2000-10-13 13:04:45 +00:00
|
|
|
|
/*-
|
2008-04-10 13:05:05 +00:00
|
|
|
|
* Copyright (c) 2000 - 2008 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.
|
|
|
|
|
|
*
|
|
|
|
|
|
* 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 */
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
#define MAX_ARRAYS 16
|
2005-12-05 17:33:57 +00:00
|
|
|
|
#define MAX_VOLUMES 4
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
#define MAX_DISKS 16
|
|
|
|
|
|
#define AR_PROXIMITY 2048 /* how many sectors is "close" */
|
|
|
|
|
|
|
|
|
|
|
|
#define ATA_MAGIC "FreeBSD ATA driver RAID "
|
|
|
|
|
|
|
|
|
|
|
|
struct ata_raid_subdisk {
|
2005-12-05 17:33:57 +00:00
|
|
|
|
struct ar_softc *raid[MAX_VOLUMES];
|
|
|
|
|
|
int disk_number[MAX_VOLUMES];
|
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
|
|
|
|
};
|
|
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
/* ATA PseudoRAID Metadata */
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct ar_softc {
|
2005-12-05 17:33:57 +00:00
|
|
|
|
int lun;
|
|
|
|
|
|
u_int8_t name[32];
|
2006-01-18 13:10:17 +00:00
|
|
|
|
int volume;
|
2005-12-05 17:33:57 +00:00
|
|
|
|
u_int64_t magic_0;
|
|
|
|
|
|
u_int64_t magic_1;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
int type;
|
|
|
|
|
|
#define AR_T_JBOD 0x0001
|
|
|
|
|
|
#define AR_T_SPAN 0x0002
|
|
|
|
|
|
#define AR_T_RAID0 0x0004
|
|
|
|
|
|
#define AR_T_RAID1 0x0008
|
|
|
|
|
|
#define AR_T_RAID01 0x0010
|
|
|
|
|
|
#define AR_T_RAID3 0x0020
|
|
|
|
|
|
#define AR_T_RAID4 0x0040
|
|
|
|
|
|
#define AR_T_RAID5 0x0080
|
|
|
|
|
|
|
|
|
|
|
|
int status;
|
|
|
|
|
|
#define AR_S_READY 0x0001
|
|
|
|
|
|
#define AR_S_DEGRADED 0x0002
|
|
|
|
|
|
#define AR_S_REBUILDING 0x0004
|
|
|
|
|
|
|
|
|
|
|
|
int format;
|
|
|
|
|
|
#define AR_F_FREEBSD_RAID 0x0001
|
|
|
|
|
|
#define AR_F_ADAPTEC_RAID 0x0002
|
|
|
|
|
|
#define AR_F_HPTV2_RAID 0x0004
|
|
|
|
|
|
#define AR_F_HPTV3_RAID 0x0008
|
|
|
|
|
|
#define AR_F_INTEL_RAID 0x0010
|
|
|
|
|
|
#define AR_F_ITE_RAID 0x0020
|
2006-02-17 13:02:10 +00:00
|
|
|
|
#define AR_F_JMICRON_RAID 0x0040
|
|
|
|
|
|
#define AR_F_LSIV2_RAID 0x0080
|
|
|
|
|
|
#define AR_F_LSIV3_RAID 0x0100
|
|
|
|
|
|
#define AR_F_NVIDIA_RAID 0x0200
|
|
|
|
|
|
#define AR_F_PROMISE_RAID 0x0400
|
|
|
|
|
|
#define AR_F_SII_RAID 0x0800
|
|
|
|
|
|
#define AR_F_SIS_RAID 0x1000
|
|
|
|
|
|
#define AR_F_VIA_RAID 0x2000
|
2006-09-12 20:37:22 +00:00
|
|
|
|
#define AR_F_FORMAT_MASK 0x3fff
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
2005-12-05 17:33:57 +00:00
|
|
|
|
u_int generation;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int64_t total_sectors;
|
|
|
|
|
|
u_int64_t offset_sectors; /* offset from start of disk */
|
|
|
|
|
|
u_int16_t heads;
|
|
|
|
|
|
u_int16_t sectors;
|
|
|
|
|
|
u_int32_t cylinders;
|
|
|
|
|
|
u_int width; /* array width in disks */
|
2005-12-05 17:33:57 +00:00
|
|
|
|
u_int interleave; /* interleave in sectors */
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int total_disks; /* number of disks in this array */
|
|
|
|
|
|
struct ar_disk {
|
|
|
|
|
|
device_t dev;
|
|
|
|
|
|
u_int8_t serial[16]; /* serial # of physical disk */
|
|
|
|
|
|
u_int64_t sectors; /* useable sectors on this disk */
|
|
|
|
|
|
off_t last_lba; /* last lba used (for performance) */
|
|
|
|
|
|
u_int flags;
|
|
|
|
|
|
#define AR_DF_PRESENT 0x0001 /* this HW pos has a disk present */
|
|
|
|
|
|
#define AR_DF_ASSIGNED 0x0002 /* this HW pos assigned to an array */
|
|
|
|
|
|
#define AR_DF_SPARE 0x0004 /* this HW pos is a spare */
|
|
|
|
|
|
#define AR_DF_ONLINE 0x0008 /* this HW pos is online and in use */
|
|
|
|
|
|
|
|
|
|
|
|
} disks[MAX_DISKS];
|
|
|
|
|
|
int toggle; /* performance hack for RAID1's */
|
|
|
|
|
|
u_int64_t rebuild_lba; /* rebuild progress indicator */
|
|
|
|
|
|
struct mtx lock; /* metadata lock */
|
|
|
|
|
|
