freebsd-skq/sys/dev/ata/ata-pci.c

706 lines
19 KiB
C
Raw Normal View History

/*-
* Copyright (c) 1998 - 2008 S<EFBFBD>ren Schmidt <sos@FreeBSD.org>
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer,
* without modification, immediately at the beginning of the file.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*/
#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");
#include "opt_ata.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/module.h>
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
#include <sys/ata.h>
#include <sys/bus.h>
#include <sys/conf.h>
#include <sys/malloc.h>
#include <sys/sema.h>
#include <sys/taskqueue.h>
#include <vm/uma.h>
#include <machine/stdarg.h>
#include <machine/resource.h>
#include <machine/bus.h>
#include <sys/rman.h>
#include <dev/pci/pcivar.h>
#include <dev/pci/pcireg.h>
#include <dev/ata/ata-all.h>
#include <dev/ata/ata-pci.h>
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
#include <ata_if.h>
/* local vars */
static MALLOC_DEFINE(M_ATAPCI, "ata_pci", "ATA driver PCI");
/* misc defines */
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
#define IOMASK 0xfffffffc
2006-01-18 13:10:17 +00:00
#define ATA_PROBE_OK -10
int
ata_legacy(device_t dev)
{
return (((pci_read_config(dev, PCIR_PROGIF, 1)&PCIP_STORAGE_IDE_MASTERDEV)&&
((pci_read_config(dev, PCIR_PROGIF, 1) &
(PCIP_STORAGE_IDE_MODEPRIM | PCIP_STORAGE_IDE_MODESEC)) !=
(PCIP_STORAGE_IDE_MODEPRIM | PCIP_STORAGE_IDE_MODESEC))) ||
(!pci_read_config(dev, PCIR_BAR(0), 4) &&
!pci_read_config(dev, PCIR_BAR(1), 4) &&
!pci_read_config(dev, PCIR_BAR(2), 4) &&
!pci_read_config(dev, PCIR_BAR(3), 4) &&
!pci_read_config(dev, PCIR_BAR(5), 4)));
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
int
ata_pci_probe(device_t dev)
{
if (pci_get_class(dev) != PCIC_STORAGE)
return ENXIO;
/* if this is an AHCI chipset grab it */
if (pci_get_subclass(dev) == PCIS_STORAGE_SATA) {
if (!ata_ahci_ident(dev))
return ATA_PROBE_OK;
}
/* run through the vendor specific drivers */
switch (pci_get_vendor(dev)) {
case ATA_ACARD_ID:
2004-06-15 11:02:09 +00:00
if (!ata_acard_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_ACER_LABS_ID:
2004-06-15 11:02:09 +00:00
if (!ata_ali_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_AMD_ID:
2004-06-15 11:02:09 +00:00
if (!ata_amd_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_ADAPTEC_ID:
if (!ata_adaptec_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_ATI_ID:
if (!ata_ati_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_CYRIX_ID:
2004-06-15 11:02:09 +00:00
if (!ata_cyrix_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_CYPRESS_ID:
2004-06-15 11:02:09 +00:00
if (!ata_cypress_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_HIGHPOINT_ID:
2004-06-15 11:02:09 +00:00
if (!ata_highpoint_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_INTEL_ID:
2004-06-15 11:02:09 +00:00
if (!ata_intel_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_ITE_ID:
if (!ata_ite_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_JMICRON_ID:
if (!ata_jmicron_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_MARVELL_ID:
if (!ata_marvell_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_NATIONAL_ID:
2004-06-15 11:02:09 +00:00
if (!ata_national_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_NETCELL_ID:
if (!ata_netcell_ident(dev))
return ATA_PROBE_OK;
break;
case ATA_NVIDIA_ID:
2004-06-15 11:02:09 +00:00
if (!ata_nvidia_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_PROMISE_ID:
2004-06-15 11:02:09 +00:00
if (!ata_promise_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_SERVERWORKS_ID:
2004-06-15 11:02:09 +00:00
