2005-04-17 02:20:36 +04:00
|
|
|
/*
|
2011-09-21 09:37:46 +04:00
|
|
|
md.h : kernel internal structure of the Linux MD driver
|
2005-04-17 02:20:36 +04:00
|
|
|
Copyright (C) 1996-98 Ingo Molnar, Gadi Oxman
|
2014-09-30 08:23:59 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
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; either version 2, or (at your option)
|
|
|
|
any later version.
|
2014-09-30 08:23:59 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
You should have received a copy of the GNU General Public License
|
|
|
|
(for example /usr/src/linux/COPYING); if not, write to the Free
|
2014-09-30 08:23:59 +04:00
|
|
|
Software Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.
|
2005-04-17 02:20:36 +04:00
|
|
|
*/
|
|
|
|
|
2009-04-14 06:01:53 +04:00
|
|
|
#ifndef _MD_MD_H
|
|
|
|
#define _MD_MD_H
|
|
|
|
|
|
|
|
#include <linux/blkdev.h>
|
|
|
|
#include <linux/kobject.h>
|
|
|
|
#include <linux/list.h>
|
|
|
|
#include <linux/mm.h>
|
|
|
|
#include <linux/mutex.h>
|
|
|
|
#include <linux/timer.h>
|
|
|
|
#include <linux/wait.h>
|
|
|
|
#include <linux/workqueue.h>
|
2014-03-29 19:01:53 +04:00
|
|
|
#include "md-cluster.h"
|
[PATCH] BLOCK: Make it possible to disable the block layer [try #6]
Make it possible to disable the block layer. Not all embedded devices require
it, some can make do with just JFFS2, NFS, ramfs, etc - none of which require
the block layer to be present.
This patch does the following:
(*) Introduces CONFIG_BLOCK to disable the block layer, buffering and blockdev
support.
(*) Adds dependencies on CONFIG_BLOCK to any configuration item that controls
an item that uses the block layer. This includes:
(*) Block I/O tracing.
(*) Disk partition code.
(*) All filesystems that are block based, eg: Ext3, ReiserFS, ISOFS.
(*) The SCSI layer. As far as I can tell, even SCSI chardevs use the
block layer to do scheduling. Some drivers that use SCSI facilities -
such as USB storage - end up disabled indirectly from this.
(*) Various block-based device drivers, such as IDE and the old CDROM
drivers.
(*) MTD blockdev handling and FTL.
(*) JFFS - which uses set_bdev_super(), something it could avoid doing by
taking a leaf out of JFFS2's book.
(*) Makes most of the contents of linux/blkdev.h, linux/buffer_head.h and
linux/elevator.h contingent on CONFIG_BLOCK being set. sector_div() is,
however, still used in places, and so is still available.
(*) Also made contingent are the contents of linux/mpage.h, linux/genhd.h and
parts of linux/fs.h.
(*) Makes a number of files in fs/ contingent on CONFIG_BLOCK.
(*) Makes mm/bounce.c (bounce buffering) contingent on CONFIG_BLOCK.
(*) set_page_dirty() doesn't call __set_page_dirty_buffers() if CONFIG_BLOCK
is not enabled.
(*) fs/no-block.c is created to hold out-of-line stubs and things that are
required when CONFIG_BLOCK is not set:
(*) Default blockdev file operations (to give error ENODEV on opening).
(*) Makes some /proc changes:
(*) /proc/devices does not list any blockdevs.
(*) /proc/diskstats and /proc/partitions are contingent on CONFIG_BLOCK.
(*) Makes some compat ioctl handling contingent on CONFIG_BLOCK.
(*) If CONFIG_BLOCK is not defined, makes sys_quotactl() return -ENODEV if
given command other than Q_SYNC or if a special device is specified.
(*) In init/do_mounts.c, no reference is made to the blockdev routines if
CONFIG_BLOCK is not defined. This does not prohibit NFS roots or JFFS2.
(*) The bdflush, ioprio_set and ioprio_get syscalls can now be absent (return
error ENOSYS by way of cond_syscall if so).
(*) The seclvl_bd_claim() and seclvl_bd_release() security calls do nothing if
CONFIG_BLOCK is not set, since they can't then happen.
Signed-Off-By: David Howells <dhowells@redhat.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>
2006-09-30 22:45:40 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
#define MaxSector (~(sector_t)0)
|
|
|
|
|
2011-07-28 05:31:46 +04:00
|
|
|
/* Bad block numbers are stored sorted in a single page.
|
|
|
|
* 64bits is used for each block or extent.
|
|
|
|
* 54 bits are sector number, 9 bits are extent size,
|
|
|
|
* 1 bit is an 'acknowledged' flag.
|
|
|
|
*/
|
|
|
|
#define MD_MAX_BADBLOCKS (PAGE_SIZE/8)
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/*
|
|
|
|
* MD's 'extended' device
|
|
|
|
*/
|
2011-10-11 09:45:26 +04:00
|
|
|
struct md_rdev {
|
2005-04-17 02:20:36 +04:00
|
|
|
struct list_head same_set; /* RAID devices within the same set */
|
|
|
|
|
2009-03-31 07:33:13 +04:00
|
|
|
sector_t sectors; /* Device size (in 512bytes sectors) */
|
2011-10-11 09:47:53 +04:00
|
|
|
struct mddev *mddev; /* RAID array if running */
|
2009-03-31 07:27:02 +04:00
|
|
|
int last_events; /* IO event timestamp */
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2011-01-14 01:14:34 +03:00
|
|
|
/*
|
|
|
|
* If meta_bdev is non-NULL, it means that a separate device is
|
|
|
|
* being used to store the metadata (superblock/bitmap) which
|
|
|
|
* would otherwise be contained on the same device as the data (bdev).
|
|
|
|
*/
|
|
|
|
struct block_device *meta_bdev;
|
2005-04-17 02:20:36 +04:00
|
|
|
struct block_device *bdev; /* block device handle */
|
|
|
|
|
2011-07-28 05:31:47 +04:00
|
|
|
struct page *sb_page, *bb_page;
|
2005-04-17 02:20:36 +04:00
|
|
|
int sb_loaded;
|
2006-06-26 11:27:57 +04:00
|
|
|
__u64 sb_events;
|
2005-04-17 02:20:36 +04:00
|
|
|
sector_t data_offset; /* start of data in array */
|
2012-05-21 03:27:00 +04:00
|
|
|
sector_t new_data_offset;/* only relevant while reshaping */
|
2014-09-30 08:23:59 +04:00
|
|
|
sector_t sb_start; /* offset of the super block (in 512byte sectors) */
|
2005-09-10 03:23:53 +04:00
|
|
|
int sb_size; /* bytes in the superblock */
|
2005-04-17 02:20:36 +04:00
|
|
|
int preferred_minor; /* autorun support */
|
|
|
|
|
2005-11-09 08:39:24 +03:00
|
|
|
struct kobject kobj;
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* A device can be in one of three states based on two flags:
|
|
|
|
* Not working: faulty==1 in_sync==0
|
|
|
|
* Fully working: faulty==0 in_sync==1
|
|
|
|
* Working, but not
|
|
|
|
* in sync with array
|
|
|
|
* faulty==0 in_sync==0
|
|
|
|
*
|
|
|
|
* It can never have faulty==1, in_sync==1
|
|
|
|
* This reduces the burden of testing multiple flags in many cases
|
|
|
|
*/
|
|
|
|
|
2011-12-23 03:17:51 +04:00
|
|
|
unsigned long flags; /* bit set of 'enum flag_bits' bits. */
|
2008-04-30 11:52:32 +04:00
|
|
|
wait_queue_head_t blocked_wait;
|
2005-09-10 03:23:45 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
int desc_nr; /* descriptor index in the superblock */
|
|
|
|
int raid_disk; /* role of device in array */
|
2010-06-15 12:36:03 +04:00
|
|
|
int new_raid_disk; /* role that the device will have in
|
|
|
|
* the array after a level-change completes.