struct disk *disk; /* disklabel/slice stuff */
|
|
|
|
|
|
struct proc *pid; /* rebuilder process id */
|
2000-10-13 13:04:45 +00:00
|
|
|
|
};
|
|
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
/* Adaptec HostRAID Metadata */
|
|
|
|
|
|
#define ADP_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 17)
|
|
|
|
|
|
|
|
|
|
|
|
/* note all entries are big endian */
|
|
|
|
|
|
struct adaptec_raid_conf {
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int32_t magic_0;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
#define ADP_MAGIC_0 0xc4650790
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t generation;
|
|
|
|
|
|
u_int16_t dummy_0;
|
|
|
|
|
|
u_int16_t total_configs;
|
|
|
|
|
|
u_int16_t dummy_1;
|
|
|
|
|
|
u_int16_t checksum;
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int32_t dummy_2;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t dummy_3;
|
|
|
|
|
|
u_int32_t flags;
|
|
|
|
|
|
u_int32_t timestamp;
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int32_t dummy_4[4];
|
|
|
|
|
|
u_int32_t dummy_5[4];
|
|
|
|
|
|
struct {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int16_t total_disks;
|
|
|
|
|
|
u_int16_t generation;
|
|
|
|
|
|
u_int32_t magic_0;
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define ADP_T_RAID0 0x00
|
|
|
|
|
|
#define ADP_T_RAID1 0x01
|
|
|
|
|
|
u_int8_t dummy_1;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t dummy_2;
|
|
|
|
|
|
u_int8_t dummy_3;
|
|
|
|
|
|
u_int8_t dummy_4;
|
|
|
|
|
|
u_int8_t dummy_5;
|
2004-06-25 21:21:59 +00:00
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t disk_number;
|
|
|
|
|
|
u_int32_t dummy_6;
|
|
|
|
|
|
u_int32_t sectors;
|
|
|
|
|
|
u_int16_t stripe_shift;
|
|
|
|
|
|
u_int16_t dummy_7;
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t dummy_8[4];
|
|
|
|
|
|
u_int8_t name[16];
|
2005-04-12 12:25:27 +00:00
|
|
|
|
} configs[127];
|
|
|
|
|
|
u_int32_t dummy_6[13];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t magic_1;
|
|
|
|
|
|
#define ADP_MAGIC_1 0x9ff85009
|
|
|
|
|
|
u_int32_t dummy_7[3];
|
|
|
|
|
|
u_int32_t magic_2;
|
|
|
|
|
|
u_int32_t dummy_8[46];
|
|
|
|
|
|
u_int32_t magic_3;
|
|
|
|
|
|
#define ADP_MAGIC_3 0x4d545044
|
|
|
|
|
|
u_int32_t magic_4;
|
|
|
|
|
|
#define ADP_MAGIC_4 0x9ff85009
|
|
|
|
|
|
u_int32_t dummy_9[62];
|
|
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/* Highpoint V2 RocketRAID Metadata */
|
|
|
|
|
|
#define HPTV2_LBA(dev) 9
|
|
|
|
|
|
|
|
|
|
|
|
struct hptv2_raid_conf {
|
|
|
|
|
|
int8_t filler1[32];
|
|
|
|
|
|
u_int32_t magic;
|
|
|
|
|
|
#define HPTV2_MAGIC_OK 0x5a7816f0
|
|
|
|
|
|
#define HPTV2_MAGIC_BAD 0x5a7816fd
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t magic_0;
|
|
|
|
|
|
u_int32_t magic_1;
|
|
|
|
|
|
u_int32_t order;
|
|
|
|
|
|
#define HPTV2_O_RAID0 0x01
|
|
|
|
|
|
#define HPTV2_O_RAID1 0x02
|
|
|
|
|
|
#define HPTV2_O_OK 0x04
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t stripe_shift;
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define HPTV2_T_RAID0 0x00
|
|
|
|
|
|
#define HPTV2_T_RAID1 0x01
|
|
|
|
|
|
#define HPTV2_T_RAID01_RAID0 0x02
|
|
|
|
|
|
#define HPTV2_T_SPAN 0x03
|
|
|
|
|
|
#define HPTV2_T_RAID_3 0x04
|
|
|
|
|
|
#define HPTV2_T_RAID_5 0x05
|
|
|
|
|
|
#define HPTV2_T_JBOD 0x06
|
|
|
|
|
|
#define HPTV2_T_RAID01_RAID1 0x07
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int32_t total_sectors;
|
|
|
|
|
|
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;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t timestamp;
|
|
|
|
|
|
u_int8_t reason;
|
|
|
|
|
|
#define HPTV2_R_REMOVED 0xfe
|
|
|
|
|
|
#define HPTV2_R_BROKEN 0xff
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t disk;
|
|
|
|
|
|
u_int8_t status;
|
|
|
|
|
|
u_int8_t sectors;
|
|
|
|
|
|
u_int32_t lba;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
} errorlog[32];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
