2001-03-06 21:43:46 +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
|
|
|
|
* Copyright (c) 1998 - 2005 S<EFBFBD>ren Schmidt <sos@FreeBSD.org>
|
2001-03-06 21:43:46 +00:00
|
|
|
|
* All rights reserved.
|
|
|
|
|
*
|
|
|
|
|
* Redistribution and use in source and binary forms, with or without
|
|
|
|
|
* modification, are permitted provided that the following conditions
|
|
|
|
|
* are met:
|
|
|
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
|
|
|
* notice, this list of conditions and the following disclaimer,
|
|
|
|
|
* without modification, immediately at the beginning of the file.
|
|
|
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
|
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
|
|
|
* documentation and/or other materials provided with the distribution.
|
|
|
|
|
* 3. The name of the author may not be used to endorse or promote products
|
|
|
|
|
* derived from this software without specific prior written permission.
|
|
|
|
|
*
|
|
|
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
|
|
|
|
|
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
|
|
|
|
|
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
|
|
|
|
|
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
|
|
|
|
|
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
|
|
|
|
|
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
|
|
|
|
|
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
|
|
|
|
|
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
|
|
|
|
|
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
|
|
|
|
|
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
|
|
|
|
|
*/
|
|
|
|
|
|
2003-08-24 17:55:58 +00:00
|
|
|
|
#include <sys/cdefs.h>
|
|
|
|
|
__FBSDID("$FreeBSD$");
|
|
|
|
|
|
2002-04-05 13:13:56 +00:00
|
|
|
|
#include "opt_ata.h"
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
#include <sys/systm.h>
|
|
|
|
|
#include <sys/kernel.h>
|
|
|
|
|
#include <sys/module.h>
|
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
|
|
|
|
#include <sys/ata.h>
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#include <sys/bus.h>
|
|
|
|
|
#include <sys/malloc.h>
|
2004-01-11 22:08:34 +00:00
|
|
|
|
#include <sys/sema.h>
|
2003-08-24 09:22:26 +00:00
|
|
|
|
#include <sys/taskqueue.h>
|
2004-01-14 21:26:35 +00:00
|
|
|
|
#include <vm/uma.h>
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#include <machine/stdarg.h>
|
|
|
|
|
#include <machine/resource.h>
|
|
|
|
|
#include <machine/bus.h>
|
|
|
|
|
#ifdef __alpha__
|
|
|
|
|
#include <machine/md_var.h>
|
|
|
|
|
#endif
|
|
|
|
|
#include <sys/rman.h>
|
2003-08-22 05:54:52 +00:00
|
|
|
|
#include <dev/pci/pcivar.h>
|
|
|
|
|
#include <dev/pci/pcireg.h>
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#include <dev/ata/ata-all.h>
|
2003-02-20 20:02:32 +00:00
|
|
|
|
#include <dev/ata/ata-pci.h>
|
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
|
|
|
|
#include <ata_if.h>
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2003-02-20 20:02:32 +00:00
|
|
|
|
/* local vars */
|
|
|
|
|
static MALLOC_DEFINE(M_ATAPCI, "ATA PCI", "ATA driver PCI");
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2001-09-25 17:10:39 +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 IOMASK 0xfffffffc
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2003-02-20 20:02:32 +00:00
|
|
|
|
/* prototypes */
|
2003-08-25 11:13:04 +00:00
|
|
|
|
static void ata_pci_dmainit(struct ata_channel *);
|
2001-09-25 17:10:39 +00:00
|
|
|
|
|
2004-04-21 20:03:26 +00:00
|
|
|
|
int
|
|
|
|
|
ata_legacy(device_t dev)
|
|
|
|
|
{
|
|
|
|
|
