freebsd-dev/sys/sparc64/pci/ofw_pci.c

392 lines
12 KiB
C
Raw Normal View History

/*
* Copyright (c) 1999, 2000 Matthew R. Green
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
* Copyright (c) 2001 - 2003 by Thomas Moestl <tmm@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.
* 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.
*
* from: NetBSD: psycho.c,v 1.35 2001/09/10 16:17:06 eeh Exp
*
* $FreeBSD$
*/
#include "opt_ofw_pci.h"
#include <sys/param.h>
#include <sys/kernel.h>
#include <sys/systm.h>
#include <sys/bus.h>
#include <dev/pci/pcireg.h>
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
#include <dev/pci/pcivar.h>
#include <dev/ofw/ofw_pci.h>
#include <dev/ofw/openfirm.h>
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
#include <machine/bus.h>
#include <machine/cache.h>
#include <machine/iommureg.h>
#include <machine/ofw_bus.h>
#include <machine/ver.h>
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
#include <sparc64/pci/ofw_pci.h>
#include "pcib_if.h"
u_int8_t pci_bus_cnt;
phandle_t *pci_bus_map;
int pci_bus_map_sz;
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
#define PCI_BUS_MAP_INC 10
#ifndef OFW_NEWPCI
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
/* Do not swizzle on a PCI bus node with no interrupt-map propery. */
#define OPQ_NO_SWIZZLE 1
/*
* INOs < 255 are really intpin numbers; use a driver method to figure out
* the real INO.
*/
#define OPQ_INO_CALLBACK 2
/*
* Do not map EBus interrupts at PCI buses, but assume that they are fully
* specified already.
*/
#define OPQ_EBUS_NOMAP 4
static struct ofw_pci_quirk {
char *opq_model;
int opq_quirks;
} ofw_pci_quirks[] = {
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
{ "SUNW,Ultra-4", OPQ_INO_CALLBACK | OPQ_EBUS_NOMAP },
{ "SUNW,Ultra-1-Engine", OPQ_NO_SWIZZLE },
};
#define OPQ_NENT (sizeof(ofw_pci_quirks) / sizeof(ofw_pci_quirks[0]))
static int pci_quirks;
#define OFW_PCI_PCIBUS "pci"
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
#define OFW_PCI_EBUS "ebus"
int
ofw_pci_orb_callback(phandle_t node, u_int8_t *pintptr, int pintsz,
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
u_int8_t *pregptr, int pregsz, u_int8_t **rintr, int *terminate,
void *cookie)
{
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
device_t dev = cookie;
struct ofw_pci_register preg;
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
ofw_pci_intr_t pintr, intr;
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
u_int slot;
char type[32];
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
int found = 0;
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
if ((pci_quirks & OPQ_EBUS_NOMAP) != 0 &&
OF_getprop(node, "name", type, sizeof(type)) != -1 &&
strcmp(type, OFW_PCI_EBUS) == 0) {
*terminate = 1;
return (-1);
}
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
if (pintsz != sizeof(pintr) || pregsz < sizeof(preg))
return (-1);
bcopy(pintptr, &pintr, sizeof(pintr));
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
bcopy(pregptr, &preg, sizeof(preg));
slot = OFW_PCI_PHYS_HI_DEVICE(preg.phys_hi);
if ((pci_quirks & OPQ_INO_CALLBACK) != 0 && pintr <= 255) {
/*
* The e450 has no interrupt maps at all, and it usually has
* full interrupt numbers, including IGN, in the interrupt
* properties. There is one exception, however: the property
* values for external PCI devices seem to always be below 255
* and describe the interrupt pin to be used on the slot, while
* we have to figure out the base INO by looking at the slot
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
* number (which we do using an ofw_pci method).
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
*
* Of course, there is an exception to that nice rule:
* in the ebus case, the interrupt property has the correct
* INO (but without IGN). This is dealt with above.
*/
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
intr = OFW_PCI_GUESS_INO(dev, node, slot, pintr);
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
found = intr != 255;
*terminate = found;
}
if (!found && (pci_quirks & OPQ_NO_SWIZZLE) == 0 &&
OF_getprop(node, "device_type", type, sizeof(type)) != -1 &&
strcmp(type, OFW_PCI_PCIBUS) == 0 && pintr >= 1 && pintr <= 4) {
/*
* Handle a quirk found on some Netra t1 models: there exist
* PCI bridges without interrupt maps, where we apparently must
* do the PCI swizzle and continue to map on at the parent.
