2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Copyright (C) 2001 Sistina Software (UK) Limited.
|
2008-04-24 14:43:52 -06:00
|
|
|
* Copyright (C) 2004-2008 Red Hat, Inc. All rights reserved.
|
2005-04-16 16:20:36 -06:00
|
|
|
*
|
|
|
|
* This file is released under the LGPL.
|
|
|
|
*/
|
|
|
|
|
|
|
|
#ifndef _LINUX_DEVICE_MAPPER_H
|
|
|
|
#define _LINUX_DEVICE_MAPPER_H
|
|
|
|
|
2008-04-24 14:43:35 -06:00
|
|
|
#include <linux/bio.h>
|
2008-07-21 05:00:37 -06:00
|
|
|
#include <linux/blkdev.h>
|
2008-04-24 14:43:35 -06:00
|
|
|
|
2009-06-22 03:12:33 -06:00
|
|
|
struct dm_dev;
|
2005-04-16 16:20:36 -06:00
|
|
|
struct dm_target;
|
|
|
|
struct dm_table;
|
2006-06-26 01:27:33 -06:00
|
|
|
struct mapped_device;
|
2008-07-21 05:00:37 -06:00
|
|
|
struct bio_vec;
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
typedef enum { STATUSTYPE_INFO, STATUSTYPE_TABLE } status_type_t;
|
|
|
|
|
|
|
|
union map_info {
|
|
|
|
void *ptr;
|
|
|
|
unsigned long long ll;
|
2009-06-22 03:12:20 -06:00
|
|
|
unsigned flush_request;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
/*
|
|
|
|
* In the constructor the target parameter will already have the
|
|
|
|
* table, type, begin and len fields filled in.
|
|
|
|
*/
|
|
|
|
typedef int (*dm_ctr_fn) (struct dm_target *target,
|
|
|
|
unsigned int argc, char **argv);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The destructor doesn't need to free the dm_target, just
|
|
|
|
* anything hidden ti->private.
|
|
|
|
*/
|
|
|
|
typedef void (*dm_dtr_fn) (struct dm_target *ti);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The map function must return:
|
|
|
|
* < 0: error
|
|
|
|
* = 0: The target will handle the io by resubmitting it later
|
2006-12-08 03:41:05 -07:00
|
|
|
* = 1: simple remap complete
|
[PATCH] dm: suspend: add noflush pushback
In device-mapper I/O is sometimes queued within targets for later processing.
For example the multipath target can be configured to store I/O when no paths
are available instead of returning it -EIO.
This patch allows the device-mapper core to instruct a target to transfer the
contents of any such in-target queue back into the core. This frees up the
resources used by the target so the core can replace that target with an
alternative one and then resend the I/O to it. Without this patch the only
way to change the target in such circumstances involves returning the I/O with
an error back to the filesystem/application. In the multipath case, this
patch will let us add new paths for existing I/O to try after all the existing
paths have failed.
DMF_NOFLUSH_SUSPENDING
----------------------
If the DM_NOFLUSH_FLAG ioctl option is specified at suspend time, the
DMF_NOFLUSH_SUSPENDING flag is set in md->flags during dm_suspend(). It
is always cleared before dm_suspend() returns.
The flag must be visible while the target is flushing pending I/Os so it
is set before presuspend where the flush starts and unset after the wait
for md->pending where the flush ends.
Target drivers can check this flag by calling dm_noflush_suspending().
DM_MAPIO_REQUEUE / DM_ENDIO_REQUEUE
-----------------------------------
A target's map() function can now return DM_MAPIO_REQUEUE to request the
device mapper core queue the bio.
Similarly, a target's end_io() function can return DM_ENDIO_REQUEUE to request
the same. This has been labelled 'pushback'.
The __map_bio() and clone_endio() functions in the core treat these return
values as errors and call dec_pending() to end the I/O.
dec_pending
-----------
dec_pending() saves the pushback request in struct dm_io->error. Once all
the split clones have ended, dec_pending() will put the original bio on
the md->pushback list. Note that this supercedes any I/O errors.
It is possible for the suspend with DM_NOFLUSH_FLAG to be aborted while
in progress (e.g. by user interrupt). dec_pending() checks for this and
returns -EIO if it happened.
pushdback list and pushback_lock
--------------------------------
The bio is queued on md->pushback temporarily in dec_pending(), and after
all pending I/Os return, md->pushback is merged into md->deferred in
dm_suspend() for re-issuing at resume time.
md->pushback_lock protects md->pushback.
The lock should be held with irq disabled because dec_pending() can be
called from interrupt context.
Queueing bios to md->pushback in dec_pending() must be done atomically
with the check for DMF_NOFLUSH_SUSPENDING flag. So md->pushback_lock is
held when checking the flag. Otherwise dec_pending() may queue a bio to
md->pushback after the interrupted dm_suspend() flushes md->pushback.
Then the bio would be left in md->pushback.
Flag setting in dm_suspend() can be done without md->pushback_lock because
the flag is checked only after presuspend and the set value is already
made visible via the target's presuspend function.
The flag can be checked without md->pushback_lock (e.g. the first part of
the dec_pending() or target drivers), because the flag is checked again
with md->pushback_lock held when the bio is really queued to md->pushback
as described above. So even if the flag is cleared after the lockless
checkings, the bio isn't left in md->pushback but returned to applications
with -EIO.
Other notes on the current patch
--------------------------------
- md->pushback is added to the struct mapped_device instead of using
md->deferred directly because md->io_lock which protects md->deferred is
rw_semaphore and can't be used in interrupt context like dec_pending(),
and md->io_lock protects the DMF_BLOCK_IO flag of md->flags too.
- Don't issue lock_fs() in dm_suspend() if the DM_NOFLUSH_FLAG
ioctl option is specified, because I/Os generated by lock_fs() would be
pushed back and never return if there were no valid devices.
