2000-10-13 13:04:45 +00:00
|
|
|
|
/*-
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
* Copyright (c) 2000,2001,2002 S<EFBFBD>ren Schmidt <sos@FreeBSD.org>
|
2000-10-13 13:04:45 +00:00
|
|
|
|
* 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,
|
|
|
|
|
|
* without modification, immediately at the beginning of the file.
|
|
|
|
|
|
* 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.
|
|
|
|
|
|
* 3. The name of the author may not be used to endorse or promote products
|
|
|
|
|
|
* derived from this software without specific prior written permission.
|
|
|
|
|
|
*
|
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``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 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.
|
|
|
|
|
|
*
|
|
|
|
|
|
* $FreeBSD$
|
|
|
|
|
|
*/
|
|
|
|
|
|
|
2002-03-07 16:32:21 +00:00
|
|
|
|
/* misc defines */
|
|
|
|
|
|
#define MAX_ARRAYS 16
|
|
|
|
|
|
#define MAX_DISKS 16
|
|
|
|
|
|
#define AR_PROXIMITY 2048
|
|
|
|
|
|
#define AR_READ 0x01
|
|
|
|
|
|
#define AR_WRITE 0x02
|
|
|
|
|
|
#define AR_WAIT 0x04
|
2002-03-08 11:33:52 +00:00
|
|
|
|
#define AR_STRATEGY(x) (x)->bio_dev->si_disk->d_devsw->d_strategy((x))
|
2002-03-07 16:32:21 +00:00
|
|
|
|
#define AD_SOFTC(x) ((struct ad_softc *)(x.device->driver))
|
2002-03-27 10:58:59 +00:00
|
|
|
|
#define ATA_MAGIC "FreeBSD ATA driver RAID "
|
2002-03-07 16:32:21 +00:00
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ar_disk {
|
|
|
|
|
|
struct ata_device *device;
|
2002-03-03 15:36:21 +00:00
|
|
|
|
u_int64_t disk_sectors; /* sectors on this disk */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
off_t last_lba; /* last lba used */
|
|
|
|
|
|
int flags;
|
|
|
|
|
|
#define AR_DF_PRESENT 0x00000001
|
|
|
|
|
|
#define AR_DF_ASSIGNED 0x00000002
|
|
|
|
|
|
#define AR_DF_SPARE 0x00000004
|
|
|
|
|
|
#define AR_DF_ONLINE 0x00000008
|
|
|
|
|
|
};
|
|
|
|
|
|
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct ar_softc {
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int lun;
|
|
|
|
|
|
int32_t magic_0; /* ident for this array */
|
|
|
|
|
|
int32_t magic_1; /* ident for this array */
|
|
|
|
|
|
int flags;
|
|
|
|
|
|
#define AR_F_RAID0 0x0001 /* STRIPE */
|
|
|
|
|
|
#define AR_F_RAID1 0x0002 /* MIRROR */
|
|
|
|
|
|
#define AR_F_SPAN 0x0004 /* SPAN */
|
|
|
|
|
|
#define AR_F_READY 0x0100
|
|
|
|
|
|
#define AR_F_DEGRADED 0x0200
|
|
|
|
|
|
#define AR_F_REBUILDING 0x0400
|
|
|
|
|
|
#define AR_F_PROMISE_RAID 0x1000
|
|
|
|
|
|
#define AR_F_HIGHPOINT_RAID 0x2000
|
2002-03-27 10:58:59 +00:00
|
|
|
|
#define AR_F_FREEBSD_RAID 0x4000
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int total_disks; /* number of disks in this array */
|
|
|
|
|
|
int generation; /* generation of this array */
|
2002-02-12 11:35:15 +00:00
|
|
|
|
struct ar_disk disks[MAX_DISKS+1]; /* ptr to each disk in array */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int width; /* array width in disks */
|
|
|
|
|
|
u_int16_t heads;
|
|
|
|
|
|
u_int16_t sectors;
|
|
|
|
|
|
u_int32_t cylinders;
|
|
|
|
|
|
u_int64_t total_sectors;
|
2002-03-27 10:58:59 +00:00
|
|
|
|
int interleave; /* interleave in blocks */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int reserved; /* sectors that are NOT to be used */
|
|
|
|
|
|
int offset; /* offset from start of disk */
|
|
|
|
|
|
u_int64_t lock_start; /* start of locked area for rebuild */
|
|
|
|
|
|