int8_t filler2[16];
|
|
|
|
|
|
u_int32_t rebuild_lba;
|
|
|
|
|
|
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];
|
2002-09-23 18:54:32 +00:00
|
|
|
|
} __packed;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
2002-02-12 11:35:15 +00:00
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
/* Highpoint V3 RocketRAID Metadata */
|
|
|
|
|
|
#define HPTV3_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 11)
|
|
|
|
|
|
|
|
|
|
|
|
struct hptv3_raid_conf {
|
|
|
|
|
|
u_int32_t magic;
|
|
|
|
|
|
#define HPTV3_MAGIC 0x5a7816f3
|
2004-06-25 21:21:59 +00:00
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t magic_0;
|
|
|
|
|
|
u_int8_t checksum_0;
|
|
|
|
|
|
u_int8_t mode;
|
|
|
|
|
|
#define HPTV3_BOOT_MARK 0x01
|
|
|
|
|
|
#define HPTV3_USER_MODE 0x02
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t user_mode;
|
|
|
|
|
|
u_int8_t config_entries;
|
|
|
|
|
|
struct {
|
|
|
|
|
|
u_int32_t total_sectors;
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define HPTV3_T_SPARE 0x00
|
|
|
|
|
|
#define HPTV3_T_JBOD 0x03
|
|
|
|
|
|
#define HPTV3_T_SPAN 0x04
|
|
|
|
|
|
#define HPTV3_T_RAID0 0x05
|
|
|
|
|
|
#define HPTV3_T_RAID1 0x06
|
|
|
|
|
|
#define HPTV3_T_RAID3 0x07
|
|
|
|
|
|
#define HPTV3_T_RAID5 0x08
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t total_disks;
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t stripe_shift;
|
|
|
|
|
|
u_int16_t status;
|
|
|
|
|
|
#define HPTV3_T_NEED_REBUILD 0x01
|
|
|
|
|
|
#define HPTV3_T_RAID5_FLAG 0x02
|
|
|
|
|
|
|
|
|
|
|
|
u_int16_t critical_disks;
|
|
|
|
|
|
u_int32_t rebuild_lba;
|
|
|
|
|
|
} __packed configs[2];
|
|
|
|
|
|
u_int8_t name[16];
|
|
|
|
|
|
u_int32_t timestamp;
|
|
|
|
|
|
u_int8_t description[64];
|
|
|
|
|
|
u_int8_t creator[16];
|
|
|
|
|
|
u_int8_t checksum_1;
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t dummy_1;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
#define HPTV3_T_ENABLE_TCQ 0x01
|
|
|
|
|
|
#define HPTV3_T_ENABLE_NCQ 0x02
|
|
|
|
|
|
#define HPTV3_T_ENABLE_WCACHE 0x04
|
|
|
|
|
|
#define HPTV3_T_ENABLE_RCACHE 0x08
|
|
|
|
|
|
|
|
|
|
|
|
struct {
|
|
|
|
|
|
u_int32_t total_sectors;
|
|
|
|
|
|
u_int32_t rebuild_lba;
|
|
|
|
|
|
} __packed configs_high[2];
|
|
|
|
|
|
u_int32_t filler[87];
|
|
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/* Intel MatrixRAID Metadata */
|
|
|
|
|
|
#define INTEL_LBA(dev) \
|
2005-12-02 10:13:53 +00:00
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 3)
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
struct intel_raid_conf {
|
|
|
|
|
|
u_int8_t intel_id[24];
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define INTEL_MAGIC "Intel Raid ISM Cfg Sig. "
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t version[6];
|
2005-12-05 17:33:57 +00:00
|
|
|
|
#define INTEL_VERSION_1100 "1.1.00"
|
|
|
|
|
|
#define INTEL_VERSION_1201 "1.2.01"
|
|
|
|
|
|
#define INTEL_VERSION_1202 "1.2.02"
|
|
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t dummy_0[2];
|
|
|
|
|
|
u_int32_t checksum;
|
|
|
|
|
|
u_int32_t config_size;
|
|
|
|
|
|
u_int32_t config_id;
|
|
|
|
|
|
u_int32_t generation;
|
|
|
|
|
|
u_int32_t dummy_1[2];
|
|
|
|
|
|
u_int8_t total_disks;
|
|
|
|
|
|
u_int8_t total_volumes;
|
|
|
|
|
|
u_int8_t dummy_2[2];
|
|
|
|
|
|
u_int32_t filler_0[39];
|
|
|
|
|
|
struct {
|
|
|
|
|
|
u_int8_t serial[16];
|
|
|
|
|
|
u_int32_t sectors;
|
|
|
|
|
|
u_int32_t id;
|
|
|
|
|
|
u_int32_t flags;
|
|
|
|
|
|
#define INTEL_F_SPARE 0x01
|
|
|
|
|
|
#define INTEL_F_ASSIGNED 0x02
|
|
|
|
|
|
#define INTEL_F_DOWN 0x04
|
|
|
|
|
|
#define INTEL_F_ONLINE 0x08
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t filler[5];
|
|
|
|
|
|
} __packed disk[1];
|
|
|
|
|
|
u_int32_t filler_1[62];
|
|
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
struct intel_raid_mapping {
|
|
|
|
|
|
