Commit a167b17e authored by David Chinner's avatar David Chinner Committed by Lachlan McIlroy

[XFS] move xfssyncd code to xfs_sync.c

Move all the xfssyncd code to the new xfs_sync.c file. This places it
closer to the actual code that it interacts with, rather than just being
associated with high level VFS code.

SGI-PV: 988139

SGI-Modid: xfs-linux-melb:xfs-kern:32283a
Signed-off-by: default avatarDavid Chinner <david@fromorbit.com>
Signed-off-by: default avatarLachlan McIlroy <lachlan@sgi.com>
Signed-off-by: default avatarChristoph Hellwig <hch@infradead.org>
parent fe4fa4b8
...@@ -979,146 +979,6 @@ xfs_fs_clear_inode( ...@@ -979,146 +979,6 @@ xfs_fs_clear_inode(
ASSERT(XFS_I(inode) == NULL); ASSERT(XFS_I(inode) == NULL);
} }
/*
* Enqueue a work item to be picked up by the vfs xfssyncd thread.
* Doing this has two advantages:
* - It saves on stack space, which is tight in certain situations
* - It can be used (with care) as a mechanism to avoid deadlocks.
* Flushing while allocating in a full filesystem requires both.
*/
STATIC void
xfs_syncd_queue_work(
struct xfs_mount *mp,
void *data,
void (*syncer)(struct xfs_mount *, void *))
{
struct bhv_vfs_sync_work *work;
work = kmem_alloc(sizeof(struct bhv_vfs_sync_work), KM_SLEEP);
INIT_LIST_HEAD(&work->w_list);
work->w_syncer = syncer;
work->w_data = data;
work->w_mount = mp;
spin_lock(&mp->m_sync_lock);
list_add_tail(&work->w_list, &mp->m_sync_list);
spin_unlock(&mp->m_sync_lock);
wake_up_process(mp->m_sync_task);
}
/*
* Flush delayed allocate data, attempting to free up reserved space
* from existing allocations. At this point a new allocation attempt
* has failed with ENOSPC and we are in the process of scratching our
* heads, looking about for more room...
*/
STATIC void
xfs_flush_inode_work(
struct xfs_mount *mp,
void *arg)
{
struct inode *inode = arg;
filemap_flush(inode->i_mapping);
iput(inode);
}
void
xfs_flush_inode(
xfs_inode_t *ip)
{
struct inode *inode = VFS_I(ip);
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work);
delay(msecs_to_jiffies(500));
}
/*
* This is the "bigger hammer" version of xfs_flush_inode_work...
* (IOW, "If at first you don't succeed, use a Bigger Hammer").
*/
STATIC void
xfs_flush_device_work(
struct xfs_mount *mp,
void *arg)
{
struct inode *inode = arg;
sync_blockdev(mp->m_super->s_bdev);
iput(inode);
}
void
xfs_flush_device(
xfs_inode_t *ip)
{
struct inode *inode = VFS_I(ip);
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work);
delay(msecs_to_jiffies(500));
xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC);
}
STATIC void
xfs_sync_worker(
struct xfs_mount *mp,
void *unused)
{
int error;
if (!(mp->m_flags & XFS_MOUNT_RDONLY))
error = xfs_sync(mp, SYNC_FSDATA | SYNC_BDFLUSH | SYNC_ATTR);
mp->m_sync_seq++;
wake_up(&mp->m_wait_single_sync_task);
}
STATIC int
xfssyncd(
void *arg)
{
struct xfs_mount *mp = arg;
long timeleft;
bhv_vfs_sync_work_t *work, *n;
LIST_HEAD (tmp);
set_freezable();
timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10);
for (;;) {
timeleft = schedule_timeout_interruptible(timeleft);
/* swsusp */
try_to_freeze();
if (kthread_should_stop() && list_empty(&mp->m_sync_list))
break;
spin_lock(&mp->m_sync_lock);
/*
* We can get woken by laptop mode, to do a sync -
* that's the (only!) case where the list would be
* empty with time remaining.
