X-Git-Url: https://git.whamcloud.com/?a=blobdiff_plain;f=lustre%2Fldlm%2Fldlm_extent.c;h=15414165916e559e5dca3b2f6d679d16c7a95518;hb=9d3ad9fba3e244735f0994c3910e9de00e2a7f4d;hp=5a84909cd9d87b7a4e5d21f03b08ec4efb4560fc;hpb=93acd158c57c4c5d0fc751d46741231490c04707;p=fs%2Flustre-release.git diff --git a/lustre/ldlm/ldlm_extent.c b/lustre/ldlm/ldlm_extent.c index 5a84909..1541416 100644 --- a/lustre/ldlm/ldlm_extent.c +++ b/lustre/ldlm/ldlm_extent.c @@ -1,99 +1,900 @@ /* -*- mode: c; c-basic-offset: 8; indent-tabs-mode: nil; -*- * vim:expandtab:shiftwidth=8:tabstop=8: * - * Copyright (C) 2002 Cluster File Systems, Inc. + * GPL HEADER START * - * This code is issued under the GNU General Public License. - * See the file COPYING in this distribution + * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER. * - * by Cluster File Systems, Inc. - * authors, Peter Braam & - * Phil Schwan + * This program is free software; you can redistribute it and/or modify + * it under the terms of the GNU General Public License version 2 only, + * as published by the Free Software Foundation. + * + * 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 version 2 for more details (a copy is included + * in the LICENSE file that accompanied this code). + * + * You should have received a copy of the GNU General Public License + * version 2 along with this program; If not, see + * http://www.sun.com/software/products/lustre/docs/GPLv2.pdf + * + * Please contact Sun Microsystems, Inc., 4150 Network Circle, Santa Clara, + * CA 95054 USA or visit www.sun.com if you need additional information or + * have any questions. + * + * GPL HEADER END + */ +/* + * Copyright 2008 Sun Microsystems, Inc. All rights reserved + * Use is subject to license terms. + */ +/* + * This file is part of Lustre, http://www.lustre.org/ + * Lustre is a trademark of Sun Microsystems, Inc. + * + * lustre/ldlm/ldlm_extent.c + * + * Author: Peter Braam + * Author: Phil Schwan */ #define DEBUG_SUBSYSTEM S_LDLM +#ifndef __KERNEL__ +# include +#else +# include +#endif -#include +#include +#include +#include +#include -/* This function will be called to judge if the granted queue of another child - * (read: another extent) is conflicting and needs its granted queue walked to - * issue callbacks. - * - * This helps to find conflicts between read and write locks on overlapping - * extents. */ -int ldlm_extent_compat(struct ldlm_lock *a, struct ldlm_lock *b) +#include "ldlm_internal.h" + +#define LDLM_MAX_GROWN_EXTENT (32 * 1024 * 1024 - 1) + +/* fixup the ldlm_extent after expanding */ +static void ldlm_extent_internal_policy_fixup(struct ldlm_lock *req, + struct ldlm_extent *new_ex, + int conflicting) { - if (MAX(a->l_extent.start, b->l_extent.start) <= - MIN(a->l_extent.end, b->l_extent.end)) - RETURN(0); + ldlm_mode_t req_mode = req->l_req_mode; + __u64 req_start = req->l_req_extent.start; + __u64 req_end = req->l_req_extent.end; + __u64 req_align, mask; + + if (conflicting > 32 && (req_mode == LCK_PW || req_mode == LCK_CW)) { + if (req_end < req_start + LDLM_MAX_GROWN_EXTENT) + new_ex->end = min(req_start + LDLM_MAX_GROWN_EXTENT, + new_ex->end); + } + + if (new_ex->start == 0 && new_ex->end == OBD_OBJECT_EOF) { + EXIT; + return; + } - RETURN(1); + /* we need to ensure that the lock extent is properly aligned to what + * the client requested. We align it to the lowest-common denominator + * of the clients requested lock start and end alignment. */ + mask = 0x1000ULL; + req_align = (req_end + 1) | req_start; + if (req_align != 0) { + while ((req_align & mask) == 0) + mask <<= 1; + } + mask -= 1; + /* We can only shrink the lock, not grow it. + * This should never cause