u_int8_t name[16];
|
|
|
|
|
|
u_int64_t total_sectors __packed;
|
|
|
|
|
|
u_int32_t state;
|
|
|
|
|
|
u_int32_t reserved;
|
2005-12-05 17:33:57 +00:00
|
|
|
|
u_int32_t filler_0[20];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t offset;
|
|
|
|
|
|
u_int32_t disk_sectors;
|
|
|
|
|
|
u_int32_t stripe_count;
|
|
|
|
|
|
u_int16_t stripe_sectors;
|
|
|
|
|
|
u_int8_t status;
|
|
|
|
|
|
#define INTEL_S_READY 0x00
|
|
|
|
|
|
#define INTEL_S_DISABLED 0x01
|
|
|
|
|
|
#define INTEL_S_DEGRADED 0x02
|
|
|
|
|
|
#define INTEL_S_FAILURE 0x03
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define INTEL_T_RAID0 0x00
|
|
|
|
|
|
#define INTEL_T_RAID1 0x01
|
2005-12-02 10:13:53 +00:00
|
|
|
|
#define INTEL_T_RAID5 0x05
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t total_disks;
|
2005-12-05 17:33:57 +00:00
|
|
|
|
u_int8_t magic[3];
|
|
|
|
|
|
u_int32_t filler_1[7];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t disk_idx[1];
|
|
|
|
|
|
} __packed;
|
2004-06-25 21:21:59 +00:00
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
/* Integrated Technology Express Metadata */
|
|
|
|
|
|
#define ITE_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 2)
|
|
|
|
|
|
|
|
|
|
|
|
struct ite_raid_conf {
|
|
|
|
|
|
u_int32_t filler_1[5];
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int8_t timestamp_0[8];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t dummy_1;
|
|
|
|
|
|
u_int32_t filler_2[5];
|
|
|
|
|
|
u_int16_t filler_3;
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int8_t ite_id[40];
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define ITE_MAGIC "Integrated Technology Express Inc "
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int16_t filler_4;
|
|
|
|
|
|
u_int32_t filler_5[6];
|
|
|
|
|
|
u_int32_t dummy_2;
|
|
|
|
|
|
u_int32_t dummy_3;
|
|
|
|
|
|
u_int32_t filler_6[12];
|
|
|
|
|
|
u_int32_t dummy_4;
|
|
|
|
|
|
u_int32_t filler_7[5];
|
|
|
|
|
|
u_int64_t total_sectors __packed;
|
|
|
|
|
|
u_int32_t filler_8[12];
|
|
|
|
|
|
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int16_t filler_9;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define ITE_T_RAID0 0x00
|
|
|
|
|
|
#define ITE_T_RAID1 0x01
|
|
|
|
|
|
#define ITE_T_RAID01 0x02
|
|
|
|
|
|
#define ITE_T_SPAN 0x03
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t filler_10;
|
|
|
|
|
|
u_int32_t dummy_5[8];
|
|
|
|
|
|
u_int8_t stripe_1kblocks;
|
|
|
|
|
|
u_int8_t filler_11[3];
|
|
|
|
|
|
u_int32_t filler_12[54];
|
|
|
|
|
|
|
2005-04-12 12:25:27 +00:00
|
|
|
|
u_int32_t dummy_6[4];
|
|
|
|
|
|
u_int8_t timestamp_1[8];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t filler_13[9];
|
|
|
|
|
|
u_int8_t stripe_sectors;
|
|
|
|
|
|
u_int8_t filler_14[3];
|
|
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t filler_15[3];
|
|
|
|
|
|
u_int32_t filler_16;
|
|
|
|
|
|
u_int8_t filler_17;
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int32_t disk_sectors;
|
|
|
|
|
|
u_int16_t filler_18;
|
|
|
|
|
|
u_int32_t dummy_7[4];
|
|
|
|
|
|
u_int32_t filler_20[104];
|
|
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
2006-02-17 13:02:10 +00:00
|
|
|
|
/* JMicron Technology Corp Metadata */
|
|
|
|
|
|
#define JMICRON_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 1)
|
|
|
|
|
|
#define JM_MAX_DISKS 8
|
|
|
|
|
|
|
|
|
|
|
|
struct jmicron_raid_conf {
|
|
|
|
|
|
u_int8_t signature[2];
|
|
|
|
|
|
#define JMICRON_MAGIC "JM"
|
|
|
|
|
|
|
|
|
|
|
|
u_int16_t version;
|
|
|
|
|
|
#define JMICRON_VERSION 0x0001
|
|
|
|
|
|
|
|
|
|
|
|
u_int16_t checksum;
|
|
|
|
|
|
u_int8_t filler_1[10];
|
|
|
|
|
|
u_int32_t disk_id;
|
|
|
|
|
|
u_int32_t offset;
|
|
|
|
|
|
u_int32_t disk_sectors_high;
|
|
|
|
|
|
u_int16_t disk_sectors_low;
|
|
|
|
|
|
u_int8_t filler_2[2];
|
|
|
|
|
|
u_int8_t name[16];
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define JM_T_RAID0 0