if (!ata_serverworks_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_SILICON_IMAGE_ID:
2004-06-15 11:02:09 +00:00
if (!ata_sii_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_SIS_ID:
2004-06-15 11:02:09 +00:00
if (!ata_sis_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_VIA_ID:
2004-06-15 11:02:09 +00:00
if (!ata_via_ident(dev))
return ATA_PROBE_OK;
2004-06-15 11:02:09 +00:00
break;
case ATA_CENATEK_ID:
if (pci_get_devid(dev) == ATA_CENATEK_ROCKET) {
ata_generic_ident(dev);
device_set_desc(dev, "Cenatek Rocket Drive controller");
return ATA_PROBE_OK;
}
break;
case ATA_MICRON_ID:
if (pci_get_devid(dev) == ATA_MICRON_RZ1000 ||
pci_get_devid(dev) == ATA_MICRON_RZ1001) {
ata_generic_ident(dev);
device_set_desc(dev,
"RZ 100? ATA controller !WARNING! data loss/corruption risk");
return ATA_PROBE_OK;
}
break;
}
/* unknown chipset, try generic DMA if it seems possible */
if (pci_get_subclass(dev) == PCIS_STORAGE_IDE) {
if (!ata_generic_ident(dev))
return ATA_PROBE_OK;
}
return ENXIO;
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
int
ata_pci_attach(device_t dev)
{
struct ata_pci_controller *ctlr = device_get_softc(dev);
u_int32_t cmd;
int unit;
/* do chipset specific setups only needed once */
2004-04-30 18:49:03 +00:00
if (ata_legacy(dev) || pci_read_config(dev, PCIR_BAR(2), 4) & IOMASK)
ctlr->channels = 2;
else
ctlr->channels = 1;
ctlr->allocate = ata_pci_allocate;
ctlr->dmainit = ata_pci_dmainit;
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
ctlr->dev = dev;
/* if needed try to enable busmastering */
cmd = pci_read_config(dev, PCIR_COMMAND, 2);
if (!(cmd & PCIM_CMD_BUSMASTEREN)) {
pci_write_config(dev, PCIR_COMMAND, cmd | PCIM_CMD_BUSMASTEREN, 2);
cmd = pci_read_config(dev, PCIR_COMMAND, 2);
}
/* if busmastering mode "stuck" use it */
if ((cmd & PCIM_CMD_BUSMASTEREN) == PCIM_CMD_BUSMASTEREN) {
ctlr->r_type1 = SYS_RES_IOPORT;
ctlr->r_rid1 = ATA_BMADDR_RID;
ctlr->r_res1 = bus_alloc_resource_any(dev, ctlr->r_type1, &ctlr->r_rid1,
RF_ACTIVE);
}
if (ctlr->chipinit(dev))
return ENXIO;
/* attach all channels on this controller */
2004-04-30 18:49:03 +00:00
for (unit = 0; unit < ctlr->channels; unit++) {
if ((unit == 0 || unit == 1) && ata_legacy(dev)) {
2004-04-30 18:49:03 +00:00
device_add_child(dev, "ata", unit);
continue;
}
device_add_child(dev, "ata", devclass_find_free_unit(ata_devclass, 2));
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
bus_generic_attach(dev);
return 0;
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
int
ata_pci_detach(device_t dev)
{
struct ata_pci_controller *ctlr = device_get_softc(dev);
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
device_t *children;
int nchildren, i;
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
/* detach & delete all children */
if (!device_get_children(dev, &children, &nchildren)) {
for (i = 0; i < nchildren; i++)
device_delete_child(dev, children[i]);
free(children, M_TEMP);
}
if (ctlr->r_irq) {
bus_teardown_intr(dev, ctlr->r_irq, ctlr->handle);
bus_release_resource(dev, SYS_RES_IRQ, ATA_IRQ_RID, ctlr->r_irq);
}
if (ctlr->r_res2)
bus_release_resource(dev, ctlr->r_type2, ctlr->r_rid2, ctlr->r_res2);
if (ctlr->r_res1)
bus_release_resource(dev, ctlr->r_type1, ctlr->r_rid1, ctlr->r_res1);
return 0;
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
struct resource *
ata_pci_alloc_resource(device_t dev, device_t child, int type, int *rid,
u_long start, u_long end, u_long count, u_int flags)
{
struct ata_pci_controller *controller = device_get_softc(dev);
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
struct resource *res = NULL;
int myrid;
if (type == SYS_RES_IOPORT) {