lock to be smaller than requested, + * since requested lock was already aligned on these boundaries. */ + new_ex->start = ((new_ex->start - 1) | mask) + 1; + new_ex->end = ((new_ex->end + 1) & ~mask) - 1; + LASSERTF(new_ex->start <= req_start, + "mask "LPX64" grant start "LPU64" req start "LPU64"\n", + mask, new_ex->start, req_start); + LASSERTF(new_ex->end >= req_end, + "mask "LPX64" grant end "LPU64" req end "LPU64"\n", + mask, new_ex->end, req_end); } /* The purpose of this function is to return: * - the maximum extent * - containing the requested extent - * - and not overlapping existing extents outside the requested one - * - * An alternative policy is to not shrink the new extent when conflicts exist. + * - and not overlapping existing conflicting extents outside the requested one * - * To reconstruct our formulas, take a deep breath. */ -static void policy_internal(struct list_head *queue, struct ldlm_extent *req_ex, - struct ldlm_extent *new_ex, ldlm_mode_t mode) + * Use interval tree to expand the lock extent for granted lock. + */ +static void ldlm_extent_internal_policy_granted(struct ldlm_lock *req, + struct ldlm_extent *new_ex) +{ + struct ldlm_resource *res = req->l_resource; + ldlm_mode_t req_mode = req->l_req_mode; + __u64 req_start = req->l_req_extent.start; + __u64 req_end = req->l_req_extent.end; + struct ldlm_interval_tree *tree; + struct interval_node_extent limiter = { new_ex->start, new_ex->end }; + int conflicting = 0; + int idx; + ENTRY; + + lockmode_verify(req_mode); + + /* using interval tree to handle the ldlm extent granted locks */ + for (idx = 0; idx < LCK_MODE_NUM; idx++) { + struct interval_node_extent ext = { req_start, req_end }; + + tree = &res->lr_itree[idx]; + if (lockmode_compat(tree->lit_mode, req_mode)) + continue; + + conflicting += tree->lit_size; + if (conflicting > 4) + limiter.start = req_start; + + if (interval_is_overlapped(tree->lit_root, &ext)) + CDEBUG(D_INFO, + "req_mode = %d, tree->lit_mode = %d, " + "tree->lit_size = %d\n", + req_mode, tree->lit_mode, tree->lit_size); + interval_expand(tree->lit_root, &ext, &limiter); + limiter.start = max(limiter.start, ext.start); + limiter.end = min(limiter.end, ext.end); + if (limiter.start == req_start && limiter.end == req_end) + break; + } + + new_ex->start = limiter.start; + new_ex->end = limiter.end; + LASSERT(new_ex->start <= req_start); + LASSERT(new_ex->end >= req_end); + + ldlm_extent_internal_policy_fixup(req, new_ex, conflicting); + EXIT; +} + +/* The purpose of this function is to return: + * - the maximum extent + * - containing the requested extent + * - and not overlapping existing conflicting extents outside the requested one + */ +static void +ldlm_extent_internal_policy_waiting(struct ldlm_lock *req, + struct ldlm_extent *new_ex) { struct list_head *tmp; + struct ldlm_resource *res = req->l_resource; + ldlm_mode_t req_mode = req->l_req_mode; + __u64 req_start = req->l_req_extent.start; + __u64 req_end = req->l_req_extent.end; + int conflicting = 0; + ENTRY; + + lockmode_verify(req_mode); - list_for_each(tmp, queue) { + /* for waiting locks */ + list_for_each(tmp, &res->lr_waiting) { struct ldlm_lock *lock; + struct ldlm_extent *l_extent; + lock = list_entry(tmp, struct ldlm_lock, l_res_link); + l_extent = &lock->l_policy_data.l_extent; + + /* We already hit the minimum requested size, search no more */ + if (new_ex->start == req_start && new_ex->end == req_end) { + EXIT; + return; + } + + /* Don't conflict with ourselves */ + if (req == lock) + continue; + + /* Locks are compatible, overlap doesn't matter */ + /* Until bug 20 is fixed, try to avoid granting overlapping + * locks on one client (they take a