*/
if (!timeleft || list_empty(&mp->m_sync_list)) {
if (!timeleft)
timeleft = xfs_syncd_centisecs *
msecs_to_jiffies(10);
INIT_LIST_HEAD(&mp->m_sync_work.w_list);
list_add_tail(&mp->m_sync_work.w_list,
&mp->m_sync_list);
}
list_for_each_entry_safe(work, n, &mp->m_sync_list, w_list)
list_move(&work->w_list, &tmp);
spin_unlock(&mp->m_sync_lock);
list_for_each_entry_safe(work, n, &tmp, w_list) {
(*work->w_syncer)(mp, work->w_data);
list_del(&work->w_list);
if (work == &mp->m_sync_work)
continue;
kmem_free(work);
}
}
return 0;
}
STATIC void STATIC void
xfs_free_fsname( xfs_free_fsname(
struct xfs_mount *mp) struct xfs_mount *mp)
...@@ -1137,8 +997,7 @@ xfs_fs_put_super( ...@@ -1137,8 +997,7 @@ xfs_fs_put_super(
int unmount_event_flags = 0; int unmount_event_flags = 0;
int error; int error;
kthread_stop(mp->m_sync_task); xfs_syncd_stop(mp);
xfs_sync(mp, SYNC_ATTR | SYNC_DELWRI); xfs_sync(mp, SYNC_ATTR | SYNC_DELWRI);
#ifdef HAVE_DMAPI #ifdef HAVE_DMAPI
...@@ -1808,13 +1667,9 @@ xfs_fs_fill_super( ...@@ -1808,13 +1667,9 @@ xfs_fs_fill_super(
goto fail_vnrele; goto fail_vnrele;
} }
mp->m_sync_work.w_syncer = xfs_sync_worker; error = xfs_syncd_init(mp);
mp->m_sync_work.w_mount = mp; if (error)
mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd");
if (IS_ERR(mp->m_sync_task)) {
error = -PTR_ERR(mp->m_sync_task);
goto fail_vnrele; goto fail_vnrele;
}
xfs_itrace_exit(XFS_I(sb->s_root->d_inode)); xfs_itrace_exit(XFS_I(sb->s_root->d_inode));
......
...@@ -101,9 +101,6 @@ struct block_device; ...@@ -101,9 +101,6 @@ struct block_device;
extern __uint64_t xfs_max_file_offset(unsigned int); extern __uint64_t xfs_max_file_offset(unsigned int);
extern void xfs_flush_inode(struct xfs_inode *);
extern void xfs_flush_device(struct xfs_inode *);
extern void xfs_blkdev_issue_flush(struct xfs_buftarg *); extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
extern const struct export_operations xfs_export_operations; extern const struct export_operations xfs_export_operations;
......
...@@ -44,6 +44,9 @@ ...@@ -44,6 +44,9 @@
#include "xfs_inode_item.h" #include "xfs_inode_item.h"
#include "xfs_rw.h" #include "xfs_rw.h"
#include <linux/kthread.h>
#include <linux/freezer.h>
/* /*
* xfs_sync flushes any pending I/O to file system vfsp. * xfs_sync flushes any pending I/O to file system vfsp.
* *
...@@ -603,3 +606,163 @@ xfs_syncsub( ...@@ -603,3 +606,163 @@ xfs_syncsub(
return XFS_ERROR(last_error); return XFS_ERROR(last_error);
} }
/*
* Enqueue a work item to be picked up by the vfs xfssyncd thread.
* Doing this has two advantages:
* - It saves on stack space, which is tight in certain situations
* - It can be used (with care) as a mechanism to avoid deadlocks.
* Flushing while allocating in a full filesystem requires both.
*/
STATIC void
xfs_syncd_queue_work(
struct xfs_mount *mp,
void *data,
void (*syncer)(struct xfs_mount *, void *))
{
struct bhv_vfs_sync_work *work;
work = kmem_alloc(sizeof(struct bhv_vfs_sync_work), KM_SLEEP);
INIT_LIST_HEAD(&work->w_list);
work->w_syncer = syncer;
work->w_data = data;
work->w_mount = mp;
spin_lock(&mp->m_sync_lock);
list_add_tail(&work->w_list, &mp->m_sync_list);
spin_unlock(&mp->m_sync_lock);
wake_up_process(mp->m_sync_task);
}
/*
* Flush delayed allocate data, attempting to free up reserved space
* from existing allocations. At this point a new allocation attempt
* has failed with ENOSPC and we are in the process of scratching our
* heads, looking about for more room...
*/
STATIC void
xfs_flush_inode_work(
struct xfs_mount *mp,
void *arg)
{
struct inode *inode = arg;
filemap_flush(inode->i_mapping);
iput(inode);
}
void
xfs_flush_inode(
xfs_inode_t *ip)
{
struct inode *inode = VFS_I(ip);
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_inode_work);
delay(msecs_to_jiffies(500));
}
/*
* This is the "bigger hammer" version of xfs_flush_inode_work...