return ((pci_read_config(dev, PCIR_PROGIF, 1)&PCIP_STORAGE_IDE_MASTERDEV) &&
|
2004-06-15 11:02:09 +00:00
|
|
|
|
((pci_read_config(dev, PCIR_PROGIF, 1) &
|
2004-04-21 20:03:26 +00:00
|
|
|
|
(PCIP_STORAGE_IDE_MODEPRIM | PCIP_STORAGE_IDE_MODESEC)) !=
|
|
|
|
|
(PCIP_STORAGE_IDE_MODEPRIM | PCIP_STORAGE_IDE_MODESEC)));
|
|
|
|
|
}
|
|
|
|
|
|
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
|
2003-02-20 20:02:32 +00:00
|
|
|
|
ata_pci_probe(device_t dev)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
{
|
|
|
|
|
if (pci_get_class(dev) != PCIC_STORAGE)
|
2003-02-20 20:02:32 +00:00
|
|
|
|
return ENXIO;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2003-02-20 20:02:32 +00:00
|
|
|
|
switch (pci_get_vendor(dev)) {
|
|
|
|
|
case ATA_ACARD_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_acard_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_ACER_LABS_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_ali_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_AMD_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_amd_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_CYRIX_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_cyrix_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_CYPRESS_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_cypress_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_HIGHPOINT_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_highpoint_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_INTEL_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_intel_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2004-12-08 11:17:38 +00:00
|
|
|
|
case ATA_ITE_ID:
|
|
|
|
|
if (!ata_ite_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-12-09 19:13:50 +00:00
|
|
|
|
case ATA_NATIONAL_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_national_ident(dev))
|
2003-12-09 19:13:50 +00:00
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_NVIDIA_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_nvidia_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_PROMISE_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_promise_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_SERVERWORKS_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_serverworks_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_SILICON_IMAGE_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_sii_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_SIS_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_sis_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
case ATA_VIA_ID:
|
2004-06-15 11:02:09 +00:00
|
|
|
|
if (!ata_via_ident(dev))
|
|
|
|
|
return 0;
|
|
|
|
|
break;
|
2004-12-08 11:17:38 +00:00
|
|
|
|
case ATA_CENATEK_ID:
|
|
|
|
|
if (pci_get_devid(dev) == ATA_CENATEK_ROCKET) {
|
2003-03-29 13:37:09 +00:00
|
|
|
|
ata_generic_ident(dev);
|
2003-02-20 20:02:32 +00:00
|
|
|
|
device_set_desc(dev, "Cenatek Rocket Drive controller");
|
|
|
|
|
return 0;
|
2002-02-11 15:48:04 +00:00
|
|
|
|
}
|
2003-10-28 19:01:48 +00:00
|
|
|
|
break;
|
2004-12-08 11:17:38 +00:00
|
|
|
|
case ATA_MICRON_ID:
|
|
|
|
|
if (pci_get_devid(dev) == ATA_MICRON_RZ1000 ||
|
|
|
|
|
pci_get_devid(dev) == ATA_MICRON_RZ1001) {
|
2003-03-29 13:37:09 +00:00
|
|
|
|
ata_generic_ident(dev);
|
2003-02-20 20:02:32 +00:00
|
|
|
|
device_set_desc(dev,
|
2004-12-08 11:17:38 +00:00
|
|
|
|
"RZ 100? ATA controller !WARNING! data loss/corruption risk");
|
2003-02-20 20:02:32 +00:00
|
|
|
|
return 0;
|
|
|
|
|
}
|
2003-10-28 19:01:48 +00:00
|
|
|
|
break;
|
|
|
|
|
}
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2003-02-20 20:02:32 +00:00