long time to cancel) */ + if (lockmode_compat(lock->l_req_mode, req_mode) && + lock->l_export != req->l_export) + continue; + + /* If this is a high-traffic lock, don't grow downwards at all + * or grow upwards too much */ + ++conflicting; + if (conflicting > 4) + new_ex->start = req_start; + + /* If lock doesn't overlap new_ex, skip it. */ + if (!ldlm_extent_overlap(l_extent, new_ex)) + continue; + + /* Locks conflicting in requested extents and we can't satisfy + * both locks, so ignore it. Either we will ping-pong this + * extent (we would regardless of what extent we granted) or + * lock is unused and it shouldn't limit our extent growth. */ + if (ldlm_extent_overlap(&lock->l_req_extent,&req->l_req_extent)) + continue; + + /* We grow extents downwards only as far as they don't overlap + * with already-granted locks, on the assumption that clients + * will be writing beyond the initial requested end and would + * then need to enqueue a new lock beyond previous request. + * l_req_extent->end strictly < req_start, checked above. */ + if (l_extent->start < req_start && new_ex->start != req_start) { + if (l_extent->end >= req_start) + new_ex->start = req_start; + else + new_ex->start = min(l_extent->end+1, req_start); + } + + /* If we need to cancel this lock anyways because our request + * overlaps the granted lock, we grow up to its requested + * extent start instead of limiting this extent, assuming that + * clients are writing forwards and the lock had over grown + * its extent downwards before we enqueued our request. */ + if (l_extent->end > req_end) { + if (l_extent->start <= req_end) + new_ex->end = max(lock->l_req_extent.start - 1, + req_end); + else + new_ex->end = max(l_extent->start - 1, req_end); + } + } + + ldlm_extent_internal_policy_fixup(req, new_ex, conflicting); + EXIT; +} + + +/* In order to determine the largest possible extent we can grant, we need + * to scan all of the queues. */ +static void ldlm_extent_policy(struct ldlm_resource *res, + struct ldlm_lock *lock, int *flags) +{ + struct ldlm_extent new_ex = { .start = 0, .end = OBD_OBJECT_EOF }; + + if (lock->l_export == NULL) + /* + * this is local lock taken by server (e.g., as a part of + * OST-side locking, or unlink handling). Expansion doesn't + * make a lot of sense for local locks, because they are + * dropped immediately on operation completion and would only + * conflict with other threads. + */ + return; + + if (lock->l_policy_data.l_extent.start == 0 && + lock->l_policy_data.l_extent.end == OBD_OBJECT_EOF) + /* fast-path whole file locks */ + return; + + ldlm_extent_internal_policy_granted(lock, &new_ex); + ldlm_extent_internal_policy_waiting(lock, &new_ex); + + if (new_ex.start != lock->l_policy_data.l_extent.start || + new_ex.end != lock->l_policy_data.l_extent.end) { + *flags |= LDLM_FL_LOCK_CHANGED; + lock->l_policy_data.l_extent.start = new_ex.start; + lock->l_policy_data.l_extent.end = new_ex.end; + } +} + +static int ldlm_check_contention(struct ldlm_lock *lock, int contended_locks) +{ + struct ldlm_resource *res = lock->l_resource; + cfs_time_t now = cfs_time_current(); + + if (OBD_FAIL_CHECK(OBD_FAIL_LDLM_SET_CONTENTION)) + return 1; + + CDEBUG(D_DLMTRACE, "contended locks = %d\n", contended_locks); + if (contended_locks > res->lr_namespace->ns_contended_locks) + res->lr_contention_time = now; + return cfs_time_before(now, cfs_time_add(res->lr_contention_time, + cfs_time_seconds(res->lr_namespace->ns_contention_time))); +} + +struct ldlm_extent_compat_args { + struct list_head *work_list; + struct ldlm_lock *lock; + ldlm_mode_t mode; + int *locks; + int *compat; +}; + +static enum interval_iter ldlm_extent_compat_cb(struct