|
|
|
|
|
|
#define JM_T_RAID1 1
|
|
|
|
|
|
#define JM_T_RAID01 2
|
|
|
|
|
|
#define JM_T_JBOD 3
|
|
|
|
|
|
#define JM_T_RAID5 5
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t stripe_shift;
|
|
|
|
|
|
u_int16_t flags;
|
|
|
|
|
|
#define JM_F_READY 0x0001
|
|
|
|
|
|
#define JM_F_BOOTABLE 0x0002
|
|
|
|
|
|
#define JM_F_BAD 0x0004
|
|
|
|
|
|
#define JM_F_ACTIVE 0c0010
|
|
|
|
|
|
#define JM_F_UNSYNC 0c0020
|
|
|
|
|
|
#define JM_F_NEWEST 0c0040
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t filler_3[4];
|
|
|
|
|
|
u_int32_t spare[2];
|
|
|
|
|
|
u_int32_t disks[JM_MAX_DISKS];
|
|
|
|
|
|
u_int8_t filler_4[32];
|
|
|
|
|
|
u_int8_t filler_5[384];
|
|
|
|
|
|
};
|
|
|
|
|
|
|
|
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
/* LSILogic V2 MegaRAID Metadata */
|
|
|
|
|
|
#define LSIV2_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 1)
|
|
|
|
|
|
|
|
|
|
|
|
struct lsiv2_raid_conf {
|
|
|
|
|
|
u_int8_t lsi_id[6];
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define LSIV2_MAGIC "$XIDE$"
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
u_int16_t version;
|
|
|
|
|
|
u_int8_t config_entries;
|
|
|
|
|
|
u_int8_t raid_count;
|
|
|
|
|
|
u_int8_t total_disks;
|
|
|
|
|
|
u_int8_t dummy_1;
|
|
|
|
|
|
u_int16_t dummy_2;
|
2004-06-25 21:21:59 +00:00
|
|
|
|
|
|
|
|
|
|
union {
|
|
|
|
|
|
struct {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define LSIV2_T_RAID0 0x01
|
|
|
|
|
|
#define LSIV2_T_RAID1 0x02
|
|
|
|
|
|
#define LSIV2_T_SPARE 0x08
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int16_t stripe_sectors;
|
|
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t disk_count;
|
|
|
|
|
|
u_int8_t config_offset;
|
|
|
|
|
|
u_int8_t dummy_1;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
#define LSIV2_R_DEGRADED 0x02
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t total_sectors;
|
|
|
|
|
|
u_int8_t filler[3];
|
2004-06-25 21:21:59 +00:00
|
|
|
|
} __packed raid;
|
|
|
|
|
|
struct {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t device;
|
|
|
|
|
|
#define LSIV2_D_MASTER 0x00
|
|
|
|
|
|
#define LSIV2_D_SLAVE 0x01
|
|
|
|
|
|
#define LSIV2_D_CHANNEL0 0x00
|
|
|
|
|
|
#define LSIV2_D_CHANNEL1 0x10
|
|
|
|
|
|
#define LSIV2_D_NONE 0xff
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int32_t disk_sectors;
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t raid_number;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
#define LSIV2_D_GONE 0x02
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t filler[7];
|
2004-06-25 21:21:59 +00:00
|
|
|
|
} __packed disk;
|
|
|
|
|
|
} configs[30];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t raid_number;
|
|
|
|
|
|
u_int32_t timestamp;
|
|
|
|
|
|
u_int8_t filler[10];
|
|
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/* LSILogic V3 MegaRAID Metadata */
|
|
|
|
|
|
#define LSIV3_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 4)
|
|
|
|
|
|
|
|
|
|
|
|
struct lsiv3_raid_conf {
|
|
|
|
|
|
u_int32_t magic_0; /* 0xa0203200 */
|
|
|
|
|
|
u_int32_t filler_0[3];
|
|
|
|
|
|
u_int8_t magic_1[4]; /* "SATA" */
|
|
|
|
|
|
u_int32_t filler_1[40];
|
|
|
|
|
|
u_int32_t dummy_0; /* 0x0d000003 */
|
|
|
|
|
|
u_int32_t filler_2[7];
|
|
|
|
|
|
u_int32_t dummy_1; /* 0x0d000003 */
|
|
|
|
|
|
u_int32_t filler_3[70];
|
|
|
|
|
|
u_int8_t magic_2[8]; /* "$_ENQ$31" */
|
|
|
|
|
|
u_int8_t filler_4[7];
|
|
|
|
|
|
u_int8_t checksum_0;
|
|
|
|
|
|
u_int8_t filler_5[512*2];
|
|
|
|
|
|
u_int8_t lsi_id[6];
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define LSIV3_MAGIC "$_IDE$"
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int16_t dummy_2; /* 0x33de for OK disk */
|
|
|
|
|
|
u_int16_t version; /* 0x0131 for this version */
|
|
|
|
|
|
u_int16_t dummy_3; /* 0x0440 always */
|
|
|
|
|
|
u_int32_t filler_6;
|
|
|
|
|
|
|
|
|
|
|
|
struct {
|
|
|
|
|
|
u_int16_t stripe_pages;
