away. * since the process that put the mle on the list still has a reference to it, * we can unhash it now, set the master and wake the process. as a result, * we will have no mle in the list to start with. now we can add an mle for * the migration and this should be the only one found for those scanning the * list. */ int dlm_migrate_request_handler(struct o2net_msg *msg, u32 len, void *data) { struct dlm_ctxt *dlm = data; struct dlm_lock_resource *res = NULL; struct dlm_migrate_request *migrate = (struct dlm_migrate_request *) msg->buf; struct dlm_master_list_entry *mle = NULL, *oldmle = NULL; const char *name; unsigned int namelen, hash; int ret = 0; if (!dlm_grab(dlm)) return -EINVAL; name = migrate->name; namelen = migrate->namelen; hash = dlm_lockid_hash(name, namelen); /* preallocate.. if this fails, abort */ mle = (struct dlm_master_list_entry *) kmem_cache_alloc(dlm_mle_cache, GFP_NOFS); if (!mle) { ret = -ENOMEM; goto leave; } /* check for pre-existing lock */ spin_lock(&dlm->spinlock); res = __dlm_lookup_lockres(dlm, name, namelen, hash); spin_lock(&dlm->master_lock); if (res) { spin_lock(&res->spinlock); if (res->state & DLM_LOCK_RES_RECOVERING) { /* if all is working ok, this can only mean that we got * a migrate request from a node that we now see as * dead. what can we do here? drop it to the floor? */ spin_unlock(&res->spinlock); mlog(ML_ERROR, "Got a migrate request, but the " "lockres is marked as recovering!"); kmem_cache_free(dlm_mle_cache, mle); ret = -EINVAL; /* need a better solution */ goto unlock; } res->state |= DLM_LOCK_RES_MIGRATING; spin_unlock(&res->spinlock); } /* ignore status. only nonzero status would BUG. */ ret = dlm_add_migration_mle(dlm, res, mle, &oldmle, name, namelen, migrate->new_master, migrate->master); unlock: spin_unlock(&dlm->master_lock); spin_unlock(&dlm->spinlock); if (oldmle) { /* master is known, detach if not already detached */ dlm_mle_detach_hb_events(dlm, oldmle); dlm_put_mle(oldmle); } if (res) dlm_lockres_put(res); leave: dlm_put(dlm); return ret; } /* must be holding dlm->spinlock and dlm->master_lock * when adding a migration mle, we can clear any other mles * in the master list because we know with certainty that * the master is "master". so we remove any old mle from * the list after setting it's master field, and then add * the new migration mle. this way we can hold with the rule * of having only one mle for a given lock name at all times. */ static int dlm_add_migration_mle(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, struct dlm_master_list_entry *mle, struct dlm_master_list_entry **oldmle, const char *name, unsigned int namelen, u8 new_master, u8 master) { int found; int ret = 0; *oldmle = NULL; mlog_entry_void(); assert_spin_locked(&dlm->spinlock); assert_spin_locked(&dlm->master_lock); /* caller is responsible for any ref taken here on oldmle */ found = dlm_find_mle(dlm, oldmle, (char *)name, namelen); if (found) { struct dlm_master_list_entry *tmp = *oldmle; spin_lock(&tmp->spinlock); if (tmp->type == DLM_MLE_MIGRATION) { if (master == dlm->node_num) { /* ah another process raced me to it */ mlog(0, "tried to migrate %.*s, but some " "process beat me to it\n", namelen, name); ret = -EEXIST; } else { /* bad. 2 NODES are trying to migrate! */ mlog(ML_ERROR, "migration error mle: " "master=%u new_master=%u // request: " "master=%u new_master=%u // " "lockres=%.*s\n", tmp->master, tmp->new_master, master, new_master, namelen, name); BUG(); } } else { /* this is essentially what assert_master does */ tmp->master = master; atomic_set(&tmp->woken, 1); wake_up(&tmp->wq); /* remove it from the list so that only one * mle will be found */ list_del_init(&tmp->list); /* this was obviously WRONG. mle is uninited here. should be tmp. */ __dlm_mle_detach_hb_events(dlm, tmp); ret = DLM_MIGRATE_RESPONSE_MASTERY_REF; mlog(0, "%s:%.