* (IOW, "If at first you don't succeed, use a Bigger Hammer").
*/
STATIC void
xfs_flush_device_work(
struct xfs_mount *mp,
void *arg)
{
struct inode *inode = arg;
sync_blockdev(mp->m_super->s_bdev);
iput(inode);
}
void
xfs_flush_device(
xfs_inode_t *ip)
{
struct inode *inode = VFS_I(ip);
igrab(inode);
xfs_syncd_queue_work(ip->i_mount, inode, xfs_flush_device_work);
delay(msecs_to_jiffies(500));
xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC);
}
STATIC void
xfs_sync_worker(
struct xfs_mount *mp,
void *unused)
{
int error;
if (!(mp->m_flags & XFS_MOUNT_RDONLY))
error = xfs_sync(mp, SYNC_FSDATA | SYNC_BDFLUSH | SYNC_ATTR);
mp->m_sync_seq++;
wake_up(&mp->m_wait_single_sync_task);
}
STATIC int
xfssyncd(
void *arg)
{
struct xfs_mount *mp = arg;
long timeleft;
bhv_vfs_sync_work_t *work, *n;
LIST_HEAD (tmp);
set_freezable();
timeleft = xfs_syncd_centisecs * msecs_to_jiffies(10);
for (;;) {
timeleft = schedule_timeout_interruptible(timeleft);
/* swsusp */
try_to_freeze();
if (kthread_should_stop() && list_empty(&mp->m_sync_list))
break;
spin_lock(&mp->m_sync_lock);
/*
* We can get woken by laptop mode, to do a sync -
* that's the (only!) case where the list would be
* empty with time remaining.
*/
if (!timeleft || list_empty(&mp->m_sync_list)) {
if (!timeleft)
timeleft = xfs_syncd_centisecs *
msecs_to_jiffies(10);
INIT_LIST_HEAD(&mp->m_sync_work.w_list);
list_add_tail(&mp->m_sync_work.w_list,
&mp->m_sync_list);
}
list_for_each_entry_safe(work, n, &mp->m_sync_list, w_list)
list_move(&work->w_list, &tmp);
spin_unlock(&mp->m_sync_lock);
list_for_each_entry_safe(work, n, &tmp, w_list) {
(*work->w_syncer)(mp, work->w_data);
list_del(&work->w_list);
if (work == &mp->m_sync_work)
continue;
kmem_free(work);
}
}
return 0;
}
int
xfs_syncd_init(
struct xfs_mount *mp)
{
mp->m_sync_work.w_syncer = xfs_sync_worker;
mp->m_sync_work.w_mount = mp;
mp->m_sync_task = kthread_run(xfssyncd, mp, "xfssyncd");
if (IS_ERR(mp->m_sync_task))
return -PTR_ERR(mp->m_sync_task);
return 0;
}
void
xfs_syncd_stop(
struct xfs_mount *mp)
{
kthread_stop(mp->m_sync_task);
}
/*
* Copyright (c) 2000-2006 Silicon Graphics, Inc.
* All Rights Reserved.
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License as
* published by the Free Software Foundation.
*
* This program is distributed in the hope that it would be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with this program; if not, write the Free Software Foundation,
* Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
*/
#ifndef XFS_SYNC_H #ifndef XFS_SYNC_H
#define XFS_SYNC_H 1 #define XFS_SYNC_H 1
struct xfs_mount;
typedef struct bhv_vfs_sync_work {
struct list_head w_list;
struct xfs_mount *w_mount;
void *w_data; /* syncer routine argument */
void (*w_syncer)(struct xfs_mount *, void *);
} bhv_vfs_sync_work_t;
#define SYNC_ATTR 0x0001 /* sync attributes */
#define SYNC_CLOSE 0x0002 /* close file system down */
#define SYNC_DELWRI 0x0004 /* look at delayed writes */
#define SYNC_WAIT 0x0008 /* wait for i/o to complete */
#define SYNC_BDFLUSH 0x0010 /* BDFLUSH is calling -- don't block */
#define SYNC_FSDATA 0x0020 /* flush fs data (e.g. superblocks) */
#define SYNC_REFCACHE 0x0040 /* prune some of the nfs ref cache */
#define SYNC_REMOUNT 0x0080 /* remount readonly, no dummy LRs */
#define SYNC_IOWAIT 0x0100 /* wait for all I/O to complete */
/*
* When remounting a filesystem read-only or freezing the filesystem,
* we have two phases to execute. This first phase is syncing the data
* before we quiesce the fielsystem, and the second is flushing all the
* inodes out after we've waited for all the transactions created by
* the first phase to complete. The second phase uses SYNC_INODE_QUIESCE
* to ensure that the inodes are written to their location on disk
* rather than just existing in transactions in the log. This means
* after a quiesce there is no log replay required to write the inodes
* to disk (this is the main difference between a sync and a quiesce).