switch (*rid) {
case ATA_IOADDR_RID:
if (ata_legacy(dev)) {
start = (unit ? ATA_SECONDARY : ATA_PRIMARY);
count = ATA_IOSIZE;
end = start + count - 1;
}
2004-04-30 18:49:03 +00:00
myrid = PCIR_BAR(0) + (unit << 3);
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
SYS_RES_IOPORT, &myrid,
start, end, count, flags);
break;
case ATA_CTLADDR_RID:
if (ata_legacy(dev)) {
start = (unit ? ATA_SECONDARY : ATA_PRIMARY) + ATA_CTLOFFSET;
count = ATA_CTLIOSIZE;
end = start + count - 1;
}
2004-04-30 18:49:03 +00:00
myrid = PCIR_BAR(1) + (unit << 3);
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), dev,
2004-06-15 11:02:09 +00:00
SYS_RES_IOPORT, &myrid,
start, end, count, flags);
break;
}
}
if (type == SYS_RES_IRQ && *rid == ATA_IRQ_RID) {
if (ata_legacy(dev)) {
int irq = (unit == 0 ? 14 : 15);
res = BUS_ALLOC_RESOURCE(device_get_parent(dev), child,
SYS_RES_IRQ, rid, irq, irq, 1, flags);
}
else
res = controller->r_irq;
}
return res;
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
int
ata_pci_release_resource(device_t dev, device_t child, int type, int rid,
struct resource *r)
{
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
if (type == SYS_RES_IOPORT) {
switch (rid) {
case ATA_IOADDR_RID:
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
2004-04-30 18:49:03 +00:00
SYS_RES_IOPORT,
PCIR_BAR(0) + (unit << 3), r);
break;
case ATA_CTLADDR_RID:
return BUS_RELEASE_RESOURCE(device_get_parent(dev), dev,
2004-04-30 18:49:03 +00:00
SYS_RES_IOPORT,
PCIR_BAR(1) + (unit << 3), r);
break;
default:
return ENOENT;
}
}
if (type == SYS_RES_IRQ) {
if (rid != ATA_IRQ_RID)
return ENOENT;
if (ata_legacy(dev)) {
return BUS_RELEASE_RESOURCE(device_get_parent(dev), child,
SYS_RES_IRQ, rid, r);
2004-06-15 11:02:09 +00:00
}
else
return 0;
}
return EINVAL;
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
int
ata_pci_setup_intr(device_t dev, device_t child, struct resource *irq,
int flags, driver_filter_t *filter, driver_intr_t *function,
void *argument, void **cookiep)
{
if (ata_legacy(dev)) {
return BUS_SETUP_INTR(device_get_parent(dev), child, irq,
flags, filter, function, argument, cookiep);
}
else {
struct ata_pci_controller *controller = device_get_softc(dev);
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
if (filter != NULL) {
printf("ata-pci.c: we cannot use a filter here\n");
return (EINVAL);
}
controller->interrupt[unit].function = function;
controller->interrupt[unit].argument = argument;
*cookiep = controller;
return 0;
}
}
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
int
ata_pci_teardown_intr(device_t dev, device_t child, struct resource *irq,
void *cookie)
{
if (ata_legacy(dev)) {
return BUS_TEARDOWN_INTR(device_get_parent(dev), child, irq, cookie);
}
else {
struct ata_pci_controller *controller = device_get_softc(dev);
int unit = ((struct ata_channel *)device_get_softc(child))->unit;
controller->interrupt[unit].function = NULL;
controller->interrupt[unit].argument = NULL;
return 0;
}
}
int
ata_pci_allocate(device_t dev)
{
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
struct ata_channel *ch = device_get_softc(dev);
struct resource *io = NULL, *ctlio = NULL;
int i, rid;
rid = ATA_IOADDR_RID;
2005-04-25 07:57:04 +00:00
if (!(io = bus_alloc_resource_any(dev, SYS_RES_IOPORT, &rid, RF_ACTIVE)))
return ENXIO;
rid = ATA_CTLADDR_RID;
2005-04-25 07:57:04 +00:00
if (!(ctlio = bus_alloc_resource_any(dev, SYS_RES_IOPORT, &rid,RF_ACTIVE))){
bus_release_resource(dev, SYS_RES_IOPORT, ATA_IOADDR_RID, io);
return ENXIO;
}
for (i = ATA_DATA; i <= ATA_COMMAND; i ++) {
ch->r_io[i].res = io;
ch->r_io[i].offset = i;
}
ch->r_io[ATA_CONTROL].res = ctlio;