|
|
|
|
*/
|
2005-06-22 04:17:25 +04:00
|
|
|
int saved_raid_disk; /* role that device used to have in the
|
|
|
|
* array and could again if we did a partial
|
|
|
|
* resync from the bitmap
|
|
|
|
*/
|
2006-06-26 11:27:40 +04:00
|
|
|
sector_t recovery_offset;/* If this device has been partially
|
|
|
|
* recovered, this is where we were
|
|
|
|
* up to.
|
|
|
|
*/
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
atomic_t nr_pending; /* number of pending requests.
|
|
|
|
* only maintained for arrays that
|
|
|
|
* support hot removal
|
|
|
|
*/
|
2005-11-09 08:39:31 +03:00
|
|
|
atomic_t read_errors; /* number of consecutive read errors that
|
|
|
|
* we have tried to ignore.
|
|
|
|
*/
|
2009-12-14 04:49:58 +03:00
|
|
|
struct timespec last_read_error; /* monotonic time since our
|
|
|
|
* last read error
|
|
|
|
*/
|
2006-01-06 11:20:52 +03:00
|
|
|
atomic_t corrected_errors; /* number of corrected read errors,
|
|
|
|
* for reporting to userspace and storing
|
|
|
|
* in superblock.
|
|
|
|
*/
|
2007-04-05 06:08:18 +04:00
|
|
|
struct work_struct del_work; /* used for delayed sysfs removal */
|
2008-10-21 06:25:28 +04:00
|
|
|
|
2013-12-11 23:11:53 +04:00
|
|
|
struct kernfs_node *sysfs_state; /* handle for 'state'
|
2008-10-21 06:25:28 +04:00
|
|
|
* sysfs entry */
|
2011-07-28 05:31:46 +04:00
|
|
|
|
|
|
|
struct badblocks {
|
|
|
|
int count; /* count of bad blocks */
|
2011-07-28 05:31:48 +04:00
|
|
|
int unacked_exist; /* there probably are unacknowledged
|
|
|
|
* bad blocks. This is only cleared
|
|
|
|
* when a read discovers none
|
|
|
|
*/
|
2011-07-28 05:31:46 +04:00
|
|
|
int shift; /* shift from sectors to block size
|
|
|
|
* a -ve shift means badblocks are
|
|
|
|
* disabled.*/
|
|
|
|
u64 *page; /* badblock list */
|
|
|
|
int changed;
|
|
|
|
seqlock_t lock;
|
2011-07-28 05:31:47 +04:00
|
|
|
|
|
|
|
sector_t sector;
|
|
|
|
sector_t size; /* in sectors */
|
2011-07-28 05:31:46 +04:00
|
|
|
} badblocks;
|
2005-04-17 02:20:36 +04:00
|
|
|
};
|
2011-12-23 03:17:51 +04:00
|
|
|
enum flag_bits {
|
|
|
|
Faulty, /* device is known to have a fault */
|
|
|
|
In_sync, /* device is in_sync with rest of array */
|
2013-12-12 03:13:33 +04:00
|
|
|
Bitmap_sync, /* ..actually, not quite In_sync. Need a
|
|
|
|
* bitmap-based recovery to get fully in sync
|
|
|
|
*/
|
2012-03-19 05:46:39 +04:00
|
|
|
Unmerged, /* device is being added to array and should
|
|
|
|
* be considerred for bvec_merge_fn but not
|
|
|
|
* yet for actual IO
|
|
|
|
*/
|
2011-12-23 03:17:51 +04:00
|
|
|
WriteMostly, /* Avoid reading if at all possible */
|
|
|
|
AutoDetected, /* added by auto-detect */
|
|
|
|
Blocked, /* An error occurred but has not yet
|
|
|
|
* been acknowledged by the metadata
|
|
|
|
* handler, so don't allow writes
|
|
|
|
* until it is cleared */
|
|
|
|
WriteErrorSeen, /* A write error has been seen on this
|
|
|
|
* device
|
|
|
|
*/
|
|
|
|
FaultRecorded, /* Intermediate state for clearing
|
|
|
|
* Blocked. The Fault is/will-be
|
|
|
|
* recorded in the metadata, but that
|
|
|
|
* metadata hasn't been stored safely
|
|
|
|
* on disk yet.
|
|
|
|
*/
|
|
|
|
BlockedBadBlocks, /* A writer is blocked because they
|
|
|
|
* found an unacknowledged bad-block.
|
|
|
|
* This can safely be cleared at any
|
|
|
|
* time, and the writer will re-check.
|
|
|
|
* It may be set at any time, and at
|
|
|
|
* worst the writer will timeout and
|
|
|
|
* re-check. So setting it as
|
|
|
|
* accurately as possible is good, but
|
|
|
|
* not absolutely critical.
|
|
|
|
*/
|
|
|
|
WantReplacement, /* This device is a candidate to be
|
|
|
|
* hot-replaced, either because it has
|
|
|
|
* reported some faults, or because
|
|
|
|
* of explicit request.
|
|
|
|
*/
|
|
|
|
Replacement, /* This device is a replacement for
|
|
|
|
* a want_replacement device with same
|
|
|
|
* raid_disk number.