|
|
|
|
/* unknown chipset, try generic DMA if it seems possible */
|
2003-10-28 19:01:48 +00:00
|
|
|
|
if ((pci_get_class(dev) == PCIC_STORAGE) &&
|
|
|
|
|
(pci_get_subclass(dev) == PCIS_STORAGE_IDE))
|
|
|
|
|
return ata_generic_ident(dev);
|
|
|
|
|
|
2003-02-20 20:02:32 +00:00
|
|
|
|
return ENXIO;
|
2001-03-06 21:43:46 +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
|
|
|
|
int
|
2001-03-06 21:43:46 +00:00
|
|
|
|
ata_pci_attach(device_t dev)
|
|
|
|
|
{
|
2003-03-29 13:37:09 +00:00
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(dev);
|
2004-03-15 12:03:48 +00:00
|
|
|
|
u_int32_t cmd;
|
2004-04-27 12:54:59 +00:00
|
|
|
|
int unit;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2003-03-29 13:37:09 +00:00
|
|
|
|
/* do chipset specific setups only needed once */
|
2004-04-30 18:49:03 +00:00
|
|
|
|
if (ata_legacy(dev) || pci_read_config(dev, PCIR_BAR(2), 4) & IOMASK)
|
2003-03-29 13:37:09 +00:00
|
|
|
|
ctlr->channels = 2;
|
|
|
|
|
else
|
|
|
|
|
ctlr->channels = 1;
|
|
|
|
|
ctlr->allocate = ata_pci_allocate;
|
|
|
|
|
ctlr->dmainit = ata_pci_dmainit;
|
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
|
|
|
|
ctlr->dev = dev;
|
2003-03-29 13:37:09 +00:00
|
|
|
|
|
2004-03-15 12:03:48 +00:00
|
|
|
|
/* if needed try to enable busmastering */
|
|
|
|
|
cmd = pci_read_config(dev, PCIR_COMMAND, 2);
|
2002-04-05 13:13:56 +00:00
|
|
|
|
if (!(cmd & PCIM_CMD_BUSMASTEREN)) {
|
|
|
|
|
pci_write_config(dev, PCIR_COMMAND, cmd | PCIM_CMD_BUSMASTEREN, 2);
|
|
|
|
|
cmd = pci_read_config(dev, PCIR_COMMAND, 2);
|
|
|
|
|
}
|
2003-03-29 13:37:09 +00:00
|
|
|
|
|
2004-03-15 12:03:48 +00:00
|
|
|
|
/* if busmastering mode "stuck" use it */
|
|
|
|
|
if ((cmd & PCIM_CMD_BUSMASTEREN) == PCIM_CMD_BUSMASTEREN) {
|
|
|
|
|
ctlr->r_type1 = SYS_RES_IOPORT;
|
|
|
|
|
ctlr->r_rid1 = ATA_BMADDR_RID;
|
2004-03-17 17:50:55 +00:00
|
|
|
|
ctlr->r_res1 = bus_alloc_resource_any(dev, ctlr->r_type1, &ctlr->r_rid1,
|
|
|
|
|
RF_ACTIVE);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
|
2003-06-07 15:19:16 +00:00
|
|
|
|
ctlr->chipinit(dev);
|
|
|
|
|
|
2003-03-29 13:37:09 +00:00
|
|
|
|
/* attach all channels on this controller */
|
2004-04-30 18:49:03 +00:00
|
|
|
|
for (unit = 0; unit < ctlr->channels; unit++) {
|
|
|
|
|
if (unit == 0 && (pci_get_progif(dev) & 0x81) == 0x80) {
|
|
|
|
|
device_add_child(dev, "ata", unit);
|
|
|
|
|
continue;
|
|
|
|
|
}
|
|
|
|
|
if (unit == 1 && (pci_get_progif(dev) & 0x84) == 0x80) {
|
|
|
|
|
device_add_child(dev, "ata", unit);
|
|
|
|
|
continue;
|
|
|
|
|
}
|
|
|
|
|
device_add_child(dev, "ata", devclass_find_free_unit(ata_devclass, 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
|
|
|
|
bus_generic_attach(dev);
|
|
|
|
|
return 0;
|
2004-04-13 09:44:20 +00:00
|
|
|
|
}
|
2001-03-06 21:43:46 +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
|
|
|
|
int
|
2004-03-15 12:03:48 +00:00
|
|
|
|
ata_pci_detach(device_t dev)
|
|
|
|
|
{
|
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(dev);
|
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
|
|
|
|
device_t *children;
|
|
|
|
|
int nchildren, i;
|
2004-03-15 12:03:48 +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
|
|
|
|
/* detach & delete all children */
|
|
|
|
|
if (!device_get_children(dev, &children, &nchildren)) {
|
|
|
|
|
for (i = 0; i < nchildren; i++)
|
|
|
|
|
device_delete_child(dev, children[i]);
|
|
|
|
|
free(children, M_TEMP);
|
2004-03-15 12:03:48 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (ctlr->r_irq) {
|
|
|
|
|
bus_teardown_intr(dev, ctlr->r_irq, ctlr->handle);
|
|
|
|
|
