From 269e780822abb2e44189afaccd6b0ee7aefa7ddd Mon Sep 17 00:00:00 2001 From: Jeff Davis Date: Fri, 5 Jul 2013 12:30:29 -0700 Subject: [PATCH] Use posix_fallocate() for new WAL files, where available. This function is more efficient than actually writing out zeroes to the new file, per microbenchmarks by Jon Nelson. Also, it may reduce the likelihood of WAL file fragmentation. Jon Nelson, with review by Andres Freund, Greg Smith and me. --- configure | 3 +- configure.in | 2 +- src/backend/access/transam/xlog.c | 93 +++++++++++++++++++------------ src/include/pg_config.h.in | 3 + src/include/pg_config.h.win32 | 3 + 5 files changed, 67 insertions(+), 37 deletions(-) diff --git a/configure b/configure index 7c662c3059..a2d0cfbba0 100755 --- a/configure +++ b/configure @@ -19695,7 +19695,8 @@ LIBS=`echo "$LIBS" | sed -e 's/-ledit//g' -e 's/-lreadline//g'` -for ac_func in cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll pstat readlink setproctitle setsid sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l + +for ac_func in cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll posix_fallocate pstat readlink setproctitle setsid sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l do as_ac_var=`$as_echo "ac_cv_func_$ac_func" | $as_tr_sh` { $as_echo "$as_me:$LINENO: checking for $ac_func" >&5 diff --git a/configure.in b/configure.in index 58ece5b391..3ecdd28006 100644 --- a/configure.in +++ b/configure.in @@ -1222,7 +1222,7 @@ PGAC_FUNC_GETTIMEOFDAY_1ARG LIBS_including_readline="$LIBS" LIBS=`echo "$LIBS" | sed -e 's/-ledit//g' -e 's/-lreadline//g'` -AC_CHECK_FUNCS([cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll pstat readlink setproctitle setsid sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l]) +AC_CHECK_FUNCS([cbrt dlopen fdatasync getifaddrs getpeerucred getrlimit mbstowcs_l memmove poll posix_fallocate pstat readlink setproctitle setsid sigprocmask symlink sync_file_range towlower utime utimes wcstombs wcstombs_l]) AC_REPLACE_FUNCS(fseeko) case $host_os in diff --git a/src/backend/access/transam/xlog.c b/src/backend/access/transam/xlog.c index 4220859c8a..f97ab1ae3a 100644 --- a/src/backend/access/transam/xlog.c +++ b/src/backend/access/transam/xlog.c @@ -2256,11 +2256,9 @@ XLogFileInit(XLogSegNo logsegno, bool *use_existent, bool use_lock) { char path[MAXPGPATH]; char tmppath[MAXPGPATH]; - char *zbuffer; XLogSegNo installed_segno; int max_advance; int fd; - int nbytes; XLogFilePath(path, ThisTimeLineID, logsegno); @@ -2294,16 +2292,6 @@ 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); @@ -2312,38 +2300,73 @@ 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) +#ifdef HAVE_POSIX_FALLOCATE { - errno = 0; - if ((int) write(fd, zbuffer, XLOG_BLCKSZ) != (int) XLOG_BLCKSZ) - { - int save_errno = errno; + errno = posix_fallocate(fd, 0, XLogSegSize); - /* - * If we fail to make the file, delete it to release disk space - */ - unlink(tmppath); + if (errno) + { + int errno_saved = errno; close(fd); - - /* if write didn't set errno, assume problem is no disk space */ - errno = save_errno ? save_errno : ENOSPC; + unlink(tmppath); + errno = errno_saved; ereport(ERROR, (errcode_for_file_access(), - errmsg("could not write to file \"%s\": %m", tmppath))); + errmsg("could not allocate space for file \"%s\" using posix_fallocate: %m", + tmppath))); + } + } +#else /* !HAVE_POSIX_FALLOCATE */ + { + /* + * 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. + */ + + char *zbuffer = (char *) palloc0(XLOG_BLCKSZ); + int nbytes; + + /* + * 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 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); +#endif /* HAVE_POSIX_FALLOCATE */ if (pg_fsync(fd) != 0) { diff --git a/src/include/pg_config.h.in b/src/include/pg_config.h.in index 8aabf3c87a..033127beff 100644 --- a/src/include/pg_config.h.in +++ b/src/include/pg_config.h.in @@ -369,6 +369,9 @@ /* Define to 1 if you have the `posix_fadvise' function. */ #undef HAVE_POSIX_FADVISE +/* Define to 1 if you have the `posix_fallocate' function. */ +#undef HAVE_POSIX_FALLOCATE + /* Define to 1 if you have the POSIX signal interface. */ #undef HAVE_POSIX_SIGNALS diff --git a/src/include/pg_config.h.win32 b/src/include/pg_config.h.win32 index 54db287aff..931551086b 100644 --- a/src/include/pg_config.h.win32 +++ b/src/include/pg_config.h.win32 @@ -276,6 +276,9 @@ /* Define to 1 if you have the header file. */ /* #undef HAVE_POLL_H */ +/* Define to 1 if you have the `posix_fallocate' function. */ +/* #undef HAVE_POSIX_FALLOCATE */ + /* Define to 1 if you have the POSIX signal interface. */ /* #undef HAVE_POSIX_SIGNALS */ -- 2.40.0