|
|
|
|
*/
|
|
|
|
};
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2011-07-28 05:31:46 +04:00
|
|
|
#define BB_LEN_MASK (0x00000000000001FFULL)
|
|
|
|
#define BB_OFFSET_MASK (0x7FFFFFFFFFFFFE00ULL)
|
|
|
|
#define BB_ACK_MASK (0x8000000000000000ULL)
|
|
|
|
#define BB_MAX_LEN 512
|
|
|
|
#define BB_OFFSET(x) (((x) & BB_OFFSET_MASK) >> 9)
|
|
|
|
#define BB_LEN(x) (((x) & BB_LEN_MASK) + 1)
|
|
|
|
#define BB_ACK(x) (!!((x) & BB_ACK_MASK))
|
|
|
|
#define BB_MAKE(a, l, ack) (((a)<<9) | ((l)-1) | ((u64)(!!(ack)) << 63))
|
|
|
|
|
|
|
|
extern int md_is_badblock(struct badblocks *bb, sector_t s, int sectors,
|
|
|
|
sector_t *first_bad, int *bad_sectors);
|
2011-10-11 09:45:26 +04:00
|
|
|
static inline int is_badblock(struct md_rdev *rdev, sector_t s, int sectors,
|
2011-07-28 05:31:46 +04:00
|
|
|
sector_t *first_bad, int *bad_sectors)
|
|
|
|
{
|
|
|
|
if (unlikely(rdev->badblocks.count)) {
|
|
|
|
int rv = md_is_badblock(&rdev->badblocks, rdev->data_offset + s,
|
|
|
|
sectors,
|
|
|
|
first_bad, bad_sectors);
|
|
|
|
if (rv)
|
|
|
|
*first_bad -= rdev->data_offset;
|
|
|
|
return rv;
|
|
|
|
}
|
|
|
|
return 0;
|
|
|
|
}
|
2011-10-11 09:45:26 +04:00
|
|
|
extern int rdev_set_badblocks(struct md_rdev *rdev, sector_t s, int sectors,
|
2012-05-21 03:27:00 +04:00
|
|
|
int is_new);
|
|
|
|
extern int rdev_clear_badblocks(struct md_rdev *rdev, sector_t s, int sectors,
|
|
|
|
int is_new);
|
2011-07-28 05:31:46 +04:00
|
|
|
extern void md_ack_all_badblocks(struct badblocks *bb);
|
|
|
|
|
2014-03-29 19:20:02 +04:00
|
|
|
struct md_cluster_info;
|
|
|
|
|
2011-10-11 09:47:53 +04:00
|
|
|
struct mddev {
|
2005-04-17 02:20:36 +04:00
|
|
|
void *private;
|
2011-10-11 09:49:58 +04:00
|
|
|
struct md_personality *pers;
|
2005-04-17 02:20:36 +04:00
|
|
|
dev_t unit;
|
|
|
|
int md_minor;
|
2013-08-27 10:28:23 +04:00
|
|
|
struct list_head disks;
|
2006-10-03 12:15:46 +04:00
|
|
|
unsigned long flags;
|
|
|
|
#define MD_CHANGE_DEVS 0 /* Some device status has changed */
|
|
|
|
#define MD_CHANGE_CLEAN 1 /* transition to or from 'clean' */
|
2010-08-30 11:33:34 +04:00
|
|
|
#define MD_CHANGE_PENDING 2 /* switch from 'clean' to 'active' in progress */
|
2013-08-27 10:28:23 +04:00
|
|
|
#define MD_UPDATE_SB_FLAGS (1 | 2 | 4) /* If these are set, md_update_sb needed */
|
2011-06-09 02:59:30 +04:00
|
|
|
#define MD_ARRAY_FIRST_USE 3 /* First use of array, needs initialization */
|
2013-08-27 10:44:13 +04:00
|
|
|
#define MD_STILL_CLOSED 4 /* If set, then array has not been opened since
|
|
|
|
* md_ioctl checked on it.
|
|
|
|
*/
|
2006-10-03 12:15:46 +04:00
|
|
|
|
2009-03-31 07:39:39 +04:00
|
|
|
int suspended;
|
|
|
|
atomic_t active_io;
|
2005-04-17 02:20:36 +04:00
|
|
|
int ro;
|
2010-08-08 15:18:03 +04:00
|
|
|
int sysfs_active; /* set when sysfs deletes
|
|
|
|
* are happening, so run/
|
|
|
|
* takeover/stop are not safe
|
|
|
|
*/
|
2013-08-27 10:28:23 +04:00
|
|
|
int ready; /* See when safe to pass
|
2011-01-14 01:14:33 +03:00
|
|
|
* IO requests down */
|
2005-04-17 02:20:36 +04:00
|
|
|
struct gendisk *gendisk;
|
|
|
|
|
2005-11-09 08:39:23 +03:00
|
|
|
struct kobject kobj;
|
md: make devices disappear when they are no longer needed.
Currently md devices, once created, never disappear until the module
is unloaded. This is essentially because the gendisk holds a
reference to the mddev, and the mddev holds a reference to the
gendisk, this a circular reference.
If we drop the reference from mddev to gendisk, then we need to ensure
that the mddev is destroyed when the gendisk is destroyed. However it
is not possible to hook into the gendisk destruction process to enable
this.
So we drop the reference from the gendisk to the mddev and destroy the
gendisk when the mddev gets destroyed. However this has a
complication.
Between the call
__blkdev_get->get_gendisk->kobj_lookup->md_probe
and the call
__blkdev_get->md_open
there is no obvious way to hold a reference on the mddev any more, so
unless something is done, it will disappear and gendisk will be
destroyed prematurely.
Also, once we decide to destroy the mddev, there will be an unlockable
moment before the gendisk is unlinked (blk_unregister_region) during
which a new reference to the gendisk can be created. We need to
ensure that this reference can not be used. i.e. the ->open must
fail.
So:
1/ in md_probe we set a flag in the mddev (hold_active) which
indicates that the array should be treated as active, even
though there are no references, and no appearance of activity.
This is cleared by md_release when the device is closed if it
is no longer needed.
This ensures that the gendisk will survive between md_probe and
md_open.
2/ In md_open we check if the mddev we expect to open matches
the gendisk that we did open.
If there is a mismatch we return -ERESTARTSYS and modify
__blkdev_get to retry from the top in that case.
In the -ERESTARTSYS sys case we make sure to wait until
the old gendisk (that we succeeded in opening) is really gone so
we loop at most once.
Some udev configurations will always open an md device when it first
appears. If we allow an md device that was just created by an open
to disappear on an immediate close, then this can race with such udev
configurations and result in an infinite loop the device being opened
and closed, then re-open due to the 'ADD' even from the first open,
and then close and so on.
So we make sure an md device, once created by an open, remains active
at least until some md 'ioctl' has been made on it. This means that
all normal usage of md devices will allow them to disappear promptly
when not needed, but the worst that an incorrect usage will do it
cause an inactive md device to be left in existence (it can easily be
removed).
As an array can be stopped by writing to a sysfs attribute
echo clear > /sys/block/mdXXX/md/array_state
we need to use scheduled work for deleting the gendisk and other
kobjects. This allows us to wait for any pending gendisk deletion to
complete by simply calling flush_scheduled_work().
Signed-off-by: NeilBrown <neilb@suse.de>
2009-01-09 00:31:10 +03:00
|
|
|
int hold_active;
|
|
|
|
#define UNTIL_IOCTL 1
|
2009-01-09 00:31:10 +03:00
|
|
|
#define UNTIL_STOP 2
|
2005-11-09 08:39:23 +03:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/* Superblock information */
|
|
|
|
int major_version,
|
|
|
|
minor_version,
|
|
|
|
patch_version;
|
|
|
|
int persistent;
|
2014-09-30 08:23:59 +04:00
|
|
|
int external; /* metadata is
|
2008-02-06 12:39:51 +03:00
|
|
|
* managed externally */
|
|
|
|
char metadata_type[17]; /* externally set*/
|
2009-06-18 02:45:01 +04:00
|
|
|
int chunk_sectors;
|
2005-04-17 02:20:36 +04:00
|
|
|
time_t ctime, utime;
|
|
|
|
int level, layout;
|
2006-01-06 11:20:51 +03:00
|
|
|
char clevel[16];
|
2005-04-17 02:20:36 +04:00
|
|
|
int raid_disks;
|
|
|
|
int max_disks;
|
2014-09-30 08:23:59 +04:00
|
|
|
sector_t dev_sectors; /* used size of
|
2009-03-31 07:33:13 +04:00
|
|
|
* component devices */
|
2008-07-21 11:05:22 +04:00
|
|
|
sector_t array_sectors; /* exported array size */
|
2009-03-31 08:00:31 +04:00
|
|
|
int external_size; /* size managed
|
|
|
|
* externally */
|
2005-04-17 02:20:36 +04:00
|
|
|
__u64 events;
|
2010-05-18 03:28:43 +04:00
|
|
|
/* If the last 'event' was simply a clean->dirty transition, and
|
|
|
|
* we didn't write it to the spares, then it is safe and simple
|
|
|
|
* to just decrement the event count on a dirty->clean transition.