u_int64_t lock_end; /* end of locked area for rebuild */
|
2002-03-15 15:39:54 +00:00
|
|
|
|
struct disk disk; /* disklabel/slice stuff */
|
|
|
|
|
|
struct proc *pid; /* rebuilder process id */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
dev_t dev; /* device place holder */
|
2000-10-13 13:04:45 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
struct ar_buf {
|
2002-03-07 16:32:21 +00:00
|
|
|
|
struct bio bp; /* must be first element! */
|
2002-03-08 21:36:49 +00:00
|
|
|
|
struct bio *org;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ar_buf *mirror;
|
|
|
|
|
|
int drive;
|
|
|
|
|
|
int flags;
|
|
|
|
|
|
#define AB_F_DONE 0x01
|
2000-10-13 13:04:45 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
#define HPT_LBA 9
|
|
|
|
|
|
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct highpoint_raid_conf {
|
|
|
|
|
|
int8_t filler1[32];
|
|
|
|
|
|
u_int32_t magic; /* 0x20 */
|
|
|
|
|
|
#define HPT_MAGIC_OK 0x5a7816f0
|
|
|
|
|
|
#define HPT_MAGIC_BAD 0x5a7816fd
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t magic_0;
|
|
|
|
|
|
u_int32_t magic_1;
|
|
|
|
|
|
u_int32_t order;
|
2002-03-27 10:58:59 +00:00
|
|
|
|
#define HPT_O_RAID0 0x01
|
2002-03-08 21:36:49 +00:00
|
|
|
|
#define HPT_O_RAID1 0x02
|
2002-03-27 10:58:59 +00:00
|
|
|
|
#define HPT_O_OK 0x04
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t stripe_shift;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
u_int8_t type;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
#define HPT_T_RAID0 0x00
|
|
|
|
|
|
#define HPT_T_RAID1 0x01
|
|
|
|
|
|
#define HPT_T_RAID01_RAID0 0x02
|
2000-10-13 13:04:45 +00:00
|
|
|
|
#define HPT_T_SPAN 0x03
|
|
|
|
|
|
#define HPT_T_RAID_3 0x04
|
|
|
|
|
|
#define HPT_T_RAID_5 0x05
|
|
|
|
|
|
#define HPT_T_SINGLEDISK 0x06
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
#define HPT_T_RAID01_RAID1 0x07
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t disk_number;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int32_t total_sectors;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
u_int32_t disk_mode;
|
|
|
|
|
|
u_int32_t boot_mode;
|
|
|
|
|
|
u_int8_t boot_disk;
|
|
|
|
|
|
u_int8_t boot_protect;
|
|
|
|
|
|
u_int8_t error_log_entries;
|
|
|
|
|
|
u_int8_t error_log_index;
|
|
|
|
|
|
struct {
|
|
|
|
|
|
u_int32_t timestamp;
|
|
|
|
|
|
u_int8_t reason;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
#define HPT_R_REMOVED 0xfe
|
|
|
|
|
|
#define HPT_R_BROKEN 0xff
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t disk;
|
|
|
|
|
|
u_int8_t status;
|
|
|
|
|
|
u_int8_t sectors;
|
|
|
|
|
|
u_int32_t lba;
|
|
|
|
|
|
} errorlog[32];
|
2002-03-08 11:33:52 +00:00
|
|
|
|
int8_t filler2[16];
|
2002-03-08 21:36:49 +00:00
|
|
|
|
u_int32_t rebuild_lba;
|
2002-03-08 11:33:52 +00:00
|
|
|
|
u_int8_t dummy_1;
|
|
|
|
|
|
u_int8_t name_1[15];
|
|
|
|
|
|
u_int8_t dummy_2;
|
|
|
|
|
|
u_int8_t name_2[15];
|
|
|
|
|
|
int8_t filler3[8];
|
2001-10-04 18:02:26 +00:00
|
|
|
|
} __attribute__((packed));
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
2002-02-12 11:35:15 +00:00
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
#define PR_LBA(adp) \
|
|
|
|
|
|
(((adp->total_secs / (adp->heads * adp->sectors)) * \
|
|
|
|
|
|
adp->heads * adp->sectors) - adp->sectors)
|
|
|
|
|
|
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct promise_raid_conf {
|
|
|
|
|
|
char promise_id[24];
|
|
|
|
|
|
#define PR_MAGIC "Promise Technology, Inc."