*/
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
intr = (slot + pintr + 3) % 4 + 1;
*terminate = 0;
found = 1;
}
if (found) {
*rintr = malloc(sizeof(intr), M_OFWPROP, M_WAITOK);
bcopy(&intr, *rintr, sizeof(intr));
return (sizeof(intr));
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
} else
return (-1);
}
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
static ofw_pci_intr_t
ofw_pci_route_intr(device_t dev, phandle_t node, ofw_pci_intr_t ign)
{
u_int32_t rv;
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
rv = ofw_bus_route_intr(node, ORIP_NOINT, ofw_pci_orb_callback, dev);
if (rv == ORIR_NOTFOUND)
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
return (PCI_INVALID_IRQ);
/*
Fix interrupt assignment for non-builtin PCI devices on e450s. This machine uses a non-standard scheme to specify the interrupts to be assigned for devices in PCI slots; instead of giving the INO or full interrupt number (which is done for the other devices in this box), the firmware interrupt properties contain intpin numbers, which have to be swizzled as usual on PCI-PCI bridges; however, the PCI host bridge nodes have no interrupt map, so we need to guess the correct INO by slot number of the device or the closest PCI-PCI bridge leading to it, and the intpin. To do this, this fix makes the following changes: - Add a newbus method for sparc64 PCI host bridges to guess the INO, and glue code in ofw_pci_orb_callback() to invoke it based on a new quirk entry. The guessing is only done for interrupt numbers too low to contain any IGN found on e450s. - Create another new quirk entry was created to prevent mapping of EBus interrupts at PCI level; the e450 has full INOs in the interrupt properties of EBus devices, so trying to remap them could cause problems. - Set both quirk entries for e450s; remove the no-swizzle entry. - Determine the psycho half (bus A or B) a driver instance manages in psycho_attach() - Implement the new guessing method for psycho, using the slot number, psycho half and property value (intpin). Thanks go to the testers, especially Brian Denehy, who tested many kernels for me until I had found the right workaround. Tested by: Brian Denehy <B.Denehy@90east.com>, jake, fenner, Marius Strobl <marius@alchemy.franken.de>, Marian Dobre <mari@onix.ro> Approved by: re (scottl)
2003-05-30 20:48:05 +00:00
* Some machines (notably the SPARCengine Ultra AX and the e450) have
* no mappings at all, but use complete interrupt vector number
* including the IGN. Catch this case and remove the IGN.
*/
if (rv > ign)
rv -= ign;
return (rv);
}
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
#endif /* !OFW_NEWCPI */
u_int8_t
ofw_pci_alloc_busno(phandle_t node)
{
phandle_t *om;
int osz;
u_int8_t n;
n = pci_bus_cnt++;
/* Establish a mapping between bus numbers and device nodes. */
if (n >= pci_bus_map_sz) {
osz = pci_bus_map_sz;
om = pci_bus_map;
pci_bus_map_sz = n + PCI_BUS_MAP_INC;
pci_bus_map = malloc(sizeof(*pci_bus_map) * pci_bus_map_sz,
M_DEVBUF, M_WAITOK | M_ZERO);
if (om != NULL) {
bcopy(om, pci_bus_map, sizeof(*om) * osz);
free(om, M_DEVBUF);
}
}
pci_bus_map[n] = node;
return (n);
}
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
#ifndef OFW_NEWPCI
/*
* Initialize bridge bus numbers for bridges that implement the primary,
* secondary and subordinate bus number registers.
*/
void
ofw_pci_binit(device_t busdev, struct ofw_pci_bdesc *obd)
{
#ifdef OFW_PCI_DEBUG
printf("PCI-PCI bridge at %u/%u/%u: setting bus #s to %u/%u/%u\n",
obd->obd_bus, obd->obd_slot, obd->obd_func, obd->obd_bus,
obd->obd_secbus, obd->obd_subbus);
#endif /* OFW_PCI_DEBUG */
PCIB_WRITE_CONFIG(busdev, obd->obd_bus, obd->obd_slot, obd->obd_func,
PCIR_PRIBUS_1, obd->obd_bus, 1);
PCIB_WRITE_CONFIG(busdev, obd->obd_bus, obd->obd_slot, obd->obd_func,
PCIR_SECBUS_1, obd->obd_secbus, 1);
PCIB_WRITE_CONFIG(busdev, obd->obd_bus, obd->obd_slot, obd->obd_func,
PCIR_SUBBUS_1, obd->obd_subbus, 1);
}
/*
* Walk the PCI bus hierarchy, starting with the root PCI bus and descending
* through bridges, and initialize the interrupt line and latency timer
* configuration registers of attached devices using firmware information,
* as well as the the bus numbers and ranges of the bridges.
