From b77b8c211af733f600853ba6e72f3d3e94b78121 Mon Sep 17 00:00:00 2001 From: Jon Nelson Date: Tue, 14 May 2013 18:34:29 -0500 Subject: [PATCH] - enhance GUC and xlog with wal_use_fallocate (boolean) to use posix_fallocate to create the WAL files instead of writing zeroes. --- src/backend/access/transam/xlog.c | 87 +++++++++++++++++++++++---------------- src/backend/utils/misc/guc.c | 10 +++++ src/include/access/xlog.h | 1 + 3 files changed, 63 insertions(+), 35 deletions(-) diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c index f7dd61c..aab5337 100644 --- a/src/backend/access/transam/xlog.c +++ b/src/backend/access/transam/xlog.c @@ -79,6 +79,7 @@ char *XLogArchiveCommand = NULL; bool EnableHotStandby = false; bool fullPageWrites = true; bool log_checkpoints = false; +bool wal_use_fallocate = false; int sync_method = DEFAULT_SYNC_METHOD; int wal_level = WAL_LEVEL_MINIMAL; int CommitDelay = 0; /* precommit delay in microseconds */ @@ -2284,16 +2285,6 @@ XLogFileInit(XLogSegNo logsegno, bool *use_existent, bool use_lock) unlink(tmppath); - /* - * Allocate a buffer full of zeros. This is done before opening the file - * so that we don't leak the file descriptor if palloc fails. - * - * Note: palloc zbuffer, instead of just using a local char array, to - * ensure it is reasonably well-aligned; this may save a few cycles - * transferring data to the kernel. - */ - zbuffer = (char *) palloc0(XLOG_BLCKSZ); - /* do not use get_sync_bit() here --- want to fsync only at end of fill */ fd = BasicOpenFile(tmppath, O_RDWR | O_CREAT | O_EXCL | PG_BINARY, S_IRUSR | S_IWUSR); @@ -2302,38 +2293,64 @@ XLogFileInit(XLogSegNo logsegno, bool *use_existent, bool use_lock) (errcode_for_file_access(), errmsg("could not create file \"%s\": %m", tmppath))); - /* - * Zero-fill the file. We have to do this the hard way to ensure that all - * the file space has really been allocated --- on platforms that allow - * "holes" in files, just seeking to the end doesn't allocate intermediate - * space. This way, we know that we have all the space and (after the - * fsync below) that all the indirect blocks are down on disk. Therefore, - * fdatasync(2) or O_DSYNC will be sufficient to sync future writes to the - * log file. - */ - for (nbytes = 0; nbytes < XLogSegSize; nbytes += XLOG_BLCKSZ) + if (wal_use_fallocate) { - errno = 0; - if ((int) write(fd, zbuffer, XLOG_BLCKSZ) != (int) XLOG_BLCKSZ) + if (posix_fallocate(fd, 0, XLogSegSize)) { - int save_errno = errno; - - /* - * If we fail to make the file, delete it to release disk space - */ + int save_errno = errno; unlink(tmppath); - - close(fd); - - /* if write didn't set errno, assume problem is no disk space */ errno = save_errno ? save_errno : ENOSPC; - ereport(ERROR, (errcode_for_file_access(), - errmsg("could not write to file \"%s\": %m", tmppath))); + errmsg("could not write to file \"%s\": %m", tmppath))); + } + } + else + { + /* + * Allocate a buffer full of zeros. This is done before opening the file + * so that we don't leak the file descriptor if palloc fails. + * + * Note: palloc zbuffer, instead of just using a local char array, to + * ensure it is reasonably well-aligned; this may save a few cycles + * transferring data to the kernel. + */ + + zbuffer = (char *) palloc0(XLOG_BLCKSZ); + + /* + * Zero-fill the file. We have to do this the hard way to ensure that all + * the file space has really been allocated --- on platforms that allow + * "holes" in files, just seeking to the end doesn't allocate intermediate + * space. This way, we know that we have all the space and (after the + * fsync below) that all the indirect blocks are down on disk. Therefore, + * fdatasync(2) or O_DSYNC will be sufficient to sync future writes to the + * log file. + */ + for (nbytes = 0; nbytes < XLogSegSize; nbytes += XLOG_BLCKSZ) + { + errno = 0; + if ((int) write(fd, zbuffer, XLOG_BLCKSZ) != (int) XLOG_BLCKSZ) + { + int save_errno = errno; + + /* + * If we fail to make the file, delete it to release disk space + */ + unlink(tmppath); + + close(fd); + + /* if write didn't set errno, assume problem is no disk space */ + errno = save_errno ? save_errno : ENOSPC; + + ereport(ERROR, + (errcode_for_file_access(), + errmsg("could not write to file \"%s\": %m", tmppath))); + } } + pfree(zbuffer); } - pfree(zbuffer); if (pg_fsync(fd) != 0) { @@ -6866,7 +6883,7 @@ CreateCheckPoint(int flags) XLogRecPtr curInsert; INSERT_RECPTR(curInsert, Insert, Insert->curridx); - if (curInsert == ControlFile->checkPoint + + if (curInsert == ControlFile->checkPoint + MAXALIGN(SizeOfXLogRecord + sizeof(CheckPoint)) && ControlFile->checkPoint == ControlFile->checkPointCopy.redo) { diff --git a/src/backend/utils/misc/guc.c b/src/backend/utils/misc/guc.c index 22ba35f..f83fcec 100644 --- a/src/backend/utils/misc/guc.c +++ b/src/backend/utils/misc/guc.c @@ -1455,6 +1455,16 @@ static struct config_bool ConfigureNamesBool[] = NULL, NULL, NULL }, + { + {"wal_use_fallocate", PGC_SIGHUP, WAL_SETTINGS, + gettext_noop("WAL writer should use posix_fallocate(3) instead of write(2)."), + NULL, + }, + &wal_use_fallocate, + false, + NULL, NULL, NULL + }, + /* End-of-list marker */ { {NULL, 0, 0, NULL, NULL}, NULL, false, NULL, NULL, NULL diff --git a/src/include/access/xlog.h b/src/include/access/xlog.h index f8f06c1..8e98e2c 100644 --- a/src/include/access/xlog.h +++ b/src/include/access/xlog.h @@ -190,6 +190,7 @@ extern char *XLogArchiveCommand; extern bool EnableHotStandby; extern bool fullPageWrites; extern bool log_checkpoints; +extern bool wal_use_fallocate; /* WAL levels */ typedef enum WalLevel -- 1.8.1.4