Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
/*-
|
2003-02-20 20:02:32 +00:00
|
|
|
|
* Copyright (c) 1998 - 2003 S<EFBFBD>ren Schmidt <sos@FreeBSD.org>
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +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.
|
|
|
|
|
|
*
|
1999-08-28 01:08:13 +00:00
|
|
|
|
* $FreeBSD$
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
*/
|
|
|
|
|
|
|
1999-10-09 19:57:13 +00:00
|
|
|
|
#include "opt_ata.h"
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
#include <sys/param.h>
|
|
|
|
|
|
#include <sys/systm.h>
|
2001-03-15 15:36:25 +00:00
|
|
|
|
#include <sys/ata.h>
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
#include <sys/kernel.h>
|
|
|
|
|
|
#include <sys/malloc.h>
|
2000-05-05 09:59:14 +00:00
|
|
|
|
#include <sys/bio.h>
|
1999-12-14 10:25:28 +00:00
|
|
|
|
#include <sys/bus.h>
|
1999-09-05 15:41:09 +00:00
|
|
|
|
#include <sys/conf.h>
|
1999-09-03 10:57:17 +00:00
|
|
|
|
#include <sys/disk.h>
|
1999-11-27 17:06:10 +00:00
|
|
|
|
#include <sys/cons.h>
|
2001-03-19 08:04:54 +00:00
|
|
|
|
#include <sys/sysctl.h>
|
2003-08-24 09:22:26 +00:00
|
|
|
|
#include <sys/taskqueue.h>
|
Nine'th update to the new ATA/ATAPI driver:
The atapi subsystem has gotten better error handeling and timeouts,
it also tries a REQUEST SENSE command when devices returns errors,
to give a little more info as to what went wrong. It might be a
little verbose for now, but I'm interested in as much feedback on
errors as possible, especially timeouts, as I'm a bit in doubt if
I've chosen resonable default values everywhere.
The disk driver has been changed a bit to prepare for tagged queing,
which is next on my list.
The disk driver has grown a dump routine, I got one implementation
from Darrell Anderson <anderson@cs.duke.edu> which also did
partial dumps (usefull on big memory machines) I left out the
partial stuff for now, and changed the rest alot to fit into the new
ad_request framework.
Some minor cleanups and rearrangements as well.
As usual USE AT YOUR OWN RISK!!, this is still pre alpha level code.
Especially the DMA support can hose your disk real bad if anything
goes wrong, again you have been warned :)
Notebook owners should be carefull that their machines dont suspend
as this might cause trouble...
But please tell me how it works for you!
Enjoy!
-Søren
1999-06-25 09:03:07 +00:00
|
|
|
|
#include <vm/vm.h>
|
|
|
|
|
|
#include <vm/pmap.h>
|
|
|
|
|
|
#include <machine/md_var.h>
|
2000-08-28 21:48:13 +00:00
|
|
|
|
#include <machine/bus.h>
|
2001-02-06 16:44:25 +00:00
|
|
|
|
#include <sys/rman.h>
|
2003-04-01 15:06:26 +00:00
|
|
|
|
#include <geom/geom_disk.h>
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
#include <dev/ata/ata-all.h>
|
2003-02-20 20:02:32 +00:00
|
|
|
|
#include <dev/ata/ata-pci.h>
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
#include <dev/ata/ata-disk.h>
|
2000-10-13 13:04:45 +00:00
|
|
|
|
#include <dev/ata/ata-raid.h>
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
|
|
|
|
|
/* prototypes */
|
2003-08-24 09:22:26 +00:00
|
|
|
|
static void ad_detach(struct ata_device *atadev);
|
|
|
|
|
|
static void ad_start(struct ata_device *atadev);
|
|
|
|
|
|
static void ad_done(struct ata_request *request);
|
2003-02-25 15:33:36 +00:00
|
|
|
|
static disk_open_t adopen;
|
|
|
|
|
|
static disk_strategy_t adstrategy;
|
|
|
|
|
|
static dumper_t addump;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
void ad_print(struct ad_softc *adp);
|
|
|
|
|
|
static int ad_version(u_int16_t version);
|
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
|
|
|
|
|
Nine'th update to the new ATA/ATAPI driver:
The atapi subsystem has gotten better error handeling and timeouts,
it also tries a REQUEST SENSE command when devices returns errors,
to give a little more info as to what went wrong. It might be a
little verbose for now, but I'm interested in as much feedback on
errors as possible, especially timeouts, as I'm a bit in doubt if
I've chosen resonable default values everywhere.
The disk driver has been changed a bit to prepare for tagged queing,
which is next on my list.