*/
void
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
ofw_pci_init(device_t dev, phandle_t bushdl, ofw_pci_intr_t ign,
struct ofw_pci_bdesc *obd)
{
struct ofw_pci_register pcir;
struct ofw_pci_bdesc subobd, *tobd;
phandle_t node;
char type[32];
int i, intr, freemap;
u_int slot, busno, func, sub, lat;
u_int8_t clnsz;
/* Initialize the quirk list. */
for (i = 0; i < OPQ_NENT; i++) {
if (strcmp(sparc64_model, ofw_pci_quirks[i].opq_model) == 0) {
pci_quirks = ofw_pci_quirks[i].opq_quirks;
break;
}
}
if ((node = OF_child(bushdl)) == 0)
return;
freemap = 0;
busno = obd->obd_secbus;
/*
* Compute the value to write into the cache line size register.
* The role of the streaming cache is unclear in write invalidate
* transfers, so it is made sure that it's line size is always reached.
*/
clnsz = imax(cache.ec_linesize, STRBUF_LINESZ);
KASSERT((clnsz / STRBUF_LINESZ) * STRBUF_LINESZ == clnsz &&
(clnsz / cache.ec_linesize) * cache.ec_linesize == clnsz &&
(clnsz / 4) * 4 == clnsz, ("bogus cache line size %d", clnsz));
do {
if (node == -1)
panic("ofw_pci_init_intr: OF_child failed");
if (OF_getprop(node, "device_type", type, sizeof(type)) == -1)
type[0] = '\0';
else
type[sizeof(type) - 1] = '\0';
if (OF_getprop(node, "reg", &pcir, sizeof(pcir)) == -1)
panic("ofw_pci_init: OF_getprop failed");
slot = OFW_PCI_PHYS_HI_DEVICE(pcir.phys_hi);
func = OFW_PCI_PHYS_HI_FUNCTION(pcir.phys_hi);
PCIB_WRITE_CONFIG(dev, busno, slot, func, PCIR_CACHELNSZ,
clnsz / 4, 1);
if (strcmp(type, OFW_PCI_PCIBUS) == 0) {
/*
* This is a pci-pci bridge, initalize the bus number and
* recurse to initialize the child bus. The hierarchy is
* usually at most 2 levels deep, so recursion is
* feasible.
*/
subobd.obd_bus = busno;
subobd.obd_slot = slot;
subobd.obd_func = func;
sub = ofw_pci_alloc_busno(node);
subobd.obd_secbus = subobd.obd_subbus = sub;
/* Assume this bridge is mostly standard conforming. */
subobd.obd_init = ofw_pci_binit;
subobd.obd_super = obd;
/*
* Need to change all subordinate bus registers of the
* bridges above this one now so that configuration
* transactions will get through.
*/
for (tobd = obd; tobd != NULL; tobd = tobd->obd_super) {
tobd->obd_subbus = sub;
tobd->obd_init(dev, tobd);
}
subobd.obd_init(dev, &subobd);
#ifdef OFW_PCI_DEBUG
device_printf(dev, "%s: descending to "
"subordinate PCI bus\n", __func__);
#endif /* OFW_PCI_DEBUG */
ofw_pci_init(dev, node, ign, &subobd);
} else {
/*
* Initialize the latency timer register for
* busmaster devices to work properly. This is another
* task which the firmware does not always perform.
* The Min_Gnt register can be used to compute it's
* recommended value: it contains the desired latency
* in units of 1/4 us. To calculate the correct latency
* timer value, a bus clock of 33 and no wait states
* should be assumed.
*/
lat = PCIB_READ_CONFIG(dev, busno, slot, func,
PCIR_MINGNT, 1) * 33 / 4;
if (lat != 0) {
#ifdef OFW_PCI_DEBUG
printf("device %d/%d/%d: latency timer %d -> "
"%d\n", busno, slot, func,
PCIB_READ_CONFIG(dev, busno, slot, func,
PCIR_LATTIMER, 1), lat);
#endif /* OFW_PCI_DEBUG */
PCIB_WRITE_CONFIG(dev, busno, slot, func,
PCIR_LATTIMER, imin(lat, 255), 1);
}
/* Initialize the intline registers. */
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
if ((intr = ofw_pci_route_intr(dev, node, ign)) !=
PCI_INVALID_IRQ) {
#ifdef OFW_PCI_DEBUG
device_printf(dev, "%s: mapping intr for "
"%d/%d/%d to %d (preset was %d)\n",
__func__, busno, slot, func, intr,
(int)PCIB_READ_CONFIG(dev, busno, slot,
func, PCIR_INTLINE, 1));
#endif /* OFW_PCI_DEBUG */
PCIB_WRITE_CONFIG(dev, busno, slot, func,
PCIR_INTLINE, intr, 1);
} else {
#ifdef OFW_PCI_DEBUG
device_printf(dev, "%s: no interrupt "
"mapping found for %d/%d/%d (preset %d)\n",
__func__, busno, slot, func,
(int)PCIB_READ_CONFIG(dev, busno, slot,
func, PCIR_INTLINE, 1));
#endif /* OFW_PCI_DEBUG */
/*
* The firmware initializes to 0 instead of
* 255.