- If an error occurs in dm_suspend() after the DMF_NOFLUSH_SUSPENDING
flag is set, md->pushback must be flushed because I/Os may be queued to
the list already. (flush_and_out label in dm_suspend())
Test results
------------
I have tested using multipath target with the next patch.
The following tests are for regression/compatibility:
- I/Os succeed when valid paths exist;
- I/Os fail when there are no valid paths and queue_if_no_path is not
set;
- I/Os are queued in the multipath target when there are no valid paths and
queue_if_no_path is set;
- The queued I/Os above fail when suspend is issued without the
DM_NOFLUSH_FLAG ioctl option. I/Os spanning 2 multipath targets also
fail.
The following tests are for the normal code path of new pushback feature:
- Queued I/Os in the multipath target are flushed from the target
but don't return when suspend is issued with the DM_NOFLUSH_FLAG
ioctl option;
- The I/Os above are queued in the multipath target again when
resume is issued without path recovery;
- The I/Os above succeed when resume is issued after path recovery
or table load;
- Queued I/Os in the multipath target succeed when resume is issued
with the DM_NOFLUSH_FLAG ioctl option after table load. I/Os
spanning 2 multipath targets also succeed.
The following tests are for the error paths of the new pushback feature:
- When the bdget_disk() fails in dm_suspend(), the
DMF_NOFLUSH_SUSPENDING flag is cleared and I/Os already queued to the
pushback list are flushed properly.
- When suspend with the DM_NOFLUSH_FLAG ioctl option is interrupted,
o I/Os which had already been queued to the pushback list
at the time don't return, and are re-issued at resume time;
o I/Os which hadn't been returned at the time return with EIO.
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: dm-devel@redhat.com
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 03:41:09 -07:00
|
|
|
* = 2: The target wants to push back the io
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
typedef int (*dm_map_fn) (struct dm_target *ti, struct bio *bio,
|
|
|
|
union map_info *map_context);
|
2009-01-05 20:05:07 -07:00
|
|
|
typedef int (*dm_map_request_fn) (struct dm_target *ti, struct request *clone,
|
|
|
|
union map_info *map_context);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Returns:
|
|
|
|
* < 0 : error (currently ignored)
|
|
|
|
* 0 : ended successfully
|
|
|
|
* 1 : for some reason the io has still not completed (eg,
|
|
|
|
* multipath target might want to requeue a failed io).
|
[PATCH] dm: suspend: add noflush pushback
In device-mapper I/O is sometimes queued within targets for later processing.
For example the multipath target can be configured to store I/O when no paths
are available instead of returning it -EIO.
This patch allows the device-mapper core to instruct a target to transfer the
contents of any such in-target queue back into the core. This frees up the
resources used by the target so the core can replace that target with an
alternative one and then resend the I/O to it. Without this patch the only
way to change the target in such circumstances involves returning the I/O with
an error back to the filesystem/application. In the multipath case, this
patch will let us add new paths for existing I/O to try after all the existing
paths have failed.
DMF_NOFLUSH_SUSPENDING
----------------------
If the DM_NOFLUSH_FLAG ioctl option is specified at suspend time, the
DMF_NOFLUSH_SUSPENDING flag is set in md->flags during dm_suspend(). It
is always cleared before dm_suspend() returns.
The flag must be visible while the target is flushing pending I/Os so it
is set before presuspend where the flush starts and unset after the wait
for md->pending where the flush ends.
Target drivers can check this flag by calling dm_noflush_suspending().
DM_MAPIO_REQUEUE / DM_ENDIO_REQUEUE
-----------------------------------
A target's map() function can now return DM_MAPIO_REQUEUE to request the
device mapper core queue the bio.
Similarly, a target's end_io() function can return DM_ENDIO_REQUEUE to request
the same. This has been labelled 'pushback'.
The __map_bio() and clone_endio() functions in the core treat these return
values as errors and call dec_pending() to end the I/O.
dec_pending
-----------
dec_pending() saves the pushback request in struct dm_io->error. Once all
the split clones have ended, dec_pending() will put the original bio on
the md->pushback list. Note that this supercedes any I/O errors.
It is possible for the suspend with DM_NOFLUSH_FLAG to be aborted while
in progress (e.g. by user interrupt). dec_pending() checks for this and
returns -EIO if it happened.
pushdback list and pushback_lock
--------------------------------
The bio is queued on md->pushback temporarily in dec_pending(), and after
all pending I/Os return, md->pushback is merged into md->deferred in
dm_suspend() for re-issuing at resume time.
md->pushback_lock protects md->pushback.
The lock should be held with irq disabled because dec_pending() can be
called from interrupt context.
Queueing bios to md->pushback in dec_pending() must be done atomically
with the check for DMF_NOFLUSH_SUSPENDING flag. So md->pushback_lock is
held when checking the flag. Otherwise dec_pending() may queue a bio to
md->pushback after the interrupted dm_suspend() flushes md->pushback.
Then the bio would be left in md->pushback.
Flag setting in dm_suspend() can be done without md->pushback_lock because
the flag is checked only after presuspend and the set value is already
made visible via the target's presuspend function.
The flag can be checked without md->pushback_lock (e.g. the first part of
the dec_pending() or target drivers), because the flag is checked again
with md->pushback_lock held when the bio is really queued to md->pushback
as described above. So even if the flag is cleared after the lockless
checkings, the bio isn't left in md->pushback but returned to applications
with -EIO.
Other notes on the current patch
--------------------------------
- md->pushback is added to the struct mapped_device instead of using
md->deferred directly because md->io_lock which protects md->deferred is
rw_semaphore and can't be used in interrupt context like dec_pending(),
and md->io_lock protects the DMF_BLOCK_IO flag of md->flags too.
- Don't issue lock_fs() in dm_suspend() if the DM_NOFLUSH_FLAG
ioctl option is specified, because I/Os generated by lock_fs() would be
pushed back and never return if there were no valid devices.
- If an error occurs in dm_suspend() after the DMF_NOFLUSH_SUSPENDING
flag is set, md->pushback must be flushed because I/Os may be queued to
the list already. (flush_and_out label in dm_suspend())
Test results
------------
I have tested using multipath target with the next patch.
The following tests are for regression/compatibility:
- I/Os succeed when valid paths exist;
- I/Os fail when there are no valid paths and queue_if_no_path is not
set;
- I/Os are queued in the multipath target when there are no valid paths and
queue_if_no_path is set;
- The queued I/Os above fail when suspend is issued without the
DM_NOFLUSH_FLAG ioctl option. I/Os spanning 2 multipath targets also
fail.
The following tests are for the normal code path of new pushback feature:
- Queued I/Os in the multipath target are flushed from the target
but don't return when suspend is issued with the DM_NOFLUSH_FLAG
ioctl option;
- The I/Os above are queued in the multipath target again when
resume is issued without path recovery;
- The I/Os above succeed when resume is issued after path recovery
or table load;
- Queued I/Os in the multipath target succeed when resume is issued
with the DM_NOFLUSH_FLAG ioctl option after table load. I/Os
spanning 2 multipath targets also succeed.
The following tests are for the error paths of the new pushback feature:
- When the bdget_disk() fails in dm_suspend(), the
DMF_NOFLUSH_SUSPENDING flag is cleared and I/Os already queued to the
pushback list are flushed properly.
- When suspend with the DM_NOFLUSH_FLAG ioctl option is interrupted,
o I/Os which had already been queued to the pushback list
at the time don't return, and are re-issued at resume time;
o I/Os which hadn't been returned at the time return with EIO.
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: dm-devel@redhat.com
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 03:41:09 -07:00
|
|
|
* 2 : The target wants to push back the io
|
2005-04-16 16:20:36 -06:00
|
|
|
*/
|
|
|
|
typedef int (*dm_endio_fn) (struct dm_target *ti,
|
|
|
|
struct bio *bio, int error,
|
|
|
|
union map_info *map_context);
|
2009-01-05 20:05:07 -07:00
|
|
|
typedef int (*dm_request_endio_fn) (struct dm_target *ti,
|
|
|
|
struct request *clone, int error,
|
|
|
|
union map_info *map_context);
|
2005-04-16 16:20:36 -06:00
|
|
|
|
2006-10-03 02:15:43 -06:00
|
|
|
typedef void (*dm_flush_fn) (struct dm_target *ti);
|
2005-04-16 16:20:36 -06:00
|
|
|
typedef void (*dm_presuspend_fn) (struct dm_target *ti);
|
|
|
|
typedef void (*dm_postsuspend_fn) (struct dm_target *ti);
|
2006-10-03 02:15:36 -06:00
|
|
|
typedef int (*dm_preresume_fn) (struct dm_target *ti);
|
2005-04-16 16:20:36 -06:00
|
|
|
typedef void (*dm_resume_fn) (struct dm_target *ti);
|
|
|
|
|
|
|
|
typedef int (*dm_status_fn) (struct dm_target *ti, status_type_t status_type,
|
|
|
|
char *result, unsigned int maxlen);
|
|
|
|
|
|
|
|
typedef int (*dm_message_fn) (struct dm_target *ti, unsigned argc, char **argv);
|
|
|
|
|
2007-08-28 20:15:59 -06:00
|
|
|
typedef int (*dm_ioctl_fn) (struct dm_target *ti, unsigned int cmd,
|
2006-10-03 02:15:15 -06:00
|
|
|
unsigned long arg);
|
|
|
|
|
2008-07-21 05:00:37 -06:00
|
|
|
typedef int (*dm_merge_fn) (struct dm_target *ti, struct bvec_merge_data *bvm,
|
|
|
|
struct bio_vec *biovec, int max_size);
|
|
|
|
|
2009-06-22 03:12:33 -06:00
|
|
|
typedef int (*iterate_devices_callout_fn) (struct dm_target *ti,
|
|
|
|
struct dm_dev *dev,
|
2009-07-23 13:30:42 -06:00
|
|
|
sector_t start, sector_t len,
|
2009-06-22 03:12:33 -06:00
|
|
|
void *data);
|
|
|
|
|
|
|
|
typedef int (*dm_iterate_devices_fn) (struct dm_target *ti,
|
|
|
|
iterate_devices_callout_fn fn,
|
|
|
|
void *data);
|
|
|
|
|
2009-09-04 13:40:25 -06:00
|
|
|
typedef void (*dm_io_hints_fn) (struct dm_target *ti,
|
|
|
|
struct queue_limits *limits);
|
|
|
|
|
2009-01-05 20:05:07 -07:00
|
|
|
/*
|
|
|
|
* Returns:
|
|
|
|
* 0: The target can handle the next I/O immediately.
|
|
|
|
* 1: The target can't handle the next I/O immediately.
|
|
|
|
*/
|
|
|
|
typedef int (*dm_busy_fn) (struct dm_target *ti);
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
void dm_error(const char *message);
|
|
|
|
|
2006-10-03 02:15:42 -06:00
|
|
|
/*
|
|
|
|
* Combine device limits.
|
|
|
|
*/
|
2009-06-22 03:12:34 -06:00
|
|
|
int dm_set_device_limits(struct dm_target *ti, struct dm_dev *dev,
|
2009-07-23 13:30:42 -06:00
|
|
|
sector_t start, sector_t len, void *data);
|
2006-10-03 02:15:42 -06:00
|
|
|
|
2008-10-10 06:37:09 -06:00
|
|
|
struct dm_dev {
|
|
|
|
struct block_device *bdev;
|
2008-09-02 13:28:45 -06:00
|
|
|
fmode_t mode;
|
2008-10-10 06:37:09 -06:00
|
|
|
char name[16];
|
|
|
|
};
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/*
|
|
|
|
* Constructors should call these functions to ensure destination devices
|
|
|
|
* are opened/closed correctly.
|
|
|
|
* FIXME: too many arguments.
|
|
|
|
*/
|
|
|
|
int dm_get_device(struct dm_target *ti, const char *path, sector_t start,
|
2008-09-02 13:28:45 -06:00
|
|
|
sector_t len, fmode_t mode, struct dm_dev **result);
|
2005-04-16 16:20:36 -06:00
|
|
|
void dm_put_device(struct dm_target *ti, struct dm_dev *d);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Information about a target type
|
|
|
|
*/
|
2009-01-05 20:05:09 -07:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Target features
|
|
|
|
*/
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
struct target_type {
|
2009-01-05 20:05:09 -07:00
|
|
|
uint64_t features;
|
2005-04-16 16:20:36 -06:00
|
|
|
const char *name;
|
|
|
|
struct module *module;
|
2006-06-26 01:27:33 -06:00
|
|
|
unsigned version[3];
|
2005-04-16 16:20:36 -06:00
|
|
|
dm_ctr_fn ctr;
|
|
|
|
dm_dtr_fn dtr;
|
|
|
|
dm_map_fn map;
|
2009-01-05 20:05:07 -07:00
|
|
|
dm_map_request_fn map_rq;
|
2005-04-16 16:20:36 -06:00
|
|
|
dm_endio_fn end_io;
|
2009-01-05 20:05:07 -07:00
|
|
|
dm_request_endio_fn rq_end_io;
|
2006-10-03 02:15:43 -06:00
|
|
|
dm_flush_fn flush;
|
2005-04-16 16:20:36 -06:00
|
|
|
dm_presuspend_fn presuspend;
|
|
|
|
dm_postsuspend_fn postsuspend;
|
2006-10-03 02:15:36 -06:00
|
|
|
dm_preresume_fn preresume;
|
2005-04-16 16:20:36 -06:00
|
|
|
dm_resume_fn resume;
|
|
|
|
dm_status_fn status;
|
|
|
|
dm_message_fn message;
|
2006-10-03 02:15:15 -06:00
|
|
|
dm_ioctl_fn ioctl;
|
2008-07-21 05:00:37 -06:00
|
|
|
dm_merge_fn merge;
|
2009-01-05 20:05:07 -07:00
|
|
|
dm_busy_fn busy;
|
2009-06-22 03:12:33 -06:00
|
|
|
dm_iterate_devices_fn iterate_devices;
|
2009-09-04 13:40:25 -06:00
|
|
|
dm_io_hints_fn io_hints;
|
2009-04-02 12:55:28 -06:00
|
|
|
|
|
|
|
/* For internal device-mapper use. */
|
|
|
|
struct list_head list;
|
2005-04-16 16:20:36 -06:00
|
|
|
};
|
|
|
|
|
|
|
|
struct dm_target {
|
|
|
|
struct dm_table *table;
|
|
|
|
struct target_type *type;
|
|
|
|
|
|
|
|
/* target limits */
|
|
|
|
sector_t begin;
|
|
|
|
sector_t len;
|
|
|
|
|
|
|
|
/* Always a power of 2 */
|
|
|
|
sector_t split_io;
|
|
|
|
|
2009-06-22 03:12:20 -06:00
|
|
|
/*
|
|
|
|
* A number of zero-length barrier requests that will be submitted
|
|
|
|
* to the target for the purpose of flushing cache.
|
|
|
|
*
|
|
|
|
* The request number will be placed in union map_info->flush_request.
|
|
|
|
* It is a responsibility of the target driver to remap these requests
|
|
|
|
* to the real underlying devices.
|
|
|
|
*/
|
|
|
|
unsigned num_flush_requests;
|
|
|
|
|
2005-04-16 16:20:36 -06:00
|
|
|
/* target specific data */
|
|
|
|
void *private;
|
|
|
|
|
|
|
|
/* Used to provide an error string from the ctr */
|
|
|
|
char *error;
|
|
|
|
};
|
|
|
|
|
|
|
|
int dm_register_target(struct target_type *t);
|
2009-01-05 20:04:58 -07:00
|
|
|
void dm_unregister_target(struct target_type *t);
|
2006-06-26 01:27:33 -06:00
|
|
|
|
|
|
|
/*-----------------------------------------------------------------
|
|
|
|
* Functions for creating and manipulating mapped devices.
|
|
|
|
* Drop the reference with dm_put when you finish with the object.
|
|
|
|
*---------------------------------------------------------------*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* DM_ANY_MINOR chooses the next available minor number.
|
|
|
|
*/
|
|
|
|
#define DM_ANY_MINOR (-1)
|
|
|
|
int dm_create(int minor, struct mapped_device **md);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Reference counting for md.
|
|
|
|
*/
|
|
|
|
struct mapped_device *dm_get_md(dev_t dev);
|
|
|
|
void dm_get(struct mapped_device *md);
|
|
|
|
void dm_put(struct mapped_device *md);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* An arbitrary pointer may be stored alongside a mapped device.
|
|
|
|
*/
|
|
|
|
void dm_set_mdptr(struct mapped_device *md, void *ptr);
|
|
|
|
void *dm_get_mdptr(struct mapped_device *md);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* A device can still be used while suspended, but I/O is deferred.
|
|
|
|
*/
|
2006-12-08 03:41:04 -07:00
|
|
|
int dm_suspend(struct mapped_device *md, unsigned suspend_flags);
|
2006-06-26 01:27:33 -06:00
|
|
|
int dm_resume(struct mapped_device *md);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Event functions.