bus_release_resource(dev, SYS_RES_IRQ, ATA_IRQ_RID, ctlr->r_irq);
|
|
|
|
|
}
|
|
|
|
|
if (ctlr->r_res2)
|
|
|
|
|
bus_release_resource(dev, ctlr->r_type2, ctlr->r_rid2, ctlr->r_res2);
|
|
|
|
|
if (ctlr->r_res1)
|
|
|
|
|
bus_release_resource(dev, ctlr->r_type1, ctlr->r_rid1, ctlr->r_res1);
|
|
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
|
}
|
2001-03-06 21:43:46 +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
|
|
|
|
struct resource *
|
2001-03-06 21:43:46 +00:00
|
|
|
|
ata_pci_alloc_resource(device_t dev, device_t child, int type, int *rid,
|
|
|
|
|
u_long start, u_long end, u_long count, u_int flags)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_pci_controller *controller = device_get_softc(dev);
|
|
|
|
|
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
struct resource *res = NULL;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
int myrid;
|
|
|
|
|
|
|
|
|
|
if (type == SYS_RES_IOPORT) {
|
|
|
|
|
switch (*rid) {
|
|
|
|
|
case ATA_IOADDR_RID:
|
2004-04-21 20:03:26 +00:00
|
|
|
|
if (ata_legacy(dev)) {
|
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
|
|
|
|
start = (unit ? ATA_SECONDARY : ATA_PRIMARY);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
count = ATA_IOSIZE;
|
2004-04-13 09:44:20 +00:00
|
|
|
|
end = start + count - 1;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
2004-04-30 18:49:03 +00:00
|
|
|
|
myrid = PCIR_BAR(0) + (unit << 3);
|
2004-04-13 09:44:20 +00:00
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
|
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
|
|
|
|
start, end, count, flags);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
|
2005-04-06 10:22:56 +00:00
|
|
|
|
case ATA_CTLADDR_RID:
|
2004-04-21 20:03:26 +00:00
|
|
|
|
if (ata_legacy(dev)) {
|
2005-04-06 10:22:56 +00:00
|
|
|
|
start = (unit ? ATA_SECONDARY : ATA_PRIMARY) + ATA_CTLOFFSET;
|
|
|
|
|
count = ATA_CTLIOSIZE;
|
2004-04-13 09:44:20 +00:00
|
|
|
|
end = start + count - 1;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
2004-04-30 18:49:03 +00:00
|
|
|
|
myrid = PCIR_BAR(1) + (unit << 3);
|
2004-04-13 09:44:20 +00:00
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
|
2004-06-15 11:02:09 +00:00
|
|
|
|
SYS_RES_IOPORT, &myrid,
|
2004-04-13 09:44:20 +00:00
|
|
|
|
start, end, count, flags);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
if (type == SYS_RES_IRQ && *rid == ATA_IRQ_RID) {
|
2004-04-21 20:03:26 +00:00
|
|
|
|
if (ata_legacy(dev)) {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#ifdef __alpha__
|
2005-04-05 14:51:43 +00:00
|
|
|
|
res = alpha_platform_alloc_ide_intr(unit);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#else
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int irq = (unit == 0 ? 14 : 15);
|
2003-02-20 20:02:32 +00:00
|
|
|
|
|
2005-04-05 14:51:43 +00:00
|
|
|
|
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IRQ, rid, irq, irq, 1, flags);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#endif
|
|
|
|
|
}
|
2005-04-05 14:51:43 +00:00
|
|
|
|
else
|
|
|
|
|
res = controller->r_irq;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
2005-04-05 14:51:43 +00:00
|
|
|
|
|
|
|
|
|
return res;
|
2001-03-06 21:43:46 +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
|
|
|
|
int
|
2001-03-06 21:43:46 +00:00
|
|
|
|
ata_pci_release_resource(device_t dev, device_t child, int type, int rid,
|
|
|
|
|
struct resource *r)
|
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
if (type == SYS_RES_IOPORT) {
|
|
|
|
|
switch (rid) {
|
|
|
|
|
case ATA_IOADDR_RID:
|
2004-04-13 09:44:20 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
2004-04-30 18:49:03 +00:00
|
|
|
|
SYS_RES_IOPORT,
|
|
|
|
|
PCIR_BAR(0) + (unit << 3), r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
|
2005-04-06 10:22:56 +00:00
|
|
|
|
case ATA_CTLADDR_RID:
|
2004-04-13 09:44:20 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
|
2004-04-30 18:49:03 +00:00
|
|
|
|
SYS_RES_IOPORT,
|
|
|
|
|
PCIR_BAR(1) + (unit << 3), r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
break;
|
|
|
|
|
default:
|
|
|
|
|
return ENOENT;
|
|
|
|
|
}
|
|
|
|
|
}
|
|
|
|
|
if (type == SYS_RES_IRQ) {
|
|
|
|
|
if (rid != ATA_IRQ_RID)
|
|
|
|
|
return ENOENT;
|
|
|
|
|
|
2004-04-21 20:03:26 +00:00
|
|
|
|
if (ata_legacy(dev)) {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#ifdef __alpha__
|
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
|
|
|
|
return alpha_platform_release_ide_intr(unit, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#else
|
2001-03-19 08:02:42 +00:00
|
|
|
|
return BUS_RELEASE_RESOURCE(device_get_parent(dev), child,
|
|
|
|
|
SYS_RES_IRQ, rid, r);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#endif
|
2004-06-15 11:02:09 +00:00
|
|
|
|
}
|
2003-02-20 20:02:32 +00:00
|
|
|
|
else
|
|
|
|
|
return 0;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
return EINVAL;
|
|
|
|
|
}
|
|
|
|
|
|
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
|
2001-03-06 21:43:46 +00:00
|
|
|
|
ata_pci_setup_intr(device_t dev, device_t child, struct resource *irq,
|
2003-02-20 20:02:32 +00:00
|
|
|
|
int flags, driver_intr_t *function, void *argument,
|
2001-03-06 21:43:46 +00:00
|
|
|
|
void **cookiep)
|
|
|
|
|
{
|
2004-04-21 20:03:26 +00:00
|
|
|
|
if (ata_legacy(dev)) {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#ifdef __alpha__
|
2003-02-20 20:02:32 +00:00
|
|
|
|
return alpha_platform_setup_ide_intr(child, irq, function, argument,
|
|
|
|
|
cookiep);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#else
|
|
|
|
|
return BUS_SETUP_INTR(device_get_parent(dev), child, irq,
|
2003-02-20 20:02:32 +00:00
|
|
|
|
flags, function, argument, cookiep);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#endif
|
|
|
|
|
}
|
2003-02-20 20:02:32 +00:00
|
|
|
|
else {
|
2003-02-25 14:46:30 +00:00
|
|
|
|
struct ata_pci_controller *controller = device_get_softc(dev);
|
|
|
|
|
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
|
2003-02-25 14:46:30 +00:00
|
|
|
|
controller->interrupt[unit].function = function;
|
|
|
|
|
controller->interrupt[unit].argument = argument;
|
|
|
|
|
*cookiep = controller;
|
|
|
|
|
return 0;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
}
|
2001-03-06 21:43:46 +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
|
|
|
|
int
|
2001-03-06 21:43:46 +00:00
|
|
|
|
ata_pci_teardown_intr(device_t dev, device_t child, struct resource *irq,
|
|
|
|
|
void *cookie)
|
|
|
|
|
{
|
2004-04-21 20:03:26 +00:00
|
|
|
|
if (ata_legacy(dev)) {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
#ifdef __alpha__
|
|
|
|
|
return alpha_platform_teardown_ide_intr(child, irq, cookie);
|
|
|
|
|
#else
|
|
|
|
|
return BUS_TEARDOWN_INTR(device_get_parent(dev), child, irq, cookie);
|
|
|
|
|
#endif
|
|
|
|
|
}
|
2003-02-25 14:46:30 +00:00
|
|
|
|
else {
|
|
|
|
|
struct ata_pci_controller *controller = device_get_softc(dev);
|
|
|
|
|
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
|
|
|
|
|
|
|
|
|
|
controller->interrupt[unit].function = NULL;
|
|
|
|
|
controller->interrupt[unit].argument = NULL;
|
2003-02-20 20:02:32 +00:00
|
|
|
|
return 0;
|
2003-02-25 14:46:30 +00:00
|
|
|
|
}
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