*/
PCIB_WRITE_CONFIG(dev, busno, slot, func,
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
PCIR_INTLINE, PCI_INVALID_IRQ, 1);
}
}
} while ((node = OF_peer(node)) != 0);
}
phandle_t
ofw_pci_find_node(int bus, int slot, int func)
{
phandle_t node, bnode;
struct ofw_pci_register pcir;
/*
* Retrieve the bus node from the mapping that was created on
* initialization. The bus numbers the firmware uses cannot be trusted,
* so they might have needed to be changed and this is necessary.
*/
if (bus >= pci_bus_map_sz)
return (0);
bnode = pci_bus_map[bus];
if (bnode == 0)
return (0);
for (node = OF_child(bnode); node != 0 && node != -1;
node = OF_peer(node)) {
if (OF_getprop(node, "reg", &pcir, sizeof(pcir)) == -1)
continue;
if (OFW_PCI_PHYS_HI_DEVICE(pcir.phys_hi) == slot &&
OFW_PCI_PHYS_HI_FUNCTION(pcir.phys_hi) == func)
return (node);
}
return (0);
}
phandle_t
ofw_pci_node(device_t dev)
{
return (ofw_pci_find_node(pci_get_bus(dev), pci_get_slot(dev),
pci_get_function(dev)));
}
Add the new sparc64 OFW PCI framework, conditional on options OFW_NEWPCI for now. It introduces a OFW PCI bus driver and a generic OFW PCI-PCI bridge driver. By utilizing these, the PCI handling is much more elegant now. The advantages of the new approach are: - Device enumeration should hopefully be more like on Solaris now, so unit numbers should match what's printed on the box more closely. - Real interrupt routing is implemented now, so cardbus bridges etc. have at least a chance to work. - The quirk tables are gone and have been replaced by (hopefully sufficient) heuristics. - Much cleaner code. There was also a report that previously bogus interrupt assignments are fixed now, which can be attributed to the new heuristics. A pitfall, and the reason why this is not the default yet, is that it changes device enumeration, as mentioned above, which can make it necessary to change the system configuration if more than one unit of a device type is present (on a system with two hme cars, for example, it is possible that hme0 becomes hme1 and vice versa after enabling the option). Systems with multiple disk controllers may need to be booted into single user (and require manual specification of the root file system on boot) to adjust the fstab. Nevertheless, I would like to encourage users to use this option, so that it can be made the default soon. In detail, the changes are: - Introduce an OFW PCI bus driver; it inherits most methods from the generic PCI bus driver, but uses the firmware for enumeration, performs additional initialization for devices and firmware-specific interrupt routing. It also implements an OFW-specific method to allow child devices to get their firmware nodes. - Introduce an OFW PCI-PCI bridge driver; again, it inherits most of the generic PCI-PCI bridge driver; it has it's own method for interrupt routing, as well as some sparc64-specific methods (one to get the node again, and one to adjust the bridge bus range, since we need to reenumerate all PCI buses). - Convert the apb driver to the new way of handling things. - Provide a common framework for OFW bridge drivers, used be the two drivers above. - Provide a small common framework for interrupt routing (for all bridge types). - Convert the psycho driver to the new framework; this gets rid of a bunch of old kludges in pci_read_config(), and the whole preinitialization (ofw_pci_init()). - Convert the ISA MD part and the EBus driver to the new way interrupts and nodes are handled. - Introduce types for firmware interrupt properties. - Rename the old sparcbus_if to ofw_pci_if by repo copy (it is only required for PCI), and move it to a more correct location (new support methodsx were also added, and an old one was deprecated). - Fix a bunch of minor bugs, perform some cleanups. In some cases, I introduced some minor code duplication to keep the new code clean, in hopes that the old code will be unifdef'ed soon. Reviewed in part by: imp Tested by: jake, Marius Strobl <marius@alchemy.franken.de>, Sergey Mokryshev <mokr@mokr.net>, Chris Jackman <cjackNOSPAM@klatsch.org> Info on u30 firmware provided by: kris
2003-07-01 14:52:47 +00:00
#endif /* OFW_NEWPCI */