Remove unnecessary <sys/malloc.h> include.

Merge thorpej-cfargs2.

branches: 1.5.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.4.4;
<sys/extent.h> not needed here.

PCI Extended Configuration stuff written by nonaka@:
- Add PCI Extended Configuration Space support into x86.
- Check register offset of pci_conf_read() in MD part. It returns (pcireg_t)-1
if it isn't accessible.
- Decode Extended Capability in PCI Extended Configuration Space.
Currently the following extended capabilities are decoded:
- Advanced Error Reporting
- Virtual Channel
- Device Serial Number
- Power Budgeting
- Root Complex Link Declaration
- Root Complex Event Collector Association
- Access Control Services
- Alternative Routing-ID Interpretation
- Address Translation Services
- Single Root IO Virtualization
- Page Request
- TPH Requester
- Latency Tolerance Reporting
- Secondary PCI Express
- Process Address Space ID
- LN Requester
- L1 PM Substates
The following extended capabilities are not decoded yet:
- Root Complex Internal Link Control
- Multi-Function Virtual Channel
- RCRB Header
- Vendor Unique
- Configuration Access Correction
- Multiple Root IO Virtualization
- Multicast
- Resizable BAR
- Dynamic Power Allocation
- Protocol Multiplexing
- Downstream Port Containment
- Precision Time Management
- M-PCIe
- Function Reading Status Queueing
- Readiness Time Reporting
- Designated Vendor-Specific

branches: 1.2.14;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.1.4; 1.1.6; 1.1.10;
Add cv3dpb, a driver for CyberVision64/3D PCI bridge, which was split from p5pb.

branches: 1.1.4; 1.1.6;
Add cv3dpb, a driver for CyberVision64/3D PCI bridge, which was split from p5pb.

branches: 1.1.4; 1.1.6;
Add cv3dpb, a driver for CyberVision64/3D PCI bridge, which was split from p5pb.

Remove unnecessary <sys/malloc.h> include.

Merge thorpej-cfargs2.

branches: 1.7.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.6.4;
Overhaul the interface to pci_configure_bus():
- Don't expose how PCI bus configuration resource management is implemented.
Provide a new resource provider API:

==> pciconf_resource_init() -- Initialize a PCI configuration resources
container.
==> pciconf_resource_add() -- Add a PCI configuration resource to the
container (I/O, MEM, or prefetchable MEM). Multiple resources of
each type may be added.
==> pciconf_resource_fini() -- Tear down the PCI configurtation resources
container once the bus has been configured.

This is much easier to use than the previous method of providing an
extent map for each kind of resource, and works better for e.g. ACPI
platforms that provide potentially multiple PCI resources in tables
provided by firmware.

- Re-implement PCI configuration resource management using vmem arenas,
rather than extent maps.

replace EX_NOWAIT with EX_WAITOK in device attach methods.
remove checks for failures that can no longer occur.

PCI Extended Configuration stuff written by nonaka@:
- Add PCI Extended Configuration Space support into x86.
- Check register offset of pci_conf_read() in MD part. It returns (pcireg_t)-1
if it isn't accessible.
- Decode Extended Capability in PCI Extended Configuration Space.
Currently the following extended capabilities are decoded:
- Advanced Error Reporting
- Virtual Channel
- Device Serial Number
- Power Budgeting
- Root Complex Link Declaration
- Root Complex Event Collector Association
- Access Control Services
- Alternative Routing-ID Interpretation
- Address Translation Services
- Single Root IO Virtualization
- Page Request
- TPH Requester
- Latency Tolerance Reporting
- Secondary PCI Express
- Process Address Space ID
- LN Requester
- L1 PM Substates
The following extended capabilities are not decoded yet:
- Root Complex Internal Link Control
- Multi-Function Virtual Channel
- RCRB Header
- Vendor Unique
- Configuration Access Correction
- Multiple Root IO Virtualization
- Multicast
- Resizable BAR
- Dynamic Power Allocation
- Protocol Multiplexing
- Downstream Port Containment
- Precision Time Management
- M-PCIe
- Function Reading Status Queueing
- Readiness Time Reporting
- Designated Vendor-Specific

branches: 1.3.4; 1.3.8;
kill sprintf
fix unused variables

branches: 1.2.2; 1.2.4;
Support PCI_NETBSD_CONFIGURE for em4k. Mediator 4000 now works completely (minus
obvious lack of DMA).

Add Mediator 4000 driver - em4k(4). Working, but PCI_NETBSD_CONFIGURE not
implemented yet. While here add some untested Mediator ZIV support to empb(4).

branches: 1.1.4; 1.1.14;
Add Mediator 4000 driver - em4k(4). Working, but PCI_NETBSD_CONFIGURE not
implemented yet. While here add some untested Mediator ZIV support to empb(4).