*/
#define SYNC_DATA_QUIESCE (SYNC_DELWRI|SYNC_FSDATA|SYNC_WAIT|SYNC_IOWAIT)
#define SYNC_INODE_QUIESCE (SYNC_REMOUNT|SYNC_ATTR|SYNC_WAIT)
int xfs_syncd_init(struct xfs_mount *mp);
void xfs_syncd_stop(struct xfs_mount *mp);
int xfs_sync(struct xfs_mount *mp, int flags); int xfs_sync(struct xfs_mount *mp, int flags);
int xfs_syncsub(struct xfs_mount *mp, int flags, int *bypassed); int xfs_syncsub(struct xfs_mount *mp, int flags, int *bypassed);
void xfs_flush_inode(struct xfs_inode *ip);
void xfs_flush_device(struct xfs_inode *ip);
#endif #endif
...@@ -33,37 +33,6 @@ struct xfs_mount_args; ...@@ -33,37 +33,6 @@ struct xfs_mount_args;
typedef struct kstatfs bhv_statvfs_t; typedef struct kstatfs bhv_statvfs_t;
typedef struct bhv_vfs_sync_work {
struct list_head w_list;
struct xfs_mount *w_mount;
void *w_data; /* syncer routine argument */
void (*w_syncer)(struct xfs_mount *, void *);
} bhv_vfs_sync_work_t;
#define SYNC_ATTR 0x0001 /* sync attributes */
#define SYNC_CLOSE 0x0002 /* close file system down */
#define SYNC_DELWRI 0x0004 /* look at delayed writes */
#define SYNC_WAIT 0x0008 /* wait for i/o to complete */
#define SYNC_BDFLUSH 0x0010 /* BDFLUSH is calling -- don't block */
#define SYNC_FSDATA 0x0020 /* flush fs data (e.g. superblocks) */
#define SYNC_REFCACHE 0x0040 /* prune some of the nfs ref cache */
#define SYNC_REMOUNT 0x0080 /* remount readonly, no dummy LRs */
#define SYNC_IOWAIT 0x0100 /* wait for all I/O to complete */
/*
* When remounting a filesystem read-only or freezing the filesystem,
* we have two phases to execute. This first phase is syncing the data
* before we quiesce the fielsystem, and the second is flushing all the
* inodes out after we've waited for all the transactions created by
* the first phase to complete. The second phase uses SYNC_INODE_QUIESCE
* to ensure that the inodes are written to their location on disk
* rather than just existing in transactions in the log. This means
* after a quiesce there is no log replay required to write the inodes
* to disk (this is the main difference between a sync and a quiesce).
*/
#define SYNC_DATA_QUIESCE (SYNC_DELWRI|SYNC_FSDATA|SYNC_WAIT|SYNC_IOWAIT)
#define SYNC_INODE_QUIESCE (SYNC_REMOUNT|SYNC_ATTR|SYNC_WAIT)
#define SHUTDOWN_META_IO_ERROR 0x0001 /* write attempt to metadata failed */ #define SHUTDOWN_META_IO_ERROR 0x0001 /* write attempt to metadata failed */
#define SHUTDOWN_LOG_IO_ERROR 0x0002 /* write attempt to the log failed */ #define SHUTDOWN_LOG_IO_ERROR 0x0002 /* write attempt to the log failed */
#define SHUTDOWN_FORCE_UMOUNT 0x0004 /* shutdown from a forced unmount */ #define SHUTDOWN_FORCE_UMOUNT 0x0004 /* shutdown from a forced unmount */
......
...@@ -18,6 +18,7 @@ ...@@ -18,6 +18,7 @@
#ifndef __XFS_MOUNT_H__ #ifndef __XFS_MOUNT_H__
#define __XFS_MOUNT_H__ #define __XFS_MOUNT_H__
#include "xfs_sync.h"
typedef struct xfs_trans_reservations { typedef struct xfs_trans_reservations {
uint tr_write; /* extent alloc trans */ uint tr_write; /* extent alloc trans */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment