2004-06-12 23:16:10 +02:00
|
|
|
/*-
|
2017-11-27 16:17:37 +01:00
|
|
|
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
|
|
|
|
|
*
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
* Copyright (c) 2004, 2007 Lukas Ertl
|
2004-06-12 23:16:10 +02:00
|
|
|
* All rights reserved.
|
|
|
|
|
*
|
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
|
* are met:
|
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
|
* notice, this list of conditions and the following disclaimer.
|
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
|
|
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
|
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
|
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
|
|
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
|
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
|
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
|
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
|
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
|
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
|
|
|
* SUCH DAMAGE.
|
|
|
|
|
*
|
|
|
|
|
* $FreeBSD$
|
|
|
|
|
*/
|
|
|
|
|
|
|
|
|
|
#ifndef _GEOM_VINUM_H_
|
|
|
|
|
#define _GEOM_VINUM_H_
|
|
|
|
|
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
/* geom_vinum_create.c */
|
|
|
|
|
void gv_concat(struct g_geom *gp, struct gctl_req *);
|
|
|
|
|
void gv_mirror(struct g_geom *gp, struct gctl_req *);
|
|
|
|
|
void gv_stripe(struct g_geom *gp, struct gctl_req *);
|
|
|
|
|
void gv_raid5(struct g_geom *gp, struct gctl_req *);
|
|
|
|
|
int gv_create_drive(struct gv_softc *, struct gv_drive *);
|
|
|
|
|
int gv_create_volume(struct gv_softc *, struct gv_volume *);
|
|
|
|
|
int gv_create_plex(struct gv_softc *, struct gv_plex *);
|
|
|
|
|
int gv_create_sd(struct gv_softc *, struct gv_sd *);
|
2004-06-12 23:16:10 +02:00
|
|
|
|
|
|
|
|
/* geom_vinum_drive.c */
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
void gv_save_config(struct gv_softc *);
|
2008-10-01 16:50:36 +02:00
|
|
|
int gv_read_header(struct g_consumer *, struct gv_hdr *);
|
|
|
|
|
int gv_write_header(struct g_consumer *, struct gv_hdr *);
|
2004-06-12 23:16:10 +02:00
|
|
|
|
|
|
|
|
/* geom_vinum_init.c */
|
|
|
|
|
void gv_start_obj(struct g_geom *, struct gctl_req *);
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
int gv_start_plex(struct gv_plex *);
|
|
|
|
|
int gv_start_vol(struct gv_volume *);
|
2004-06-12 23:16:10 +02:00
|
|
|
|
|
|
|
|
/* geom_vinum_list.c */
|
|
|
|
|
void gv_ld(struct g_geom *, struct gctl_req *, struct sbuf *);
|
|
|
|
|
void gv_lp(struct g_geom *, struct gctl_req *, struct sbuf *);
|
|
|
|
|
void gv_ls(struct g_geom *, struct gctl_req *, struct sbuf *);
|
|
|
|
|
void gv_lv(struct g_geom *, struct gctl_req *, struct sbuf *);
|
|
|
|
|
void gv_list(struct g_geom *, struct gctl_req *);
|
|
|
|
|
|
2005-11-19 21:25:18 +01:00
|
|
|
/* geom_vinum_move.c */
|
|
|
|
|
void gv_move(struct g_geom *, struct gctl_req *);
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
int gv_move_sd(struct gv_softc *, struct gv_sd *, struct gv_drive *, int);
|
2005-11-19 21:25:18 +01:00
|
|
|
|
|
|
|
|
/* geom_vinum_rename.c */
|
|
|
|
|
void gv_rename(struct g_geom *, struct gctl_req *);
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
int gv_rename_drive(struct gv_softc *, struct gv_drive *, char *, int);
|
|
|
|
|
int gv_rename_plex(struct gv_softc *, struct gv_plex *, char *, int);
|
|
|
|
|
int gv_rename_sd(struct gv_softc *, struct gv_sd *, char *, int);
|
|
|
|
|
int gv_rename_vol(struct gv_softc *, struct gv_volume *, char *, int);
|
2005-11-19 21:25:18 +01:00
|
|
|
|
2004-06-12 23:16:10 +02:00
|
|
|
/* geom_vinum_rm.c */
|
|
|
|
|
