From 92f1da4da04a7a86ddee91be5eaf0b10c333ac64 Mon Sep 17 00:00:00 2001 From: Ulrich Drepper Date: Wed, 27 Aug 1997 20:26:10 +0000 Subject: Update. 1997-08-10 19:17 Philip Blundell * nss/nss_db/db-XXX.c: Include not . Somebody should update this to use the new db API. * nss/nss_db/db-netgrp.c: Likewise. * nss/nss_db/db-alias.c: Likewise. * db2/Makefile: Makefile for db-2.x in glibc. 1997-08-27 21:20 Ulrich Drepper * csu/Makefile (before-compile): New goal. Make sure abi-tag.h is generated. [$(elf)=yes] (asm-CPPFLAGS): Make sure abi-tag.h file can be found. * Makeconfig [$(build-omitfp)=yes] (CFLAGS-.o): Add -D__USE_STRING_INLINES. * string/string.f: Move strnlen optimization after inclusion of . Include only if __USE_STRING_INLINES is defined. * sysdeps/generic/memcpy.c: Undef memcpy to allow macro of this name in . * sysdeps/generic/memset.c: Likewise. * sysdeps/i386/string.h: i386 optimized string functions. * sysdeps/i386/i486string.h: i486+ optimized string functions. * Makefile (subdirs): Change db to db2. * shlib-versions: Bump libdb verion number to 3. * include/db.h: Include from db2 directory. * include/db_185.h: New file. * sysdeps/i386/Makefile [$(subdirs)=db2] (CPPFLAGS): Add macros to provide spinlock information for db2. * sysdeps/m68k/m68020/Makefile: New file. Likewise. * sysdeps/sparc/Makefile: New file. Likewise. * sysdeps/unix/sysv/linux/Makefile [$(subdirs)=db2] (CPPFLAGS): Add -DHAVE_LLSEEK. * db2/config.h: Hand-edited config file for db2 in glibc. * db2/compat.h: New file from db-2.3.4. * db2/db.h: Likewise. * db2/db_185.h: Likewise. * db2/db_int.h: Likewise. * db2/makedb.c: Likewise. * db2/btree/bt_close.c: Likewise. * db2/btree/bt_compare.c: Likewise. * db2/btree/bt_conv.c: Likewise. * db2/btree/bt_cursor.c: Likewise. * db2/btree/bt_delete.c: Likewise. * db2/btree/bt_open.c: Likewise. * db2/btree/bt_page.c: Likewise. * db2/btree/bt_put.c: Likewise. * db2/btree/bt_rec.c: Likewise. * db2/btree/bt_recno.c: Likewise. * db2/btree/btree_auto.c: Likewise. * db2/btree/bt_rsearch.c: Likewise. * db2/btree/bt_search.c: Likewise. * db2/btree/bt_split.c: Likewise. * db2/btree/bt_stat.c: Likewise. * db2/btree/btree.src: Likewise. * db2/common/db_appinit.c: Likewise. * db2/common/db_err.c: Likewise. * db2/common/db_byteorder.c: Likewise. * db2/common/db_apprec.c: Likewise. * db2/common/db_salloc.c: Likewise. * db2/common/db_log2.c: Likewise. * db2/common/db_region.c: Likewise. * db2/common/db_shash.c: Likewise. * db2/db/db.c: Likewise. * db2/db/db.src: Likewise. * db2/db/db_conv.c: Likewise. * db2/db/db_dispatch.c: Likewise. * db2/db/db_dup.c: Likewise. * db2/db/db_overflow.c: Likewise. * db2/db/db_pr.c: Likewise. * db2/db/db_rec.c: Likewise. * db2/db/db_ret.c: Likewise. * db2/db/db_thread.c: Likewise. * db2/db/db_auto.c: Likewise. * db2/db185/db185.c: Likewise. * db2/db185/db185_int.h: Likewise. * db2/dbm/dbm.c: Likewise. * db2/hash/hash.c: Likewise. * db2/hash/hash.src: Likewise. * db2/hash/hash_page.c: Likewise. * db2/hash/hash_conv.c: Likewise. * db2/hash/hash_debug.c: Likewise. * db2/hash/hash_stat.c: Likewise. * db2/hash/hash_rec.c: Likewise. * db2/hash/hash_dup.c: Likewise. * db2/hash/hash_func.c: Likewise. * db2/hash/hash_auto.c: Likewise. * db2/include/mp.h: Likewise. * db2/include/btree.h: Likewise. * db2/include/db.h.src: Likewise. * db2/include/db_int.h.src: Likewise. * db2/include/db_shash.h: Likewise. * db2/include/db_swap.h: Likewise. * db2/include/db_185.h.src: Likewise. * db2/include/txn.h: Likewise. * db2/include/db_am.h: Likewise. * db2/include/shqueue.h: Likewise. * db2/include/hash.h: Likewise. * db2/include/db_dispatch.h: Likewise. * db2/include/lock.h: Likewise. * db2/include/db_page.h: Likewise. * db2/include/log.h: Likewise. * db2/include/db_auto.h: Likewise. * db2/include/btree_auto.h: Likewise. * db2/include/hash_auto.h: Likewise. * db2/include/log_auto.h: Likewise. * db2/include/txn_auto.h: Likewise. * db2/include/db_ext.h: Likewise. * db2/include/btree_ext.h: Likewise. * db2/include/clib_ext.h: Likewise. * db2/include/common_ext.h: Likewise. * db2/include/hash_ext.h: Likewise. * db2/include/lock_ext.h: Likewise. * db2/include/log_ext.h: Likewise. * db2/include/mp_ext.h: Likewise. * db2/include/mutex_ext.h: Likewise. * db2/include/os_ext.h: Likewise. * db2/include/txn_ext.h: Likewise. * db2/include/cxx_int.h: Likewise. * db2/include/db_cxx.h: Likewise. * db2/include/queue.h: Likewise. * db2/lock/lock.c: Likewise. * db2/lock/lock_conflict.c: Likewise. * db2/lock/lock_util.c: Likewise. * db2/lock/lock_deadlock.c: Likewise. * db2/log/log.c: Likewise. * db2/log/log_get.c: Likewise. * db2/log/log.src: Likewise. * db2/log/log_compare.c: Likewise. * db2/log/log_put.c: Likewise. * db2/log/log_rec.c: Likewise. * db2/log/log_archive.c: Likewise. * db2/log/log_register.c: Likewise. * db2/log/log_auto.c: Likewise. * db2/log/log_findckp.c: Likewise. * db2/mp/mp_bh.c: Likewise. * db2/mp/mp_fget.c: Likewise. * db2/mp/mp_fopen.c: Likewise. * db2/mp/mp_fput.c: Likewise. * db2/mp/mp_fset.c: Likewise. * db2/mp/mp_open.c: Likewise. * db2/mp/mp_region.c: Likewise. * db2/mp/mp_pr.c: Likewise. * db2/mp/mp_sync.c: Likewise. * db2/mutex/68020.gcc: Likewise. * db2/mutex/mutex.c: Likewise. * db2/mutex/README: Likewise. * db2/mutex/x86.gcc: Likewise. * db2/mutex/sparc.gcc: Likewise. * db2/mutex/uts4.cc.s: Likewise. * db2/mutex/alpha.dec: Likewise. * db2/mutex/alpha.gcc: Likewise. * db2/mutex/parisc.gcc: Likewise. * db2/mutex/parisc.hp: Likewise. * db2/os/db_os_abs.c: Likewise. * db2/os/db_os_dir.c: Likewise. * db2/os/db_os_fid.c: Likewise. * db2/os/db_os_lseek.c: Likewise. * db2/os/db_os_mmap.c: Likewise. * db2/os/db_os_open.c: Likewise. * db2/os/db_os_rw.c: Likewise. * db2/os/db_os_sleep.c: Likewise. * db2/os/db_os_stat.c: Likewise. * db2/os/db_os_unlink.c: Likewise. * db2/txn/txn.c: Likewise. * db2/txn/txn.src: Likewise. * db2/txn/txn_rec.c: Likewise. * db2/txn/txn_auto.c: Likewise. * db2/clib/getlong.c: Likewise. * db2/progs/db_archive/db_archive.c: Likewise. * db2/progs/db_checkpoint/db_checkpoint.c: Likewise. * db2/progs/db_deadlock/db_deadlock.c: Likewise. * db2/progs/db_dump/db_dump.c: Likewise. * db2/progs/db_dump185/db_dump185.c: Likewise. * db2/progs/db_load/db_load.c: Likewise. * db2/progs/db_printlog/db_printlog.c: Likewise. * db2/progs/db_recover/db_recover.c: Likewise. * db2/progs/db_stat/db_stat.c: Likewise. * libio/stdio.h [__cplusplus] (__STDIO_INLINE): Define as inline. * po/de.po, po/sv.po: Update from 2.0.5 translations. * sysdeps/unix/sysv/linux/netinet/tcp.h: Pretty print. * sunrpc/rpc/xdr.h (XDR): Don't define argument of x_destroy callback as const. * sunrpc/xdr_mem.c (xdrmem_destroy): Don't define argument as const. * sunrpx/xdr_rec.c (xdrrec_destroy): Likewise. * sunrpx/xdr_stdio.c (xdrstdio_destroy): Likewise. 1997-08-27 18:47 Ulrich Drepper * sysdeps/unix/sysv/linux/if_index.c: Include . Reported by Benjamin Kosnik . 1997-08-27 02:27 Roland McGrath * abi-tags: New file. * csu/Makefile (distribute): Remove abi-tag.h. ($(objpfx)abi-tag.h): New target. * Makefile (distribute): Add abi-tags. * sysdeps/unix/sysv/linux/abi-tag.h: File removed. * sysdeps/mach/hurd/abi-tag.h: File removed. * sysdeps/stub/abi-tag.h: File removed. 1997-08-25 Andreas Schwab * sysdeps/unix/make-syscalls.sh: Change output so that it generates compilation rules only for the currently selected object suffixes. 