2003-03-29 13:37:09 +00:00
|
|
|
|
|
2005-04-08 09:37:47 +00:00
|
|
|
|
int
|
2005-04-05 14:51:43 +00:00
|
|
|
|
ata_pci_allocate(device_t dev)
|
2003-03-29 13:37:09 +00:00
|
|
|
|
{
|
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
|
2005-04-05 14:51:43 +00:00
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
2005-04-06 10:22:56 +00:00
|
|
|
|
struct resource *io = NULL, *ctlio = NULL;
|
2003-03-29 13:37:09 +00:00
|
|
|
|
int i, rid;
|
|
|
|
|
|
|
|
|
|
rid = ATA_IOADDR_RID;
|
2005-03-31 15:05:40 +00:00
|
|
|
|
io = bus_alloc_resource_any(dev, SYS_RES_IOPORT, &rid, RF_ACTIVE);
|
2003-03-29 13:37:09 +00:00
|
|
|
|
if (!io)
|
|
|
|
|
return ENXIO;
|
|
|
|
|
|
2005-04-06 10:22:56 +00:00
|
|
|
|
rid = ATA_CTLADDR_RID;
|
|
|
|
|
ctlio = bus_alloc_resource_any(dev, SYS_RES_IOPORT, &rid, RF_ACTIVE);
|
|
|
|
|
if (!ctlio) {
|
2003-03-29 13:37:09 +00:00
|
|
|
|
bus_release_resource(dev, SYS_RES_IOPORT, ATA_IOADDR_RID, io);
|
|
|
|
|
return ENXIO;
|
|
|
|
|
}
|
|
|
|
|
|
2005-04-06 10:22:56 +00:00
|
|
|
|
for (i = ATA_DATA; i <= ATA_COMMAND; i ++) {
|
2003-03-29 13:37:09 +00:00
|
|
|
|
ch->r_io[i].res = io;
|
|
|
|
|
ch->r_io[i].offset = i;
|
|
|
|
|
}
|
2005-04-06 10:22:56 +00:00
|
|
|
|
ch->r_io[ATA_CONTROL].res = ctlio;
|
|
|
|
|
ch->r_io[ATA_CONTROL].offset = ata_legacy(device_get_parent(dev)) ? 0 : 2;
|
2003-04-07 14:12:12 +00:00
|
|
|
|
ch->r_io[ATA_IDX_ADDR].res = io;
|
2005-04-06 10:22:56 +00:00
|
|
|
|
ata_default_registers(ch);
|
2004-03-15 12:03:48 +00:00
|
|
|
|
if (ctlr->r_res1) {
|
2003-03-29 13:37:09 +00:00
|
|
|
|
for (i = ATA_BMCMD_PORT; i <= ATA_BMDTP_PORT; i++) {
|
2004-03-15 12:03:48 +00:00
|
|
|
|
ch->r_io[i].res = ctlr->r_res1;
|
2005-04-06 10:22:56 +00:00
|
|
|
|
ch->r_io[i].offset = (i - ATA_BMCMD_PORT) + (ch->unit*ATA_BMIOSIZE);
|
2003-03-29 13:37:09 +00:00
|
|
|
|
}
|
|
|
|
|
}
|
2004-04-13 09:44:20 +00:00
|
|
|
|
|
|
|
|
|
ata_generic_hw(ch);
|
|
|
|
|
|
2003-03-29 13:37:09 +00:00
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
2003-10-21 19:20:37 +00:00
|
|
|
|
ata_pci_dmastart(struct ata_channel *ch)
|
2003-03-29 13:37:09 +00:00
|
|
|
|
{
|
|
|
|
|
ATA_IDX_OUTB(ch, ATA_BMSTAT_PORT, (ATA_IDX_INB(ch, ATA_BMSTAT_PORT) |
|
2003-04-07 14:12:12 +00:00
|
|
|
|
(ATA_BMSTAT_INTERRUPT | ATA_BMSTAT_ERROR)));
|
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_IDX_OUTL(ch, ATA_BMDTP_PORT, ch->dma->sg_bus);
|
2004-09-26 11:42:42 +00:00
|
|
|
|
ch->dma->flags |= ATA_DMA_ACTIVE;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
ATA_IDX_OUTB(ch, ATA_BMCMD_PORT,
|
2004-05-10 20:23:25 +00:00
|
|
|
|
(ATA_IDX_INB(ch, ATA_BMCMD_PORT) & ~ATA_BMCMD_WRITE_READ) |
|
2003-10-21 19:20:37 +00:00
|
|
|
|
((ch->dma->flags & ATA_DMA_READ) ? ATA_BMCMD_WRITE_READ : 0) |
|
|
|
|
|
ATA_BMCMD_START_STOP);
|
2003-03-29 13:37:09 +00:00
|
|
|
|
return 0;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
|
|
|
|
ata_pci_dmastop(struct ata_channel *ch)
|
|
|
|
|
{
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
ATA_IDX_OUTB(ch, ATA_BMCMD_PORT,
|
2003-04-07 14:12:12 +00:00
|
|
|
|
ATA_IDX_INB(ch, ATA_BMCMD_PORT) & ~ATA_BMCMD_START_STOP);
|
2004-08-05 21:13:41 +00:00
|
|
|
|
ch->dma->flags &= ~ATA_DMA_ACTIVE;
|
2004-09-26 11:42:42 +00:00
|
|
|
|
error = ATA_IDX_INB(ch, ATA_BMSTAT_PORT) & ATA_BMSTAT_MASK;
|
2003-03-29 13:37:09 +00:00
|
|
|
|
ATA_IDX_OUTB(ch, ATA_BMSTAT_PORT, ATA_BMSTAT_INTERRUPT | ATA_BMSTAT_ERROR);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
return error;
|
2003-03-29 13:37:09 +00:00
|
|
|
|
}
|
|
|
|
|
|
2003-08-25 11:13:04 +00:00
|
|
|
|
static void
|
2003-03-29 13:37:09 +00:00
|
|
|
|