The disk driver has grown a dump routine, I got one implementation
from Darrell Anderson <anderson@cs.duke.edu> which also did
partial dumps (usefull on big memory machines) I left out the
partial stuff for now, and changed the rest alot to fit into the new
ad_request framework.
Some minor cleanups and rearrangements as well.
As usual USE AT YOUR OWN RISK!!, this is still pre alpha level code.
Especially the DMA support can hose your disk real bad if anything
goes wrong, again you have been warned :)
Notebook owners should be carefull that their machines dont suspend
as this might cause trouble...
But please tell me how it works for you!
Enjoy!
-Søren
1999-06-25 09:03:07 +00:00
|
|
|
|
/* internal vars */
|
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
|
|
|
|
static MALLOC_DEFINE(M_AD, "AD driver", "ATA disk driver");
|
2003-08-24 09:22:26 +00:00
|
|
|
|
static u_int32_t adp_lun_map = 0;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
2001-03-19 08:04:54 +00:00
|
|
|
|
/* sysctl vars */
|
|
|
|
|
|
SYSCTL_DECL(_hw_ata);
|
|
|
|
|
|
SYSCTL_INT(_hw_ata, OID_AUTO, ata_dma, CTLFLAG_RD, &ata_dma, 0,
|
|
|
|
|
|
"ATA disk DMA mode control");
|
2001-04-05 11:13:07 +00:00
|
|
|
|
SYSCTL_INT(_hw_ata, OID_AUTO, wc, CTLFLAG_RD, &ata_wc, 0,
|
2001-03-19 08:04:54 +00:00
|
|
|
|
"ATA disk write caching");
|
|
|
|
|
|
|
2000-01-18 21:02:59 +00:00
|
|
|
|
void
|
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
|
|
|
|
ad_attach(struct ata_device *atadev)
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
{
|
|
|
|
|
|
struct ad_softc *adp;
|
2002-04-05 13:13:56 +00:00
|
|
|
|
u_int32_t lbasize;
|
|
|
|
|
|
u_int64_t lbasize48;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
2000-12-26 12:05:49 +00:00
|
|
|
|
if (!(adp = malloc(sizeof(struct ad_softc), M_AD, M_NOWAIT | M_ZERO))) {
|
2003-08-24 09:22:26 +00:00
|
|
|
|
ata_prtdev(atadev, "FAILURE - could not allocate driver storage\n");
|
|
|
|
|
|
atadev->attach = NULL;
|
2000-01-18 21:02:59 +00:00
|
|
|
|
return;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +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
|
|
|
|
adp->device = atadev;
|
2000-02-18 20:57:33 +00:00
|
|
|
|
#ifdef ATA_STATIC_ID
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
adp->lun = (device_get_unit(atadev->channel->dev)<<1)+ATA_DEV(atadev->unit);
|
2000-02-18 20:57:33 +00:00
|
|
|
|
#else
|
|
|
|
|
|
adp->lun = ata_get_lun(&adp_lun_map);
|
|
|
|
|
|
#endif
|
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
|
|
|
|
ata_set_name(atadev, "ad", adp->lun);
|
|
|
|
|
|
adp->heads = atadev->param->heads;
|
|
|
|
|
|
adp->sectors = atadev->param->sectors;
|
|
|
|
|
|
adp->total_secs = atadev->param->cylinders * adp->heads * adp->sectors;
|
2002-12-17 16:26:22 +00:00
|
|
|
|
if (adp->device->channel->flags & ATA_USE_PC98GEOM &&
|
|
|
|
|
|
adp->total_secs < 17 * 8 * 65536) {
|
|
|
|
|
|
adp->sectors = 17;
|
|
|
|
|
|
adp->heads = 8;
|
|
|
|
|
|
}
|
2003-08-24 09:22:26 +00:00
|
|
|
|
mtx_init(&adp->queue_mtx, "ATA disk bioqueue lock", MTX_DEF, 0);
|
2002-03-11 21:04:32 +00:00
|
|
|
|
bioq_init(&adp->queue);
|
2000-01-18 21:02:59 +00:00
|
|
|
|
|
2002-04-05 18:04:05 +00:00
|
|
|
|
lbasize = (u_int32_t)atadev->param->lba_size_1 |
|
|
|
|
|
|
((u_int32_t)atadev->param->lba_size_2 << 16);
|
2002-04-05 13:13:56 +00:00
|
|
|
|
|
2001-10-06 11:07:04 +00:00
|
|
|
|
/* does this device need oldstyle CHS addressing */
|
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
|
|
|
|
if (!ad_version(atadev->param->version_major) ||
|
2002-04-05 13:13:56 +00:00
|
|
|
|
!(atadev->param->atavalid & ATA_FLAG_54_58) || !lbasize)
|
2003-08-24 09:22:26 +00:00