*s: master=%u, newmaster=%u, " "telling master to get ref for cleared out mle " "during migration\n", dlm->name, namelen, name, master, new_master); } spin_unlock(&tmp->spinlock); } /* now add a migration mle to the tail of the list */ dlm_init_mle(mle, DLM_MLE_MIGRATION, dlm, res, name, namelen); mle->new_master = new_master; /* the new master will be sending an assert master for this. * at that point we will get the refmap reference */ mle->master = master; /* do this for consistency with other mle types */ set_bit(new_master, mle->maybe_map); list_add(&mle->list, &dlm->master_list); return ret; } void dlm_clean_master_list(struct dlm_ctxt *dlm, u8 dead_node) { struct list_head *iter, *iter2; struct dlm_master_list_entry *mle; struct dlm_lock_resource *res; unsigned int hash; mlog_entry("dlm=%s, dead node=%u\n", dlm->name, dead_node); top: assert_spin_locked(&dlm->spinlock); /* clean the master list */ spin_lock(&dlm->master_lock); list_for_each_safe(iter, iter2, &dlm->master_list) { mle = list_entry(iter, struct dlm_master_list_entry, list); BUG_ON(mle->type != DLM_MLE_BLOCK && mle->type != DLM_MLE_MASTER && mle->type != DLM_MLE_MIGRATION); /* MASTER mles are initiated locally. the waiting * process will notice the node map change * shortly. let that happen as normal. */ if (mle->type == DLM_MLE_MASTER) continue; /* BLOCK mles are initiated by other nodes. * need to clean up if the dead node would have * been the master. */ if (mle->type == DLM_MLE_BLOCK) { int bit; spin_lock(&mle->spinlock); bit = find_next_bit(mle->maybe_map, O2NM_MAX_NODES, 0); if (bit != dead_node) { mlog(0, "mle found, but dead node %u would " "not have been master\n", dead_node); spin_unlock(&mle->spinlock); } else { /* must drop the refcount by one since the * assert_master will never arrive. this * may result in the mle being unlinked and * freed, but there may still be a process * waiting in the dlmlock path which is fine. */ mlog(0, "node %u was expected master\n", dead_node); atomic_set(&mle->woken, 1); spin_unlock(&mle->spinlock); wake_up(&mle->wq); /* do not need events any longer, so detach * from heartbeat */ __dlm_mle_detach_hb_events(dlm, mle); __dlm_put_mle(mle); } continue; } /* everything else is a MIGRATION mle */ /* the rule for MIGRATION mles is that the master * becomes UNKNOWN if *either* the original or * the new master dies. all UNKNOWN lockreses * are sent to whichever node becomes the recovery * master. the new master is responsible for * determining if there is still a master for * this lockres, or if he needs to take over * mastery. either way, this node should expect * another message to resolve this. */ if (mle->master != dead_node && mle->new_master != dead_node) continue; /* if we have reached this point, this mle needs to * be removed from the list and freed. */ /* remove from the list early. NOTE: unlinking * list_head while in list_for_each_safe */ __dlm_mle_detach_hb_events(dlm, mle); spin_lock(&mle->spinlock); list_del_init(&mle->list); atomic_set(&mle->woken, 1); spin_unlock(&mle->spinlock); wake_up(&mle->wq); mlog(0, "%s: node %u died during migration from " "%u to %u!\n", dlm->name, dead_node, mle->master, mle->new_master); /* if there is a lockres associated with this * mle, find it and set its owner to UNKNOWN */ hash = dlm_lockid_hash(mle->u.name.name, mle->u.name.len); res = __dlm_lookup_lockres(dlm, mle->u.name.name, mle->u.name.len, hash); if (res) { /* unfortunately if we hit this rare case, our * lock ordering is messed. we need to drop * the master lock so that we can take the * lockres lock, meaning that we will have to * restart from the head of list. */ spin_unlock(&dlm->master_lock); /* move lockres onto recovery list */ spin_lock(&res->spinlock); dlm_set_lockres_owner(dlm, res, DLM_LOCK_RES_OWNER_UNKNOWN); dlm_move_lockres_to_recovery_list(dlm, res); spin_unlock(&res->spinlock); dlm_lockres_put(res); /* about to get rid of mle, detach from heartbeat */ __dlm_mle_detach_hb_events(dlm, mle); /* dump the mle */ spin_lock(&dlm->master_lock); __dlm_put_mle(mle); spin_unlock(&dlm->master_lock); /* restart */ goto top; } /* this may be the last reference */ __dlm_put_mle(mle); } spin_unlock(&dlm->master_lock); } int dlm_finish_migration(struct dlm_ctxt *dlm, struct dlm_lock_resource *res, u8 old_master) { struct dlm_node_iter iter; int ret = 0; spin_lock(&dlm->spinlock); dlm_node_iter_init(dlm->domain_map, &iter); clear_bit(old_master, iter.node_map); clear_bit(dlm->node_num, iter.node_map); spin_unlock(&dlm->spinlock); /* ownership of the lockres is changing. account for the * mastery reference here since old_master will briefly have * a reference after the migration completes */ spin_lock(&res->spinlock); dlm_lockres_set_refmap_bit(old_master, res); spin_unlock(&res->spinlock); mlog(0, "now time to do a migrate request to other nodes\n"); ret = dlm_do_migrate_request(dlm, res, old_master, dlm->node_num, &iter); if (ret < 0) { mlog_errno(ret); goto leave; } mlog(0, "doing assert master of %.*s to all except the original node\n", res->lockname.len, res->lockname.name); /* this call now finishes out the nodemap * even if one or more nodes die */ ret = dlm_do_assert_master(dlm, res, iter.node_map, DLM_ASSERT_MASTER_FINISH_MIGRATION); if (ret < 0) { /* no longer need to retry. all living nodes contacted. */ mlog_errno(ret); ret = 0; } memset(iter.node_map, 0, sizeof(iter.node_map)); set_bit(old_master, iter.node_map); mlog(0, "doing assert master of %.*s back to %u\n", res->lockname.len, res->lockname.name, old_master); ret = dlm_do_assert_master(dlm, res, iter.node_map, DLM_ASSERT_MASTER_FINISH_MIGRATION); if (ret < 0) { mlog(0, "assert master to original master failed " "with %d.\n", ret); /* the only nonzero status here would be because of * a dead original node. we're done. */ ret = 0; } /* all done, set the owner, clear the flag */ spin_lock(&res->spinlock); dlm_set_lockres_owner(dlm, res, dlm->node_num); res->state &= ~DLM_LOCK_RES_MIGRATING; spin_unlock(&res->spinlock); /* re-dirty it on the new master */ dlm_kick_thread(dlm, res); wake_up(&res->wq); leave: return ret; } /* * LOCKRES AST REFCOUNT * this is integral to migration */ /* for future intent to call an ast, reserve one ahead of time. * this should be called only after waiting on the lockres * with dlm_wait_on_lockres, and while still holding the * spinlock after the call. */ void __dlm_lockres_reserve_ast(struct dlm_lock_resource *res) { assert_spin_locked(&res->spinlock); if (res->state & DLM_LOCK_RES_MIGRATING) { __dlm_print_one_lock_resource(res); } BUG_ON(res->state & DLM_LOCK_RES_MIGRATING); atomic_inc(&res->asts_reserved); } /* * used to drop the reserved ast, either because it went unused, * or because the ast/bast was actually called. * * also, if there is a pending migration on this lockres, * and this was the last pending ast on the lockres, * atomically set the MIGRATING flag before we drop the lock. * this is how we ensure that migration can proceed with no * asts in progress. note that it is ok if the state of the * queues is such that a lock should be granted in the future * or that a bast should be fired, because the new master will * shuffle the lists on this lockres as soon as it is migrated. */ void dlm_lockres_release_ast(struct dlm_ctxt *dlm, struct dlm_lock_resource *res) { if (!atomic_dec_and_lock(&res->asts_reserved, &res->spinlock)) return; if (!res->migration_pending) { spin_unlock(&res->spinlock); return; } BUG_ON(res->state & DLM_LOCK_RES_MIGRATING); res->migration_pending = 0; res->state |= DLM_LOCK_RES_MIGRATING; spin_unlock(&res->spinlock); wake_up(&res->wq); wake_up(&dlm->migration_wq); }