ata_pci_dmainit(struct ata_channel *ch)
|
|
|
|
|
{
|
2003-08-25 11:13:04 +00:00
|
|
|
|
ata_dmainit(ch);
|
|
|
|
|
if (ch->dma) {
|
|
|
|
|
ch->dma->start = ata_pci_dmastart;
|
|
|
|
|
ch->dma->stop = ata_pci_dmastop;
|
|
|
|
|
}
|
2003-03-29 13:37:09 +00:00
|
|
|
|
}
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
static device_method_t ata_pci_methods[] = {
|
|
|
|
|
/* device interface */
|
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
|
|
|
|
DEVMETHOD(device_probe, ata_pci_probe),
|
|
|
|
|
DEVMETHOD(device_attach, ata_pci_attach),
|
|
|
|
|
DEVMETHOD(device_detach, ata_pci_detach),
|
|
|
|
|
DEVMETHOD(device_shutdown, bus_generic_shutdown),
|
|
|
|
|
DEVMETHOD(device_suspend, bus_generic_suspend),
|
|
|
|
|
DEVMETHOD(device_resume, bus_generic_resume),
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
/* bus methods */
|
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
|
|
|
|
DEVMETHOD(bus_alloc_resource, ata_pci_alloc_resource),
|
|
|
|
|
DEVMETHOD(bus_release_resource, ata_pci_release_resource),
|
|
|
|
|
DEVMETHOD(bus_activate_resource, bus_generic_activate_resource),
|
|
|
|
|
DEVMETHOD(bus_deactivate_resource, bus_generic_deactivate_resource),
|
|
|
|
|
DEVMETHOD(bus_setup_intr, ata_pci_setup_intr),
|
|
|
|
|
DEVMETHOD(bus_teardown_intr, ata_pci_teardown_intr),
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
{ 0, 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
|
|
|
|
devclass_t atapci_devclass;
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
static driver_t ata_pci_driver = {
|
|
|
|
|
"atapci",
|
|
|
|
|
ata_pci_methods,
|
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
|
|
|
|
sizeof(struct ata_pci_controller),
|
2001-03-06 21:43:46 +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
|
|
|
|
DRIVER_MODULE(atapci, pci, ata_pci_driver, atapci_devclass, 0, 0);
|
|
|
|
|
MODULE_VERSION(atapci, 1);
|
|
|
|
|
MODULE_DEPEND(atapci, ata, 1, 1, 1);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
|
|
|
|
static int
|
2005-03-31 15:05:40 +00:00
|
|
|
|
ata_pcichannel_probe(device_t dev)
|
2001-03-06 21:43:46 +00:00
|
|
|
|
{
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
2001-03-06 21:43:46 +00:00
|
|
|
|
device_t *children;
|
2004-08-12 08:20:36 +00:00
|
|
|
|
int count, i;
|
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 buffer[32];
|
2001-03-06 21:43:46 +00:00
|
|
|
|
|
2004-03-15 12:03:48 +00:00
|
|
|
|
/* take care of green memory */
|
|
|
|
|
bzero(ch, sizeof(struct ata_channel));
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* find channel number on this controller */
|
|
|
|
|
device_get_children(device_get_parent(dev), &children, &count);
|
|
|
|
|
for (i = 0; i < count; i++) {
|
|
|
|
|
if (children[i] == dev)
|
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
|
|
|
|
ch->unit = i;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
free(children, M_TEMP);
|
2002-12-03 20:20:44 +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
|
|
|
|
sprintf(buffer, "ATA channel %d", ch->unit);
|
|
|
|
|
device_set_desc_copy(dev, buffer);
|
|
|
|
|
|
2004-08-12 08:20:36 +00:00
|
|
|
|
return ata_probe(dev);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
2005-03-31 15:05:40 +00:00
|
|
|
|
ata_pcichannel_attach(device_t dev)
|
2004-08-12 08:20:36 +00:00
|
|
|
|
{
|
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
|
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
|
|
|
|
int error;
|
|
|
|
|
|
2004-08-20 06:19:25 +00:00
|
|
|
|
if (ctlr->r_res1)
|
|
|
|
|
ctlr->dmainit(ch);
|
2004-08-12 08:20:36 +00:00
|
|
|
|
if (ch->dma)
|
|
|
|
|
ch->dma->alloc(ch);