Remove unnecessary <sys/malloc.h> include.

<sys/extent.h> not needed here.

Add Mediator 4000 driver - em4k(4). Working, but PCI_NETBSD_CONFIGURE not
implemented yet. While here add some untested Mediator ZIV support to empb(4).

branches: 1.3.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.2.2; 1.2.4;
Add empb(4) functions to switch memory window position.

Add a driver for Elbox Mediator 1200 PCI bridges. This is still a work in
progress, but now it is good enough to drive a NE2000 card. Support for
memory space accesses will be added soon(ish).

branches: 1.1.2; 1.1.6;
Add a driver for Elbox Mediator 1200 PCI bridges. This is still a work in
progress, but now it is good enough to drive a NE2000 card. Support for
memory space accesses will be added soon(ish).

Remove unnecessary <sys/malloc.h> include.

Merge thorpej-cfargs2.

branches: 1.14.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.13.4;
Overhaul the interface to pci_configure_bus():
- Don't expose how PCI bus configuration resource management is implemented.
Provide a new resource provider API:

==> pciconf_resource_init() -- Initialize a PCI configuration resources
container.
==> pciconf_resource_add() -- Add a PCI configuration resource to the
container (I/O, MEM, or prefetchable MEM). Multiple resources of
each type may be added.
==> pciconf_resource_fini() -- Tear down the PCI configurtation resources
container once the bus has been configured.

This is much easier to use than the previous method of providing an
extent map for each kind of resource, and works better for e.g. ACPI
platforms that provide potentially multiple PCI resources in tables
provided by firmware.

- Re-implement PCI configuration resource management using vmem arenas,
rather than extent maps.

replace EX_NOWAIT with EX_WAITOK in device attach methods.
remove checks for failures that can no longer occur.

PCI Extended Configuration stuff written by nonaka@:
- Add PCI Extended Configuration Space support into x86.
- Check register offset of pci_conf_read() in MD part. It returns (pcireg_t)-1
if it isn't accessible.
- Decode Extended Capability in PCI Extended Configuration Space.
Currently the following extended capabilities are decoded:
- Advanced Error Reporting
- Virtual Channel
- Device Serial Number
- Power Budgeting
- Root Complex Link Declaration
- Root Complex Event Collector Association
- Access Control Services
- Alternative Routing-ID Interpretation
- Address Translation Services
- Single Root IO Virtualization
- Page Request
- TPH Requester
- Latency Tolerance Reporting
- Secondary PCI Express
- Process Address Space ID
- LN Requester
- L1 PM Substates
The following extended capabilities are not decoded yet:
- Root Complex Internal Link Control
- Multi-Function Virtual Channel
- RCRB Header
- Vendor Unique
- Configuration Access Correction
- Multiple Root IO Virtualization
- Multicast
- Resizable BAR
- Dynamic Power Allocation
- Protocol Multiplexing
- Downstream Port Containment
- Precision Time Management
- M-PCIe
- Function Reading Status Queueing
- Readiness Time Reporting
- Designated Vendor-Specific

branches: 1.10.14;
Add Mediator 4000 driver - em4k(4). Working, but PCI_NETBSD_CONFIGURE not
implemented yet. While here add some untested Mediator ZIV support to empb(4).

Add support for power management on Mediator 1200 SX and TX models. Allows
software power-off. Feels so modern.

branches: 1.8.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.7.2;
Fix Mediator 1200 memory space support. Now it's good enough to drive a Voodoo 3.

Finally add support for memory space access for empb(4). Buggy but better than nothing.

Fix PCI interrupt handling on empb. Also, continue work on memory accesses. Now empb is good enough to handle satalink!

branches: 1.4.2;
Add support for PCI_NETBSD_CONFIGURE in empb(4). Continue work on memory window.

More work on PCI memory accesses on empb(4).

Add empb(4) functions to switch memory window position.

Add a driver for Elbox Mediator 1200 PCI bridges. This is still a work in
progress, but now it is good enough to drive a NE2000 card. Support for
memory space accesses will be added soon(ish).

branches: 1.5.4;
Fix Mediator 1200 memory space support. Now it's good enough to drive a Voodoo 3.

Finally add support for memory space access for empb(4). Buggy but better than nothing.

Fix PCI interrupt handling on empb. Also, continue work on memory accesses. Now empb is good enough to handle satalink!

branches: 1.2.2;
Add support for PCI_NETBSD_CONFIGURE in empb(4). Continue work on memory window.