|
|
|
|
* So we record that possibility here.
|
|
|
|
*/
|
|
|
|
int can_decrease_events;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
char uuid[16];
|
|
|
|
|
2006-03-27 13:18:11 +04:00
|
|
|
/* If the array is being reshaped, we need to record the
|
|
|
|
* new shape and an indication of where we are up to.
|
|
|
|
* This is written to the superblock.
|
|
|
|
* If reshape_position is MaxSector, then no reshape is happening (yet).
|
|
|
|
*/
|
|
|
|
sector_t reshape_position;
|
2009-06-18 02:45:27 +04:00
|
|
|
int delta_disks, new_level, new_layout;
|
|
|
|
int new_chunk_sectors;
|
2012-05-21 03:27:00 +04:00
|
|
|
int reshape_backwards;
|
2006-03-27 13:18:11 +04:00
|
|
|
|
2011-10-11 09:48:23 +04:00
|
|
|
struct md_thread *thread; /* management thread */
|
|
|
|
struct md_thread *sync_thread; /* doing resync or reconstruct */
|
2013-06-25 10:23:59 +04:00
|
|
|
|
|
|
|
/* 'last_sync_action' is initialized to "none". It is set when a
|
|
|
|
* sync operation (i.e "data-check", "requested-resync", "resync",
|
|
|
|
* "recovery", or "reshape") is started. It holds this value even
|
|
|
|
* when the sync thread is "frozen" (interrupted) or "idle" (stopped
|
|
|
|
* or finished). It is overwritten when a new sync operation is begun.
|
|
|
|
*/
|
|
|
|
char *last_sync_action;
|
2006-07-10 15:44:16 +04:00
|
|
|
sector_t curr_resync; /* last block scheduled */
|
2009-03-31 07:33:13 +04:00
|
|
|
/* As resync requests can complete out of order, we cannot easily track
|
|
|
|
* how much resync has been completed. So we occasionally pause until
|
|
|
|
* everything completes, then set curr_resync_completed to curr_resync.
|
|
|
|
* As such it may be well behind the real resync mark, but it is a value
|
|
|
|
* we are certain of.
|
|
|
|
*/
|
|
|
|
sector_t curr_resync_completed;
|
2005-04-17 02:20:36 +04:00
|
|
|
unsigned long resync_mark; /* a recent timestamp */
|
|
|
|
sector_t resync_mark_cnt;/* blocks written at resync_mark */
|
2006-07-10 15:44:16 +04:00
|
|
|
sector_t curr_mark_cnt; /* blocks scheduled now */
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
sector_t resync_max_sectors; /* may be set by personality */
|
2005-11-09 08:39:26 +03:00
|
|
|
|
2012-10-11 07:17:59 +04:00
|
|
|
atomic64_t resync_mismatches; /* count of sectors where
|
2005-11-09 08:39:26 +03:00
|
|
|
* parity/replica mismatch found
|
|
|
|
*/
|
2006-03-27 13:18:14 +04:00
|
|
|
|
|
|
|
/* allow user-space to request suspension of IO to regions of the array */
|
|
|
|
sector_t suspend_lo;
|
|
|
|
sector_t suspend_hi;
|
2006-01-06 11:21:36 +03:00
|
|
|
/* if zero, use the system-wide default */
|
|
|
|
int sync_speed_min;
|
|
|
|
int sync_speed_max;
|
|
|
|
|
2008-05-24 00:04:38 +04:00
|
|
|
/* resync even though the same disks are shared among md-devices */
|
|
|
|
int parallel_resync;
|
|
|
|
|
2006-01-06 11:20:15 +03:00
|
|
|
int ok_start_degraded;
|
2014-09-30 08:23:59 +04:00
|
|
|
/* recovery/resync flags
|
2005-04-17 02:20:36 +04:00
|
|
|
* NEEDED: we might need to start a resync/recover
|
|
|
|
* RUNNING: a thread is running, or about to be started
|
|
|
|
* SYNC: actually doing a resync, not a recovery
|
2008-06-28 02:31:41 +04:00
|
|
|
* RECOVER: doing recovery, or need to try it.
|
md: restart recovery cleanly after device failure.
When we get any IO error during a recovery (rebuilding a spare), we abort
the recovery and restart it.
For RAID6 (and multi-drive RAID1) it may not be best to restart at the
beginning: when multiple failures can be tolerated, the recovery may be
able to continue and re-doing all that has already been done doesn't make
sense.
We already have the infrastructure to record where a recovery is up to
and restart from there, but it is not being used properly.
This is because:
- We sometimes abort with MD_RECOVERY_ERR rather than just MD_RECOVERY_INTR,
which causes the recovery not be be checkpointed.
- We remove spares and then re-added them which loses important state
information.
The distinction between MD_RECOVERY_ERR and MD_RECOVERY_INTR really isn't
needed. If there is an error, the relevant drive will be marked as
Faulty, and that is enough to ensure correct handling of the error. So we
first remove MD_RECOVERY_ERR, changing some of the uses of it to
MD_RECOVERY_INTR.
Then we cause the attempt to remove a non-faulty device from an array to
fail (unless recovery is impossible as the array is too degraded). Then
when remove_and_add_spares attempts to remove the devices on which
recovery can continue, it will fail, they will remain in place, and
recovery will continue on them as desired.
Issue: If we are halfway through rebuilding a spare and another drive
fails, and a new spare is immediately available, do we want to:
1/ complete the current rebuild, then go back and rebuild the new spare or
2/ restart the rebuild from the start and rebuild both devices in
parallel.
Both options can be argued for. The code currently takes option 2 as
a/ this requires least code change
b/ this results in a minimally-degraded array in minimal time.
Cc: "Eivind Sarto" <ivan@kasenna.com>
Signed-off-by: Neil Brown <neilb@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2008-05-24 00:04:39 +04:00
|
|
|
* INTR: resync needs to be aborted for some reason
|
2005-04-17 02:20:36 +04:00
|
|
|
* DONE: thread is done and is waiting to be reaped
|
2005-11-09 08:39:26 +03:00
|
|
|
* REQUEST: user-space has requested a sync (used with SYNC)
|
2009-07-07 13:54:23 +04:00
|
|
|
* CHECK: user-space request for check-only, no repair
|
2006-03-27 13:18:09 +04:00
|
|
|
* RESHAPE: A reshape is happening
|
2012-11-19 15:57:34 +04:00
|
|
|
* ERROR: sync-action interrupted because io-error
|
2006-03-27 13:18:09 +04:00
|
|
|
*
|
|
|
|
* If neither SYNC or RESHAPE are set, then it is a recovery.
