/* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*- * vim:expandtab:shiftwidth=8:tabstop=8: * * Copyright (C) 2002, 2003 Cluster File Systems, Inc. * Author: Phil Schwan * Author: Peter Braam * * 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_LDLM #ifdef __KERNEL__ # include #else # include #endif #include #include "ldlm_internal.h" kmem_cache_t *ldlm_resource_slab, *ldlm_lock_slab; DECLARE_MUTEX(ldlm_namespace_lock); struct list_head ldlm_namespace_list = LIST_HEAD_INIT(ldlm_namespace_list); struct proc_dir_entry *ldlm_type_proc_dir = NULL; struct proc_dir_entry *ldlm_ns_proc_dir = NULL; struct proc_dir_entry *ldlm_svc_proc_dir = NULL; #ifdef __KERNEL__ static int ldlm_proc_dump_ns(struct file *file, const char *buffer, unsigned long count, void *data) { ldlm_dump_all_namespaces(D_DLMTRACE); RETURN(count); } int ldlm_proc_setup(void) { int rc; struct lprocfs_vars list[] = { { "dump_namespaces", NULL, ldlm_proc_dump_ns, NULL }, { NULL }}; ENTRY; LASSERT(ldlm_ns_proc_dir == NULL); ldlm_type_proc_dir = lprocfs_register(OBD_LDLM_DEVICENAME, proc_lustre_root, NULL, NULL); if (IS_ERR(ldlm_type_proc_dir)) { CERROR("LProcFS failed in ldlm-init\n"); rc = PTR_ERR(ldlm_type_proc_dir); GOTO(err, rc); } ldlm_ns_proc_dir = lprocfs_register("namespaces", ldlm_type_proc_dir, NULL, NULL); if (IS_ERR(ldlm_ns_proc_dir)) { CERROR("LProcFS failed in ldlm-init\n"); rc = PTR_ERR(ldlm_ns_proc_dir); GOTO(err_type, rc); } ldlm_svc_proc_dir = lprocfs_register("services", ldlm_type_proc_dir, NULL, NULL); if (IS_ERR(ldlm_svc_proc_dir)) { CERROR("LProcFS failed in ldlm-init\n"); rc = PTR_ERR(ldlm_svc_proc_dir); GOTO(err_ns, rc); } rc = lprocfs_add_vars(ldlm_type_proc_dir, list, NULL); RETURN(0); err_ns: lprocfs_remove(ldlm_ns_proc_dir); err_type: lprocfs_remove(ldlm_type_proc_dir); err: ldlm_type_proc_dir = NULL; ldlm_ns_proc_dir = NULL; ldlm_svc_proc_dir = NULL; RETURN(rc); } void ldlm_proc_cleanup(void) { if (ldlm_svc_proc_dir) { lprocfs_remove(ldlm_svc_proc_dir); ldlm_svc_proc_dir = NULL; } if (ldlm_ns_proc_dir) { lprocfs_remove(ldlm_ns_proc_dir); ldlm_ns_proc_dir = NULL; } if (ldlm_type_proc_dir) { lprocfs_remove(ldlm_type_proc_dir); ldlm_type_proc_dir = NULL; } } static int lprocfs_uint_rd(char *page, char **start, off_t off, int count, int *eof, void *data) { unsigned int *temp = (unsigned int *)data; return snprintf(page, count, "%u\n", *temp); } static int lprocfs_read_lru_size(char *page, char **start, off_t off, int count, int *eof, void *data) { struct ldlm_namespace *ns = data; return lprocfs_uint_rd(page, start, off, count, eof, &ns->ns_max_unused); } #define MAX_STRING_SIZE 128 static int lprocfs_write_lru_size(struct file *file, const char *buffer, unsigned long count, void *data) { struct ldlm_namespace *ns = data; char dummy[MAX_STRING_SIZE + 1], *end; unsigned long tmp; dummy[MAX_STRING_SIZE] = '\0'; copy_from_user(dummy, buffer, MAX_STRING_SIZE); if (count == 6 && memcmp(dummy, "clear", 5) == 0) { CDEBUG(D_DLMTRACE, "dropping all unused locks from namespace %s\n", ns->ns_name); tmp = ns->ns_max_unused; ns->ns_max_unused = 0; ldlm_cancel_lru(ns, LDLM_SYNC); ns->ns_max_unused = tmp; return count; } tmp = simple_strtoul(dummy, &end, 0); if (tmp == 0 && *end) { CERROR("invalid value written\n"); return -EINVAL; } CDEBUG(D_DLMTRACE, "changing namespace %s max_unused from %u to %u\n", ns->ns_name, ns->ns_max_unused, (unsigned int)tmp); ns->ns_max_unused = (unsigned int)tmp; ldlm_cancel_lru(ns, LDLM_ASYNC); return count; } void ldlm_proc_namespace(struct ldlm_namespace *ns) { struct lprocfs_vars lock_vars[2]; char lock_name[MAX_STRING_SIZE + 1]; LASSERT(ns != NULL); LASSERT(ns->ns_name != NULL); lock_name[MAX_STRING_SIZE] = '\0'; memset(lock_vars, 0, sizeof(lock_vars)); lock_vars[0].read_fptr = lprocfs_rd_u64; lock_vars[0].name = lock_name; snprintf(lock_name, MAX_STRING_SIZE, "%s/resource_count", ns->ns_name); lock_vars[0].data = &ns->ns_resources; lprocfs_add_vars(ldlm_ns_proc_dir, lock_vars, 0); snprintf(lock_name, MAX_STRING_SIZE, "%s/lock_count", ns->ns_name); lock_vars[0].data = &ns->ns_locks; lprocfs_add_vars(ldlm_ns_proc_dir, lock_vars, 0); if (ns->ns_client) { snprintf(lock_name, MAX_STRING_SIZE, "%s/lock_unused_count", ns->ns_name); lock_vars[0].data = &ns->ns_nr_unused; lock_vars[0].read_fptr = lprocfs_uint_rd; lprocfs_add_vars(ldlm_ns_proc_dir, lock_vars, 0); snprintf(lock_name, MAX_STRING_SIZE, "%s/lru_size", ns->ns_name); lock_vars[0].data = ns; lock_vars[0].read_fptr = lprocfs_read_lru_size; lock_vars[0].write_fptr = lprocfs_write_lru_size; lprocfs_add_vars(ldlm_ns_proc_dir, lock_vars, 0); } } #endif #undef MAX_STRING_SIZE struct ldlm_namespace *ldlm_namespace_new(char *name, __u32 client) { struct ldlm_namespace *ns = NULL; struct list_head *bucket; int rc; ENTRY; rc = ldlm_get_ref(); if (rc) { CERROR("ldlm_get_ref failed: %d\n", rc); RETURN(NULL); } OBD_ALLOC(ns, sizeof(*ns)); if (!ns) GOTO(out_ref, NULL); OBD_VMALLOC(ns->ns_hash, sizeof(*ns->ns_hash) * RES_HASH_SIZE); if (!ns->ns_hash) GOTO(out_ns, NULL); OBD_ALLOC(ns->ns_name, strlen(name) + 1); if (!ns->ns_name) GOTO(out_hash, NULL); strcpy(ns->ns_name, name); INIT_LIST_HEAD(&ns->ns_root_list); l_lock_init(&ns->ns_lock); ns->ns_refcount = 0; ns->ns_client = client; spin_lock_init(&ns->ns_counter_lock); ns->ns_locks = 0; ns->ns_resources = 0; init_waitqueue_head(&ns->ns_waitq); for (bucket = ns->ns_hash + RES_HASH_SIZE - 1; bucket >= ns->ns_hash; bucket--) INIT_LIST_HEAD(bucket); INIT_LIST_HEAD(&ns->ns_unused_list); ns->ns_nr_unused = 0; ns->ns_max_unused = LDLM_DEFAULT_LRU_SIZE; down(&ldlm_namespace_lock); list_add(&ns->ns_list_chain, &ldlm_namespace_list); up(&ldlm_namespace_lock); #ifdef __KERNEL__ ldlm_proc_namespace(ns); #endif RETURN(ns); out_hash: POISON(ns->ns_hash, 0x5a, sizeof(*ns->ns_hash) * RES_HASH_SIZE); OBD_VFREE(ns->ns_hash, sizeof(*ns->ns_hash) * RES_HASH_SIZE); out_ns: OBD_FREE(ns, sizeof(*ns)); out_ref: ldlm_put_ref(0); RETURN(NULL); } extern struct ldlm_lock *ldlm_lock_get(struct ldlm_lock *lock); /* If flags contains FL_LOCAL_ONLY, don't try to tell the server, just cleanup. * This is currently only used for recovery, and we make certain assumptions * as a result--notably, that we shouldn't cancel locks with refs. -phil * * Called with the ns_lock held. */ static void cleanup_resource(struct ldlm_resource *res, struct list_head *q, int flags) { struct list_head *tmp, *pos; int rc = 0, client = res->lr_namespace->ns_client; int local_only = (flags & LDLM_FL_LOCAL_ONLY); ENTRY; list_for_each_safe(tmp, pos, q) { struct ldlm_lock *lock; lock = list_entry(tmp, struct ldlm_lock, l_res_link); LDLM_LOCK_GET(lock); /* Set CBPENDING so nothing in the cancellation path * can match this lock */ lock->l_flags |= LDLM_FL_CBPENDING; lock->l_flags |= LDLM_FL_FAILED; lock->l_flags |= flags; if (local_only && (lock->l_readers || lock->l_writers)) { /* This is a little bit gross, but much better than the * alternative: pretend that we got a blocking AST from * the server, so that when the lock is decref'd, it * will go away ... */ /* ... without sending a CANCEL message. */ lock->l_flags |= LDLM_FL_LOCAL_ONLY; LDLM_DEBUG(lock, "setting FL_LOCAL_ONLY"); if (lock->l_completion_ast) lock->l_completion_ast(lock, 0, NULL); LDLM_LOCK_PUT(lock); continue; } if (client) { struct lustre_handle lockh; ldlm_lock2handle(lock, &lockh); if (!local_only) { rc = ldlm_cli_cancel(&lockh); if (rc) CERROR("ldlm_cli_cancel: %d\n", rc); } /* Force local cleanup on errors, too. */ if (local_only || rc != ELDLM_OK) ldlm_lock_cancel(lock); } else { LDLM_DEBUG(lock, "Freeing a lock still held by a " "client node"); ldlm_resource_unlink_lock(lock); ldlm_lock_destroy(lock); } LDLM_LOCK_PUT(lock); } EXIT; } int ldlm_namespace_cleanup(struct ldlm_namespace *ns, int flags) { int i; if (ns == NULL) { CDEBUG(D_INFO, "NULL ns, skipping cleanup\n"); return ELDLM_OK; } l_lock(&ns->ns_lock); for (i = 0; i < RES_HASH_SIZE; i++) { struct list_head *tmp, *pos; list_for_each_safe(tmp, pos, &(ns->ns_hash[i])) { struct ldlm_resource *res; res = list_entry(tmp, struct ldlm_resource, lr_hash); ldlm_resource_getref(res); cleanup_resource(res, &res->lr_granted, flags); cleanup_resource(res, &res->lr_converting, flags); cleanup_resource(res, &res->lr_waiting, flags); /* XXX what a mess: don't force cleanup if we're * local_only (which is only used by recovery). In that * case, we probably still have outstanding lock refs * which reference these resources. -phil */ if (!ldlm_resource_putref(res) && !(flags & LDLM_FL_LOCAL_ONLY)) { CERROR("Resource refcount nonzero (%d) after " "lock cleanup; forcing cleanup.