void gv_remove(struct g_geom *, struct gctl_req *);
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
int gv_resetconfig(struct gv_softc *);
|
|
|
|
|
void gv_rm_sd(struct gv_softc *sc, struct gv_sd *s);
|
|
|
|
|
void gv_rm_drive(struct gv_softc *, struct gv_drive *, int);
|
|
|
|
|
void gv_rm_plex(struct gv_softc *, struct gv_plex *);
|
|
|
|
|
void gv_rm_vol(struct gv_softc *, struct gv_volume *);
|
|
|
|
|
|
2004-06-12 23:16:10 +02:00
|
|
|
/* geom_vinum_state.c */
|
|
|
|
|
int gv_sdstatemap(struct gv_plex *);
|
2004-11-26 13:31:36 +01:00
|
|
|
void gv_setstate(struct g_geom *, struct gctl_req *);
|
2004-06-12 23:16:10 +02:00
|
|
|
int gv_set_drive_state(struct gv_drive *, int, int);
|
|
|
|
|
int gv_set_sd_state(struct gv_sd *, int, int);
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
int gv_set_vol_state(struct gv_volume *, int, int);
|
|
|
|
|
int gv_set_plex_state(struct gv_plex *, int, int);
|
2004-06-12 23:16:10 +02:00
|
|
|
void gv_update_sd_state(struct gv_sd *);
|
|
|
|
|
void gv_update_plex_state(struct gv_plex *);
|
|
|
|
|
void gv_update_vol_state(struct gv_volume *);
|
|
|
|
|
|
|
|
|
|
/* geom_vinum_subr.c */
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
void gv_adjust_freespace(struct gv_sd *, off_t);
|
|
|
|
|
void gv_free_sd(struct gv_sd *);
|
|
|
|
|
struct gv_drive *gv_find_drive(struct gv_softc *, char *);
|
|
|
|
|
struct gv_drive *gv_find_drive_device(struct gv_softc *, char *);
|
|
|
|
|
struct gv_plex *gv_find_plex(struct gv_softc *, char *);
|
|
|
|
|
struct gv_sd *gv_find_sd(struct gv_softc *, char *);
|
|
|
|
|
struct gv_volume *gv_find_vol(struct gv_softc *, char *);
|
|
|
|
|
void gv_format_config(struct gv_softc *, struct sbuf *, int,
|
|
|
|
|
char *);
|
|
|
|
|
int gv_is_striped(struct gv_plex *);
|
|
|
|
|
int gv_consumer_is_open(struct g_consumer *);
|
|
|
|
|
int gv_provider_is_open(struct g_provider *);
|
|
|
|
|
int gv_object_type(struct gv_softc *, char *);
|
|
|
|
|
void gv_parse_config(struct gv_softc *, char *,
|
|
|
|
|
struct gv_drive *);
|
|
|
|
|
int gv_sd_to_drive(struct gv_sd *, struct gv_drive *);
|
|
|
|
|
int gv_sd_to_plex(struct gv_sd *, struct gv_plex *);
|
|
|
|
|
int gv_sdcount(struct gv_plex *, int);
|
|
|
|
|
void gv_update_plex_config(struct gv_plex *);
|
|
|
|
|
void gv_update_vol_size(struct gv_volume *, off_t);
|
|
|
|
|
off_t gv_vol_size(struct gv_volume *);
|
|
|
|
|
off_t gv_plex_size(struct gv_plex *);
|
|
|
|
|
int gv_plexdown(struct gv_volume *);
|
|
|
|
|
int gv_attach_plex(struct gv_plex *, struct gv_volume *,
|
|
|
|
|
int);
|
|
|
|
|
int gv_attach_sd(struct gv_sd *, struct gv_plex *, off_t,
|
|
|
|
|
int);
|
|
|
|
|
int gv_detach_plex(struct gv_plex *, int);
|
|
|
|
|
int gv_detach_sd(struct gv_sd *, int);
|
|
|
|
|
|
|
|
|
|
/* geom_vinum.c */
|
|
|
|
|
void gv_worker(void *);
|
|
|
|
|
void gv_post_event(struct gv_softc *, int, void *, void *, intmax_t,
|
|
|
|
|
intmax_t);
|
2010-05-10 21:12:23 +02:00
|
|
|
void gv_worker_exit(struct gv_softc *);
|
2009-05-06 20:21:48 +02:00
|
|
|
struct gv_event *gv_get_event(struct gv_softc *);
|
|
|
|
|
void gv_remove_event(struct gv_softc *, struct gv_event *);
|
2019-12-27 02:36:53 +01:00
|
|
|
void gv_drive_done(struct gv_drive *);
|
Import the gvinum work that have been done during and after Summer of Code 2007.
The work have been under testing and fixing since then, and it is mature enough
to be put into HEAD for further testing.
A lot have changed in this time, and here are the most important:
- Gvinum now uses one single workerthread instead of one thread for each
volume and each plex. The reason for this is that the previous scheme was
very complex, and was the cause of many of the bugs discovered in gvinum.