|
2005-04-17 02:20:36 +04:00
|
|
|
*/
|
|
|
|
#define MD_RECOVERY_RUNNING 0
|
|
|
|
#define MD_RECOVERY_SYNC 1
|
2008-06-28 02:31:41 +04:00
|
|
|
#define MD_RECOVERY_RECOVER 2
|
2005-04-17 02:20:36 +04:00
|
|
|
#define MD_RECOVERY_INTR 3
|
|
|
|
#define MD_RECOVERY_DONE 4
|
|
|
|
#define MD_RECOVERY_NEEDED 5
|
2005-11-09 08:39:26 +03:00
|
|
|
#define MD_RECOVERY_REQUESTED 6
|
|
|
|
#define MD_RECOVERY_CHECK 7
|
2006-03-27 13:18:09 +04:00
|
|
|
#define MD_RECOVERY_RESHAPE 8
|
2006-06-26 11:27:40 +04:00
|
|
|
#define MD_RECOVERY_FROZEN 9
|
2012-11-19 15:57:34 +04:00
|
|
|
#define MD_RECOVERY_ERROR 10
|
2006-06-26 11:27:40 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
unsigned long recovery;
|
2011-07-27 05:00:36 +04:00
|
|
|
/* If a RAID personality determines that recovery (of a particular
|
|
|
|
* device) will fail due to a read error on the source device, it
|
|
|
|
* takes a copy of this number and does not attempt recovery again
|
|
|
|
* until this number changes.
|
|
|
|
*/
|
|
|
|
int recovery_disabled;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
int in_sync; /* know to not need resync */
|
2009-08-10 06:50:52 +04:00
|
|
|
/* 'open_mutex' avoids races between 'md_open' and 'do_md_stop', so
|
|
|
|
* that we are never stopping an array while it is open.
|
|
|
|
* 'reconfig_mutex' protects all other reconfiguration.
|
|
|
|
* These locks are separate due to conflicting interactions
|
|
|
|
* with bdev->bd_mutex.
|
|
|
|
* Lock ordering is:
|
|
|
|
* reconfig_mutex -> bd_mutex : e.g. do_md_run -> revalidate_disk
|
|
|
|
* bd_mutex -> open_mutex: e.g. __blkdev_get -> md_open
|
|
|
|
*/
|
|
|
|
struct mutex open_mutex;
|
2006-03-27 13:18:20 +04:00
|
|
|
struct mutex reconfig_mutex;
|
2008-07-21 11:05:25 +04:00
|
|
|
atomic_t active; /* general refcount */
|
|
|
|
atomic_t openers; /* number of active opens */
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2011-02-24 09:26:41 +03:00
|
|
|
int changed; /* True if we might need to
|
|
|
|
* reread partition info */
|
2005-04-17 02:20:36 +04:00
|
|
|
int degraded; /* whether md should consider
|
|
|
|
* adding a spare
|
|
|
|
*/
|
2012-03-19 05:46:39 +04:00
|
|
|
int merge_check_needed; /* at least one
|
|
|
|
* member device
|
|
|
|
* has a
|
|
|
|
* merge_bvec_fn */
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
atomic_t recovery_active; /* blocks scheduled, but not written */
|
|
|
|
wait_queue_head_t recovery_wait;
|
|
|
|
sector_t recovery_cp;
|
2008-06-28 02:31:24 +04:00
|
|
|
sector_t resync_min; /* user requested sync
|
|
|
|
* starts here */
|
2008-02-06 12:39:52 +03:00
|
|
|
sector_t resync_max; /* resync should pause
|
|
|
|
* when it gets here */
|
2005-06-22 04:17:12 +04:00
|
|
|
|
2013-12-11 23:11:53 +04:00
|
|
|
struct kernfs_node *sysfs_state; /* handle for 'array_state'
|
2008-10-21 06:25:21 +04:00
|
|
|
* file in sysfs.
|
|
|
|
*/
|
2013-12-11 23:11:53 +04:00
|
|
|
struct kernfs_node *sysfs_action; /* handle for 'sync_action' */
|
2008-10-21 06:25:21 +04:00
|
|
|
|
md: make devices disappear when they are no longer needed.
Currently md devices, once created, never disappear until the module
is unloaded. This is essentially because the gendisk holds a
reference to the mddev, and the mddev holds a reference to the
gendisk, this a circular reference.
If we drop the reference from mddev to gendisk, then we need to ensure
that the mddev is destroyed when the gendisk is destroyed. However it
is not possible to hook into the gendisk destruction process to enable
this.
So we drop the reference from the gendisk to the mddev and destroy the
gendisk when the mddev gets destroyed. However this has a
complication.
Between the call
__blkdev_get->get_gendisk->kobj_lookup->md_probe
and the call
__blkdev_get->md_open
there is no obvious way to hold a reference on the mddev any more, so
unless something is done, it will disappear and gendisk will be
destroyed prematurely.
Also, once we decide to destroy the mddev, there will be an unlockable
moment before the gendisk is unlinked (blk_unregister_region) during
which a new reference to the gendisk can be created. We need to
ensure that this reference can not be used. i.e. the ->open must
fail.
So:
1/ in md_probe we set a flag in the mddev (hold_active) which
indicates that the array should be treated as active, even
though there are no references, and no appearance of activity.
This is cleared by md_release when the device is closed if it
is no longer needed.
This ensures that the gendisk will survive between md_probe and
md_open.
2/ In md_open we check if the mddev we expect to open matches
the gendisk that we did open.
If there is a mismatch we return -ERESTARTSYS and modify
__blkdev_get to retry from the top in that case.
In the -ERESTARTSYS sys case we make sure to wait until
the old gendisk (that we succeeded in opening) is really gone so
we loop at most once.
Some udev configurations will always open an md device when it first
appears. If we allow an md device that was just created by an open
to disappear on an immediate close, then this can race with such udev
configurations and result in an infinite loop the device being opened
and closed, then re-open due to the 'ADD' even from the first open,
and then close and so on.
So we make sure an md device, once created by an open, remains active
at least until some md 'ioctl' has been made on it. This means that
all normal usage of md devices will allow them to disappear promptly
when not needed, but the worst that an incorrect usage will do it
cause an inactive md device to be left in existence (it can easily be
removed).
As an array can be stopped by writing to a sysfs attribute
echo clear > /sys/block/mdXXX/md/array_state
we need to use scheduled work for deleting the gendisk and other
kobjects. This allows us to wait for any pending gendisk deletion to
complete by simply calling flush_scheduled_work().