1997-08-25 Andreas Schwab * sysdeps/m68k/dl-machine.h (RTLD_START): Switch back to previous section to avoid confusing the compiler. * sysdeps/alpha/dl-machine.h (RTLD_START): Likewise. * sysdeps/i386/dl-machine.h (RTLD_START): Likewise. * sysdeps/mips/dl-machine.h (RTLD_START): Likewise. * sysdeps/mips/mips64/dl-machine.h (RTLD_START): Likewise. * sysdeps/sparc/sparc32/dl-machine.h (RTLD_START): Likewise. * sysdeps/m68k/dl-machine.h (elf_machine_load_address): Use a GOT relocation instead of a constant to avoid text relocation. (ELF_MACHINE_BEFORE_RTLD_RELOC): Removed. (RTLD_START): Declare global labels as functions and add size directive. 1997-08-25 17:01 Ulrich Drepper * sysdeps/i386/bits/select.h: Correct assembler versions to work even for descriptors >= 32. * stdlib/alloca.h: Don't define alloca to __alloca since if gcc is used __alloca is not defined to __builtin_alloca and so might not be available. Reported by Uwe Ohse . * sysdeps/unix/sysv/linux/sys/sysmacros.h: Define macros in a special way if gcc is not used and so dev_t is an array. Reported by Uwe Ohse . 1997-08-23 Andreas Schwab * manual/libc.texinfo: Reorder chapters to match logical order. 1997-08-25 12:22 Ulrich Drepper * sunrpc/rpc/xdr.h: Change name of parameters in prototypes of xdr_reference, xdrmem_create, and xdrstdio_create because of clash with g++ internal symbols. Patch by Sudish Joseph . * elf/dl-deps.c: Implement handling of DT_FILTER. --- db2/txn/txn.c | 809 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 809 insertions(+) create mode 100644 db2/txn/txn.c (limited to 'db2/txn/txn.c') diff --git a/db2/txn/txn.c b/db2/txn/txn.c new file mode 100644 index 0000000000..b20697be3b --- /dev/null +++ b/db2/txn/txn.c @@ -0,0 +1,809 @@ +/*- + * See the file LICENSE for redistribution information. + * + * Copyright (c) 1996, 1997 + * Sleepycat Software. All rights reserved. + */ +/* + * Copyright (c) 1995, 1996 + * The President and Fellows of Harvard University. All rights reserved. + * + * This code is derived from software contributed to Berkeley by + * Margo Seltzer. + * + * Redistribution and use in source and binary forms, with or without + * modification, are permitted provided that the following conditions + * are met: + * 1. Redistributions of source code must retain the above copyright + * notice, this list of conditions and the following disclaimer. + * 2. Redistributions in binary form must reproduce the above copyright + * notice, this list of conditions and the following disclaimer in the + * documentation and/or other materials provided with the distribution. + * 3. All advertising materials mentioning features or use of this software + * must display the following acknowledgement: + * This product includes software developed by the University of + * California, Berkeley and its contributors. + * 4. Neither the name of the University nor the names of its contributors + * may be used to endorse or promote products derived from this software + * without specific prior written permission. + * + * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND + * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE + * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE + * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE + * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL + * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS + * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) + * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT + * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY + * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF + * SUCH DAMAGE. + */ + +#include "config.h" + +#ifndef lint +static const char sccsid[] = "@(#)txn.c 10.20 (Sleepycat) 8/24/97"; +#endif /* not lint */ + + +/* + * This file contains the top level routines of the transaction library. + * It assumes that a lock manager and log manager that conform to the db_log(3) + * and db_lock(3) interfaces exist. + */ + +#ifndef NO_SYSTEM_INCLUDES +#include +#include +#include + +#include +#include +#include +#include +#include +#include +#include +#include +#endif + +#include "shqueue.h" +#include "db_int.h" +#include "db_page.h" +#include "db_shash.h" +#include "txn.h" +#include "db_dispatch.h" +#include "lock.h" +#include "log.h" +#include "db_am.h" +#include "common_ext.h" + +static int __txn_check_running __P((const DB_TXN *)); + +static int __txn_create __P((DB_ENV *, const char *, u_int)); +static int __txn_grow_region __P((DB_TXNMGR *)); +static int __txn_validate_region __P((DB_TXNMGR *)); +static int __txn_end __P((DB_TXN *, int)); +static int __txn_undo __P((DB_TXN *)); + +/* + * Create and initialize a transaction region in shared memory. + * 0 means, success. + * +1 means that the db_create failed, so we did not create the region. + * -1 means that we got some sort of system error. + */ +static int +__txn_create(dbenv, path, mode) + DB_ENV *dbenv; + const char *path; + u_int mode; +{ + DB_TXNREGION *txn_region; + TXN_DETAIL *txnp; + time_t now; + int fd, i, maxtxns, ret; + + maxtxns = dbenv->tx_max != 0 ? dbenv->tx_max : 1000; + (void)time(&now); + + ret = __db_rcreate(dbenv, DB_APP_NONE, path, + DEFAULT_TXN_FILE, mode, TXN_REGION_SIZE(maxtxns), &fd, &txn_region); + + /* Region may have existed. If it didn't, the open will fail. */ + if (ret != 0) + return (ret); + + txn_region->magic = DB_TXNMAGIC; + txn_region->version = DB_TXNVERSION; + txn_region->maxtxns = maxtxns; + txn_region->last_txnid = TXN_MINIMUM; + /* XXX If we ever do more types of locking and logging, this changes. */ + txn_region->logtype = 0; + txn_region->locktype = 0; + txn_region->free_txn = 0; + txn_region->time_ckp = now; + ZERO_LSN(txn_region->last_ckp); + ZERO_LSN(txn_region->pending_ckp); + + for (txnp = &txn_region->table[0], i = 0; i < maxtxns; i++, txnp++) { + ZERO_LSN(txnp->begin_lsn); + txnp->status = TXN_UNALLOC; + txnp->txnid = i + 1; + } + txn_region->table[maxtxns - 1].txnid = TXN_INVALID; + + /* Unlock the region. */ + (void)__db_mutex_unlock(&txn_region->hdr.lock, fd); + + /* Now unmap and close the region. */ + if ((ret = __db_rclose(dbenv, fd, txn_region)) != 0) { + (void)txn_unlink(path, 1 /* force */, dbenv); + return (ret); + } + + return (0); +} + +int +txn_open(path, flags, mode, dbenv, mgrpp) + const char *path; + int flags, mode; + DB_ENV *dbenv; + DB_TXNMGR **mgrpp; +{ + DB_TXNMGR *tmgrp; + DB_TXNREGION *txn_regionp; + int fd, ret, retry_cnt; + + tmgrp = NULL; + txn_regionp = NULL; + fd = -1; + + /* Validate arguments. */ + if (dbenv == NULL) + return (EINVAL); +#ifdef HAVE_SPINLOCKS +#define OKFLAGS (DB_CREATE | DB_THREAD | DB_TXN_NOSYNC) +#else +#define OKFLAGS (DB_CREATE | DB_TXN_NOSYNC) +#endif + if ((ret = __db_fchk(dbenv, "txn_open", flags, OKFLAGS)) != 0) + return (ret); + + retry_cnt = 0; +retry: if (LF_ISSET(DB_CREATE) && (ret = __txn_create(dbenv, path, mode)) != 0) + if (ret == EAGAIN && ++retry_cnt < 0) { + (void)__db_sleep(1, 0); + goto retry; + } else /* We did not really create the region */ + flags &= ~DB_CREATE; + + retry_cnt = 0; +retry1: if ((ret = __db_ropen(dbenv, DB_APP_NONE, path, DEFAULT_TXN_FILE, + flags & ~(DB_CREATE | DB_THREAD | DB_TXN_NOSYNC), + &fd, &txn_regionp)) != 0) { + if (ret == EAGAIN && ++retry_cnt < 3) { + (void)__db_sleep(1, 0); + goto retry1; + } + goto