More work on PCI memory accesses on empb(4).

Overhaul the interface to pci_configure_bus():
- Don't expose how PCI bus configuration resource management is implemented.
Provide a new resource provider API:

==> pciconf_resource_init() -- Initialize a PCI configuration resources
container.
==> pciconf_resource_add() -- Add a PCI configuration resource to the
container (I/O, MEM, or prefetchable MEM). Multiple resources of
each type may be added.
==> pciconf_resource_fini() -- Tear down the PCI configurtation resources
container once the bus has been configured.

This is much easier to use than the previous method of providing an
extent map for each kind of resource, and works better for e.g. ACPI
platforms that provide potentially multiple PCI resources in tables
provided by firmware.

- Re-implement PCI configuration resource management using vmem arenas,
rather than extent maps.

Add Mediator 4000 driver - em4k(4). Working, but PCI_NETBSD_CONFIGURE not
implemented yet. While here add some untested Mediator ZIV support to empb(4).

Add support for power management on Mediator 1200 SX and TX models. Allows
software power-off. Feels so modern.

branches: 1.4.2; 1.4.4;
Fix PCI interrupt handling on empb. Also, continue work on memory accesses. Now empb is good enough to handle satalink!

branches: 1.3.2;
Add support for PCI_NETBSD_CONFIGURE in empb(4). Continue work on memory window.

Add empb(4) functions to switch memory window position.

Add a driver for Elbox Mediator 1200 PCI bridges. This is still a work in
progress, but now it is good enough to drive a NE2000 card. Support for
memory space accesses will be added soon(ish).

Add support for power management on Mediator 1200 SX and TX models. Allows
software power-off. Feels so modern.

branches: 1.2.2; 1.2.4;
Fix PCI interrupt handling on empb. Also, continue work on memory accesses. Now empb is good enough to handle satalink!

branches: 1.1.2;
More work on PCI memory accesses on empb(4).

branches: 1.1.4; 1.1.14;
Add support for power management on Mediator 1200 SX and TX models. Allows
software power-off. Feels so modern.

branches: 1.1.4; 1.1.14;
Add support for power management on Mediator 1200 SX and TX models. Allows
software power-off. Feels so modern.

Remove unnecessary <sys/malloc.h> include.

Merge thorpej-cfargs2.

branches: 1.11.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.10.4;
Overhaul the interface to pci_configure_bus():
- Don't expose how PCI bus configuration resource management is implemented.
Provide a new resource provider API:

==> pciconf_resource_init() -- Initialize a PCI configuration resources
container.
==> pciconf_resource_add() -- Add a PCI configuration resource to the
container (I/O, MEM, or prefetchable MEM). Multiple resources of
each type may be added.
==> pciconf_resource_fini() -- Tear down the PCI configurtation resources
container once the bus has been configured.

This is much easier to use than the previous method of providing an
extent map for each kind of resource, and works better for e.g. ACPI
platforms that provide potentially multiple PCI resources in tables
provided by firmware.

- Re-implement PCI configuration resource management using vmem arenas,
rather than extent maps.

replace EX_NOWAIT with EX_WAITOK in device attach methods.
remove checks for failures that can no longer occur.

PCI Extended Configuration stuff written by nonaka@:
- Add PCI Extended Configuration Space support into x86.
- Check register offset of pci_conf_read() in MD part. It returns (pcireg_t)-1
if it isn't accessible.
- Decode Extended Capability in PCI Extended Configuration Space.
Currently the following extended capabilities are decoded:
- Advanced Error Reporting
- Virtual Channel
- Device Serial Number
- Power Budgeting
- Root Complex Link Declaration
- Root Complex Event Collector Association
- Access Control Services
- Alternative Routing-ID Interpretation
- Address Translation Services
- Single Root IO Virtualization
- Page Request
- TPH Requester
- Latency Tolerance Reporting
- Secondary PCI Express
- Process Address Space ID
- LN Requester
- L1 PM Substates
The following extended capabilities are not decoded yet:
- Root Complex Internal Link Control
- Multi-Function Virtual Channel
- RCRB Header
- Vendor Unique
- Configuration Access Correction
- Multiple Root IO Virtualization
- Multicast
- Resizable BAR
- Dynamic Power Allocation
- Protocol Multiplexing
- Downstream Port Containment
- Precision Time Management
- M-PCIe
- Function Reading Status Queueing
- Readiness Time Reporting
- Designated Vendor-Specific

branches: 1.7.14;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.6.6;
adapt to extent(9) changes