|
|
|
|
*/
|
|
|
|
uint32_t dm_get_event_nr(struct mapped_device *md);
|
|
|
|
int dm_wait_event(struct mapped_device *md, int event_nr);
|
2007-10-19 15:48:01 -06:00
|
|
|
uint32_t dm_next_uevent_seq(struct mapped_device *md);
|
|
|
|
void dm_uevent_add(struct mapped_device *md, struct list_head *elist);
|
2006-06-26 01:27:33 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Info functions.
|
|
|
|
*/
|
2006-06-26 01:27:35 -06:00
|
|
|
const char *dm_device_name(struct mapped_device *md);
|
2007-10-19 15:47:59 -06:00
|
|
|
int dm_copy_name_and_uuid(struct mapped_device *md, char *name, char *uuid);
|
2006-06-26 01:27:33 -06:00
|
|
|
struct gendisk *dm_disk(struct mapped_device *md);
|
2009-12-10 16:52:27 -07:00
|
|
|
int dm_suspended(struct dm_target *ti);
|
[PATCH] dm: suspend: add noflush pushback
In device-mapper I/O is sometimes queued within targets for later processing.
For example the multipath target can be configured to store I/O when no paths
are available instead of returning it -EIO.
This patch allows the device-mapper core to instruct a target to transfer the
contents of any such in-target queue back into the core. This frees up the
resources used by the target so the core can replace that target with an
alternative one and then resend the I/O to it. Without this patch the only
way to change the target in such circumstances involves returning the I/O with
an error back to the filesystem/application. In the multipath case, this
patch will let us add new paths for existing I/O to try after all the existing
paths have failed.
DMF_NOFLUSH_SUSPENDING
----------------------
If the DM_NOFLUSH_FLAG ioctl option is specified at suspend time, the
DMF_NOFLUSH_SUSPENDING flag is set in md->flags during dm_suspend(). It
is always cleared before dm_suspend() returns.
The flag must be visible while the target is flushing pending I/Os so it
is set before presuspend where the flush starts and unset after the wait
for md->pending where the flush ends.
Target drivers can check this flag by calling dm_noflush_suspending().
DM_MAPIO_REQUEUE / DM_ENDIO_REQUEUE
-----------------------------------
A target's map() function can now return DM_MAPIO_REQUEUE to request the
device mapper core queue the bio.
Similarly, a target's end_io() function can return DM_ENDIO_REQUEUE to request
the same. This has been labelled 'pushback'.
The __map_bio() and clone_endio() functions in the core treat these return
values as errors and call dec_pending() to end the I/O.
dec_pending
-----------
dec_pending() saves the pushback request in struct dm_io->error. Once all
the split clones have ended, dec_pending() will put the original bio on
the md->pushback list. Note that this supercedes any I/O errors.
It is possible for the suspend with DM_NOFLUSH_FLAG to be aborted while
in progress (e.g. by user interrupt). dec_pending() checks for this and
returns -EIO if it happened.
pushdback list and pushback_lock
--------------------------------
The bio is queued on md->pushback temporarily in dec_pending(), and after
all pending I/Os return, md->pushback is merged into md->deferred in
dm_suspend() for re-issuing at resume time.
md->pushback_lock protects md->pushback.
The lock should be held with irq disabled because dec_pending() can be
called from interrupt context.
Queueing bios to md->pushback in dec_pending() must be done atomically
with the check for DMF_NOFLUSH_SUSPENDING flag. So md->pushback_lock is
held when checking the flag. Otherwise dec_pending() may queue a bio to
md->pushback after the interrupted dm_suspend() flushes md->pushback.
Then the bio would be left in md->pushback.
Flag setting in dm_suspend() can be done without md->pushback_lock because
the flag is checked only after presuspend and the set value is already
made visible via the target's presuspend function.
The flag can be checked without md->pushback_lock (e.g. the first part of
the dec_pending() or target drivers), because the flag is checked again
with md->pushback_lock held when the bio is really queued to md->pushback
as described above. So even if the flag is cleared after the lockless
checkings, the bio isn't left in md->pushback but returned to applications
with -EIO.
Other notes on the current patch
--------------------------------
- md->pushback is added to the struct mapped_device instead of using
md->deferred directly because md->io_lock which protects md->deferred is
rw_semaphore and can't be used in interrupt context like dec_pending(),
and md->io_lock protects the DMF_BLOCK_IO flag of md->flags too.
- Don't issue lock_fs() in dm_suspend() if the DM_NOFLUSH_FLAG
ioctl option is specified, because I/Os generated by lock_fs() would be
pushed back and never return if there were no valid devices.
- If an error occurs in dm_suspend() after the DMF_NOFLUSH_SUSPENDING
flag is set, md->pushback must be flushed because I/Os may be queued to
the list already. (flush_and_out label in dm_suspend())
Test results
------------
I have tested using multipath target with the next patch.
The following tests are for regression/compatibility:
- I/Os succeed when valid paths exist;
- I/Os fail when there are no valid paths and queue_if_no_path is not
set;
- I/Os are queued in the multipath target when there are no valid paths and
queue_if_no_path is set;
- The queued I/Os above fail when suspend is issued without the
DM_NOFLUSH_FLAG ioctl option. I/Os spanning 2 multipath targets also
fail.
The following tests are for the normal code path of new pushback feature:
- Queued I/Os in the multipath target are flushed from the target
but don't return when suspend is issued with the DM_NOFLUSH_FLAG
ioctl option;
- The I/Os above are queued in the multipath target again when
resume is issued without path recovery;
- The I/Os above succeed when resume is issued after path recovery
or table load;
- Queued I/Os in the multipath target succeed when resume is issued
with the DM_NOFLUSH_FLAG ioctl option after table load. I/Os
spanning 2 multipath targets also succeed.
The following tests are for the error paths of the new pushback feature:
- When the bdget_disk() fails in dm_suspend(), the
DMF_NOFLUSH_SUSPENDING flag is cleared and I/Os already queued to the
pushback list are flushed properly.