\n", atomic_read(&res->lr_refcount)); ldlm_resource_dump(D_ERROR, res); atomic_set(&res->lr_refcount, 1); ldlm_resource_putref(res); } } } l_unlock(&ns->ns_lock); return ELDLM_OK; } /* Cleanup, but also free, the namespace */ int ldlm_namespace_free(struct ldlm_namespace *ns, int force) { if (!ns) RETURN(ELDLM_OK); down(&ldlm_namespace_lock); list_del(&ns->ns_list_chain); up(&ldlm_namespace_lock); /* At shutdown time, don't call the cancellation callback */ ldlm_namespace_cleanup(ns, 0); #ifdef __KERNEL__ { struct proc_dir_entry *dir; dir = lprocfs_srch(ldlm_ns_proc_dir, ns->ns_name); if (dir == NULL) { CERROR("dlm namespace %s has no procfs dir?\n", ns->ns_name); } else { lprocfs_remove(dir); } } #endif POISON(ns->ns_hash, 0x5a, sizeof(*ns->ns_hash) * RES_HASH_SIZE); OBD_VFREE(ns->ns_hash, sizeof(*ns->ns_hash) * RES_HASH_SIZE); OBD_FREE(ns->ns_name, strlen(ns->ns_name) + 1); OBD_FREE(ns, sizeof(*ns)); ldlm_put_ref(force); return ELDLM_OK; } static __u32 ldlm_hash_fn(struct ldlm_resource *parent, struct ldlm_res_id name) { __u32 hash = 0; int i; for (i = 0; i < RES_NAME_SIZE; i++) hash += name.name[i]; hash += (__u32)((unsigned long)parent >> 4); return (hash & RES_HASH_MASK); } static struct ldlm_resource *ldlm_resource_new(void) { struct ldlm_resource *res; OBD_SLAB_ALLOC(res, ldlm_resource_slab, SLAB_NOFS, sizeof *res); if (res == NULL) return NULL; memset(res, 0, sizeof(*res)); INIT_LIST_HEAD(&res->lr_children); INIT_LIST_HEAD(&res->lr_childof); INIT_LIST_HEAD(&res->lr_granted); INIT_LIST_HEAD(&res->lr_converting); INIT_LIST_HEAD(&res->lr_waiting); sema_init(&res->lr_lvb_sem, 1); atomic_set(&res->lr_refcount, 1); return res; } /* Args: locked namespace * Returns: newly-allocated, referenced, unlocked resource */ static struct ldlm_resource * ldlm_resource_add(struct ldlm_namespace *ns, struct ldlm_resource *parent, struct ldlm_res_id name, __u32 type) { struct list_head *bucket; struct ldlm_resource *res; ENTRY; LASSERTF(type >= LDLM_MIN_TYPE && type <= LDLM_MAX_TYPE, "type: %d", type); res = ldlm_resource_new(); if (!res) RETURN(NULL); spin_lock(&ns->ns_counter_lock); ns->ns_resources++; spin_unlock(&ns->ns_counter_lock); l_lock(&ns->ns_lock); memcpy(&res->lr_name, &name, sizeof(res->lr_name)); res->lr_namespace = ns; ns->ns_refcount++; res->lr_type = type; res->lr_most_restr = LCK_NL; bucket = ns->ns_hash + ldlm_hash_fn(parent, name); list_add(&res->lr_hash, bucket); if (parent == NULL) { list_add(&res->lr_childof, &ns->ns_root_list); } else { res->lr_parent = parent; list_add(&res->lr_childof, &parent->lr_children); } l_unlock(&ns->ns_lock); RETURN(res); } /* Args: unlocked namespace * Locks: takes and releases ns->ns_lock and res->lr_lock * Returns: referenced, unlocked ldlm_resource or NULL */ struct ldlm_resource * ldlm_resource_get(struct ldlm_namespace *ns, struct ldlm_resource *parent, struct ldlm_res_id name, __u32 type, int create) { struct list_head *bucket, *tmp; struct ldlm_resource *res = NULL; ENTRY; LASSERT(ns != NULL); LASSERT(ns->ns_hash != NULL); LASSERT(name.name[0] != 