ch->r_io[ATA_CONTROL].offset = ata_legacy(device_get_parent(dev)) ? 0 : 2;
ch->r_io[ATA_IDX_ADDR].res = io;
ata_default_registers(dev);
if (ctlr->r_res1) {
for (i = ATA_BMCMD_PORT; i <= ATA_BMDTP_PORT; i++) {
ch->r_io[i].res = ctlr->r_res1;
ch->r_io[i].offset = (i - ATA_BMCMD_PORT) + (ch->unit*ATA_BMIOSIZE);
}
}
ata_pci_hw(dev);
return 0;
}
int
ata_pci_status(device_t dev)
{
struct ata_channel *ch = device_get_softc(dev);
if ((dumping || !ata_legacy(device_get_parent(dev))) &&
((ch->flags & ATA_ALWAYS_DMASTAT) ||
(ch->dma.flags & ATA_DMA_ACTIVE))) {
int bmstat = ATA_IDX_INB(ch, ATA_BMSTAT_PORT) & ATA_BMSTAT_MASK;
2006-01-18 13:10:17 +00:00
if ((bmstat & (ATA_BMSTAT_ACTIVE | ATA_BMSTAT_INTERRUPT)) !=
ATA_BMSTAT_INTERRUPT)
return 0;
ATA_IDX_OUTB(ch, ATA_BMSTAT_PORT, bmstat & ~ATA_BMSTAT_ERROR);
DELAY(1);
}
if (ATA_IDX_INB(ch, ATA_ALTSTAT) & ATA_S_BUSY) {
2006-01-18 13:10:17 +00:00
DELAY(100);
if (ATA_IDX_INB(ch, ATA_ALTSTAT) & ATA_S_BUSY)
return 0;
}
return 1;
}
void
ata_pci_hw(device_t dev)
{
struct ata_channel *ch = device_get_softc(dev);
ata_generic_hw(dev);
ch->hw.status = ata_pci_status;
}
static int
ata_pci_dmastart(struct ata_request *request)
{
struct ata_channel *ch = device_get_softc(request->parent);
ATA_DEBUG_RQ(request, "dmastart");
ATA_IDX_OUTB(ch, ATA_BMSTAT_PORT, (ATA_IDX_INB(ch, ATA_BMSTAT_PORT) |
(ATA_BMSTAT_INTERRUPT | ATA_BMSTAT_ERROR)));
ATA_IDX_OUTL(ch, ATA_BMDTP_PORT, request->dma->sg_bus);
ch->dma.flags |= ATA_DMA_ACTIVE;
ATA_IDX_OUTB(ch, ATA_BMCMD_PORT,
2004-05-10 20:23:25 +00:00
(ATA_IDX_INB(ch, ATA_BMCMD_PORT) & ~ATA_BMCMD_WRITE_READ) |
((request->flags & ATA_R_READ) ? ATA_BMCMD_WRITE_READ : 0)|
ATA_BMCMD_START_STOP);
return 0;
}
static int
ata_pci_dmastop(struct ata_request *request)
{
struct ata_channel *ch = device_get_softc(request->parent);
int error;
ATA_DEBUG_RQ(request, "dmastop");
ATA_IDX_OUTB(ch, ATA_BMCMD_PORT,
ATA_IDX_INB(ch, ATA_BMCMD_PORT) & ~ATA_BMCMD_START_STOP);
ch->dma.flags &= ~ATA_DMA_ACTIVE;
error = ATA_IDX_INB(ch, ATA_BMSTAT_PORT) & ATA_BMSTAT_MASK;
ATA_IDX_OUTB(ch, ATA_BMSTAT_PORT, ATA_BMSTAT_INTERRUPT | ATA_BMSTAT_ERROR);
return error;
}
static void
ata_pci_dmareset(device_t dev)
{
struct ata_channel *ch = device_get_softc(dev);
struct ata_request *request;
ATA_IDX_OUTB(ch, ATA_BMCMD_PORT,
ATA_IDX_INB(ch, ATA_BMCMD_PORT) & ~ATA_BMCMD_START_STOP);
ch->dma.flags &= ~ATA_DMA_ACTIVE;
ATA_IDX_OUTB(ch, ATA_BMSTAT_PORT, ATA_BMSTAT_INTERRUPT | ATA_BMSTAT_ERROR);
if ((request = ch->running)) {
device_printf(request->dev, "DMA reset calling unload\n");
ch->dma.unload(request);
}
}
void
ata_pci_dmainit(device_t dev)
{
struct ata_channel *ch = device_get_softc(dev);
ata_dmainit(dev);
ch->dma.start = ata_pci_dmastart;
ch->dma.stop = ata_pci_dmastop;
ch->dma.reset = ata_pci_dmareset;
}
char *
ata_pcivendor2str(device_t dev)
{
switch (pci_get_vendor(dev)) {
case ATA_ACARD_ID: return "Acard";
case ATA_ACER_LABS_ID: return "AcerLabs";
case ATA_AMD_ID: return "AMD";
case ATA_ADAPTEC_ID: return "Adaptec";
case ATA_ATI_ID: return "ATI";
case ATA_CYRIX_ID: return "Cyrix";
case ATA_CYPRESS_ID: return "Cypress";
case ATA_HIGHPOINT_ID: return "HighPoint";
case ATA_INTEL_ID: return "Intel";
case ATA_ITE_ID: return "ITE";
case ATA_JMICRON_ID: return "JMicron";
case ATA_MARVELL_ID: return "Marvell";
case ATA_NATIONAL_ID: return "National";
case ATA_NETCELL_ID: return "Netcell";
case ATA_NVIDIA_ID: return "nVidia";
case ATA_PROMISE_ID: return "Promise";
case ATA_SERVERWORKS_ID: return "ServerWorks";
case ATA_SILICON_IMAGE_ID: return "SiI";
case ATA_SIS_ID: return "SiS";
case ATA_VIA_ID: return "VIA";
case ATA_CENATEK_ID: return "Cenatek";
case ATA_MICRON_ID: return "Micron";