interval_node *n, + void *data) +{ + struct ldlm_extent_compat_args *priv = data; + struct ldlm_interval *node = to_ldlm_interval(n); + struct ldlm_extent *extent; + struct list_head *work_list = priv->work_list; + struct ldlm_lock *lock, *enq = priv->lock; + ldlm_mode_t mode = priv->mode; + int count = 0; + ENTRY; + + LASSERT(!list_empty(&node->li_group)); + + list_for_each_entry(lock, &node->li_group, l_sl_policy) { + /* interval tree is for granted lock */ + LASSERTF(mode == lock->l_granted_mode, + "mode = %s, lock->l_granted_mode = %s\n", + ldlm_lockname[mode], + ldlm_lockname[lock->l_granted_mode]); + count++; + if (lock->l_blocking_ast) + ldlm_add_ast_work_item(lock, enq, work_list); + } + + /* don't count conflicting glimpse locks */ + extent = ldlm_interval_extent(node); + if (!(mode == LCK_PR && + extent->start == 0 && extent->end == OBD_OBJECT_EOF)) + *priv->locks += count; + + if (priv->compat) + *priv->compat = 0; + + RETURN(INTERVAL_ITER_CONT); +} + +/* Determine if the lock is compatible with all locks on the queue. + * We stop walking the queue if we hit ourselves so we don't take + * conflicting locks enqueued after us into accound, or we'd wait forever. + * + * 0 if the lock is not compatible + * 1 if the lock is compatible + * 2 if this group lock is compatible and requires no further checking + * negative error, such as EWOULDBLOCK for group locks + */ +static int +ldlm_extent_compat_queue(struct list_head *queue, struct ldlm_lock *req, + int *flags, ldlm_error_t *err, + struct list_head *work_list, int *contended_locks) +{ + struct list_head *tmp; + struct ldlm_lock *lock; + struct ldlm_resource *res = req->l_resource; + ldlm_mode_t req_mode = req->l_req_mode; + __u64 req_start = req->l_req_extent.start; + __u64 req_end = req->l_req_extent.end; + int compat = 1; + int scan = 0; + int check_contention; + ENTRY; + + lockmode_verify(req_mode); + + /* Using interval tree for granted lock */ + if (queue == &res->lr_granted) { + struct ldlm_interval_tree *tree; + struct ldlm_extent_compat_args data = {.work_list = work_list, + .lock = req, + .locks = contended_locks, + .compat = &compat }; + struct interval_node_extent ex = { .start = req_start, + .end = req_end }; + int idx, rc; + + for (idx = 0; idx < LCK_MODE_NUM; idx++) { + tree = &res->lr_itree[idx]; + if (tree->lit_root == NULL) /* empty tree, skipped */ + continue; + + data.mode = tree->lit_mode; + if (lockmode_compat(req_mode, tree->lit_mode)) { + struct ldlm_interval *node; + struct ldlm_extent *extent; + + if (req_mode != LCK_GROUP) + continue; + + /* group lock, grant it immediately if + * compatible */ + node = to_ldlm_interval(tree->lit_root); + extent = ldlm_interval_extent(node); + if (req->l_policy_data.l_extent.gid == + extent->gid) + RETURN(2); + } + + if (tree->lit_mode == LCK_GROUP) { + if (*flags & LDLM_FL_BLOCK_NOWAIT) { + compat = -EWOULDBLOCK; + goto destroylock; + } + + *flags |= LDLM_FL_NO_TIMEOUT; + if (!work_list) + RETURN(0); + + /* if work list is not NULL,add all + locks in the tree to work list */ + compat = 0; + interval_iterate(tree->lit_root, + ldlm_extent_compat_cb, &data); + continue; + } + + if (!work_list) { + rc = interval_is_overlapped(tree->lit_root,&ex); + if (rc) + RETURN(0); + } else { + interval_search(tree->lit_root, &ex, + ldlm_extent_compat_cb, &data); + if (!list_empty(work_list) && compat) + compat = 0; + } + } + } else { /* for waiting queue */ + list_for_each(tmp, queue) { + check_contention = 1; + + lock = list_entry(tmp, struct ldlm_lock, l_res_link); + + if (req == lock) + break; + + if (unlikely(scan)) { + /* We only get here if we are queuing GROUP lock + and met some incompatible one. The