- When suspend with the DM_NOFLUSH_FLAG ioctl option is interrupted,
o I/Os which had already been queued to the pushback list
at the time don't return, and are re-issued at resume time;
o I/Os which hadn't been returned at the time return with EIO.
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Cc: dm-devel@redhat.com
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-12-08 03:41:09 -07:00
|
|
|
int dm_noflush_suspending(struct dm_target *ti);
|
2008-10-10 06:37:10 -06:00
|
|
|
union map_info *dm_get_mapinfo(struct bio *bio);
|
dm: prepare for request based option
This patch adds core functions for request-based dm.
When struct mapped device (md) is initialized, md->queue has
an I/O scheduler and the following functions are used for
request-based dm as the queue functions:
make_request_fn: dm_make_request()
pref_fn: dm_prep_fn()
request_fn: dm_request_fn()
softirq_done_fn: dm_softirq_done()
lld_busy_fn: dm_lld_busy()
Actual initializations are done in another patch (PATCH 2).
Below is a brief summary of how request-based dm behaves, including:
- making request from bio
- cloning, mapping and dispatching request
- completing request and bio
- suspending md
- resuming md
bio to request
==============
md->queue->make_request_fn() (dm_make_request()) calls __make_request()
for a bio submitted to the md.
Then, the bio is kept in the queue as a new request or merged into
another request in the queue if possible.
Cloning and Mapping
===================
Cloning and mapping are done in md->queue->request_fn() (dm_request_fn()),
when requests are dispatched after they are sorted by the I/O scheduler.
dm_request_fn() checks busy state of underlying devices using
target's busy() function and stops dispatching requests to keep them
on the dm device's queue if busy.
It helps better I/O merging, since no merge is done for a request
once it is dispatched to underlying devices.
Actual cloning and mapping are done in dm_prep_fn() and map_request()
called from dm_request_fn().
dm_prep_fn() clones not only request but also bios of the request
so that dm can hold bio completion in error cases and prevent
the bio submitter from noticing the error.
(See the "Completion" section below for details.)
After the cloning, the clone is mapped by target's map_rq() function
and inserted to underlying device's queue using
blk_insert_cloned_request().
Completion
==========
Request completion can be hooked by rq->end_io(), but then, all bios
in the request will have been completed even error cases, and the bio
submitter will have noticed the error.
To prevent the bio completion in error cases, request-based dm clones
both bio and request and hooks both bio->bi_end_io() and rq->end_io():
bio->bi_end_io(): end_clone_bio()
rq->end_io(): end_clone_request()
Summary of the request completion flow is below:
blk_end_request() for a clone request
=> blk_update_request()
=> bio->bi_end_io() == end_clone_bio() for each clone bio
=> Free the clone bio
=> Success: Complete the original bio (blk_update_request())
Error: Don't complete the original bio
=> blk_finish_request()
=> rq->end_io() == end_clone_request()
=> blk_complete_request()
=> dm_softirq_done()
=> Free the clone request
=> Success: Complete the original request (blk_end_request())
Error: Requeue the original request
end_clone_bio() completes the original request on the size of
the original bio in successful cases.
Even if all bios in the original request are completed by that
completion, the original request must not be completed yet to keep
the ordering of request completion for the stacking.
So end_clone_bio() uses blk_update_request() instead of
blk_end_request().
In error cases, end_clone_bio() doesn't complete the original bio.
It just frees the cloned bio and gives over the error handling to
end_clone_request().
end_clone_request(), which is called with queue lock held, completes
the clone request and the original request in a softirq context
(dm_softirq_done()), which has no queue lock, to avoid a deadlock
issue on submission of another request during the completion:
- The submitted request may be mapped to the same device
- Request submission requires queue lock, but the queue lock
has been held by itself and it doesn't know that
The clone request has no clone bio when dm_softirq_done() is called.
So target drivers can't resubmit it again even error cases.
Instead, they can ask dm core for requeueing and remapping
the original request in that cases.
suspend
=======
Request-based dm uses stopping md->queue as suspend of the md.
For noflush suspend, just stops md->queue.
For flush suspend, inserts a marker request to the tail of md->queue.
And dispatches all requests in md->queue until the marker comes to
the front of md->queue. Then, stops dispatching request and waits
for the all dispatched requests to complete.
After that, completes the marker request, stops md->queue and
wake up the waiter on the suspend queue, md->wait.
resume
======
Starts md->queue.
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
2009-06-22 03:12:35 -06:00
|
|
|
union map_info *dm_get_rq_mapinfo(struct request *rq);
|
2006-06-26 01:27:33 -06:00
|
|
|
|
|
|
|
/*
|
|
|
|
* Geometry functions.
|
|
|
|
*/
|
|
|
|
int dm_get_geometry(struct mapped_device *md, struct hd_geometry *geo);
|
|
|
|
int dm_set_geometry(struct mapped_device *md, struct hd_geometry *geo);
|
|
|
|
|
|
|
|
|
|
|
|
/*-----------------------------------------------------------------
|
|
|
|
* Functions for manipulating device-mapper tables.
|
|
|
|
*---------------------------------------------------------------*/
|
|
|
|
|
|
|
|
/*
|
|
|
|
* First create an empty table.
|
|
|
|
*/
|
2008-09-02 13:28:45 -06:00
|
|
|
int dm_table_create(struct dm_table **result, fmode_t mode,
|
2006-06-26 01:27:33 -06:00
|
|
|
unsigned num_targets, struct mapped_device *md);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Then call this once for each target.
|
|
|
|
*/
|
|
|
|
int dm_table_add_target(struct dm_table *t, const char *type,
|
|
|
|
sector_t start, sector_t len, char *params);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Finally call this to make the table ready for use.