|
|
|
|
|
|
u_int8_t type;
|
|
|
|
|
|
#define LSIV3_T_RAID0 0x00
|
|
|
|
|
|
#define LSIV3_T_RAID1 0x01
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t total_disks;
|
|
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t filler_0[10];
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t sectors;
|
|
|
|
|
|
u_int16_t dummy_1;
|
|
|
|
|
|
u_int32_t offset;
|
|
|
|
|
|
u_int16_t dummy_2;
|
|
|
|
|
|
u_int8_t device;
|
|
|
|
|
|
#define LSIV3_D_DEVICE 0x01
|
|
|
|
|
|
#define LSIV3_D_CHANNEL 0x10
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t dummy_3;
|
|
|
|
|
|
u_int8_t dummy_4;
|
|
|
|
|
|
u_int8_t dummy_5;
|
|
|
|
|
|
u_int8_t filler_1[16];
|
|
|
|
|
|
} __packed raid[8];
|
|
|
|
|
|
struct {
|
|
|
|
|
|
u_int32_t disk_sectors;
|
|
|
|
|
|
u_int32_t dummy_0;
|
|
|
|
|
|
u_int32_t dummy_1;
|
|
|
|
|
|
u_int8_t dummy_2;
|
|
|
|
|
|
u_int8_t dummy_3;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
#define LSIV3_D_MIRROR 0x00
|
|
|
|
|
|
#define LSIV3_D_STRIPE 0xff
|
|
|
|
|
|
u_int8_t dummy_4;
|
|
|
|
|
|
} __packed disk[6];
|
|
|
|
|
|
u_int8_t filler_7[7];
|
|
|
|
|
|
u_int8_t device;
|
|
|
|
|
|
u_int32_t timestamp;
|
|
|
|
|
|
u_int8_t filler_8[3];
|
|
|
|
|
|
u_int8_t checksum_1;
|
2004-06-25 21:21:59 +00:00
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
2005-06-06 13:35:24 +00:00
|
|
|
|
/* nVidia MediaShield Metadata */
|
|
|
|
|
|
#define NVIDIA_LBA(dev) \
|
|
|
|
|
|
(((struct ad_softc *)device_get_ivars(dev))->total_secs - 2)
|
|
|
|
|
|
|
|
|
|
|
|
struct nvidia_raid_conf {
|
|
|
|
|
|
u_int8_t nvidia_id[8];
|
|
|
|
|
|
#define NV_MAGIC "NVIDIA "
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t config_size;
|
|
|
|
|
|
u_int32_t checksum;
|
|
|
|
|
|
u_int16_t version;
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int32_t total_sectors;
|
|
|
|
|
|
u_int32_t sector_size;
|
|
|
|
|
|
u_int8_t serial[16];
|
|
|
|
|
|
u_int8_t revision[4];
|
|
|
|
|
|
u_int32_t dummy_1;
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t magic_0;
|
|
|
|
|
|
#define NV_MAGIC0 0x00640044
|
|
|
|
|
|
|
|
|
|
|
|
u_int64_t magic_1;
|
|
|
|
|
|
u_int64_t magic_2;
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t total_disks;
|
|
|
|
|
|
u_int8_t dummy_2;
|
|
|
|
|
|
u_int16_t type;
|
|
|
|
|
|
#define NV_T_RAID0 0x00000080
|
|
|
|
|
|
#define NV_T_RAID1 0x00000081
|
|
|
|
|
|
#define NV_T_RAID3 0x00000083
|
|
|
|
|
|
#define NV_T_RAID5 0x00000085
|
|
|
|
|
|
#define NV_T_RAID01 0x00008180
|
|
|
|
|
|
#define NV_T_SPAN 0x000000ff
|
|
|
|
|
|
|
|
|
|
|
|
u_int16_t dummy_3;
|
|
|
|
|
|
u_int32_t stripe_sectors;
|
|
|
|
|
|
u_int32_t stripe_bytes;
|
|
|
|
|
|
u_int32_t stripe_shift;
|
|
|
|
|
|
u_int32_t stripe_mask;
|
|
|
|
|
|
u_int32_t stripe_sizesectors;
|
|
|
|
|
|
u_int32_t stripe_sizebytes;
|
|
|
|
|
|
u_int32_t rebuild_lba;
|
|
|
|
|
|
u_int32_t dummy_4;
|
|
|
|
|
|
u_int32_t dummy_5;
|
|
|
|
|
|
u_int32_t status;
|
|
|
|
|
|
#define NV_S_BOOTABLE 0x00000001
|
|
|
|
|
|
#define NV_S_DEGRADED 0x00000002
|
|
|
|
|
|
|
|
|
|
|
|
u_int32_t filler[98];
|
|
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
/* Promise FastTrak Metadata */
|
2005-12-15 13:30:23 +00:00
|
|
|
|
#define PROMISE_LBA(dev) \
|
2005-05-01 08:45:12 +00:00
|
|
|
|
(((((struct ad_softc *)device_get_ivars(dev))->total_secs / (((struct ad_softc *)device_get_ivars(dev))->heads * ((struct ad_softc *)device_get_ivars(dev))->sectors)) * ((struct ad_softc *)device_get_ivars(dev))->heads * ((struct ad_softc *)device_get_ivars(dev))->sectors) - ((struct ad_softc *)device_get_ivars(dev))->sectors)
|
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
|
|
|
|
|
2000-10-13 13:04:45 +00:00
|
|
|
|
struct promise_raid_conf {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
char promise_id[24];
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define PR_MAGIC "Promise Technology, Inc."
