mirror_zfs/include/sys/zfs_rlock.h

118 lines
3.3 KiB
C
Raw Normal View History

2008-11-20 23:01:55 +03:00
/*
* 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 2006 Sun Microsystems, Inc. All rights reserved.
* Use is subject to license terms.
*/
#ifndef _SYS_FS_ZFS_RLOCK_H
#define _SYS_FS_ZFS_RLOCK_H
#ifdef __cplusplus
extern "C" {
#endif
#include <sys/list.h>
#include <sys/avl.h>
#ifdef _KERNEL
#include <sys/condvar.h>
#else
#include <sys/zfs_context.h>
#endif
2008-11-20 23:01:55 +03:00
typedef enum {
RL_READER,
RL_WRITER,
RL_APPEND
} rl_type_t;
typedef struct zfs_rlock {
kmutex_t zr_mutex; /* protects changes to zr_avl */
avl_tree_t zr_avl; /* avl tree of range locks */
uint64_t *zr_size; /* points to znode->z_size */
uint_t *zr_blksz; /* points to znode->z_blksz */
uint64_t *zr_max_blksz; /* points to zsb->z_max_blksz */
} zfs_rlock_t;
2008-11-20 23:01:55 +03:00
typedef struct rl {
zfs_rlock_t *r_zrl;
2008-11-20 23:01:55 +03:00
avl_node_t r_node; /* avl node link */
uint64_t r_off; /* file range offset */
uint64_t r_len; /* file range length */
uint_t r_cnt; /* range reference count in tree */
rl_type_t r_type; /* range type */
kcondvar_t r_wr_cv; /* cv for waiting writers */
kcondvar_t r_rd_cv; /* cv for waiting readers */
uint8_t r_proxy; /* acting for original range */
uint8_t r_write_wanted; /* writer wants to lock this range */
uint8_t r_read_wanted; /* reader wants to lock this range */
Range lock performance improvements The original range lock implementation had to be modified by commit 8926ab7 because it was unsafe on Linux. In particular, calling cv_destroy() immediately after cv_broadcast() is dangerous because the waiters may still be asleep. Thus the following cv_destroy() will free memory which may still be in use. This was fixed by updating cv_destroy() to block on waiters but this in turn introduced a deadlock. The deadlock was resolved with the use of a taskq to move the offending free outside the range lock. This worked well but using the taskq for the free resulted in a serious performace hit. This is somewhat ironic because at the time I felt using the taskq might improve things by making the free asynchronous. This patch refines the original fix and moves the free from the taskq to a private free list. Then items which must be free'd are simply inserted in to the list. When the range lock is dropped it's safe to free the items. The list is walked and all rl_t entries are freed. This change improves small cached read performance by 26x. This was expected because for small reads the number of locking calls goes up significantly. More surprisingly this change significantly improves large cache read performance. This probably attributable to better cpu/memory locality. Very likely the same processor which allocated the memory is now freeing it. bs ext3 zfs zfs+fix faster ---------------------------------------------- 512 435 3 79 26x 1k 820 7 160 22x 2k 1536 14 305 21x 4k 2764 28 572 20x 8k 3788 50 1024 20x 16k 4300 86 1843 21x 32k 4505 138 2560 18x 64k 5324 252 3891 15x 128k 5427 276 4710 17x 256k 5427 413 5017 12x 512k 5427 497 5324 10x 1m 5427 521 5632 10x Closes #142
2011-03-08 23:17:35 +03:00
list_node_t rl_node; /* used for deferred release */
2008-11-20 23:01:55 +03:00
} rl_t;
/*
* Lock a range (offset, length) as either shared (RL_READER)
* or exclusive (RL_WRITER or RL_APPEND). RL_APPEND is a special type that
* is converted to RL_WRITER that specified to lock from the start of the
* end of file. Returns the range lock structure.
2008-11-20 23:01:55 +03:00
*/
rl_t *zfs_range_lock(zfs_rlock_t *zrl, uint64_t off, uint64_t len,
rl_type_t type);
2008-11-20 23:01:55 +03:00
/* Unlock range and destroy range lock structure. */
2008-11-20 23:01:55 +03:00
void zfs_range_unlock(rl_t *rl);
/*
* Reduce range locked as RW_WRITER from whole file to specified range.
* Asserts the whole file was previously locked.
*/
void zfs_range_reduce(rl_t *rl, uint64_t off, uint64_t len);
/*
* AVL comparison function used to order range locks
* Locks are ordered on the start offset of the range.
2008-11-20 23:01:55 +03:00
*/
int zfs_range_compare(const void *arg1, const void *arg2);
static inline void
zfs_rlock_init(zfs_rlock_t *zrl)
{
mutex_init(&zrl->zr_mutex, NULL, MUTEX_DEFAULT, NULL);
avl_create(&zrl->zr_avl, zfs_range_compare,
sizeof (rl_t), offsetof(rl_t, r_node));
zrl->zr_size = NULL;
zrl->zr_blksz = NULL;
zrl->zr_max_blksz = NULL;
}
static inline void
zfs_rlock_destroy(zfs_rlock_t *zrl)
{
avl_destroy(&zrl->zr_avl);
mutex_destroy(&zrl->zr_mutex);
}
2008-11-20 23:01:55 +03:00
#ifdef __cplusplus
}
#endif
#endif /* _SYS_FS_ZFS_RLOCK_H */