out; + } + + + /* Check if valid region. */ + if (txn_regionp->magic != DB_TXNMAGIC) { + __db_err(dbenv, "txn_open: Bad magic number"); + ret = EINVAL; + goto out; + } + + /* Now, create the transaction manager structure and set its fields. */ + if ((tmgrp = (DB_TXNMGR *)malloc(sizeof(DB_TXNMGR))) == NULL) { + __db_err(dbenv, "txn_open: %s", strerror(errno)); + ret = ENOMEM; + goto out; + } + + tmgrp->dbenv = dbenv; + tmgrp->recover = + dbenv->tx_recover == NULL ? __db_dispatch : dbenv->tx_recover; + tmgrp->region = txn_regionp; + tmgrp->reg_size = txn_regionp->hdr.size; + tmgrp->fd = fd; + tmgrp->flags = LF_ISSET(DB_TXN_NOSYNC | DB_THREAD); + TAILQ_INIT(&tmgrp->txn_chain); + if (LF_ISSET(DB_THREAD)) + __db_mutex_init(&tmgrp->mutex, -1); + *mgrpp = tmgrp; + return (0); + +out: if (txn_regionp != NULL) + (void)__db_rclose(dbenv, fd, txn_regionp); + if (flags & DB_CREATE) + (void)txn_unlink(path, 1, dbenv); + if (tmgrp != NULL) + free(tmgrp); + return (ret); +} + +/* + * Internally, we use TXN_DETAIL structures, but we allocate and return + * DB_TXN structures that provide access to the transaction ID and the + * offset in the transaction region of the TXN_DETAIL structure. + */ +int +txn_begin(tmgrp, parent, txnpp) + DB_TXNMGR *tmgrp; + DB_TXN *parent; + DB_TXN **txnpp; +{ + TXN_DETAIL *txnp; + DB_TXN *retp; + int id, index, ret; + + LOCK_TXNREGION(tmgrp); + + if ((ret = __txn_validate_region(tmgrp)) != 0) { + UNLOCK_TXNREGION(tmgrp); + return (ret); + } + + /* Remove element from free list. */ + if (tmgrp->region->free_txn == TXN_INVALID && + (ret = __txn_grow_region(tmgrp)) != 0) { + UNLOCK_TXNREGION(tmgrp); + return (ret); + } + + index = tmgrp->region->free_txn; + txnp = &tmgrp->region->table[index]; + tmgrp->region->free_txn = txnp->txnid; + + if (txnp->status != TXN_UNALLOC) { + UNLOCK_TXNREGION(tmgrp); + return (EINVAL); + } + + /* Make sure that last_txnid is not going to wrap around. */ + if (tmgrp->region->last_txnid == TXN_INVALID) + return (EINVAL); + + if ((retp = (DB_TXN *)malloc(sizeof(DB_TXN))) == NULL) { + __db_err(tmgrp->dbenv, "txn_begin : %s", strerror(ENOMEM)); + UNLOCK_TXNREGION(tmgrp); + return (ENOMEM); + } + + id = ++tmgrp->region->last_txnid; + tmgrp->region->nbegins++; + + txnp->txnid = id; + txnp->last_lock = 0; + txnp->status = TXN_RUNNING; + ZERO_LSN(txnp->last_lsn); + ZERO_LSN(txnp->begin_lsn); + + UNLOCK_TXNREGION(tmgrp); + + ZERO_LSN(retp->last_lsn); + retp->txnid = id; + retp->parent = parent; + retp->off = (u_int8_t *)txnp - (u_int8_t *)tmgrp->region; + retp->mgrp = tmgrp; + + if (tmgrp->dbenv->lg_info != NULL && + (ret = __txn_regop_log(tmgrp->dbenv->lg_info, + retp, &txnp->begin_lsn, 0, TXN_BEGIN)) != 0) { + + /* Deallocate transaction. */ + LOCK_TXNREGION(tmgrp); + txnp->txnid = tmgrp->region->free_txn; + tmgrp->region->free_txn = txnp - &tmgrp->region->table[0]; + UNLOCK_TXNREGION(tmgrp); + free (retp); + return (ret); + } + + LOCK_TXNTHREAD(tmgrp); + TAILQ_INSERT_TAIL(&tmgrp->txn_chain, retp, links); + UNLOCK_TXNTHREAD(tmgrp); + + *txnpp = retp; + return (0); +} + +/* The db_txn(3) man page describes txn_commit. */ +int +txn_commit(txnp) + DB_TXN *txnp; +{ + DB_LOG *logp; + int ret; + + if ((ret = __txn_check_running(txnp)) != 0) + return (ret); + + /* Sync the log. */ + if ((logp = txnp->mgrp->dbenv->lg_info) != NULL && + (ret = __txn_regop_log(logp, + txnp, &txnp->last_lsn, + F_ISSET(txnp->mgrp, DB_TXN_NOSYNC) ? 