|
|
|
|
atadev->flags |= ATA_D_USE_CHS;
|
2001-10-06 11:07:04 +00:00
|
|
|
|
|
2003-05-01 06:17:30 +00:00
|
|
|
|
/* use the 28bit LBA size if valid or bigger than the CHS mapping */
|
|
|
|
|
|
if (atadev->param->cylinders == 16383 || adp->total_secs < lbasize)
|
2002-04-05 13:13:56 +00:00
|
|
|
|
adp->total_secs = lbasize;
|
|
|
|
|
|
|
2002-04-05 18:04:05 +00:00
|
|
|
|
lbasize48 = ((u_int64_t)atadev->param->lba_size48_1) |
|
|
|
|
|
|
((u_int64_t)atadev->param->lba_size48_2 << 16) |
|
|
|
|
|
|
((u_int64_t)atadev->param->lba_size48_3 << 32) |
|
|
|
|
|
|
((u_int64_t)atadev->param->lba_size48_4 << 48);
|
2001-10-06 11:07:04 +00:00
|
|
|
|
|
|
|
|
|
|
/* use the 48bit LBA size if valid */
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if ((atadev->param->support.command2 & ATA_SUPPORT_ADDRESS48) &&
|
|
|
|
|
|
lbasize48 > 268435455)
|
2002-04-05 13:13:56 +00:00
|
|
|
|
adp->total_secs = lbasize48;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
|
|
|
|
|
|
/* enable read caching */
|
|
|
|
|
|
ata_controlcmd(atadev, ATA_SETFEATURES, ATA_SF_ENAB_RCACHE, 0, 0);
|
|
|
|
|
|
|
|
|
|
|
|
/* enable write caching if enabled */
|
|
|
|
|
|
if (ata_wc)
|
|
|
|
|
|
ata_controlcmd(atadev, ATA_SETFEATURES, ATA_SF_ENAB_WCACHE, 0, 0);
|
|
|
|
|
|
else
|
|
|
|
|
|
ata_controlcmd(atadev, ATA_SETFEATURES, ATA_SF_DIS_WCACHE, 0, 0);
|
2002-07-22 18:35:01 +00:00
|
|
|
|
|
2000-01-18 21:02:59 +00:00
|
|
|
|
/* use multiple sectors/interrupt if device supports it */
|
2003-08-24 09:22:26 +00:00
|
|
|
|
adp->max_iosize = DEV_BSIZE;
|
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
|
|
|
|
if (ad_version(atadev->param->version_major)) {
|
|
|
|
|
|
int secsperint = max(1, min(atadev->param->sectors_intr, 16));
|
2001-10-06 11:07:04 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if (!ata_controlcmd(atadev, ATA_SET_MULTI, 0, 0, secsperint))
|
|
|
|
|
|
adp->max_iosize = secsperint * DEV_BSIZE;
|
2001-03-15 15:36:25 +00:00
|
|
|
|
}
|
2000-01-18 21:02:59 +00:00
|
|
|
|
|
2001-03-15 15:36:25 +00:00
|
|
|
|
/* use DMA if allowed and if drive/controller supports it */
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if (ata_dma && atadev->channel->dma)
|
2003-02-20 20:02:32 +00:00
|
|
|
|
atadev->setmode(atadev, ATA_DMA_MAX);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
else
|
2003-02-20 20:02:32 +00:00
|
|
|
|
atadev->setmode(atadev, ATA_PIO_MAX);
|
2001-03-15 15:36:25 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
/* setup the function ptrs */
|
|
|
|
|
|
atadev->detach = ad_detach;
|
|
|
|
|
|
atadev->start = ad_start;
|
|
|
|
|
|
atadev->softc = adp;
|
2002-03-11 21:04:32 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
/* lets create the disk device */
|
2003-02-25 15:33:36 +00:00
|
|
|
|
adp->disk.d_open = adopen;
|
|
|
|
|
|
adp->disk.d_strategy = adstrategy;
|
|
|
|
|
|
adp->disk.d_dump = addump;
|
|
|
|
|
|
adp->disk.d_name = "ad";
|
|
|
|
|
|
adp->disk.d_drv1 = adp;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if (atadev->channel->dma)
|
|
|
|
|
|
adp->disk.d_maxsize = atadev->channel->dma->max_iosize;
|
|
|
|
|
|
else
|
|
|
|
|
|
adp->disk.d_maxsize = DFLTPHYS;
|
2002-09-20 19:36:05 +00:00
|
|
|
|
adp->disk.d_sectorsize = DEV_BSIZE;
|
|
|
|
|
|
adp->disk.d_mediasize = DEV_BSIZE * (off_t)adp->total_secs;
|
|
|
|
|
|
adp->disk.d_fwsectors = adp->sectors;
|
|
|
|
|
|
adp->disk.d_fwheads = adp->heads;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
disk_create(adp->lun, &adp->disk, DISKFLAG_NOGIANT, NULL, NULL);
|
2001-04-27 19:47:33 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
/* announce we are here */
|
|
|
|
|
|
ad_print(adp);