|
|
|
|
*/
|
|
|
|
int dm_table_complete(struct dm_table *t);
|
|
|
|
|
2008-10-10 06:37:11 -06:00
|
|
|
/*
|
|
|
|
* Unplug all devices in a table.
|
|
|
|
*/
|
|
|
|
void dm_table_unplug_all(struct dm_table *t);
|
|
|
|
|
2006-06-26 01:27:33 -06:00
|
|
|
/*
|
|
|
|
* Table reference counting.
|
|
|
|
*/
|
2009-12-10 16:52:19 -07:00
|
|
|
struct dm_table *dm_get_live_table(struct mapped_device *md);
|
2006-06-26 01:27:33 -06:00
|
|
|
void dm_table_get(struct dm_table *t);
|
|
|
|
void dm_table_put(struct dm_table *t);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Queries
|
|
|
|
*/
|
|
|
|
sector_t dm_table_get_size(struct dm_table *t);
|
|
|
|
unsigned int dm_table_get_num_targets(struct dm_table *t);
|
2008-09-02 13:28:45 -06:00
|
|
|
fmode_t dm_table_get_mode(struct dm_table *t);
|
2006-06-26 01:27:33 -06:00
|
|
|
struct mapped_device *dm_table_get_md(struct dm_table *t);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Trigger an event.
|
|
|
|
*/
|
|
|
|
void dm_table_event(struct dm_table *t);
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The device must be suspended before calling this method.
|
2009-12-10 16:52:24 -07:00
|
|
|
* Returns the previous table, which the caller must destroy.
|
2006-06-26 01:27:33 -06:00
|
|
|
*/
|
2009-12-10 16:52:24 -07:00
|
|
|
struct dm_table *dm_swap_table(struct mapped_device *md,
|
|
|
|
struct dm_table *t);
|
2006-06-26 01:27:33 -06:00
|
|
|
|
2008-10-10 06:37:12 -06:00
|
|
|
/*
|
|
|
|
* A wrapper around vmalloc.
|
|
|
|
*/
|
|
|
|
void *dm_vcalloc(unsigned long nmemb, unsigned long elem_size);
|
|
|
|
|
2008-04-24 14:43:52 -06:00
|
|
|
/*-----------------------------------------------------------------
|
|
|
|
* Macros.
|
|
|
|
*---------------------------------------------------------------*/
|
|
|
|
#define DM_NAME "device-mapper"
|
|
|
|
|
2009-01-05 20:04:58 -07:00
|
|
|
#define DMCRIT(f, arg...) \
|
|
|
|
printk(KERN_CRIT DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
|
|
|
|
|
2008-04-24 14:43:52 -06:00
|
|
|
#define DMERR(f, arg...) \
|
|
|
|
printk(KERN_ERR DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
|
|
|
|
#define DMERR_LIMIT(f, arg...) \
|
|
|
|
do { \
|
|
|
|
if (printk_ratelimit()) \
|
|
|
|
printk(KERN_ERR DM_NAME ": " DM_MSG_PREFIX ": " \
|
|
|
|
f "\n", ## arg); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
#define DMWARN(f, arg...) \
|
|
|
|
printk(KERN_WARNING DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
|
|
|
|
#define DMWARN_LIMIT(f, arg...) \
|
|
|
|
do { \
|
|
|
|
if (printk_ratelimit()) \
|
|
|
|
printk(KERN_WARNING DM_NAME ": " DM_MSG_PREFIX ": " \
|
|
|
|
f "\n", ## arg); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
#define DMINFO(f, arg...) \
|
|
|
|
printk(KERN_INFO DM_NAME ": " DM_MSG_PREFIX ": " f "\n", ## arg)
|
|
|
|
#define DMINFO_LIMIT(f, arg...) \
|
|
|
|
do { \
|
|
|
|
if (printk_ratelimit()) \
|
|
|
|
printk(KERN_INFO DM_NAME ": " DM_MSG_PREFIX ": " f \
|
|
|
|
"\n", ## arg); \
|
|
|
|
} while (0)
|
|
|
|
|
|
|
|
#ifdef CONFIG_DM_DEBUG
|
|
|
|
# define DMDEBUG(f, arg...) \
|
|
|
|
printk(KERN_DEBUG DM_NAME ": " DM_MSG_PREFIX " DEBUG: " f "\n", ## arg)
|
|
|
|
# define DMDEBUG_LIMIT(f, arg...) \
|
|
|
|
do { \
|
|
|
|
if (printk_ratelimit()) \
|
|
|
|
printk(KERN_DEBUG DM_NAME ": " DM_MSG_PREFIX ": " f \
|
|
|
|
"\n", ## arg); \
|
|
|
|
} while (0)
|
|
|
|
#else
|
|
|
|
# define DMDEBUG(f, arg...) do {} while (0)
|
|
|
|
# define DMDEBUG_LIMIT(f, arg...) do {} while (0)
|
|
|
|
#endif
|
|
|
|
|
|
|
|
#define DMEMIT(x...) sz += ((sz >= maxlen) ? \
|
|
|
|
0 : scnprintf(result + sz, maxlen - sz, x))
|
|
|
|
|
|
|
|
#define SECTOR_SHIFT 9
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Definitions of return values from target end_io function.
|
|
|
|
*/
|
|
|
|
#define DM_ENDIO_INCOMPLETE 1
|
|
|
|
#define DM_ENDIO_REQUEUE 2
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Definitions of return values from target map function.