Instead, gvinum now uses one worker thread with an event queue, quite
similar to what used in gmirror.
- The rebuild/grow/initialize/parity check routines no longer runs in
separate threads, but are run as regular I/O requests with special flags.
This made it easier to support mounted growing and parity rebuild.
- Support for growing striped and raid5-plexes, meaning that one can extend the
volumes for these plex types in addition to the concat type. Also works while
the volume is mounted.
- Implementation of many of the missing commands from the old vinum:
attach/detach, start (was partially implemented), stop (was partially
implemented), concat, mirror, stripe, raid5 (shortcuts for creating volumes
with one plex of these organizations).
- The parity check and rebuild no longer goes between userland/kernel, meaning
that the gvinum command will not stay and wait forever for the rebuild to
finish. You can instead watch the status with the list command.
- Many problems with gvinum have been reported since 5.x, and some has been hard
to fix due to the complicated architecture. Hopefully, it should be more
stable and better handle edge cases that previously made gvinum crash.
- Failed drives no longer disappears entirely, but now leave behind a dummy
drive that makes sure the original state is not forgotten in case the system
is rebooted between drive failures/swaps.
- Update manpage to reflect new commands and extend it with some examples.
Sponsored by: Google Summer of Code 2007
Mentored by: le
Tested by: Rick C. Petty <rick-freebsd2008 -at- kiwi-computer.com>
2009-03-28 18:20:08 +01:00
|
|
|
void gv_drive_tasted(struct gv_softc *, struct g_provider *);
|
|
|
|
|
void gv_drive_lost(struct gv_softc *, struct gv_drive *);
|
|
|
|
|
void gv_setup_objects(struct gv_softc *);
|
|
|
|
|
void gv_start(struct bio *);
|
|
|
|
|
int gv_access(struct g_provider *, int, int, int);
|
|
|
|
|
void gv_cleanup(struct gv_softc *);
|
|
|
|
|
|
|
|
|
|
/* geom_vinum_volume.c */
|
|
|
|
|
void gv_done(struct bio *);
|
|
|
|
|
void gv_volume_start(struct gv_softc *, struct bio *);
|
|
|
|
|
void gv_volume_flush(struct gv_volume *);
|
|
|
|
|
void gv_bio_done(struct gv_softc *, struct bio *);
|
|
|
|
|
|
|
|
|
|
/* geom_vinum_plex.c */
|
|
|
|
|
void gv_plex_start(struct gv_plex *, struct bio *);
|
|
|
|
|
void gv_plex_raid5_done(struct gv_plex *, struct bio *);
|
|
|
|
|
void gv_plex_normal_done(struct gv_plex *, struct bio *);
|
|
|
|
|
int gv_grow_request(struct gv_plex *, off_t, off_t, int, caddr_t);
|
|
|
|
|
void gv_grow_complete(struct gv_plex *, struct bio *);
|
|
|
|
|
void gv_init_request(struct gv_sd *, off_t, caddr_t, off_t);
|
|
|
|
|
void gv_init_complete(struct gv_plex *, struct bio *);
|
|
|
|
|
void gv_parity_request(struct gv_plex *, int, off_t);
|
|
|
|
|
void gv_parity_complete(struct gv_plex *, struct bio *);
|
|
|
|
|
void gv_rebuild_complete(struct gv_plex *, struct bio *);
|
|
|
|
|
int gv_sync_request(struct gv_plex *, struct gv_plex *, off_t, off_t, int,
|
|
|
|
|
caddr_t);
|
|
|
|
|
int gv_sync_complete(struct gv_plex *, struct bio *);
|
2004-06-12 23:16:10 +02:00
|
|
|
|
2008-10-26 18:20:37 +01:00
|
|
|
extern u_int g_vinum_debug;
|
|
|
|
|
|
2019-08-07 21:28:35 +02:00
|
|
|
#define G_VINUM_DEBUG(lvl, ...) \
|
|
|
|
|
_GEOM_DEBUG("GEOM_VINUM", g_vinum_debug, (lvl), NULL, __VA_ARGS__)
|
|
|
|
|
#define G_VINUM_LOGREQ(lvl, bp, ...) \
|
|
|
|
|
_GEOM_DEBUG("GEOM_VINUM", g_vinum_debug, (lvl), (bp), __VA_ARGS__)
|
2008-10-26 18:20:37 +01:00
|
|
|
|
2004-06-12 23:16:10 +02:00
|
|
|
#endif /* !_GEOM_VINUM_H_ */
|