default: return "Generic";
}
}
static device_method_t ata_pci_methods[] = {
/* device interface */
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
DEVMETHOD(device_probe, ata_pci_probe),
DEVMETHOD(device_attach, ata_pci_attach),
DEVMETHOD(device_detach, ata_pci_detach),
DEVMETHOD(device_shutdown, bus_generic_shutdown),
DEVMETHOD(device_suspend, bus_generic_suspend),
DEVMETHOD(device_resume, bus_generic_resume),
/* bus methods */
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
DEVMETHOD(bus_alloc_resource, ata_pci_alloc_resource),
DEVMETHOD(bus_release_resource, ata_pci_release_resource),
DEVMETHOD(bus_activate_resource, bus_generic_activate_resource),
DEVMETHOD(bus_deactivate_resource, bus_generic_deactivate_resource),
DEVMETHOD(bus_setup_intr, ata_pci_setup_intr),
DEVMETHOD(bus_teardown_intr, ata_pci_teardown_intr),
{ 0, 0 }
};
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
devclass_t atapci_devclass;
static driver_t ata_pci_driver = {
"atapci",
ata_pci_methods,
sizeof(struct ata_pci_controller),
};
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
DRIVER_MODULE(atapci, pci, ata_pci_driver, atapci_devclass, 0, 0);
MODULE_VERSION(atapci, 1);
MODULE_DEPEND(atapci, ata, 1, 1, 1);
static int
ata_pcichannel_probe(device_t dev)
{
struct ata_channel *ch = device_get_softc(dev);
device_t *children;
int count, i;
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
char buffer[32];
/* take care of green memory */
bzero(ch, sizeof(struct ata_channel));
/* find channel number on this controller */
device_get_children(device_get_parent(dev), &children, &count);
for (i = 0; i < count; i++) {
if (children[i] == dev)
ch->unit = i;
}
free(children, M_TEMP);
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
sprintf(buffer, "ATA channel %d", ch->unit);
device_set_desc_copy(dev, buffer);
return ata_probe(dev);
}
static int
ata_pcichannel_attach(device_t dev)
{
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
int error;
if (ctlr->dmainit)
ctlr->dmainit(dev);
if ((error = ctlr->allocate(dev)))
2004-06-15 11:02:09 +00:00
return error;
return ata_attach(dev);
}
static int
ata_pcichannel_detach(device_t dev)
{
struct ata_channel *ch = device_get_softc(dev);
int error;
if ((error = ata_detach(dev)))
return error;
ch->dma.free(dev);
2005-05-13 10:25:19 +00:00
/* XXX SOS free resources for io and ctlio ?? */
return 0;
}
static int
ata_pcichannel_locking(device_t dev, int mode)
{
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
struct ata_channel *ch = device_get_softc(dev);
if (ctlr->locking)
return ctlr->locking(dev, mode);
else
return ch->unit;
}
static void
ata_pcichannel_reset(device_t dev)
{
struct ata_pci_controller *ctlr = device_get_softc(device_get_parent(dev));
struct ata_channel *ch = device_get_softc(dev);
/* if DMA engine present reset it */
if (ch->dma.reset)
ch->dma.reset(dev);
/* reset the controller HW */
if (ctlr->reset)
ctlr->reset(dev);
else
ata_generic_reset(dev);
}
static void
ata_pcichannel_setmode(device_t parent, device_t dev)
{
struct ata_pci_controller *ctlr = device_get_softc(GRANDPARENT(dev));
struct ata_device *atadev = device_get_softc(dev);
int mode = atadev->mode;
ctlr->setmode(dev, ATA_PIO_MAX);
if (mode >= ATA_DMA)
ctlr->setmode(dev, mode);
}
static device_method_t ata_pcichannel_methods[] = {
/* device interface */
DEVMETHOD(device_probe, ata_pcichannel_probe),
DEVMETHOD(device_attach, ata_pcichannel_attach),
DEVMETHOD(device_detach, ata_pcichannel_detach),