main idea of this + code is to insert GROUP lock past compatible GROUP + lock in the waiting queue or if there is not any, + then in front of first non-GROUP lock */ + if (lock->l_req_mode != LCK_GROUP) { + /* Ok, we hit non-GROUP lock, there should + * be no more GROUP locks later on, queue in + * front of first non-GROUP lock */ + + ldlm_resource_insert_lock_after(lock, req); + list_del_init(&lock->l_res_link); + ldlm_resource_insert_lock_after(req, lock); + compat = 0; + break; + } + if (req->l_policy_data.l_extent.gid == + lock->l_policy_data.l_extent.gid) { + /* found it */ + ldlm_resource_insert_lock_after(lock, req); + compat = 0; + break; + } + continue; + } + + /* locks are compatible, overlap doesn't matter */ + if (lockmode_compat(lock->l_req_mode, req_mode)) { + if (req_mode == LCK_PR && + ((lock->l_policy_data.l_extent.start <= + req->l_policy_data.l_extent.start) && + (lock->l_policy_data.l_extent.end >= + req->l_policy_data.l_extent.end))) { + /* If we met a PR lock just like us or wider, + and nobody down the list conflicted with + it, that means we can skip processing of + the rest of the list and safely place + ourselves at the end of the list, or grant + (dependent if we met an conflicting locks + before in the list). + In case of 1st enqueue only we continue + traversing if there is something conflicting + down the list because we need to make sure + that something is marked as AST_SENT as well, + in cse of empy worklist we would exit on + first conflict met. */ + /* There IS a case where such flag is + not set for a lock, yet it blocks + something. Luckily for us this is + only during destroy, so lock is + exclusive. So here we are safe */ + if (!(lock->l_flags & LDLM_FL_AST_SENT)) { + RETURN(compat); + } + } + + /* non-group locks are compatible, overlap doesn't + matter */ + if (likely(req_mode != LCK_GROUP)) + continue; + + /* If we are trying to get a GROUP lock and there is + another one of this kind, we need to compare gid */ + if (req->l_policy_data.l_extent.gid == + lock->l_policy_data.l_extent.gid) { + /* If existing lock with matched gid is granted, + we grant new one too. */ + if (lock->l_req_mode == lock->l_granted_mode) + RETURN(2); + + /* Otherwise we are scanning queue of waiting + * locks and it means current request would + * block along with existing lock (that is + * already blocked. + * If we are in nonblocking mode - return + * immediately */ + if (*flags & LDLM_FL_BLOCK_NOWAIT) { + compat = -EWOULDBLOCK; + goto destroylock; + } + /* If this group lock is compatible with another + * group lock on the waiting list, they must be + * together in the list, so they can be granted + * at the same time. Otherwise the later lock + * can get stuck behind another, incompatible, + * lock. */ + ldlm_resource_insert_lock_after(lock, req); + /* Because 'lock' is not granted, we can stop + * processing this queue and return immediately. + * There is no need to check the rest of the + * list. */ + RETURN(0); + } + } + + if (unlikely(req_mode == LCK_GROUP && + (lock->l_req_mode != lock->l_granted_mode))) { + scan = 1; + compat = 0; + if (lock->l_req_mode != LCK_GROUP) { + /* Ok, we hit non-GROUP lock, there should be no + more GROUP locks later on, queue in front of + first non-GROUP lock */ + + ldlm_resource_insert_lock_after(lock, req); + list_del_init(&lock->l_res_link); + ldlm_resource_insert_lock_after(req, lock); + break; + } + if (req->l_policy_data.l_extent.gid == + lock->l_policy_data.l_extent.gid) { + /* found it */ + ldlm_resource_insert_lock_after(lock, req); + break; + } + continue; + } + + if (unlikely(lock->l_req_mode == LCK_GROUP)) { + /* If compared lock is GROUP, then