Tons of enchancements to MD PCI code:
- The G-REX 4000 and G-REX 1200 are now really supported and working (still only 1st slot but the rest are a matter of time).
- Add bridge identification to p5pb, simplifies the driver and make it more user-friendly.
- Retire P5PB_GENFB and P5PB_OLD_FIRMWARE options, these are now handled automatically, depending on used hardware and firmware.
- Add PCI bus reconfiguration support to p5pb (PCI_NETBSD_CONFIGURE), allows using G-REX without boot to AmigaOS.
- Bring p5pb registers desription up to date with my knowledge.
- Add a lot of debugging code to p5pb (enabled by P5PB_DEBUG kernel option).
- Fix PCI_NETBSD_CONFIGURE in mppb (now it really works).
- Fix quirk handling in zbus.
- Update GENERIC.in and regen the kernel configs.
- Misc cleanup.

Make it actually build without PCI_NETBSD_CONFIGURE too.

Make it build without PCI_NETBSD_CONFIGURE too.

branches: 1.2.2; 1.2.6;
Minor changes to Amiga PCI MD code:
- Add bus enumeration and resource assignment (PCI_NETBSD_CONFIGURE) to mppb(4) - no more need to run AmigaOS to configure the bus.
- Add PCI configuration address register to p5pb(4).
- Add PCI I/O space support to p5pb(4).
- Move common code from p5pb(4) and mppb(4) to new pci_machdep.c file.
- Fix style.

Add driver for Prometheus PCI bridge.
Change approved by phx.

Add driver for Prometheus PCI bridge.
Change approved by phx.

Remove unnecessary <sys/malloc.h> include.

<sys/extent.h> not needed here.

split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.2.4; 1.2.6; 1.2.10;
Tons of enchancements to MD PCI code:
- The G-REX 4000 and G-REX 1200 are now really supported and working (still only 1st slot but the rest are a matter of time).
- Add bridge identification to p5pb, simplifies the driver and make it more user-friendly.
- Retire P5PB_GENFB and P5PB_OLD_FIRMWARE options, these are now handled automatically, depending on used hardware and firmware.
- Add PCI bus reconfiguration support to p5pb (PCI_NETBSD_CONFIGURE), allows using G-REX without boot to AmigaOS.
- Bring p5pb registers desription up to date with my knowledge.
- Add a lot of debugging code to p5pb (enabled by P5PB_DEBUG kernel option).
- Fix PCI_NETBSD_CONFIGURE in mppb (now it really works).
- Fix quirk handling in zbus.
- Update GENERIC.in and regen the kernel configs.
- Misc cleanup.

Changes to Phase5 hardware support:
- Rework p5pb driver - simplify, cleanup, make more flexible.
- Add p5membar driver, which handles PCI resources autoconfigured by the firmware.
- Introduce intermediate p5bus layer, between zbus and CSPPC/BPPC on-board devices (p5pb, cbiiisc, bppcsc).
- Add experimental G-REX support to p5pb (first slot support only).
- Split CV64/3D PCI bridge support into separate cv3dpb driver (to be committed later).

Approved by phx.

branches: 1.1.4; 1.1.6;
Changes to Phase5 hardware support:
- Rework p5pb driver - simplify, cleanup, make more flexible.
- Add p5membar driver, which handles PCI resources autoconfigured by the firmware.
- Introduce intermediate p5bus layer, between zbus and CSPPC/BPPC on-board devices (p5pb, cbiiisc, bppcsc).
- Add experimental G-REX support to p5pb (first slot support only).
- Split CV64/3D PCI bridge support into separate cv3dpb driver (to be committed later).

Approved by phx.

Remove unnecessary <sys/malloc.h> include.

amiga/p5pb(4): move parent_dict definition/assignment to #if NGENFB > 0 block,
where it is only used in P5PB_CONSOLE code.

fixes P5PB_CONSOLE enabled build, no functional changes intended.

Merge thorpej-cfargs2.

branches: 1.18.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.17.4;
Overhaul the interface to pci_configure_bus():
- Don't expose how PCI bus configuration resource management is implemented.
Provide a new resource provider API:

==> pciconf_resource_init() -- Initialize a PCI configuration resources
container.
==> pciconf_resource_add() -- Add a PCI configuration resource to the
container (I/O, MEM, or prefetchable MEM). Multiple resources of
each type may be added.
==> pciconf_resource_fini() -- Tear down the PCI configurtation resources
container once the bus has been configured.

This is much easier to use than the previous method of providing an
extent map for each kind of resource, and works better for e.g. ACPI
platforms that provide potentially multiple PCI resources in tables
provided by firmware.