|
|
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int32_t dummy_0;
|
2002-02-12 11:35:15 +00:00
|
|
|
|
u_int64_t magic_0;
|
|
|
|
|
|
#define PR_MAGIC0(x) ((u_int64_t)x.device->channel->unit << 48) | \
|
|
|
|
|
|
((u_int64_t)(x.device->unit != 0) << 56)
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int16_t magic_1;
|
|
|
|
|
|
u_int32_t magic_2;
|
|
|
|
|
|
u_int8_t filler1[470];
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct {
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int32_t integrity; /* 0x200 */
|
|
|
|
|
|
#define PR_I_VALID 0x00000080
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t flags;
|
2001-10-04 18:02:26 +00:00
|
|
|
|
#define PR_F_VALID 0x00000001
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
#define PR_F_ONLINE 0x00000002
|
|
|
|
|
|
#define PR_F_ASSIGNED 0x00000004
|
|
|
|
|
|
#define PR_F_SPARE 0x00000008
|
|
|
|
|
|
#define PR_F_DUPLICATE 0x00000010
|
|
|
|
|
|
#define PR_F_REDIR 0x00000020
|
|
|
|
|
|
#define PR_F_DOWN 0x00000040
|
|
|
|
|
|
#define PR_F_READY 0x00000080
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t channel;
|
|
|
|
|
|
u_int8_t device;
|
2002-03-16 15:55:20 +00:00
|
|
|
|
u_int64_t magic_0 __attribute__((packed));
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int32_t disk_offset; /* 0x210 */
|
|
|
|
|
|
u_int32_t disk_sectors;
|
2002-02-12 11:35:15 +00:00
|
|
|
|
u_int32_t rebuild_lba;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int16_t generation;
|
|
|
|
|
|
u_int8_t status;
|
|
|
|
|
|
#define PR_S_VALID 0x01
|
|
|
|
|
|
#define PR_S_ONLINE 0x02
|
|
|
|
|
|
#define PR_S_INITED 0x04
|
|
|
|
|
|
#define PR_S_READY 0x08
|
|
|
|
|
|
#define PR_S_DEGRADED 0x10
|
|
|
|
|
|
#define PR_S_MARKED 0x20
|
2002-02-12 11:35:15 +00:00
|
|
|
|
#define PR_S_FUNCTIONAL 0x80
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define PR_T_RAID0 0x00
|
|
|
|
|
|
#define PR_T_RAID1 0x01
|
|
|
|
|
|
#define PR_T_RAID3 0x02
|
|
|
|
|
|
#define PR_T_RAID5 0x04
|
|
|
|
|
|
#define PR_T_SPAN 0x08
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t total_disks; /* 0x220 */
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int8_t stripe_shift;
|
|
|
|
|
|
u_int8_t array_width;
|
2000-11-01 17:35:44 +00:00
|
|
|
|
u_int8_t array_number;
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int32_t total_sectors;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
u_int16_t cylinders;
|
|
|
|
|
|
u_int8_t heads;
|
|
|
|
|
|
u_int8_t sectors;
|
2002-03-16 15:55:20 +00:00
|
|
|
|
int64_t magic_1 __attribute__((packed));
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct { /* 0x240 */
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t channel;
|
|
|
|
|
|
u_int8_t device;
|
2002-03-16 15:55:20 +00:00
|
|
|
|
u_int64_t magic_0 __attribute__((packed));
|
2000-10-13 13:04:45 +00:00
|
|
|
|
} disk[8];
|
2000-11-01 17:35:44 +00:00
|
|
|
|
} raid;
|
|
|
|
|
|
int32_t filler2[346];
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
u_int32_t checksum;
|
2001-10-04 18:02:26 +00:00
|
|
|
|
} __attribute__((packed));
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
2002-03-03 15:36:21 +00:00
|
|
|
|
int ata_raiddisk_probe(struct ad_softc *);
|
|
|
|
|
|
int ata_raiddisk_attach(struct ad_softc *);
|
|
|
|
|
|
int ata_raiddisk_detach(struct ad_softc *);
|
2002-02-12 11:35:15 +00:00
|
|
|
|
void ata_raid_attach(void);
|
2002-03-27 10:58:59 +00:00
|
|
|
|
int ata_raid_create(struct raid_setup *);
|
|
|
|
|
|
int ata_raid_delete(int);
|
2002-04-02 13:48:17 +00:00
|
|
|
|
int ata_raid_status(int array, struct raid_status *);
|
2002-03-03 15:36:21 +00:00
|
|
|
|
int ata_raid_rebuild(int);
|