|
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
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
#ifdef DEV_ATARAID
|
2003-05-02 12:41:44 +00:00
|
|
|
|
ata_raiddisk_attach(adp);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
#endif
|
1999-03-05 09:43:30 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
static void
|
2003-05-02 13:47:44 +00:00
|
|
|
|
ad_detach(struct ata_device *atadev)
|
2000-02-18 20:57:33 +00:00
|
|
|
|
{
|
2003-08-24 09:22:26 +00:00
|
|
|
|
struct ad_softc *adp = atadev->softc;
|
2001-03-14 12:05:44 +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
|
|
|
|
atadev->flags |= ATA_D_DETACHING;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
#ifdef DEV_ATARAID
|
2003-03-29 13:37:09 +00:00
|
|
|
|
if (adp->flags & AD_F_RAID_SUBDISK)
|
|
|
|
|
|
ata_raiddisk_detach(adp);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
#endif
|
|
|
|
|
|
mtx_lock(&adp->queue_mtx);
|
|
|
|
|
|
bioq_flush(&adp->queue, NULL, ENXIO);
|
|
|
|
|
|
mtx_unlock(&adp->queue_mtx);
|
|
|
|
|
|
disk_destroy(&adp->disk);
|
|
|
|
|
|
ata_prtdev(atadev, "WARNING - removed from configuration\n");
|
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
|
|
|
|
ata_free_name(atadev);
|
2000-02-18 20:57:33 +00:00
|
|
|
|
ata_free_lun(&adp_lun_map, adp->lun);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
atadev->attach = NULL;
|
|
|
|
|
|
atadev->detach = NULL;
|
|
|
|
|
|
atadev->start = NULL;
|
|
|
|
|
|
atadev->softc = NULL;
|
2002-02-12 11:35:15 +00:00
|
|
|
|
atadev->flags = 0;
|
2000-02-18 20:57:33 +00:00
|
|
|
|
free(adp, M_AD);
|
|
|
|
|
|
}
|
|
|
|
|
|
|
1999-03-05 09:43:30 +00:00
|
|
|
|
static int
|
2003-02-25 15:33:36 +00:00
|
|
|
|
adopen(struct disk *dp)
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
{
|
2003-02-25 15:33:36 +00:00
|
|
|
|
struct ad_softc *adp = dp->d_drv1;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if (adp->device->flags & ATA_D_DETACHING)
|
|
|
|
|
|
return ENXIO;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
return 0;
|
|
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static void
|
2000-04-15 05:54:02 +00:00
|
|
|
|
adstrategy(struct bio *bp)
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
{
|
2003-02-25 15:33:36 +00:00
|
|
|
|
struct ad_softc *adp = bp->bio_disk->d_drv1;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +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
|
|
|
|
if (adp->device->flags & ATA_D_DETACHING) {
|
2001-05-06 20:00:03 +00:00
|
|
|
|
biofinish(bp, NULL, ENXIO);
|
2001-03-14 12:05:44 +00:00
|
|
|
|
return;
|
|
|
|
|
|
}
|
2003-08-24 09:22:26 +00:00
|
|
|
|
mtx_lock(&adp->queue_mtx);
|
2003-04-01 15:06:26 +00:00
|
|
|
|
bioq_disksort(&adp->queue, bp);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
mtx_unlock(&adp->queue_mtx);
|
2002-03-11 21:04:32 +00:00
|
|
|
|
ata_start(adp->device->channel);
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
static void
|
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
|
|
|
|
ad_start(struct ata_device *atadev)
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
{
|
2003-08-24 09:22:26 +00:00
|
|
|
|
struct ad_softc *adp = atadev->softc;
|
|
|
|
|
|
struct bio *bp;
|
|
|
|
|
|
struct ata_request *request;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
/* remove request from drive queue */
|
|
|
|
|
|
mtx_lock(&adp->queue_mtx);
|
|
|
|
|
|
bp = bioq_first(&adp->queue);
|
|
|
|
|
|
if (!bp) {
|
|
|
|
|
|
mtx_unlock(&adp->queue_mtx);
|
Ten'th update to the new ATA/ATAPI driver:
It been awhile since the last major update, as a benefit there
are some cool things in this one (and new bugs probably :) )...
The ATA driver has grown "real" timeout support for all devices.
This means that it should be possible to get in contact with
(especially) lost ATAPI devices. It also means that the ATA
driver is now usable on notebooks as it will DTRT on resume.
An experimental hack at utilizing the Promise66's at UDMA66 is
in there, but I cant test it. If someone feels like sending
me one, give me a ping.
The ATAPI DMA enableling scheme has been changed, also better DMA
support for the Aladdin chipset has been implemented for ATAPI
devices. Note that the Aladdin apparently only can do DMA reads
on ATAPI devices, and the Promise cant do ATAPI DMA at all.
I have seen problems on some ATAPI devices that should be able
to run in DMA mode, so if you encounter problems with hanging
atapi devices during the probe, or during access, disable DMA
in atapi-all.c, and let me know. It might be nessesary to do this
via a "white list" for known good devices...
The ATAPI CDROM driver can now use eject/close without hanging and
the bug that caused reading beyond the end of a CD has been fixed.
Media change is also handled proberly. DVD drives are identified
and are usable as CDROM devices at least, I dont have the HW to
test this further, see above :).
The ATAPI tape driver has gotten some support for using the DSC
method for not blocking the IDE channel during read/write when
the device has full buffers. It knows about the OnStream DI-30
device, support is not completed yet, but it can function as a
primitive backup medium, without filemarks, and without bad media
handeling. This is because the OnStream device doesn't handle this
(like everybody else) in HW. It also now supports getting/setting
the record position on devices that supports it.
Some rather major cleanups and rearrangements as well (cvs -b diff
is your freind). I'm closing in on declaring this for beta code,
most of the infrastruture is in place by now.
As usual USE AT YOUR OWN RISK!!, this is still alpha level code.
This driver can hose your disk real bad if anything goes wrong, but
now you have been warned :)
But please tell me how it works for you!
Enjoy!
-Søren
1999-09-21 19:50:40 +00:00
|
|
|
|
return;
|
2000-09-19 11:08:39 +00:00
|
|
|
|
}
|
2003-03-27 10:48:21 +00:00
|
|
|
|
bioq_remove(&adp->queue, bp);
|
2003-08-24 09:22:26 +00:00
|
|
|
|
mtx_unlock(&adp->queue_mtx);
|
2003-03-27 10:48:21 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if (!(request = ata_alloc_request())) {
|
|
|
|
|
|
ata_prtdev(atadev, "FAILURE - out of memory in start\n");
|
2003-03-27 10:48:21 +00:00
|
|
|
|
biofinish(bp, NULL, ENOMEM);
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
return;
|
1999-03-07 21:49:14 +00:00
|
|
|
|
}
|
Nine'th update to the new ATA/ATAPI driver:
The atapi subsystem has gotten better error handeling and timeouts,
it also tries a REQUEST SENSE command when devices returns errors,
to give a little more info as to what went wrong. It might be a
little verbose for now, but I'm interested in as much feedback on
errors as possible, especially timeouts, as I'm a bit in doubt if
I've chosen resonable default values everywhere.
The disk driver has been changed a bit to prepare for tagged queing,
which is next on my list.
The disk driver has grown a dump routine, I got one implementation
from Darrell Anderson <anderson@cs.duke.edu> which also did
partial dumps (usefull on big memory machines) I left out the
partial stuff for now, and changed the rest alot to fit into the new
ad_request framework.
Some minor cleanups and rearrangements as well.
As usual USE AT YOUR OWN RISK!!, this is still pre alpha level code.
Especially the DMA support can hose your disk real bad if anything
goes wrong, again you have been warned :)
Notebook owners should be carefull that their machines dont suspend
as this might cause trouble...
But please tell me how it works for you!
Enjoy!
-Søren
1999-06-25 09:03:07 +00:00
|
|
|
|
|
|
|
|
|
|
/* setup request */
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request->device = atadev;
|
|
|
|
|
|
request->driver = bp;
|
|
|
|
|
|
request->callback = ad_done;
|
|
|
|
|
|
request->timeout = 5;
|
|
|
|
|
|
request->retries = 2;
|
2000-04-15 05:54:02 +00:00
|
|
|
|
request->data = bp->bio_data;
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request->bytecount = bp->bio_bcount;
|
2000-01-27 19:00:51 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
/* convert LBA contents if this is an old non-LBA device */
|
|
|
|
|
|
if (atadev->flags & ATA_D_USE_CHS) {
|
|
|
|
|
|
struct ata_params *param = atadev->param;
|
|
|
|
|
|
int sector = (bp->bio_pblkno % param->sectors) + 1;
|
|
|
|
|
|
int cylinder = bp->bio_pblkno / (param->sectors * param->heads);
|
|
|
|
|
|
int head = (bp->bio_pblkno %
|
|
|
|
|
|
(param->sectors * param->heads)) / param->sectors;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request->u.ata.lba =
|
|
|
|
|
|
(sector & 0xff) | (cylinder & 0xffff) << 8 | (head & 0xf) << 24;
|
2000-05-23 19:05:56 +00:00
|
|
|
|
}
|
2000-01-17 02:04:19 +00:00
|
|
|
|
else
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request->u.ata.lba = bp->bio_pblkno;
|
2000-09-19 11:08:39 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request->u.ata.count = request->bytecount / DEV_BSIZE;
|
|
|
|
|
|
request->transfersize = min(bp->bio_bcount, adp->max_iosize);
|
2000-09-19 11:08:39 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
switch (bp->bio_cmd) {
|
|
|
|
|
|
case BIO_READ:
|
|
|
|
|
|
request->flags |= ATA_R_READ;
|
|
|
|
|
|
if (atadev->mode >= ATA_DMA) {
|
|
|
|
|
|
request->u.ata.command = ATA_READ_DMA;
|
|
|
|
|
|
request->flags |= ATA_R_DMA;
|
2000-09-19 11:08:39 +00:00
|
|
|
|
}
|
2003-08-24 09:22:26 +00:00
|
|
|
|
else if (adp->max_iosize > DEV_BSIZE)
|
|
|
|
|
|
request->u.ata.command = ATA_READ_MUL;
|
|
|
|
|
|
else
|
|
|
|
|
|
request->u.ata.command = ATA_READ;
|
|
|
|
|
|
|
|
|
|
|
|
break;
|
|
|
|
|
|
case BIO_WRITE:
|
|
|
|
|
|
request->flags |= ATA_R_WRITE;
|
|
|
|
|
|
if (atadev->mode >= ATA_DMA) {
|
|
|
|
|
|
request->u.ata.command = ATA_WRITE_DMA;
|
|
|
|
|
|
request->flags |= ATA_R_DMA;
|
2000-09-19 11:08:39 +00:00
|
|
|
|
}
|
2003-08-24 09:22:26 +00:00
|
|
|
|
else if (adp->max_iosize > DEV_BSIZE)
|
|
|
|
|
|
request->u.ata.command = ATA_WRITE_MUL;
|
|
|
|
|
|
else
|
|
|
|
|
|
request->u.ata.command = ATA_WRITE;
|
|
|
|
|
|
break;
|
|
|
|
|
|
default:
|
|
|
|
|
|
ata_prtdev(atadev, "FAILURE - unknown BIO operation\n");
|
|
|
|
|
|
ata_free_request(request);
|
|
|
|
|
|
biofinish(bp, NULL, EIO);
|
|
|
|
|
|
return;
|
2000-09-19 11:08:39 +00:00
|
|
|
|
}
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request->flags |= ATA_R_SKIPSTART;
|
|
|
|
|
|
ata_queue_request(request);
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
2000-09-19 11:08:39 +00:00
|
|
|
|
static void
|
2003-08-24 09:22:26 +00:00
|
|
|
|
ad_done(struct ata_request *request)
|
1999-10-09 19:57:13 +00:00
|
|
|
|
{
|
2003-08-24 09:22:26 +00:00
|
|
|
|
struct bio *bp = request->driver;
|
2000-09-19 11:08:39 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
/* finish up transfer */
|
|
|
|
|
|
if ((bp->bio_error = request->result))
|
|
|
|
|
|
bp->bio_flags |= BIO_ERROR;
|
|
|
|
|
|
bp->bio_resid = bp->bio_bcount - request->donecount;
|
|
|
|
|
|
biodone(bp);
|
|
|
|
|
|
ata_free_request(request);
|
2000-09-19 11:08:39 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
static int
|
2003-08-24 09:22:26 +00:00
|
|
|
|
addump(void *arg, void *virtual, vm_offset_t physical,
|
|
|
|
|
|
off_t offset, size_t length)