|
|
|
|
|
|
2005-04-05 14:51:43 +00:00
|
|
|
|
if ((error = ctlr->allocate(dev)))
|
2004-06-15 11:02:09 +00:00
|
|
|
|
return error;
|
|
|
|
|
|
2004-08-12 08:20:36 +00:00
|
|
|
|
return ata_attach(dev);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static int
|
2005-03-31 15:05:40 +00:00
|
|
|
|
ata_pcichannel_detach(device_t dev)
|
2004-08-12 08:20:36 +00:00
|
|
|
|
{
|
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
|
|
|
|
int error;
|
|
|
|
|
|
|
|
|
|
if ((error = ata_detach(dev)))
|
|
|
|
|
return error;
|
|
|
|
|
|
|
|
|
|
if (ch->dma)
|
|
|
|
|
ch->dma->free(ch);
|
|
|
|
|
|
2005-04-06 10:22:56 +00:00
|
|
|
|
/* free resources for io and ctlio XXX SOS */
|
2005-04-05 14:51:43 +00:00
|
|
|
|
|
2004-08-12 08:20:36 +00:00
|
|
|
|
return 0;
|
2001-03-06 21:43:46 +00:00
|
|
|
|
}
|
|
|
|
|
|
2005-03-31 15:05:40 +00:00
|
|
|
|
static int
|
|
|
|
|
ata_pcichannel_locking(device_t dev, int mode)
|
|
|
|
|
{
|
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
|
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
|
|
|
|
|
|
|
|
|
if (ctlr->locking)
|
|
|
|
|
return ctlr->locking(ch, mode);
|
|
|
|
|
else
|
|
|
|
|
return ch->unit;
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
ata_pcichannel_reset(device_t dev)
|
|
|
|
|
{
|
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
|
|
|
|
|
struct ata_channel *ch = device_get_softc(dev);
|
|
|
|
|
|
2005-04-08 09:37:47 +00:00
|
|
|
|
/* if DMA functionality present stop it */
|
|
|
|
|
if (ch->dma) {
|
|
|
|
|
if (ch->dma->stop)
|
|
|
|
|
ch->dma->stop(ch);
|
|
|
|
|
if (ch->dma->flags & ATA_DMA_LOADED)
|
|
|
|
|
ch->dma->unload(ch);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
/* reset the controller HW */
|
2005-03-31 15:05:40 +00:00
|
|
|
|
if (ctlr->reset)
|
|
|
|
|
ctlr->reset(ch);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static void
|
|
|
|
|
ata_pcichannel_setmode(device_t parent, device_t dev)
|
|
|
|
|
{
|
|
|
|
|
struct ata_pci_controller *ctlr = device_get_softc(GRANDPARENT(dev));
|
|
|
|
|
struct ata_device *atadev = device_get_softc(dev);
|
|
|
|
|
int mode = atadev->mode;
|
|
|
|
|
|
|
|
|
|
ctlr->setmode(atadev, ATA_PIO_MAX);
|
|
|
|
|
if (mode >= ATA_DMA)
|
|
|
|
|
ctlr->setmode(atadev, mode);
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
static device_method_t ata_pcichannel_methods[] = {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
/* device interface */
|
2005-03-31 15:05:40 +00:00
|
|
|
|
DEVMETHOD(device_probe, ata_pcichannel_probe),
|
|
|
|
|
DEVMETHOD(device_attach, ata_pcichannel_attach),
|
|
|
|
|
DEVMETHOD(device_detach, ata_pcichannel_detach),
|
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
|
|
|
|
DEVMETHOD(device_shutdown, bus_generic_shutdown),
|
|
|
|
|
DEVMETHOD(device_suspend, bus_generic_suspend),
|
|
|
|
|
DEVMETHOD(device_resume, bus_generic_resume),
|
2005-03-31 15:05:40 +00:00
|
|
|
|
|
|
|
|
|
/* ATA methods */
|
|
|
|
|
DEVMETHOD(ata_setmode, ata_pcichannel_setmode),
|
|
|
|
|
DEVMETHOD(ata_locking, ata_pcichannel_locking),
|
|
|
|
|
DEVMETHOD(ata_reset, ata_pcichannel_reset),
|
|
|
|
|
|
2001-03-06 21:43:46 +00:00
|
|
|
|
{ 0, 0 }
|
|
|
|
|
};
|
|
|
|
|
|
2005-03-31 15:05:40 +00:00
|
|
|
|
driver_t ata_pcichannel_driver = {
|
2001-03-06 21:43:46 +00:00
|
|
|
|
"ata",
|
2005-03-31 15:05:40 +00:00
|
|
|
|
ata_pcichannel_methods,
|
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
|
|
|
|
sizeof(struct ata_channel),
|
2001-03-06 21:43:46 +00:00
|
|
|
|
};
|
|
|
|
|
|
2005-03-31 15:05:40 +00:00
|
|
|
|
DRIVER_MODULE(ata, atapci, ata_pcichannel_driver, ata_devclass, 0, 0);
|