Signed-off-by: NeilBrown <neilb@suse.de>
2009-01-09 00:31:10 +03:00
|
|
|
struct work_struct del_work; /* used for delayed sysfs removal */
|
|
|
|
|
2014-12-15 04:56:56 +03:00
|
|
|
/* "lock" protects:
|
|
|
|
* flush_bio transition from NULL to !NULL
|
|
|
|
* rdev superblocks, events
|
|
|
|
* clearing MD_CHANGE_*
|
|
|
|
* in_sync - and related safemode and MD_CHANGE changes
|
2014-12-15 04:56:58 +03:00
|
|
|
* pers (also protected by reconfig_mutex and pending IO).
|
2014-12-15 04:56:58 +03:00
|
|
|
* clearing ->bitmap
|
2014-12-15 04:57:00 +03:00
|
|
|
* clearing ->bitmap_info.file
|
2014-12-15 04:57:01 +03:00
|
|
|
* changing ->resync_{min,max}
|
|
|
|
* setting MD_RECOVERY_RUNNING (which interacts with resync_{min,max})
|
2014-12-15 04:56:56 +03:00
|
|
|
*/
|
|
|
|
spinlock_t lock;
|
2005-06-22 04:17:26 +04:00
|
|
|
wait_queue_head_t sb_wait; /* for waiting on superblock updates */
|
2005-06-22 04:17:28 +04:00
|
|
|
atomic_t pending_writes; /* number of active superblock writes */
|
2005-06-22 04:17:12 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
unsigned int safemode; /* if set, update "clean" superblock
|
|
|
|
* when no writes pending.
|
2014-09-30 08:23:59 +04:00
|
|
|
*/
|
2005-04-17 02:20:36 +04:00
|
|
|
unsigned int safemode_delay;
|
|
|
|
struct timer_list safemode_timer;
|
2014-09-30 08:23:59 +04:00
|
|
|
atomic_t writes_pending;
|
2007-07-24 11:28:11 +04:00
|
|
|
struct request_queue *queue; /* for plugging ... */
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2014-09-30 08:23:59 +04:00
|
|
|
struct bitmap *bitmap; /* the bitmap for the device */
|
2009-12-14 04:49:52 +03:00
|
|
|
struct {
|
|
|
|
struct file *file; /* the bitmap file */
|
2009-12-14 04:49:54 +03:00
|
|
|
loff_t offset; /* offset from superblock of
|
2009-12-14 04:49:52 +03:00
|
|
|
* start of bitmap. May be
|
|
|
|
* negative, but not '0'
|
2009-12-14 04:49:54 +03:00
|
|
|
* For external metadata, offset
|
2014-09-30 08:23:59 +04:00
|
|
|
* from start of device.
|
2009-12-14 04:49:52 +03:00
|
|
|
*/
|
2012-05-22 07:55:07 +04:00
|
|
|
unsigned long space; /* space available at this offset */
|
2009-12-14 04:49:54 +03:00
|
|
|
loff_t default_offset; /* this is the offset to use when
|
2009-12-14 04:49:52 +03:00
|
|
|
* hot-adding a bitmap. It should
|
|
|
|
* eventually be settable by sysfs.
|
|
|
|
*/
|
2012-05-22 07:55:07 +04:00
|
|
|
unsigned long default_space; /* space available at
|
|
|
|
* default offset */
|
2009-12-14 04:49:52 +03:00
|
|
|
struct mutex mutex;
|
2009-12-14 04:49:53 +03:00
|
|
|
unsigned long chunksize;
|
2010-06-01 13:37:31 +04:00
|
|
|
unsigned long daemon_sleep; /* how many jiffies between updates? */
|
2009-12-14 04:49:53 +03:00
|
|
|
unsigned long max_write_behind; /* write-behind mode */
|
2009-12-14 04:49:56 +03:00
|
|
|
int external;
|
2014-03-29 19:20:02 +04:00
|
|
|
int nodes; /* Maximum number of nodes in the cluster */
|
2009-12-14 04:49:52 +03:00
|
|
|
} bitmap_info;
|
2005-06-22 04:17:14 +04:00
|
|
|
|
2014-09-30 08:23:59 +04:00
|
|
|
atomic_t max_corr_read_errors; /* max read retries */
|
2005-04-17 02:20:36 +04:00
|
|
|
struct list_head all_mddevs;
|
2009-12-14 04:49:49 +03:00
|
|
|
|
2010-04-14 11:15:37 +04:00
|
|
|
struct attribute_group *to_remove;
|
2010-06-01 13:37:29 +04:00
|
|
|
|
2010-10-26 11:31:13 +04:00
|
|
|
struct bio_set *bio_set;
|
|
|
|
|
2010-09-03 13:56:18 +04:00
|
|
|
/* Generic flush handling.
|
|
|
|
* The last to finish preflush schedules a worker to submit
|
|
|
|
* the rest of the request (without the REQ_FLUSH flag).
|
2009-12-14 04:49:49 +03:00
|
|
|
*/
|
2010-09-03 13:56:18 +04:00
|
|
|
struct bio *flush_bio;
|
2009-12-14 04:49:49 +03:00
|
|
|
atomic_t flush_pending;
|
2010-09-03 13:56:18 +04:00
|
|
|
struct work_struct flush_work;
|
2010-07-26 05:49:55 +04:00
|
|
|
struct work_struct event_work; /* used by dm to report failure event */
|
2011-10-11 09:47:53 +04:00
|
|
|
void (*sync_super)(struct mddev *mddev, struct md_rdev *rdev);
|
2014-03-29 19:20:02 +04:00
|
|
|
struct md_cluster_info *cluster_info;
|
2005-04-17 02:20:36 +04:00
|
|
|
};
|
|
|
|
|
2014-12-15 04:57:01 +03:00
|
|
|
static inline int __must_check mddev_lock(struct mddev *mddev)
|
|
|
|
{
|
|
|
|
return mutex_lock_interruptible(&mddev->reconfig_mutex);
|
|
|
|
}
|
|
|
|
|
|
|
|
/* Sometimes we need to take the lock in a situation where
|
|
|
|
* failure due to interrupts is not acceptable.
|
|
|
|
*/
|
|
|
|
static inline void mddev_lock_nointr(struct mddev *mddev)
|
|
|
|
{
|
|
|
|
mutex_lock(&mddev->reconfig_mutex);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int mddev_is_locked(struct mddev *mddev)
|
|
|
|
{
|
|
|
|
return mutex_is_locked(&mddev->reconfig_mutex);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline int mddev_trylock(struct mddev *mddev)
|
|
|
|
{
|
|
|
|
return mutex_trylock(&mddev->reconfig_mutex);
|
|
|
|
}
|
|
|
|
extern void mddev_unlock(struct mddev *mddev);
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
static inline void md_sync_acct(struct block_device *bdev, unsigned long nr_sectors)
|
|
|
|
{
|
2014-09-30 08:23:59 +04:00
|
|
|
atomic_add(nr_sectors, &bdev->bd_contains->bd_disk->sync_io);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
2011-10-11 09:49:58 +04:00
|
|
|
struct md_personality
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
char *name;
|
2006-01-06 11:20:36 +03:00
|
|
|
int level;
|
|
|
|
struct list_head list;
|
2005-04-17 02:20:36 +04:00
|
|
|
struct module *owner;
|
2011-11-05 04:06:58 +04:00
|
|
|
void (*make_request)(struct mddev *mddev, struct bio *bio);
|
2011-10-11 09:47:53 +04:00
|
|
|
int (*run)(struct mddev *mddev);
|
2014-12-15 04:56:58 +03:00
|
|
|
void (*free)(struct mddev *mddev, void *priv);
|
2011-10-11 09:47:53 +04:00
|
|
|
void (*status)(struct seq_file *seq, struct mddev *mddev);
|
2005-04-17 02:20:36 +04:00
|
|
|
/* error_handler must set ->faulty and clear ->in_sync
|
2014-09-30 08:23:59 +04:00
|
|
|
* if appropriate, and should abort recovery if needed
|
2005-04-17 02:20:36 +04:00
|
|
|
*/
|
2011-10-11 09:47:53 +04:00
|
|
|
void (*error_handler)(struct mddev *mddev, struct md_rdev *rdev);
|
|
|
|
int (*hot_add_disk) (struct mddev *mddev, struct md_rdev *rdev);
|
2011-12-23 03:17:51 +04:00
|
|
|
int (*hot_remove_disk) (struct mddev *mddev, struct md_rdev *rdev);
|
2011-10-11 09:47:53 +04:00
|
|
|
int (*spare_active) (struct mddev *mddev);
|
|
|
|
sector_t (*sync_request)(struct mddev *mddev, sector_t sector_nr, int *skipped, int go_faster);
|
|
|
|
int (*resize) (struct mddev *mddev, sector_t sectors);
|
|
|
|
sector_t (*size) (struct mddev *mddev, sector_t sectors, int raid_disks);
|
|
|
|
int (*check_reshape) (struct mddev *mddev);
|
|
|
|
int (*start_reshape) (struct mddev *mddev);
|
|
|
|
void (*finish_reshape) (struct mddev *mddev);
|
2005-09-10 03:23:45 +04:00
|
|
|
/* quiesce moves between quiescence states
|
|
|
|
* 0 - fully active
|
|
|
|
* 1 - no new requests allowed
|
|
|
|
* others - reserved
|
|
|
|
*/
|
2011-10-11 09:47:53 +04:00
|
|
|
void (*quiesce) (struct mddev *mddev, int state);
|
2009-03-31 07:39:39 +04:00
|
|
|
/* takeover is used to transition an array from one
|
|
|
|
* personality to another. The new personality must be able
|
|
|
|
* to handle the data in the current layout.