|
2000-09-19 11:08:39 +00:00
|
|
|
|
{
|
2003-08-24 09:22:26 +00:00
|
|
|
|
struct ata_request request;
|
|
|
|
|
|
struct disk *dp = arg;
|
|
|
|
|
|
struct ad_softc *adp = dp->d_drv1;
|
2002-07-22 18:35:01 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
if (!adp)
|
|
|
|
|
|
return ENXIO;
|
2002-07-22 18:35:01 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
bzero(&request, sizeof(struct ata_request));
|
|
|
|
|
|
request.data = virtual;
|
|
|
|
|
|
request.bytecount = length;
|
|
|
|
|
|
request.transfersize = min(length, adp->max_iosize);
|
2002-07-22 18:35:01 +00:00
|
|
|
|
|
2003-08-24 09:22:26 +00:00
|
|
|
|
request.flags |= ATA_R_WRITE;
|
|
|
|
|
|
if (adp->max_iosize > DEV_BSIZE)
|
|
|
|
|
|
request.u.ata.command = ATA_WRITE_MUL;
|
|
|
|
|
|
else
|
|
|
|
|
|
request.u.ata.command = ATA_WRITE;
|
|
|
|
|
|
request.u.ata.lba = offset / DEV_BSIZE;
|
|
|
|
|
|
request.u.ata.count = request.bytecount / DEV_BSIZE;
|
|
|
|
|
|
|
|
|
|
|
|
while (request.bytecount > request.donecount) {
|
|
|
|
|
|
if (adp->device->channel->hw.transaction(&request) == ATA_OP_FINISHED)
|
|
|
|
|
|
return EIO;
|
|
|
|
|
|
DELAY(20);
|
2000-09-19 11:08:39 +00:00
|
|
|
|
}
|
|
|
|
|
|
return 0;
|
1999-10-09 19:57:13 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
2001-03-19 08:04:54 +00:00
|
|
|
|
void
|
2002-03-03 15:36:21 +00:00
|
|
|
|
ad_print(struct ad_softc *adp)
|
2001-03-19 08:04:54 +00:00
|
|
|
|
{
|
|
|
|
|
|
if (bootverbose) {
|
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
|
|
|
|
ata_prtdev(adp->device, "<%.40s/%.8s> ATA-%d disk at ata%d-%s\n",
|
|
|
|
|
|
adp->device->param->model, adp->device->param->revision,
|
|
|
|
|
|
ad_version(adp->device->param->version_major),
|
|
|
|
|
|
device_get_unit(adp->device->channel->dev),
|
|
|
|
|
|
(adp->device->unit == ATA_MASTER) ? "master" : "slave");
|
2001-03-19 08:04:54 +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
|
|
|
|
ata_prtdev(adp->device,
|
2001-10-06 11:07:04 +00:00
|
|
|
|
"%lluMB (%llu sectors), %llu C, %u H, %u S, %u B\n",
|
2001-10-14 19:44:15 +00:00
|
|
|
|
(unsigned long long)(adp->total_secs /
|
2002-07-22 18:35:01 +00:00
|
|
|
|
((1024L*1024L)/DEV_BSIZE)),
|
|
|
|
|
|
(unsigned long long)adp->total_secs,
|
|
|
|
|
|
(unsigned long long)(adp->total_secs /
|
|
|
|
|
|
(adp->heads * adp->sectors)),
|
2001-03-19 08:04:54 +00:00
|
|
|
|
adp->heads, adp->sectors, DEV_BSIZE);
|
|
|
|
|
|
|
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
|
|
|
|
ata_prtdev(adp->device, "%d secs/int, %d depth queue, %s%s\n",
|
2003-08-24 09:22:26 +00:00
|
|
|
|
adp->max_iosize / DEV_BSIZE, adp->num_tags + 1,
|
2001-03-19 08:04:54 +00:00
|
|
|
|
(adp->flags & AD_F_TAG_ENABLED) ? "tagged " : "",
|
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
|
|
|
|
ata_mode2str(adp->device->mode));
|
2001-03-19 08:04:54 +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
|
|
|
|
ata_prtdev(adp->device,"%lluMB <%.40s> [%lld/%d/%d] at ata%d-%s %s%s\n",
|
2001-10-14 19:44:15 +00:00
|
|
|
|
(unsigned long long)(adp->total_secs /
|
2002-07-22 18:35:01 +00:00
|
|
|
|
((1024L * 1024L) / DEV_BSIZE)),
|
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
|
|
|
|
adp->device->param->model,
|
2001-10-14 19:44:15 +00:00
|
|
|
|
(unsigned long long)(adp->total_secs /
|
2002-07-22 18:35:01 +00:00
|
|
|
|
(adp->heads * adp->sectors)),
|
2001-03-19 08:04:54 +00:00
|
|
|
|
adp->heads, adp->sectors,
|
Major update of the ATA RAID code, part 1:
Overhaul of the attach/detach code and structures, there were some nasty
bugs in the old implementation. This made it possible to collapse the
ATA/ATAPI device control structures into one generic structure.
