mirror of
https://git.proxmox.com/git/mirror_zfs.git
synced 2024-11-18 02:20:59 +03:00
aa755b3549
Mixing ZIL and normal allocations has several problems:
1. The ZIL allocations are allocated, written to disk, and then a few
seconds later freed. This leaves behind holes (free segments) where the
ZIL blocks used to be, which increases fragmentation, which negatively
impacts performance.
2. When under moderate load, ZIL allocations are of 128KB. If the pool
is fairly fragmented, there may not be many free chunks of that size.
This causes ZFS to load more metaslabs to locate free segments of 128KB
or more. The loading happens synchronously (from zil_commit()), and can
take around a second even if the metaslab's spacemap is cached in the
ARC. All concurrent synchronous operations on this filesystem must wait
while the metaslab is loading. This can cause a significant performance
impact.
3. If the pool is very fragmented, there may be zero free chunks of
128KB or more. In this case, the ZIL falls back to txg_wait_synced(),
which has an enormous performance impact.
These problems can be eliminated by using a dedicated log device
("slog"), even one with the same performance characteristics as the
normal devices.
This change sets aside one metaslab from each top-level vdev that is
preferentially used for ZIL allocations (vdev_log_mg,
spa_embedded_log_class). From an allocation perspective, this is
similar to having a dedicated log device, and it eliminates the
above-mentioned performance problems.
Log (ZIL) blocks can be allocated from the following locations. Each
one is tried in order until the allocation succeeds:
1. dedicated log vdevs, aka "slog" (spa_log_class)
2. embedded slog metaslabs (spa_embedded_log_class)
3. other metaslabs in normal vdevs (spa_normal_class)
The space required for the embedded slog metaslabs is usually between
0.5% and 1.0% of the pool, and comes out of the existing 3.2% of "slop"
space that is not available for user data.
On an all-ssd system with 4TB storage, 87% fragmentation, 60% capacity,
and recordsize=8k, testing shows a ~50% performance increase on random
8k sync writes. On even more fragmented systems (which hit problem #3
above and call txg_wait_synced()), the performance improvement can be
arbitrarily large (>100x).
Reviewed-by: Serapheim Dimitropoulos <serapheim@delphix.com>
Reviewed-by: George Wilson <gwilson@delphix.com>
Reviewed-by: Don Brady <don.brady@delphix.com>
Reviewed-by: Mark Maybee <mark.maybee@delphix.com>
Signed-off-by: Matthew Ahrens <mahrens@delphix.com>
Closes #11389
113 lines
3.3 KiB
C
113 lines
3.3 KiB
C
/*
|
|
* CDDL HEADER START
|
|
*
|
|
* The contents of this file are subject to the terms of the
|
|
* Common Development and Distribution License (the "License").
|
|
* You may not use this file except in compliance with the License.
|
|
*
|
|
* You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
|
|
* or http://www.opensolaris.org/os/licensing.
|
|
* See the License for the specific language governing permissions
|
|
* and limitations under the License.
|
|
*
|
|
* When distributing Covered Code, include this CDDL HEADER in each
|
|
* file and include the License file at usr/src/OPENSOLARIS.LICENSE.
|
|
* If applicable, add the following below this CDDL HEADER, with the
|
|
* fields enclosed by brackets "[]" replaced with your own identifying
|
|
* information: Portions Copyright [yyyy] [name of copyright owner]
|
|
*
|
|
* CDDL HEADER END
|
|
*/
|
|
/*
|
|
* Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
|
|
* Copyright (c) 2012, 2019 by Delphix. All rights reserved.