|
|
|
|
* e.g. 2drive raid1 -> 2drive raid5
|
|
|
|
* ndrive raid5 -> degraded n+1drive raid6 with special layout
|
|
|
|
* If the takeover succeeds, a new 'private' structure is returned.
|
|
|
|
* This needs to be installed and then ->run used to activate the
|
|
|
|
* array.
|
|
|
|
*/
|
2011-10-11 09:47:53 +04:00
|
|
|
void *(*takeover) (struct mddev *mddev);
|
2014-12-15 04:56:56 +03:00
|
|
|
/* congested implements bdi.congested_fn().
|
|
|
|
* Will not be called while array is 'suspended' */
|
|
|
|
int (*congested)(struct mddev *mddev, int bits);
|
2014-12-15 04:56:57 +03:00
|
|
|
/* mergeable_bvec is use to implement ->merge_bvec_fn */
|
|
|
|
int (*mergeable_bvec)(struct mddev *mddev,
|
|
|
|
struct bvec_merge_data *bvm,
|
|
|
|
struct bio_vec *biovec);
|
2005-04-17 02:20:36 +04:00
|
|
|
};
|
|
|
|
|
2005-11-09 08:39:30 +03:00
|
|
|
struct md_sysfs_entry {
|
|
|
|
struct attribute attr;
|
2011-10-11 09:47:53 +04:00
|
|
|
ssize_t (*show)(struct mddev *, char *);
|
|
|
|
ssize_t (*store)(struct mddev *, const char *, size_t);
|
2005-11-09 08:39:30 +03:00
|
|
|
};
|
2009-12-14 04:49:55 +03:00
|
|
|
extern struct attribute_group md_bitmap_group;
|
2005-11-09 08:39:30 +03:00
|
|
|
|
2013-12-11 23:11:53 +04:00
|
|
|
static inline struct kernfs_node *sysfs_get_dirent_safe(struct kernfs_node *sd, char *name)
|
2010-06-01 13:37:23 +04:00
|
|
|
{
|
|
|
|
if (sd)
|
2013-09-12 07:19:13 +04:00
|
|
|
return sysfs_get_dirent(sd, name);
|
2010-06-01 13:37:23 +04:00
|
|
|
return sd;
|
|
|
|
}
|
2013-12-11 23:11:53 +04:00
|
|
|
static inline void sysfs_notify_dirent_safe(struct kernfs_node *sd)
|
2010-06-01 13:37:23 +04:00
|
|
|
{
|
|
|
|
if (sd)
|
|
|
|
sysfs_notify_dirent(sd);
|
|
|
|
}
|
|
|
|
|
2011-10-11 09:47:53 +04:00
|
|
|
static inline char * mdname (struct mddev * mddev)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
return mddev->gendisk ? mddev->gendisk->disk_name : "mdX";
|
|
|
|
}
|
|
|
|
|
2011-10-11 09:47:53 +04:00
|
|
|
static inline int sysfs_link_rdev(struct mddev *mddev, struct md_rdev *rdev)
|
2011-07-27 05:00:36 +04:00
|
|
|
{
|
|
|
|
char nm[20];
|
2013-03-08 02:24:26 +04:00
|
|
|
if (!test_bit(Replacement, &rdev->flags) && mddev->kobj.sd) {
|
2011-12-23 03:17:51 +04:00
|
|
|
sprintf(nm, "rd%d", rdev->raid_disk);
|
|
|
|
return sysfs_create_link(&mddev->kobj, &rdev->kobj, nm);
|
|
|
|
} else
|
|
|
|
return 0;
|
2011-07-27 05:00:36 +04:00
|
|
|
}
|
|
|
|
|
2011-10-11 09:47:53 +04:00
|
|
|
static inline void sysfs_unlink_rdev(struct mddev *mddev, struct md_rdev *rdev)
|
2011-07-27 05:00:36 +04:00
|
|
|
{
|
|
|
|
char nm[20];
|
2013-03-08 02:24:26 +04:00
|
|
|
if (!test_bit(Replacement, &rdev->flags) && mddev->kobj.sd) {
|
2011-12-23 03:17:51 +04:00
|
|
|
sprintf(nm, "rd%d", rdev->raid_disk);
|
|
|
|
sysfs_remove_link(&mddev->kobj, nm);
|
|
|
|
}
|
2011-07-27 05:00:36 +04:00
|
|
|
}
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/*
|
|
|
|
* iterates through some rdev ringlist. It's safe to remove the
|
|
|
|
* current 'rdev'. Dont touch 'tmp' though.