|
|
|
|
*/
|
|
|
|
#define DM_MAPIO_SUBMITTED 0
|
|
|
|
#define DM_MAPIO_REMAPPED 1
|
|
|
|
#define DM_MAPIO_REQUEUE DM_ENDIO_REQUEUE
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Ceiling(n / sz)
|
|
|
|
*/
|
|
|
|
#define dm_div_up(n, sz) (((n) + (sz) - 1) / (sz))
|
|
|
|
|
|
|
|
#define dm_sector_div_up(n, sz) ( \
|
|
|
|
{ \
|
|
|
|
sector_t _r = ((n) + (sz) - 1); \
|
|
|
|
sector_div(_r, (sz)); \
|
|
|
|
_r; \
|
|
|
|
} \
|
|
|
|
)
|
|
|
|
|
|
|
|
/*
|
|
|
|
* ceiling(n / size) * size
|
|
|
|
*/
|
|
|
|
#define dm_round_up(n, sz) (dm_div_up((n), (sz)) * (sz))
|
|
|
|
|
2008-10-21 10:44:57 -06:00
|
|
|
#define dm_array_too_big(fixed, obj, num) \
|
|
|
|
((num) > (UINT_MAX - (fixed)) / (obj))
|
|
|
|
|
2008-04-24 14:43:52 -06:00
|
|
|
static inline sector_t to_sector(unsigned long n)
|
|
|
|
{
|
|
|
|
return (n >> SECTOR_SHIFT);
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline unsigned long to_bytes(sector_t n)
|
|
|
|
{
|
|
|
|
return (n << SECTOR_SHIFT);
|
|
|
|
}
|
|
|
|
|
dm: prepare for request based option
This patch adds core functions for request-based dm.
When struct mapped device (md) is initialized, md->queue has
an I/O scheduler and the following functions are used for
request-based dm as the queue functions:
make_request_fn: dm_make_request()
pref_fn: dm_prep_fn()
request_fn: dm_request_fn()
softirq_done_fn: dm_softirq_done()
lld_busy_fn: dm_lld_busy()
Actual initializations are done in another patch (PATCH 2).
Below is a brief summary of how request-based dm behaves, including:
- making request from bio
- cloning, mapping and dispatching request
- completing request and bio
- suspending md
- resuming md
bio to request
==============
md->queue->make_request_fn() (dm_make_request()) calls __make_request()
for a bio submitted to the md.
Then, the bio is kept in the queue as a new request or merged into
another request in the queue if possible.
Cloning and Mapping
===================
Cloning and mapping are done in md->queue->request_fn() (dm_request_fn()),
when requests are dispatched after they are sorted by the I/O scheduler.
dm_request_fn() checks busy state of underlying devices using
target's busy() function and stops dispatching requests to keep them
on the dm device's queue if busy.
It helps better I/O merging, since no merge is done for a request
once it is dispatched to underlying devices.
Actual cloning and mapping are done in dm_prep_fn() and map_request()
called from dm_request_fn().
dm_prep_fn() clones not only request but also bios of the request
so that dm can hold bio completion in error cases and prevent
the bio submitter from noticing the error.
(See the "Completion" section below for details.)
After the cloning, the clone is mapped by target's map_rq() function
and inserted to underlying device's queue using
blk_insert_cloned_request().
Completion
==========
Request completion can be hooked by rq->end_io(), but then, all bios
in the request will have been completed even error cases, and the bio
submitter will have noticed the error.
To prevent the bio completion in error cases, request-based dm clones
both bio and request and hooks both bio->bi_end_io() and rq->end_io():
bio->bi_end_io(): end_clone_bio()
rq->end_io(): end_clone_request()
Summary of the request completion flow is below:
blk_end_request() for a clone request
=> blk_update_request()
=> bio->bi_end_io() == end_clone_bio() for each clone bio
=> Free the clone bio
=> Success: Complete the original bio (blk_update_request())
Error: Don't complete the original bio
=> blk_finish_request()
=> rq->end_io() == end_clone_request()
=> blk_complete_request()
=> dm_softirq_done()
=> Free the clone request
=> Success: Complete the original request (blk_end_request())
Error: Requeue the original request
end_clone_bio() completes the original request on the size of
the original bio in successful cases.
Even if all bios in the original request are completed by that
completion, the original request must not be completed yet to keep
the ordering of request completion for the stacking.
So end_clone_bio() uses blk_update_request() instead of
blk_end_request().
In error cases, end_clone_bio() doesn't complete the original bio.
It just frees the cloned bio and gives over the error handling to
end_clone_request().
end_clone_request(), which is called with queue lock held, completes
the clone request and the original request in a softirq context
(dm_softirq_done()), which has no queue lock, to avoid a deadlock
issue on submission of another request during the completion:
- The submitted request may be mapped to the same device
- Request submission requires queue lock, but the queue lock
has been held by itself and it doesn't know that
The clone request has no clone bio when dm_softirq_done() is called.
So target drivers can't resubmit it again even error cases.
Instead, they can ask dm core for requeueing and remapping
the original request in that cases.
suspend
=======
Request-based dm uses stopping md->queue as suspend of the md.
For noflush suspend, just stops md->queue.
For flush suspend, inserts a marker request to the tail of md->queue.
And dispatches all requests in md->queue until the marker comes to
the front of md->queue. Then, stops dispatching request and waits
for the all dispatched requests to complete.
After that, completes the marker request, stops md->queue and
wake up the waiter on the suspend queue, md->wait.
resume
======
Starts md->queue.
Signed-off-by: Kiyoshi Ueda <k-ueda@ct.jp.nec.com>
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
2009-06-22 03:12:35 -06:00
|
|
|
/*-----------------------------------------------------------------
|
|
|
|
* Helper for block layer and dm core operations
|
|
|
|
*---------------------------------------------------------------*/
|
|
|
|
void dm_dispatch_request(struct request *rq);
|
|
|
|
void dm_requeue_unmapped_request(struct request *rq);
|
|
|
|
void dm_kill_unmapped_request(struct request *rq, int error);
|
|
|
|
int dm_underlying_device_busy(struct request_queue *q);
|
|
|
|
|
2006-06-26 01:27:33 -06:00
|
|
|
#endif /* _LINUX_DEVICE_MAPPER_H */
|