0 : DB_FLUSH, TXN_COMMIT)) + != 0) + return (ret); + + return (__txn_end(txnp, 1)); +} + +/* The db_txn(3) man page describes txn_abort. */ +int +txn_abort(txnp) + DB_TXN *txnp; +{ + int ret; + + if ((ret = __txn_check_running(txnp)) != 0) + return (ret); + + if ((ret = __txn_undo(txnp)) != 0) { + __db_err(txnp->mgrp->dbenv, + "txn_abort: Log undo failed %s", strerror(ret)); + return (ret); + } + return (__txn_end(txnp, 0)); +} + +/* + * Flush the log so a future commit is guaranteed to succeed. + */ +int +txn_prepare(txnp) + DB_TXN *txnp; +{ + int ret; + TXN_DETAIL *tp; + + ret = 0; + if ((ret = __txn_check_running(txnp)) != 0) + return (ret); + + if (txnp->mgrp->dbenv->lg_info) { + ret = log_flush(txnp->mgrp->dbenv->lg_info, &txnp->last_lsn); + if (ret) + __db_err(txnp->mgrp->dbenv, + "txn_prepare: log_flush failed %s\n", + strerror(errno)); + return (ret); + } + + LOCK_TXNTHREAD(txnp->mgrp); + tp = (TXN_DETAIL *)((u_int8_t *)txnp->mgrp->region + txnp->off); + tp->status = TXN_PREPARED; + UNLOCK_TXNTHREAD(txnp->mgrp); + return (ret); +} + +/* + * Return the transaction ID associated with a particular transaction + */ +u_int32_t +txn_id(txnp) + DB_TXN *txnp; +{ + return (txnp->txnid); +} + +/* + * The db_txn(3) man page describes txn_close. Currently the caller should + * arrange a checkpoint before calling txn_close. + */ +int +txn_close(tmgrp) + DB_TXNMGR *tmgrp; +{ + DB_TXN *txnp; + int ret, t_ret; + + /* + * This function had better only be called once per process + * (i.e., not per thread), so there should be no synchronization + * required. + */ + for (ret = 0, txnp = TAILQ_FIRST(&tmgrp->txn_chain); + txnp != TAILQ_END(&tmgrp->txn_chain); + txnp = TAILQ_FIRST(&tmgrp->txn_chain)) { + if ((t_ret = txn_abort(txnp)) != 0 && ret == 0) + ret = t_ret; + } + + if (tmgrp->dbenv->lg_info && (t_ret = + log_flush(tmgrp->dbenv->lg_info, NULL)) != 0 && + ret == 0) + ret = t_ret; + + if ((t_ret = __db_rclose(tmgrp->dbenv, tmgrp->fd, tmgrp->region)) != 0 + && ret == 0) + ret = t_ret; + + if (ret == 0) + free (tmgrp); + return (ret); +} + +/* + * The db_txn(3) man page describes txn_unlink. Right now it is up to + * txn_close to write the final checkpoint record. + */ +int +txn_unlink(path, force, dbenv) + const char *path; + int force; + DB_ENV *dbenv; +{ + return (__db_runlink(dbenv, + DB_APP_NONE, path, DEFAULT_TXN_FILE, force)); +} + +/* Internal routines. */ + +/* + * Return 0 if the txnp is reasonable, otherwise returns EINVAL. + */ +static int +__txn_check_running(txnp) + const DB_TXN *txnp; +{ + TXN_DETAIL *tp; + + tp = NULL; + if (txnp != NULL && txnp->mgrp != NULL && txnp->mgrp->region != NULL) { + tp = (TXN_DETAIL *)((u_int8_t *)txnp->mgrp->region + txnp->off); + if (tp->status != TXN_RUNNING) + tp = NULL; + } + + return (tp == NULL ? EINVAL : 0); +} + +static int +__txn_end(txnp, is_commit) + DB_TXN *txnp; + int is_commit; +{ + DB_TXNMGR *mgr; + TXN_DETAIL *tp; + DB_LOCKREQ request; + int ret; + u_int32_t locker; + + mgr = txnp->mgrp; + + LOCK_TXNTHREAD(mgr); + TAILQ_REMOVE(&mgr->txn_chain, txnp, links); + UNLOCK_TXNTHREAD(mgr); + + /* Release the locks. */ + locker = txnp->txnid; + request.op = DB_LOCK_PUT_ALL; + + if (mgr->dbenv->lk_info) { + ret = lock_vec(mgr->dbenv->lk_info, locker, 0, + &request, 1, NULL); + if (ret != 0 && (ret != DB_LOCK_DEADLOCK || is_commit)) { + __db_err(mgr->dbenv, "%s: release locks failed %s", + is_commit ? "txn_commit" : "txn_abort", + strerror(ret)); + return (ret); + } + } + + /* End the transaction. */ + LOCK_TXNREGION(mgr); + tp = (TXN_DETAIL *)((u_int8_t *)mgr->region + txnp->off); + tp->status = TXN_UNALLOC; + tp->txnid = mgr->region->free_txn; + mgr->region->free_txn = tp - &mgr->region->table[0]; + if (is_commit) + mgr->region->ncommits++; + else + mgr->region->naborts++; + UNLOCK_TXNREGION(mgr); + + FREE(txnp, sizeof(*txnp)); + + return (0); +} + + +/* + * Undo the transaction with id txnid. Returns 0 