database/postgres
1
0
Fork 0
mirror of https://github.com/postgres/postgres.git synced 2025-06-30 21:42:05 +03:00
Code Activity

Don't use O_DIRECT when writing WAL files if archiving or streaming is

Browse Source 
enabled. Bypassing the kernel cache is counter-productive in that case,
because the archiver/walsender process will read from the WAL file
soon after it's written, and if it's not cached the read will cause
a physical read, eating I/O bandwidth available on the WAL drive.

Also, walreceiver process does unaligned writes, so disable O_DIRECT
in walreceiver process for that reason too.
This commit is contained in:
Heikki Linnakangas
2010-02-19 10:51:04 +00:00
parent 94f610b163
commit ad458cfe81
4 changed files with 48 additions and 27 deletions
Download Patch File Download Diff File Expand all files Collapse all files

32
src/backend/access/transam/xlog.c
View File

@ -7,7 +7,7 @@
* Portions Copyright (c) 1996-2010, PostgreSQL Global Development Group
* Portions Copyright (c) 1994, Regents of the University of California
*
* $PostgreSQL: pgsql/src/backend/access/transam/xlog.c,v 1.376 2010/02/19 01:04:03 itagaki Exp $
* $PostgreSQL: pgsql/src/backend/access/transam/xlog.c,v 1.377 2010/02/19 10:51:03 heikki Exp $
*
*-------------------------------------------------------------------------
*/
@ -2686,13 +2686,10 @@ XLogFileClose(void)
* WAL segment files will not be re-read in normal operation, so we advise
* the OS to release any cached pages. But do not do so if WAL archiving
* or streaming is active, because archiver and walsender process could use
* the cache to read the WAL segment. Also, don't bother with it if we
* are using O_DIRECT, since the kernel is presumably not caching in that
* case.
* the cache to read the WAL segment.
*/
#if defined(USE_POSIX_FADVISE) && defined(POSIX_FADV_DONTNEED)
if (!XLogIsNeeded() &&
(get_sync_bit(sync_method) & PG_O_DIRECT) == 0)
if (!XLogIsNeeded())
(void) posix_fadvise(openLogFile, 0, 0, POSIX_FADV_DONTNEED);
#endif
@ -7652,10 +7649,29 @@ xlog_outrec(StringInfo buf, XLogRecord *record)
static int
get_sync_bit(int method)
{
int o_direct_flag = 0;
/* If fsync is disabled, never open in sync mode */
if (!enableFsync)
return 0;
/*
* Optimize writes by bypassing kernel cache with O_DIRECT when using
* O_SYNC, O_DSYNC or O_FSYNC. But only if archiving and streaming are
* disabled, otherwise the archive command or walsender process will
* read the WAL soon after writing it, which is guaranteed to cause a
* physical read if we bypassed the kernel cache. We also skip the
* posix_fadvise(POSIX_FADV_DONTNEED) call in XLogFileClose() for the
* same reason.
*
* Never use O_DIRECT in walreceiver process for similar reasons; the WAL
* written by walreceiver is normally read by the startup process soon
* after its written. Also, walreceiver performs unaligned writes, which
* don't work with O_DIRECT, so it is required for correctness too.
*/
if (!XLogIsNeeded() && !am_walreceiver)
o_direct_flag = PG_O_DIRECT;
switch (method)
{
/*
@ -7670,11 +7686,11 @@ get_sync_bit(int method)
return 0;
#ifdef OPEN_SYNC_FLAG
case SYNC_METHOD_OPEN:
return OPEN_SYNC_FLAG;
return OPEN_SYNC_FLAG | o_direct_flag;
#endif
#ifdef OPEN_DATASYNC_FLAG
case SYNC_METHOD_OPEN_DSYNC:
return OPEN_DATASYNC_FLAG;
return OPEN_DATASYNC_FLAG | o_direct_flag;
#endif
default:
/* can't happen (unless we are out of sync with option array) */