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
u_int32_t dummy_0;
|
|
|
|
|
|
u_int64_t magic_0;
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define PR_MAGIC0(x) (((u_int64_t)(x.channel) << 48) | \
|
2005-04-12 12:25:27 +00:00
|
|
|
|
((u_int64_t)(x.device != 0) << 56))
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +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 {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int32_t integrity;
|
|
|
|
|
|
#define PR_I_VALID 0x00000080
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
#define PR_F_VALID 0x00000001
|
|
|
|
|
|
#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
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t channel;
|
|
|
|
|
|
u_int8_t device;
|
|
|
|
|
|
u_int64_t magic_0 __packed;
|
|
|
|
|
|
u_int32_t disk_offset;
|
|
|
|
|
|
u_int32_t disk_sectors;
|
|
|
|
|
|
u_int32_t rebuild_lba;
|
|
|
|
|
|
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
|
|
|
|
|
|
#define PR_S_FUNCTIONAL 0x80
|
|
|
|
|
|
|
|
|
|
|
|
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
|
|
|
|
|
|
#define PR_T_JBOD 0x10
|
|
|
|
|
|
|
|
|
|
|
|
u_int8_t total_disks;
|
|
|
|
|
|
u_int8_t stripe_shift;
|
|
|
|
|
|
u_int8_t array_width;
|
|
|
|
|
|
u_int8_t array_number;
|
|
|
|
|
|
u_int32_t total_sectors;
|
|
|
|
|
|
u_int16_t cylinders;
|
|
|
|
|
|
u_int8_t heads;
|
|
|
|
|
|
u_int8_t sectors;
|
|
|
|
|
|
u_int64_t magic_1 __packed;
|
2004-06-25 21:21:59 +00:00
|
|
|
|
struct {
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
u_int8_t flags;
|
|
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t channel;
|
|
|
|
|
|
u_int8_t device;
|
|
|
|
|
|
u_int64_t magic_0 __packed;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
} disk[8];
|
2000-11-01 17:35:44 +00:00
|
|
|
|
} raid;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
int32_t filler2[346];
|
|
|
|
|
|
u_int32_t checksum;
|
2002-09-23 18:54:32 +00:00
|
|
|
|
} __packed;
|
2000-10-13 13:04:45 +00:00
|
|
|
|
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
|
|
|
|
|
|
/* Silicon Image Medley Metadata */
|
|
|
|
|
|
#define SII_LBA(dev) \
|
|
|
|
|
|
( ((struct ad_softc *)device_get_ivars(dev))->total_secs - 1)
|
|
|
|
|
|
|
|
|
|
|
|
struct sii_raid_conf {
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int16_t ata_params_00_53[54];
|
|
|
|
|
|
u_int64_t total_sectors;
|
|
|
|
|
|
u_int16_t ata_params_58_79[70];
|
|
|
|
|
|
u_int16_t dummy_0;
|
|
|
|
|
|
u_int16_t dummy_1;
|
|
|
|
|
|
u_int32_t controller_pci_id;
|
|
|
|
|
|
u_int16_t version_minor;
|
|
|
|
|
|
u_int16_t version_major;
|
|
|
|
|
|
u_int8_t timestamp[6];
|
|
|
|
|
|
u_int16_t stripe_sectors;
|
|
|
|
|
|
u_int16_t dummy_2;
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t type;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
#define SII_T_RAID0 0x00
|
|
|
|
|
|
#define SII_T_RAID1 0x01
|
|
|
|
|
|
#define SII_T_RAID01 0x02
|
|
|
|
|
|
#define SII_T_SPARE 0x03
|
|
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t raid0_disks;
|
|
|
|
|
|
u_int8_t raid0_ident;
|
|
|
|
|
|
u_int8_t raid1_disks;
|
|
|
|
|
|
u_int8_t raid1_ident;
|
|
|
|
|
|
u_int64_t rebuild_lba;
|
|
|
|
|
|
u_int32_t generation;
|
|
|
|
|
|
u_int8_t status;
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
#define SII_S_READY 0x01
|
|
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t base_raid1_position;
|
|
|
|
|
|
u_int8_t base_raid0_position;
|
|
|
|
|
|
u_int8_t position;
|
|
|
|
|
|
u_int16_t dummy_3;
|
|
|
|
|
|
u_int8_t name[16];
|
|
|
|
|
|
u_int16_t checksum_0;
|
|
|
|
|
|
int8_t filler1[190];
|
|
|
|
|
|
u_int16_t checksum_1;
|
2005-11-28 23:08:37 +00:00
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
/* Silicon Integrated Systems RAID Metadata */
|
|
|
|
|
|
#define SIS_LBA(dev) \
|
|
|
|
|
|
( ((struct ad_softc *)device_get_ivars(dev))->total_secs - 16)
|
|
|
|
|
|
|
|
|
|
|
|
struct sis_raid_conf {
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int16_t magic;
|
2005-11-28 23:08:37 +00:00
|
|
|
|
#define SIS_MAGIC 0x0010
|
|
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t disks;
|
|
|
|
|
|
#define SIS_D_MASTER 0xf0
|
|
|
|
|
|
#define SIS_D_MIRROR 0x0f
|
2005-11-28 23:08:37 +00:00
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t type_total_disks;
|
|
|
|
|
|
#define SIS_D_MASK 0x0f
|
2005-11-28 23:08:37 +00:00
|
|
|
|
#define SIS_T_MASK 0xf0
|
|
|
|
|
|
#define SIS_T_JBOD 0x10
|
2006-01-18 13:10:17 +00:00
|
|
|
|
#define SIS_T_RAID0 0x20
|
2005-11-28 23:08:37 +00:00
|
|
|
|
#define SIS_T_RAID1 0x30
|
|
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int32_t dummy_0;
|
|
|
|
|
|
u_int32_t controller_pci_id;
|
|
|
|
|
|
u_int16_t stripe_sectors;
|
|
|
|
|
|
u_int16_t dummy_1;
|
|
|
|
|
|
u_int32_t timestamp;
|
|
|
|
|
|
u_int8_t model[40];
|
|
|
|
|
|
u_int8_t disk_number;
|
|
|
|
|
|
u_int8_t dummy_2[3];
|
|
|
|
|
|
int8_t filler1[448];
|
2005-04-12 12:25:27 +00:00
|
|
|
|
} __packed;
|
|
|
|
|
|
|
|
|
|
|
|
|