on success and sets + * errno and returns -1 on failure. + */ +static int +__txn_undo(txnp) + DB_TXN *txnp; +{ + DB_TXNMGR *mgr; + DB_LOG *logp; + DBT rdbt; + DB_LSN key_lsn; + int ret; + + mgr = txnp->mgrp; + logp = mgr->dbenv->lg_info; + if (logp == NULL) + return (0); + + /* + * This is the simplest way to code this, but if the mallocs during + * recovery turn out to be a performance issue, we can do the + * allocation here and use DB_DBT_USERMEM. + */ + memset(&rdbt, 0, sizeof(rdbt)); + if (F_ISSET(logp, DB_AM_THREAD)) + F_SET(&rdbt, DB_DBT_MALLOC); + + key_lsn = txnp->last_lsn; /* structure assignment */ + for (ret = 0; ret == 0 && !IS_ZERO_LSN(key_lsn);) { + /* + * The dispatch routine returns the lsn of the record + * before the current one in the key_lsn argument. + */ + if ((ret = log_get(logp, &key_lsn, &rdbt, DB_SET)) == 0) { + ret = + mgr->recover(logp, &rdbt, &key_lsn, TXN_UNDO, NULL); + if (F_ISSET(logp, DB_AM_THREAD) && rdbt.data != NULL) { + free(rdbt.data); + rdbt.data = NULL; + } + } + if (ret != 0) + return (ret); + } + + return (ret); +} + +/* + * Transaction checkpoint. + * If either kbytes or minutes is non-zero, then we only take the checkpoint + * more than "minutes" minutes have passed since the last checkpoint or if + * more than "kbytes" of log data have been written since the last checkpoint. + * When taking a checkpoint, find the oldest active transaction and figure out + * its first LSN. This is the lowest LSN we can checkpoint, since any record + * written after since that point may be involved in a transaction and may + * therefore need to be undone in the case of an abort. + */ +int +txn_checkpoint(mgr, kbytes, minutes) + const DB_TXNMGR *mgr; + long kbytes, minutes; +{ + TXN_DETAIL *txnp; + DB_LSN ckp_lsn, last_ckp; + DB_LOG *dblp; + u_int32_t bytes_written, i; + time_t last_ckp_time, now; + int ret; + + /* Check usage. */ + if (kbytes < 0 || minutes < 0) + return (EINVAL); + + /* + * Check if we need to run recovery. + */ + ZERO_LSN(ckp_lsn); + if (minutes != 0) { + (void)time(&now); + + LOCK_TXNREGION(mgr); + last_ckp_time = mgr->region->time_ckp; + UNLOCK_TXNREGION(mgr); + + if (now - last_ckp_time >= (time_t)(minutes * 60)) + goto do_ckp; + } + + if (kbytes != 0) { + dblp = mgr->dbenv->lg_info; + LOCK_LOGREGION(dblp); + bytes_written = dblp->lp->written; + ckp_lsn = dblp->lp->lsn; + UNLOCK_LOGREGION(dblp); + if (bytes_written >= (u_int32_t)(kbytes * 1024)) + goto do_ckp; + } + + /* + * If we checked time and data and didn't go to checkpoint, + * we're done. + */ + if (minutes != 0 || kbytes != 0) + return (0); + + if (IS_ZERO_LSN(ckp_lsn)) { + dblp = mgr->dbenv->lg_info; + LOCK_LOGREGION(dblp); + ckp_lsn = dblp->lp->lsn; + UNLOCK_LOGREGION(dblp); + } + + /* + * We have to find an LSN such that all transactions begun + * before that LSN are complete. + */ +do_ckp: + LOCK_TXNREGION(mgr); + + if (!IS_ZERO_LSN(mgr->region->pending_ckp)) + ckp_lsn = mgr->region->pending_ckp; + else + for (txnp = &mgr->region->table[0], i = 0; + i < mgr->region->maxtxns; i++, txnp++) { + + /* + * Look through the transaction table for the LSN of + * the transaction that is in-use (e.g., not + * TXN_UNALLOC) and whose begin lsn is the lowest. + */ + if (txnp->status != TXN_UNALLOC && + !IS_ZERO_LSN(txnp->begin_lsn) && + log_compare(&txnp->begin_lsn, &ckp_lsn) < 0) + ckp_lsn = txnp->begin_lsn; + } + + mgr->region->pending_ckp = ckp_lsn; + UNLOCK_TXNREGION(mgr); + + ret = memp_sync(mgr->dbenv->mp_info, &ckp_lsn); + if (ret > 0) { + __db_err(mgr->dbenv, + "txn_checkpoint: system failure in memp_sync %s\n", + strerror(ret)); + } else if (ret == 0 && mgr->dbenv->lg_info != NULL) { + LOCK_TXNREGION(mgr); + last_ckp = mgr->region->last_ckp; + ZERO_LSN(mgr->region->pending_ckp); + UNLOCK_TXNREGION(mgr); + + if ((ret = __txn_ckp_log(mgr->dbenv->lg_info, + NULL, &ckp_lsn, DB_CHECKPOINT, &ckp_lsn, &last_ckp)) != 0) { + __db_err(mgr->dbenv, + "txn_checkpoint: log failed at LSN [%ld %ld] %s\n", + (long)ckp_lsn.file, (long)ckp_lsn.offset, + strerror(ret)); + return (ret); + } + + LOCK_TXNREGION(mgr); + mgr->region->last_ckp = ckp_lsn; + (void)time(&mgr->region->time_ckp); + UNLOCK_TXNREGION(mgr); + } + /* + * ret < 0 means that there are still buffers to flush; the + * checkpoint is not complete. Back off and try again. + */ + return (ret); +} + +/* + * This is called at every interface to verify if the region + * has changed size, and if so, to remap the region in and + * reset the process pointers. + */ +static int +__txn_validate_region(tp) + DB_TXNMGR *tp; +{ + int ret; + + if (tp->reg_size == tp->region->hdr.size) + return (0); + + /* Grow the region. */ + if ((ret = __db_rremap(tp->dbenv, tp->region, + tp->reg_size, tp->region->hdr.size, tp->fd, &tp->region)) != 0) + return (ret); + + tp->reg_size = tp->region->hdr.size; + + return (0); +} + +static int +__txn_grow_region(tp) + DB_TXNMGR *tp; +{ + TXN_DETAIL *tx; + size_t incr; + u_int32_t i, oldmax; + int ret; + + oldmax = tp->region->maxtxns; + incr = oldmax * sizeof(DB_TXN); + + if ((ret = __db_rgrow(tp->dbenv, tp->fd, incr)) != 0) + return (ret); + + if ((ret = __db_rremap(tp->dbenv, tp->region, + tp->reg_size, tp->reg_size + incr, tp->fd, &tp->region)) != 0) + return (ret); + tp->reg_size += incr; + + /* + * Initialize all the new transactions and up the transaction count. + */ + for (i = 0, tx = &tp->region->table[oldmax]; i < oldmax; i++, tx++) { + ZERO_LSN(tx->begin_lsn); + tx->status = TXN_UNALLOC; + tx->txnid = oldmax + i + 1; + } + tp->region->free_txn = oldmax; + tp->region->maxtxns = 2 * oldmax; + tp->region->table[tp->region->maxtxns - 1].txnid = TXN_INVALID; + + return (0); +} + +int +txn_stat(mgr, statp, db_malloc) + DB_TXNMGR *mgr; + DB_TXN_STAT **statp; + void *(*db_malloc) __P((size_t)); +{ + DB_TXN_STAT *stats; + size_t nbytes; + u_int32_t nactive; + unsigned int i, ndx; + + LOCK_TXNREGION(mgr); + nactive = mgr->region->nbegins - + mgr->region->naborts - mgr->region->ncommits; + UNLOCK_TXNREGION(mgr); + + /* + * Allocate a bunch of extra active structures to handle any + * that have been created since we unlocked the region. + */ + nbytes = sizeof(DB_TXN_STAT) + sizeof(DB_TXN_ACTIVE) * (nactive + 200); + if (db_malloc == NULL) + stats = (DB_TXN_STAT *)malloc(nbytes); + else + stats = (DB_TXN_STAT *)db_malloc(nbytes); + + if (stats == NULL) + return (ENOMEM); + + LOCK_TXNREGION(mgr); + stats->st_last_txnid = mgr->region->last_txnid; + stats->st_last_ckp = mgr->region->last_ckp; + stats->st_maxtxns = mgr->region->maxtxns; + stats->st_naborts = mgr->region->naborts; + stats->st_nbegins = mgr->region->nbegins; + stats->st_ncommits = mgr->region->ncommits; + stats->st_pending_ckp = mgr->region->pending_ckp; + stats->st_time_ckp = mgr->region->time_ckp; + stats->st_nactive = stats->st_nbegins - + stats->st_naborts - stats->st_ncommits; + if (stats->st_nactive > nactive + 200) + stats->st_nactive = nactive + 200; + stats->st_txnarray = (DB_TXN_ACTIVE *)&stats[1]; + + for (ndx = 0, i = 0; i < mgr->region->maxtxns; i++) + if (mgr->region->table[i].status != TXN_UNALLOC) { + stats->st_txnarray[ndx].txnid = + mgr->region->table[i].txnid; + stats->st_txnarray[ndx].lsn = + mgr->region->table[i].begin_lsn; + ndx++; + + if (ndx >= stats->st_nactive) + break; + } + + UNLOCK_TXNREGION(mgr); + *statp = stats; + return (0); +} -- cgit 1.4.1