This is the much rumoured ATA mkIII update that I've been working on. o ATA is now fully newbus'd and split into modules. This means that on a modern system you just load "atapci and ata" to get the base support, and then one or more of the device subdrivers "atadisk atapicd atapifd atapist ataraid". All can be loaded/unloaded anytime, but for obvious reasons you dont want to unload atadisk when you have mounted filesystems. o The device identify part of the probe has been rewritten to fix the problems with odd devices the old had, and to try to remove so of the long delays some HW could provoke. Also probing is done without the need for interrupts, making earlier probing possible. o SATA devices can be hot inserted/removed and devices will be created/ removed in /dev accordingly. NOTE: only supported on controllers that has this feature: Promise and Silicon Image for now. On other controllers the usual atacontrol detach/attach dance is still needed. o Support for "atomic" composite ATA requests used for RAID. o ATA RAID support has been rewritten and and now supports these metadata formats: "Adaptec HostRAID" "Highpoint V2 RocketRAID" "Highpoint V3 RocketRAID" "Intel MatrixRAID" "Integrated Technology Express" "LSILogic V2 MegaRAID" "LSILogic V3 MegaRAID" "Promise FastTrak" "Silicon Image Medley" "FreeBSD PseudoRAID" o Update the ioctl API to match new RAID levels etc. o Update atacontrol to know about the new RAID levels etc NOTE: you need to recompile atacontrol with the new sys/ata.h, make world will take care of that. NOTE2: that rebuild is done differently from the old system as the rebuild is now done piggybacked on read requests to the array, so atacontrol simply starts a background "dd" to rebuild the array. o The reinit code has been worked over to be much more robust. o The timeout code has been overhauled for races. o Support of new chipsets. o Lots of fixes for bugs found while doing the modulerization and reviewing the old code. Missing or changed features from current ATA: o atapi-cd no longer has support for ATAPI changers. Todays its much cheaper and alot faster to copy those CD images to disk and serve them from there. Besides they dont seem to be made anymore, maybe for that exact reason. o ATA RAID can only read metadata from all the above metadata formats, not write all of them (Promise and Highpoint V2 so far). This means that arrays can be picked up from the BIOS, but they cannot be created from FreeBSD. There is more to it than just the missing write metadata support, those formats are not unique to a given controller like Promise and Highpoint formats, instead they exist for several types, and even worse, some controllers can have different formats and its impossible to tell which one. The outcome is that we cannot reliably create the metadata of those formats and be sure the controller BIOS will understand it. However write support is needed to update/fail/rebuild the arrays properly so it sits fairly high on the TODO list. o So far atapicam is not supported with these changes. When/if this will change is up to the maintainer of atapi-cam so go there for questions. HW donated by: Webveveriet AS HW donated by: Frode Nordahl HW donated by: Yahoo! HW donated by: Sentex Patience by: Vife and my boys (and even the cats)
2005-03-30 12:03:40 +00:00
DEVMETHOD(device_shutdown, bus_generic_shutdown),
DEVMETHOD(device_suspend, ata_suspend),
DEVMETHOD(device_resume, ata_resume),
/* ATA methods */
DEVMETHOD(ata_setmode, ata_pcichannel_setmode),
DEVMETHOD(ata_locking, ata_pcichannel_locking),
DEVMETHOD(ata_reset, ata_pcichannel_reset),
{ 0, 0 }
};
driver_t ata_pcichannel_driver = {
"ata",
ata_pcichannel_methods,
sizeof(struct ata_channel),
};
DRIVER_MODULE(ata, atapci, ata_pcichannel_driver, ata_devclass, 0, 0);