|
|
|
|
*/
|
2009-01-09 00:31:08 +03:00
|
|
|
#define rdev_for_each_list(rdev, tmp, head) \
|
|
|
|
list_for_each_entry_safe(rdev, tmp, head, same_set)
|
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
/*
|
|
|
|
* iterates through the 'same array disks' ringlist
|
|
|
|
*/
|
2012-03-19 05:46:39 +04:00
|
|
|
#define rdev_for_each(rdev, mddev) \
|
|
|
|
list_for_each_entry(rdev, &((mddev)->disks), same_set)
|
|
|
|
|
|
|
|
#define rdev_for_each_safe(rdev, tmp, mddev) \
|
2009-01-09 00:31:08 +03:00
|
|
|
list_for_each_entry_safe(rdev, tmp, &((mddev)->disks), same_set)
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2008-07-21 11:05:25 +04:00
|
|
|
#define rdev_for_each_rcu(rdev, mddev) \
|
|
|
|
list_for_each_entry_rcu(rdev, &((mddev)->disks), same_set)
|
|
|
|
|
2011-10-11 09:48:23 +04:00
|
|
|
struct md_thread {
|
2012-10-11 06:34:00 +04:00
|
|
|
void (*run) (struct md_thread *thread);
|
2011-10-11 09:47:53 +04:00
|
|
|
struct mddev *mddev;
|
2005-04-17 02:20:36 +04:00
|
|
|
wait_queue_head_t wqueue;
|
2014-09-30 08:23:59 +04:00
|
|
|
unsigned long flags;
|
2005-04-17 02:20:36 +04:00
|
|
|
struct task_struct *tsk;
|
2005-06-22 04:17:14 +04:00
|
|
|
unsigned long timeout;
|
2012-10-11 06:34:00 +04:00
|
|
|
void *private;
|
2011-10-11 09:48:23 +04:00
|
|
|
};
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
#define THREAD_WAKEUP 0
|
|
|
|
|
2006-01-06 11:20:40 +03:00
|
|
|
static inline void safe_put_page(struct page *p)
|
|
|
|
{
|
|
|
|
if (p) put_page(p);
|
|
|
|
}
|
|
|
|
|
2011-10-11 09:49:58 +04:00
|
|
|
extern int register_md_personality(struct md_personality *p);
|
|
|
|
extern int unregister_md_personality(struct md_personality *p);
|
2014-03-29 19:01:53 +04:00
|
|
|
extern int register_md_cluster_operations(struct md_cluster_operations *ops,
|
|
|
|
struct module *module);
|
|
|
|
extern int unregister_md_cluster_operations(void);
|
|
|
|
extern int md_setup_cluster(struct mddev *mddev, int nodes);
|
|
|
|
extern void md_cluster_stop(struct mddev *mddev);
|
2011-10-11 09:48:23 +04:00
|
|
|
extern struct md_thread *md_register_thread(
|
2012-10-11 06:34:00 +04:00
|
|
|
void (*run)(struct md_thread *thread),
|
2011-10-11 09:48:23 +04:00
|
|
|
struct mddev *mddev,
|
|
|
|
const char *name);
|
|
|
|
extern void md_unregister_thread(struct md_thread **threadp);
|
|
|
|
extern void md_wakeup_thread(struct md_thread *thread);
|
2011-10-11 09:47:53 +04:00
|
|
|
extern void md_check_recovery(struct mddev *mddev);
|
2013-04-24 05:42:43 +04:00
|
|
|
extern void md_reap_sync_thread(struct mddev *mddev);
|
2011-10-11 09:47:53 +04:00
|
|
|
extern void md_write_start(struct mddev *mddev, struct bio *bi);
|
|
|
|
extern void md_write_end(struct mddev *mddev);
|
|
|
|
extern void md_done_sync(struct mddev *mddev, int blocks, int ok);
|
|
|
|
extern void md_error(struct mddev *mddev, struct md_rdev *rdev);
|
2012-05-21 03:27:00 +04:00
|
|
|
extern void md_finish_reshape(struct mddev *mddev);
|
2011-10-11 09:47:53 +04:00
|
|
|
|
|
|
|
extern int mddev_congested(struct mddev *mddev, int bits);
|
|
|
|
extern void md_flush_request(struct mddev *mddev, struct bio *bio);
|
|
|
|
extern void md_super_write(struct mddev *mddev, struct md_rdev *rdev,
|
2009-03-31 07:33:13 +04:00
|
|
|
sector_t sector, int size, struct page *page);
|
2011-10-11 09:47:53 +04:00
|
|
|
extern void md_super_wait(struct mddev *mddev);
|
2014-09-30 08:23:59 +04:00
|
|
|
extern int sync_page_io(struct md_rdev *rdev, sector_t sector, int size,
|
2011-01-14 01:14:33 +03:00
|
|
|
struct page *page, int rw, bool metadata_op);
|
2012-10-11 06:34:00 +04:00
|
|
|
extern void md_do_sync(struct md_thread *thread);
|
2011-10-11 09:47:53 +04:00
|
|
|
extern void md_new_event(struct mddev *mddev);
|
|
|
|
extern int md_allow_write(struct mddev *mddev);
|
|
|
|
extern void md_wait_for_blocked_rdev(struct md_rdev *rdev, struct mddev *mddev);
|
|
|
|
extern void md_set_array_sectors(struct mddev *mddev, sector_t array_sectors);
|
|
|
|
extern int md_check_no_bitmap(struct mddev *mddev);
|
|
|
|
extern int md_integrity_register(struct mddev *mddev);
|
|
|
|
extern void md_integrity_add_rdev(struct md_rdev *rdev, struct mddev *mddev);
|
2009-12-14 04:49:55 +03:00
|
|
|
extern int strict_strtoul_scaled(const char *cp, unsigned long *res, int scale);
|
2009-04-14 06:01:53 +04:00
|
|
|
|
2011-10-11 09:47:53 +04:00
|
|
|
extern void mddev_init(struct mddev *mddev);
|
|
|
|
extern int md_run(struct mddev *mddev);
|
|
|
|
extern void md_stop(struct mddev *mddev);
|
|
|
|
extern void md_stop_writes(struct mddev *mddev);
|
2011-10-11 09:45:26 +04:00
|
|
|
extern int md_rdev_init(struct md_rdev *rdev);
|
2012-05-22 07:54:30 +04:00
|
|
|
extern void md_rdev_clear(struct md_rdev *rdev);
|
2009-04-14 06:01:53 +04:00
|
|
|
|
2011-10-11 09:47:53 +04:00
|
|
|
extern void mddev_suspend(struct mddev *mddev);
|
|
|
|
extern void mddev_resume(struct mddev *mddev);
|
2010-10-26 11:31:13 +04:00
|
|
|
extern struct bio *bio_clone_mddev(struct bio *bio, gfp_t gfp_mask,
|
2011-10-11 09:47:53 +04:00
|
|
|
struct mddev *mddev);
|
2010-10-26 11:31:13 +04:00
|
|
|
extern struct bio *bio_alloc_mddev(gfp_t gfp_mask, int nr_iovecs,
|
2011-10-11 09:47:53 +04:00
|
|
|
struct mddev *mddev);
|
2012-07-31 11:08:14 +04:00
|
|
|
|
2012-07-31 11:08:15 +04:00
|
|
|
extern void md_unplug(struct blk_plug_cb *cb, bool from_schedule);
|
2012-07-31 11:08:14 +04:00
|
|
|
static inline int mddev_check_plugged(struct mddev *mddev)
|
|
|
|
{
|
|
|
|
return !!blk_check_plugged(md_unplug, mddev,
|
|
|
|
sizeof(struct blk_plug_cb));
|
|
|
|
}
|
2012-07-26 13:12:18 +04:00
|
|
|
|
|
|
|
static inline void rdev_dec_pending(struct md_rdev *rdev, struct mddev *mddev)
|
|
|
|
{
|
|
|
|
int faulty = test_bit(Faulty, &rdev->flags);
|
|
|
|
if (atomic_dec_and_test(&rdev->nr_pending) && faulty) {
|
|
|
|
set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
|
|
|
|
md_wakeup_thread(mddev->thread);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2014-03-29 19:01:53 +04:00
|
|
|
extern struct md_cluster_operations *md_cluster_ops;
|
2014-03-29 19:20:02 +04:00
|
|
|
static inline int mddev_is_clustered(struct mddev *mddev)
|
|
|
|
{
|
|
|
|
return mddev->cluster_info && mddev->bitmap_info.nodes > 1;
|
|
|
|
}
|
2009-04-14 06:01:53 +04:00
|
|
|
#endif /* _MD_MD_H */
|