requested is PR/PW/ + * so this is not compatible; extent range does not + * matter */ + if (*flags & LDLM_FL_BLOCK_NOWAIT) { + compat = -EWOULDBLOCK; + goto destroylock; + } else { + *flags |= LDLM_FL_NO_TIMEOUT; + } + } else if (lock->l_policy_data.l_extent.end < req_start || + lock->l_policy_data.l_extent.start > req_end) { + /* if a non group lock doesn't overlap skip it */ + continue; + } else if (lock->l_req_extent.end < req_start || + lock->l_req_extent.start > req_end) { + /* false contention, the requests doesn't really overlap */ + check_contention = 0; + } + + if (!work_list) + RETURN(0); + + /* don't count conflicting glimpse locks */ + if (lock->l_req_mode == LCK_PR && + lock->l_policy_data.l_extent.start == 0 && + lock->l_policy_data.l_extent.end == OBD_OBJECT_EOF) + check_contention = 0; + + *contended_locks += check_contention; + + compat = 0; + if (lock->l_blocking_ast) + ldlm_add_ast_work_item(lock, req, work_list); + } + } + + if (ldlm_check_contention(req, *contended_locks) && + compat == 0 && + (*flags & LDLM_FL_DENY_ON_CONTENTION) && + req->l_req_mode != LCK_GROUP && + req_end - req_start <= + req->l_resource->lr_namespace->ns_max_nolock_size) + GOTO(destroylock, compat = -EUSERS); + + RETURN(compat); +destroylock: + list_del_init(&req->l_res_link); + ldlm_lock_destroy_nolock(req); + *err = compat; + RETURN(compat); +} + +static void discard_bl_list(struct list_head *bl_list) +{ + struct list_head *tmp, *pos; + ENTRY; + + list_for_each_safe(pos, tmp, bl_list) { + struct ldlm_lock *lock = + list_entry(pos, struct ldlm_lock, l_bl_ast); + + list_del_init(&lock->l_bl_ast); + LASSERT(lock->l_flags & LDLM_FL_AST_SENT); + lock->l_flags &= ~LDLM_FL_AST_SENT; + LASSERT(lock->l_bl_ast_run == 0); + LASSERT(lock->l_blocking_lock); + LDLM_LOCK_RELEASE(lock->l_blocking_lock); + lock->l_blocking_lock = NULL; + LDLM_LOCK_RELEASE(lock); + } + EXIT; +} - if (lock->l_extent.end < req_ex->start) - new_ex->start = MIN(lock->l_extent.end, new_ex->start); - else { - if (lock->l_extent.start < req_ex->start && - !lockmode_compat(lock->l_req_mode, mode)) - /* Policy: minimize conflict overlap */ - new_ex->start = req_ex->start; +/* If first_enq is 0 (ie, called from ldlm_reprocess_queue): + * - blocking ASTs have already been sent + * - must call this function with the ns lock held + * + * If first_enq is 1 (ie, called from ldlm_lock_enqueue): + * - blocking ASTs have not been sent + * - must call this function with the ns lock held once */ +int ldlm_process_extent_lock(struct ldlm_lock *lock, int *flags, int first_enq, + ldlm_error_t *err, struct list_head *work_list) +{ + struct ldlm_resource *res = lock->l_resource; + CFS_LIST_HEAD(rpc_list); + int rc, rc2; + int contended_locks = 0; + ENTRY; + + LASSERT(list_empty(&res->lr_converting)); + LASSERT(!(*flags & LDLM_FL_DENY_ON_CONTENTION) || + !(lock->l_flags & LDLM_AST_DISCARD_DATA)); + check_res_locked(res); + *err = ELDLM_OK; + + if (!first_enq) { + /* Careful observers will note that we don't handle -EWOULDBLOCK + * here, but it's ok for a non-obvious reason -- compat_queue + * can only return -EWOULDBLOCK if (flags & BLOCK_NOWAIT). + * flags should always be zero here, and if that ever stops + * being true, we want to find out. */ + LASSERT(*flags == 0); + rc = ldlm_extent_compat_queue(&res->lr_granted, lock, flags, + err, NULL, &contended_locks); + if (rc == 1) { + rc = ldlm_extent_compat_queue(&res->lr_waiting, lock, + flags, err, NULL, + &contended_locks); } - if (lock->l_extent.start > req_ex->end) - new_ex->end = MAX(lock->l_extent.start, new_ex->end); - else { - if (lock->l_extent.end > req_ex->end && - !lockmode_compat(lock->l_req_mode, mode)) - /* Policy: minimize conflict overlap */ - new_ex->end = req_ex->end; + if (rc == 0) + RETURN(LDLM_ITER_STOP); + + ldlm_resource_unlink_lock(lock); + + if (!OBD_FAIL_CHECK(OBD_FAIL_LDLM_CANCEL_EVICT_RACE)) + ldlm_extent_policy(res, lock, flags); + ldlm_grant_lock(lock, work_list); + RETURN(LDLM_ITER_CONTINUE); + } + + restart: + contended_locks = 0; + rc = ldlm_extent_compat_queue(&res->lr_granted, lock, flags, err, + &rpc_list, &contended_locks); + if (rc < 0) + GOTO(out, rc); /* lock was destroyed */ + if (rc == 2) + goto grant; + + rc2 = ldlm_extent_compat_queue(&res->lr_waiting, lock, flags, err, + &rpc_list, &contended_locks); + if (rc2 < 0) + GOTO(out, rc = rc2); /* lock was destroyed */ + + if (rc + rc2 == 2) { + grant: + ldlm_extent_policy(res, lock, flags); + ldlm_resource_unlink_lock(lock); + ldlm_grant_lock(lock, NULL); + } else { + /* If either of the compat_queue()s returned failure, then we + * have ASTs to send and must go onto the waiting list. + * + * bug 2322: we used to unlink and re-add here, which was a + * terrible folly -- if we goto restart, we could get + * re-ordered! Causes deadlock, because ASTs aren't sent! */ + if (list_empty(&lock->l_res_link)) + ldlm_resource_add_lock(res, &res->lr_waiting, lock); + unlock_res(res); + rc = ldlm_run_ast_work(&rpc_list, LDLM_WORK_BL_AST); + lock_res(res); + + if (rc == -ERESTART) { + /* lock was granted while resource was unlocked. */ + if (lock->l_granted_mode == lock->l_req_mode) { + /* bug 11300: if the lock has been granted, + * break earlier because otherwise, we will go + * to restart and ldlm_resource_unlink will be + * called and it causes the interval node to be + * freed. Then we will fail at + * ldlm_extent_add_lock() */ + *flags &= ~(LDLM_FL_BLOCK_GRANTED | LDLM_FL_BLOCK_CONV | + LDLM_FL_BLOCK_WAIT); + GOTO(out, rc = 0); + } + + GOTO(restart, -ERESTART); } + + *flags |= LDLM_FL_BLOCK_GRANTED; + /* this way we force client to wait for the lock + * endlessly once the lock is enqueued -bzzz */ + *flags |= LDLM_FL_NO_TIMEOUT; + + } + RETURN(0); +out: + if (!list_empty(&rpc_list)) { + LASSERT(!(lock->l_flags & LDLM_AST_DISCARD_DATA)); + discard_bl_list(&rpc_list); } + RETURN(rc); } -/* apply the internal policy by walking all the lists */ -int ldlm_extent_policy(struct ldlm_namespace *ns, struct ldlm_lock **lockp, - void *req_cookie, ldlm_mode_t mode, int flags, - void *data) +/* When a lock is cancelled by a client, the KMS may undergo change if this + * is the "highest lock". This function returns the new KMS value. + * Caller must hold ns_lock already. + * + * NB: A lock on [x,y] protects a KMS of up to y + 1 bytes! */ +__u64 ldlm_extent_shift_kms(struct ldlm_lock *lock, __u64 old_kms) { - struct ldlm_lock *lock = *lockp; struct ldlm_resource *res = lock->l_resource; - struct ldlm_extent *req_ex = req_cookie; - struct ldlm_extent new_ex; - new_ex.start = 0; - new_ex.end = ~0; + struct list_head *tmp; + struct ldlm_lock *lck; + __u64 kms = 0; + ENTRY; - if (!res) - LBUG(); + /* don't let another thread in ldlm_extent_shift_kms race in + * just after we finish and take our lock into account in its + * calculation of the kms */ + lock->l_flags |= LDLM_FL_KMS_IGNORE; - l_lock(&ns->ns_lock); - policy_internal(&res->lr_granted, req_ex, &new_ex, mode); - policy_internal(&res->lr_converting, req_ex, &new_ex, mode); - policy_internal(&res->lr_waiting, req_ex, &new_ex, mode); - l_unlock(&ns->ns_lock); + list_for_each(tmp, &res->lr_granted) { + lck = list_entry(tmp, struct ldlm_lock, l_res_link); - memcpy(&lock->l_extent, &new_ex, sizeof(new_ex)); + if (lck->l_flags & LDLM_FL_KMS_IGNORE) + continue; - LDLM_DEBUG(lock, "new extent "LPU64" -> "LPU64, new_ex.start, - new_ex.end); + if (lck->l_policy_data.l_extent.end >= old_kms) + RETURN(old_kms); - if (new_ex.end != req_ex->end || new_ex.start != req_ex->start) - return ELDLM_LOCK_CHANGED; - else - return 0; + /* This extent _has_ to be smaller than old_kms (checked above) + * so kms can only ever be smaller or the same as old_kms. */ + if (lck->l_policy_data.l_extent.end + 1 > kms) + kms = lck->l_policy_data.l_extent.end + 1; + } + LASSERTF(kms <= old_kms, "kms "LPU64" old_kms "LPU64"\n", kms, old_kms); + + RETURN(kms); +} + +cfs_mem_cache_t *ldlm_interval_slab; +struct ldlm_interval *ldlm_interval_alloc(struct ldlm_lock *lock) +{ + struct ldlm_interval *node; + ENTRY; + + LASSERT(lock->l_resource->lr_type == LDLM_EXTENT); + OBD_SLAB_ALLOC(node, ldlm_interval_slab, CFS_ALLOC_IO, sizeof(*node)); + if (node == NULL) + RETURN(NULL); + + CFS_INIT_LIST_HEAD(&node->li_group); + ldlm_interval_attach(node, lock); + RETURN(node); +} + +void ldlm_interval_free(struct ldlm_interval *node) +{ + if (node) { + LASSERT(list_empty(&node->li_group)); + OBD_SLAB_FREE(node, ldlm_interval_slab, sizeof(*node)); + } +} + +/* interval tree, for LDLM_EXTENT. */ +void ldlm_interval_attach(struct ldlm_interval *n, + struct ldlm_lock *l) +{ + LASSERT(l->l_tree_node == NULL); + LASSERT(l->l_resource->lr_type == LDLM_EXTENT); + + list_add_tail(&l->l_sl_policy, &n->li_group); + l->l_tree_node = n; +} + +struct ldlm_interval *ldlm_interval_detach(struct ldlm_lock *l) +{ + struct ldlm_interval *n = l->l_tree_node; + + if (n == NULL) + return NULL; + + LASSERT(!list_empty(&n->li_group)); + l->l_tree_node = NULL; + list_del_init(&l->l_sl_policy); + + return (list_empty(&n->li_group) ? n : NULL); +} + +static inline int lock_mode_to_index(ldlm_mode_t mode) +{ + int index; + + LASSERT(mode != 0); + LASSERT(IS_PO2(mode)); + for (index = -1; mode; index++, mode >>= 1) ; + LASSERT(index < LCK_MODE_NUM); + return index; +} + +void ldlm_extent_add_lock(struct ldlm_resource *res, + struct ldlm_lock *lock) +{ + struct interval_node *found, **root; + struct ldlm_interval *node; + struct ldlm_extent *extent; + int idx; + + LASSERT(lock->l_granted_mode == lock->l_req_mode); + + node = lock->l_tree_node; + LASSERT(node != NULL); + + idx = lock_mode_to_index(lock->l_granted_mode); + LASSERT(lock->l_granted_mode == 1 << idx); + LASSERT(lock->l_granted_mode == res->lr_itree[idx].lit_mode); + + /* node extent initialize */ + extent = &lock->l_policy_data.l_extent; + interval_set(&node->li_node, extent->start, extent->end); + + root = &res->lr_itree[idx].lit_root; + found = interval_insert(&node->li_node, root); + if (found) { /* The policy group found. */ + struct ldlm_interval *tmp = ldlm_interval_detach(lock); + LASSERT(tmp != NULL); + ldlm_interval_free(tmp); + ldlm_interval_attach(to_ldlm_interval(found), lock); + } + res->lr_itree[idx].lit_size++; + + /* even though we use interval tree to manage the extent lock, we also + * add the locks into grant list, for debug purpose, .. */ + ldlm_resource_add_lock(res, &res->lr_granted, lock); +} + +void ldlm_extent_unlink_lock(struct ldlm_lock *lock) +{ + struct ldlm_resource *res = lock->l_resource; + struct ldlm_interval *node; + struct ldlm_interval_tree *tree; + int idx; + + if (lock->l_granted_mode != lock->l_req_mode) + return; + + LASSERT(lock->l_tree_node != NULL); + idx = lock_mode_to_index(lock->l_granted_mode); + LASSERT(lock->l_granted_mode == 1 << idx); + tree = &res->lr_itree[idx]; + + LASSERT(tree->lit_root != NULL); /* assure the tree is not null */ + + tree->lit_size--; + node = ldlm_interval_detach(lock); + if (node) { + interval_erase(&node->li_node, &tree->lit_root); + ldlm_interval_free(node); + } }