0); l_lock(&ns->ns_lock); bucket = ns->ns_hash + ldlm_hash_fn(parent, name); list_for_each(tmp, bucket) { res = list_entry(tmp, struct ldlm_resource, lr_hash); if (memcmp(&res->lr_name, &name, sizeof(res->lr_name)) == 0) { ldlm_resource_getref(res); l_unlock(&ns->ns_lock); RETURN(res); } } if (create) { res = ldlm_resource_add(ns, parent, name, type); if (res == NULL) GOTO(out, NULL); } else { res = NULL; } if (create && ns->ns_lvbo && ns->ns_lvbo->lvbo_init) { int rc; /* Although this is technically a lock inversion risk (lvb_sem * should be taken before DLM lock), this resource was just * created, so nobody else can take the lvb_sem yet. -p */ down(&res->lr_lvb_sem); /* Drop the dlm lock, because lvbo_init can touch the disk */ l_unlock(&ns->ns_lock); OBD_FAIL_TIMEOUT(OBD_FAIL_LDLM_CREATE_RESOURCE, 2); rc = ns->ns_lvbo->lvbo_init(res); up(&res->lr_lvb_sem); if (rc) CERROR("lvbo_init failed for resource "LPU64": rc %d\n", name.name[0], rc); } else { out: l_unlock(&ns->ns_lock); } RETURN(res); } struct ldlm_resource *ldlm_resource_getref(struct ldlm_resource *res) { LASSERT(res != NULL); LASSERT(res != LP_POISON); atomic_inc(&res->lr_refcount); CDEBUG(D_INFO, "getref res: %p count: %d\n", res, atomic_read(&res->lr_refcount)); return res; } /* Returns 1 if the resource was freed, 0 if it remains. */ int ldlm_resource_putref(struct ldlm_resource *res) { int rc = 0; ENTRY; CDEBUG(D_INFO, "putref res: %p count: %d\n", res, atomic_read(&res->lr_refcount) - 1); LASSERT(atomic_read(&res->lr_refcount) > 0); LASSERT(atomic_read(&res->lr_refcount) < LI_POISON); if (atomic_dec_and_test(&res->lr_refcount)) { struct ldlm_namespace *ns = res->lr_namespace; ENTRY; l_lock(&ns->ns_lock); if (atomic_read(&res->lr_refcount) != 0) { /* We lost the race. */ l_unlock(&ns->ns_lock); RETURN(rc); } if (!list_empty(&res->lr_granted)) { ldlm_resource_dump(D_ERROR, res); LBUG(); } if (!list_empty(&res->lr_converting)) { ldlm_resource_dump(D_ERROR, res); LBUG(); } if (!list_empty(&res->lr_waiting)) { ldlm_resource_dump(D_ERROR, res); LBUG(); } if (!list_empty(&res->lr_children)) { ldlm_resource_dump(D_ERROR, res); LBUG(); } ns->ns_refcount--; list_del_init(&res->lr_hash); list_del_init(&res->lr_childof); if (res->lr_lvb_data) OBD_FREE(res->lr_lvb_data, res->lr_lvb_len); l_unlock(&ns->ns_lock); OBD_SLAB_FREE(res, ldlm_resource_slab, sizeof *res); spin_lock(&ns->ns_counter_lock); ns->ns_resources--; if (ns->ns_resources == 0) wake_up(&ns->ns_waitq); spin_unlock(&ns->ns_counter_lock); rc = 1; EXIT; } RETURN(rc); } void ldlm_resource_add_lock(struct ldlm_resource *res, struct list_head *head, struct ldlm_lock *lock) { l_lock(&res->lr_namespace->ns_lock); ldlm_resource_dump(D_OTHER, res); CDEBUG(D_OTHER, "About to add this lock:\n"); ldlm_lock_dump(D_OTHER, lock, 0); if (lock->l_destroyed) { CDEBUG(D_OTHER, "Lock destroyed, not adding to resource\n"); goto out; } LASSERT(list_empty(&lock->l_res_link)); list_add_tail(&lock->l_res_link, head); out: l_unlock(&res->lr_namespace->ns_lock); } void ldlm_resource_insert_lock_after(struct ldlm_lock *original, struct ldlm_lock *new) { struct ldlm_resource *res = original->l_resource; l_lock(&res->lr_namespace->ns_lock); ldlm_resource_dump(D_OTHER, res); CDEBUG(D_OTHER, "About to insert this lock after %p:\n", original); ldlm_lock_dump(D_OTHER, new, 0); if (new->l_destroyed) { CDEBUG(D_OTHER, "Lock destroyed, not adding to resource\n"); goto out; } LASSERT(list_empty(&new->l_res_link)); list_add(&new->l_res_link, &original->l_res_link); out: l_unlock(&res->lr_namespace->ns_lock); } void ldlm_resource_unlink_lock(struct ldlm_lock *lock) { l_lock(&lock->l_resource->lr_namespace->ns_lock); list_del_init(&lock->l_res_link); l_unlock(&lock->l_resource->lr_namespace->ns_lock); } void ldlm_res2desc(struct ldlm_resource *res, struct ldlm_resource_desc *desc) { desc->lr_type = res->lr_type; memcpy(&desc->lr_name, &res->lr_name, sizeof(desc->lr_name)); } void ldlm_dump_all_namespaces(int level) { struct list_head *tmp; down(&ldlm_namespace_lock); list_for_each(tmp, &ldlm_namespace_list) { struct ldlm_namespace *ns; ns = list_entry(tmp, struct ldlm_namespace, ns_list_chain); ldlm_namespace_dump(level, ns); } up(&ldlm_namespace_lock); } void ldlm_namespace_dump(int level, struct ldlm_namespace *ns) { struct list_head *tmp; CDEBUG(level, "--- Namespace: %s (rc: %d, client: %d)\n", ns->ns_name, ns->ns_refcount, ns->ns_client); l_lock(&ns->ns_lock); if (time_after(jiffies, ns->ns_next_dump)) { list_for_each(tmp, &ns->ns_root_list) { struct ldlm_resource *res; res = list_entry(tmp, struct ldlm_resource, lr_childof); /* Once we have resources with children, this should * really dump them recursively. */ ldlm_resource_dump(level, res); } ns->ns_next_dump = jiffies + 10 * HZ; } l_unlock(&ns->ns_lock); } void ldlm_resource_dump(int level, struct ldlm_resource *res) { struct list_head *tmp; int pos; if (RES_NAME_SIZE != 4) LBUG(); CDEBUG(level, "--- Resource: %p ("LPU64"/"LPU64"/"LPU64"/"LPU64 ") (rc: %d)\n", res, res->lr_name.name[0], res->lr_name.name[1], res->lr_name.name[2], res->lr_name.name[3], atomic_read(&res->lr_refcount)); if (!list_empty(&res->lr_granted)) { pos = 0; CDEBUG(level, "Granted locks:\n"); list_for_each(tmp, &res->lr_granted) { struct ldlm_lock *lock; lock = list_entry(tmp, struct ldlm_lock, l_res_link); ldlm_lock_dump(level, lock, ++pos); } } if (!list_empty(&res->lr_converting)) { pos = 0; CDEBUG(level, "Converting locks:\n"); list_for_each(tmp, &res->lr_converting) { struct ldlm_lock *lock; lock = list_entry(tmp, struct ldlm_lock, l_res_link); ldlm_lock_dump(level, lock, ++pos); } } if (!list_empty(&res->lr_waiting)) { pos = 0; CDEBUG(level, "Waiting locks:\n"); list_for_each(tmp, &res->lr_waiting) { struct ldlm_lock *lock; lock = list_entry(tmp, struct ldlm_lock, l_res_link); ldlm_lock_dump(level, lock, ++pos); } } }