A note here, the kernel is NOT ready for detach of active devices,
it fails all over in random places, but for inactive devices it works.
However for ATA RAID this works, since the RAID abstration layer
insulates the buggy^H^H^H^H^H^Hfragile device subsystem from the
physical disks.
Proberly detect the RAID's from the BIOS, and mark critical RAID1
arrays as such, but continue if there is enough of the mirror left
to do so.
Properly fail arrays on a live system. For RAID0 that means return EIO,
and for RAID1 it means continue on the still working part of the mirror
if possible, else return EIO.
If the state changes, log this to the console.
Allow for Promise & Highpoint controllers/arrays to coexist on the
same machine. It is not possible to distribute arrays over different
makes of controllers though.
If Promise SuperSwap enclosures are used, signal disk state on the
status LED on the front.
Misc fixes that I had lying around for various minor bugs.
Sponsored by: Advanis Inc.
2002-02-04 19:23:40 +00:00
|
|
|
|
device_get_unit(adp->device->channel->dev),
|
|
|
|
|
|
(adp->device->unit == ATA_MASTER) ? "master" : "slave",
|
2001-03-19 08:04:54 +00:00
|
|
|
|
(adp->flags & AD_F_TAG_ENABLED) ? "tagged " : "",
|
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
|
|
|
|
ata_mode2str(adp->device->mode));
|
2001-03-19 08:04:54 +00:00
|
|
|
|
}
|
|
|
|
|
|
|
2000-09-19 11:08:39 +00:00
|
|
|
|
static int
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
ad_version(u_int16_t version)
|
|
|
|
|
|
{
|
2000-09-19 11:08:39 +00:00
|
|
|
|
int bit;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
|
|
|
|
|
|
if (version == 0xffff)
|
1999-12-13 10:19:39 +00:00
|
|
|
|
return 0;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
for (bit = 15; bit >= 0; bit--)
|
|
|
|
|
|
if (version & (1<<bit))
|
1999-12-13 10:19:39 +00:00
|
|
|
|
return bit;
|
|
|
|
|
|
return 0;
|
Finally!!
The much roumored replacement for our current IDE/ATA/ATAPI is
materialising in the CVS repositories around the globe.
So what does this bring us:
A new reengineered ATA/ATAPI subsystem, that tries to overcome
most of the deficiencies with the current drivers.
It supports PCI as well as ISA devices without all the hackery
in ide_pci.c to make PCI devices look like ISA counterparts.
It doesn't have the excessive wait problem on probe, in fact you
shouldn't notice any delay when your devices are getting probed.
Probing and attaching of devices are postponed until interrupts
are enabled (well almost, not finished yet for disks), making
things alot cleaner.
Improved performance, although DMA support is still WIP and not
in this pre alpha release, worldstone is faster with the new
driver compared to the old even with DMA.
So what does it take away:
There is NO support for old MFM/RLL/ESDI disks.
There is NO support for bad144, if your disk is bad, ditch it, it has
already outgrown its internal spare sectors, and is dying.
For you to try this out, you will have to modify your kernel config
file to use the "ata" controller instead of all wdc? entries.
example:
# for a PCI only system (most modern machines)
controller ata0
device atadisk0 # ATA disks
device atapicd0 # ATAPI CDROM's
device atapist0 # ATAPI tapes
#You should add the following on ISA systems:
controller ata1 at isa? port "IO_WD1" bio irq 14
controller ata2 at isa? port "IO_WD2" bio irq 15
You can leave it all in there, the system knows how to manage.
For now this driver reuses the device entries from the old system
(that will probably change later), but remember that disks are
now numbered in the sequence they are found (like the SCSI system)
not as absolute positions as the old system.
Although I have tested this on all the systems I can get my hands on,
there might very well be gremlins in there, so use AT YOU OWN RISK!!
This is still WIP, so there are lots of rough edges and unfinished
things in there, and what I have in my lab might look very different
from whats in CVS at any given time. So please have all eventual
changes go through me, or chances are they just dissapears...
I would very much like to hear from you, both good and bad news
are very welcome.
Enjoy!!
-Søren
1999-03-01 21:19:19 +00:00
|
|
|
|
}
|