2005-04-12 13:29:47 +00:00
|
|
|
|
/* VIA Tech V-RAID Metadata */
|
2005-04-12 12:25:27 +00:00
|
|
|
|
#define VIA_LBA(dev) \
|
|
|
|
|
|
( ((struct ad_softc *)device_get_ivars(dev))->total_secs - 1)
|
|
|
|
|
|
|
|
|
|
|
|
struct via_raid_conf {
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int16_t magic;
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define VIA_MAGIC 0xaa55
|
|
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t dummy_0;
|
|
|
|
|
|
u_int8_t type;
|
2005-12-14 13:07:49 +00:00
|
|
|
|
#define VIA_T_MASK 0x7e
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define VIA_T_BOOTABLE 0x01
|
|
|
|
|
|
#define VIA_T_RAID0 0x04
|
|
|
|
|
|
#define VIA_T_RAID1 0x0c
|
2005-12-14 13:07:49 +00:00
|
|
|
|
#define VIA_T_RAID01 0x4c
|
|
|
|
|
|
#define VIA_T_RAID5 0x2c
|
2005-04-30 16:22:07 +00:00
|
|
|
|
#define VIA_T_SPAN 0x44
|
2005-12-14 13:07:49 +00:00
|
|
|
|
#define VIA_T_UNKNOWN 0x80
|
2005-04-30 16:22:07 +00:00
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t disk_index;
|
|
|
|
|
|
#define VIA_D_MASK 0x0f
|
|
|
|
|
|
#define VIA_D_DEGRADED 0x10
|
|
|
|
|
|
#define VIA_D_HIGH_IDX 0x20
|
2005-09-06 19:17:48 +00:00
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int8_t stripe_layout;
|
2005-12-15 13:30:23 +00:00
|
|
|
|
#define VIA_L_DISKS 0x07
|
|
|
|
|
|
#define VIA_L_MASK 0xf0
|
2006-01-18 13:10:17 +00:00
|
|
|
|
#define VIA_L_SHIFT 4
|
2005-04-30 16:22:07 +00:00
|
|
|
|
|
2006-01-18 13:10:17 +00:00
|
|
|
|
u_int64_t disk_sectors;
|
|
|
|
|
|
u_int32_t disk_id;
|
|
|
|
|
|
u_int32_t disks[8];
|
|
|
|
|
|
u_int8_t checksum;
|
|
|
|
|
|
u_int8_t filler_1[461];
|
This is the much rumoured ATA mkIII update that I've been working on.
o ATA is now fully newbus'd and split into modules.
This means that on a modern system you just load "atapci and ata"
to get the base support, and then one or more of the device
subdrivers "atadisk atapicd atapifd atapist ataraid".
All can be loaded/unloaded anytime, but for obvious reasons you
dont want to unload atadisk when you have mounted filesystems.
o The device identify part of the probe has been rewritten to fix
the problems with odd devices the old had, and to try to remove
so of the long delays some HW could provoke. Also probing is done
without the need for interrupts, making earlier probing possible.
o SATA devices can be hot inserted/removed and devices will be created/
removed in /dev accordingly.
NOTE: only supported on controllers that has this feature:
Promise and Silicon Image for now.
On other controllers the usual atacontrol detach/attach dance is
still needed.
o Support for "atomic" composite ATA requests used for RAID.
o ATA RAID support has been rewritten and and now supports these
metadata formats:
"Adaptec HostRAID"
"Highpoint V2 RocketRAID"
"Highpoint V3 RocketRAID"
"Intel MatrixRAID"
"Integrated Technology Express"
"LSILogic V2 MegaRAID"
"LSILogic V3 MegaRAID"
"Promise FastTrak"
"Silicon Image Medley"
"FreeBSD PseudoRAID"
o Update the ioctl API to match new RAID levels etc.
o Update atacontrol to know about the new RAID levels etc
NOTE: you need to recompile atacontrol with the new sys/ata.h,
make world will take care of that.
NOTE2: that rebuild is done differently from the old system as
the rebuild is now done piggybacked on read requests to the
array, so atacontrol simply starts a background "dd" to rebuild
the array.
o The reinit code has been worked over to be much more robust.
o The timeout code has been overhauled for races.
o Support of new chipsets.
o Lots of fixes for bugs found while doing the modulerization and
reviewing the old code.
Missing or changed features from current ATA:
o atapi-cd no longer has support for ATAPI changers. Todays its
much cheaper and alot faster to copy those CD images to disk
and serve them from there. Besides they dont seem to be made
anymore, maybe for that exact reason.
o ATA RAID can only read metadata from all the above metadata formats,
not write all of them (Promise and Highpoint V2 so far). This means
that arrays can be picked up from the BIOS, but they cannot be
created from FreeBSD. There is more to it than just the missing
write metadata support, those formats are not unique to a given
controller like Promise and Highpoint formats, instead they exist
for several types, and even worse, some controllers can have
different formats and its impossible to tell which one.
The outcome is that we cannot reliably create the metadata of those
formats and be sure the controller BIOS will understand it.
However write support is needed to update/fail/rebuild the arrays
properly so it sits fairly high on the TODO list.
o So far atapicam is not supported with these changes. When/if this
will change is up to the maintainer of atapi-cam so go there for
questions.
HW donated by: Webveveriet AS
HW donated by: Frode Nordahl
HW donated by: Yahoo!
HW donated by: Sentex
Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
|
|
|
|
} __packed;
|