|
|
*/
|
|
|
|
#ifndef _SYS_ZFS_DEBUG_H
|
|
#define _SYS_ZFS_DEBUG_H
|
|
|
|
#ifdef __cplusplus
|
|
extern "C" {
|
|
#endif
|
|
|
|
#ifndef TRUE
|
|
#define TRUE 1
|
|
#endif
|
|
|
|
#ifndef FALSE
|
|
#define FALSE 0
|
|
#endif
|
|
|
|
extern int zfs_flags;
|
|
extern int zfs_recover;
|
|
extern int zfs_free_leak_on_eio;
|
|
extern int zfs_dbgmsg_enable;
|
|
|
|
#define ZFS_DEBUG_DPRINTF (1 << 0)
|
|
#define ZFS_DEBUG_DBUF_VERIFY (1 << 1)
|
|
#define ZFS_DEBUG_DNODE_VERIFY (1 << 2)
|
|
#define ZFS_DEBUG_SNAPNAMES (1 << 3)
|
|
#define ZFS_DEBUG_MODIFY (1 << 4)
|
|
/* 1<<5 was previously used, try not to reuse */
|
|
#define ZFS_DEBUG_ZIO_FREE (1 << 6)
|
|
#define ZFS_DEBUG_HISTOGRAM_VERIFY (1 << 7)
|
|
#define ZFS_DEBUG_METASLAB_VERIFY (1 << 8)
|
|
#define ZFS_DEBUG_SET_ERROR (1 << 9)
|
|
#define ZFS_DEBUG_INDIRECT_REMAP (1 << 10)
|
|
#define ZFS_DEBUG_TRIM (1 << 11)
|
|
#define ZFS_DEBUG_LOG_SPACEMAP (1 << 12)
|
|
#define ZFS_DEBUG_METASLAB_ALLOC (1 << 13)
|
|
|
|
extern void __set_error(const char *file, const char *func, int line, int err);
|
|
extern void __zfs_dbgmsg(char *buf);
|
|
extern void __dprintf(boolean_t dprint, const char *file, const char *func,
|
|
int line, const char *fmt, ...);
|
|
|
|
/*
|
|
* Some general principles for using zfs_dbgmsg():
|
|
* 1. We don't want to pollute the log with typically-irrelevant messages,
|
|
* so don't print too many messages in the "normal" code path - O(1)
|
|
* per txg.
|
|
* 2. We want to know for sure what happened, so make the message specific
|
|
* (e.g. *which* thing am I operating on).
|
|
* 3. Do print a message when something unusual or unexpected happens
|
|
* (e.g. error cases).
|
|
* 4. Print a message when making user-initiated on-disk changes.
|
|
*
|
|
* Note that besides principle 1, another reason that we don't want to
|
|
* use zfs_dbgmsg in high-frequency routines is the potential impact
|
|
* that it can have on performance.
|
|
*/
|
|
#define zfs_dbgmsg(...) \
|
|
if (zfs_dbgmsg_enable) \
|
|
__dprintf(B_FALSE, __FILE__, __func__, __LINE__, __VA_ARGS__)
|
|
|
|
#ifdef ZFS_DEBUG
|
|
/*
|
|
* To enable this:
|
|
*
|
|
* $ echo 1 >/sys/module/zfs/parameters/zfs_flags
|
|
*/
|
|
#define dprintf(...) \
|
|
if (zfs_flags & ZFS_DEBUG_DPRINTF) \
|
|
__dprintf(B_TRUE, __FILE__, __func__, __LINE__, __VA_ARGS__)
|
|
#else
|
|
#define dprintf(...) ((void)0)
|
|
#endif /* ZFS_DEBUG */
|
|
|
|
extern void zfs_panic_recover(const char *fmt, ...);
|
|
|
|
extern void zfs_dbgmsg_init(void);
|
|
extern void zfs_dbgmsg_fini(void);
|
|
|
|
#ifndef _KERNEL
|
|
extern int dprintf_find_string(const char *string);
|
|
extern void zfs_dbgmsg_print(const char *tag);
|
|
#endif
|
|
|
|
#ifdef __cplusplus
|
|
}
|
|
#endif
|
|
|
|
#endif /* _SYS_ZFS_DEBUG_H */
|