- Re-implement PCI configuration resource management using vmem arenas,
rather than extent maps.

replace EX_NOWAIT with EX_WAITOK in device attach methods.
remove checks for failures that can no longer occur.

Include genfb.h for NGENFB.

PCI Extended Configuration stuff written by nonaka@:
- Add PCI Extended Configuration Space support into x86.
- Check register offset of pci_conf_read() in MD part. It returns (pcireg_t)-1
if it isn't accessible.
- Decode Extended Capability in PCI Extended Configuration Space.
Currently the following extended capabilities are decoded:
- Advanced Error Reporting
- Virtual Channel
- Device Serial Number
- Power Budgeting
- Root Complex Link Declaration
- Root Complex Event Collector Association
- Access Control Services
- Alternative Routing-ID Interpretation
- Address Translation Services
- Single Root IO Virtualization
- Page Request
- TPH Requester
- Latency Tolerance Reporting
- Secondary PCI Express
- Process Address Space ID
- LN Requester
- L1 PM Substates
The following extended capabilities are not decoded yet:
- Root Complex Internal Link Control
- Multi-Function Virtual Channel
- RCRB Header
- Vendor Unique
- Configuration Access Correction
- Multiple Root IO Virtualization
- Multicast
- Resizable BAR
- Dynamic Power Allocation
- Protocol Multiplexing
- Downstream Port Containment
- Precision Time Management
- M-PCIe
- Function Reading Status Queueing
- Readiness Time Reporting
- Designated Vendor-Specific

branches: 1.13.6;
kill sprintf
fix unused variables

branches: 1.12.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.11.2;
Clean up p5pb console glue a bit.

Fix incomplete bus_dma tag.

Add support for all 4 slots of G-REX 4000D and G-REX 1200. Also add
(untested yet) DMA support for G-REX bridges.

adapt to extent(9) changes

Add voodoofb support. Allows using 3Dfx Voodoo 3 in G-REX as a console.

Tons of enchancements to MD PCI code:
- The G-REX 4000 and G-REX 1200 are now really supported and working (still only 1st slot but the rest are a matter of time).
- Add bridge identification to p5pb, simplifies the driver and make it more user-friendly.
- Retire P5PB_GENFB and P5PB_OLD_FIRMWARE options, these are now handled automatically, depending on used hardware and firmware.
- Add PCI bus reconfiguration support to p5pb (PCI_NETBSD_CONFIGURE), allows using G-REX without boot to AmigaOS.
- Bring p5pb registers desription up to date with my knowledge.
- Add a lot of debugging code to p5pb (enabled by P5PB_DEBUG kernel option).
- Fix PCI_NETBSD_CONFIGURE in mppb (now it really works).
- Fix quirk handling in zbus.
- Update GENERIC.in and regen the kernel configs.
- Misc cleanup.

Changes to Phase5 hardware support:
- Rework p5pb driver - simplify, cleanup, make more flexible.
- Add p5membar driver, which handles PCI resources autoconfigured by the firmware.
- Introduce intermediate p5bus layer, between zbus and CSPPC/BPPC on-board devices (p5pb, cbiiisc, bppcsc).
- Add experimental G-REX support to p5pb (first slot support only).
- Split CV64/3D PCI bridge support into separate cv3dpb driver (to be committed later).

Approved by phx.

branches: 1.4.2; 1.4.6;
P5PB_MATCH_CV643D is supposed to be disabled by default

Added experimental support for CV64/3D PCI bridge to p5pb(4). Approved by phx.

Minor changes to Amiga PCI MD code:
- Add bus enumeration and resource assignment (PCI_NETBSD_CONFIGURE) to mppb(4) - no more need to run AmigaOS to configure the bus.
- Add PCI configuration address register to p5pb(4).
- Add PCI I/O space support to p5pb(4).
- Move common code from p5pb(4) and mppb(4) to new pci_machdep.c file.
- Fix style.

Add basic PCI support for amiga port.
Add missing bus_space(9) methods needed for MI PCI.
Add p5pb(4) - Phase5 PCI bridge driver (and the man page).
Add quirks table to zbus(4) - needed for p5pb.
Change approved by phx.

Add support for all 4 slots of G-REX 4000D and G-REX 1200. Also add
(untested yet) DMA support for G-REX bridges.

Tons of enchancements to MD PCI code:
- The G-REX 4000 and G-REX 1200 are now really supported and working (still only 1st slot but the rest are a matter of time).
- Add bridge identification to p5pb, simplifies the driver and make it more user-friendly.
- Retire P5PB_GENFB and P5PB_OLD_FIRMWARE options, these are now handled automatically, depending on used hardware and firmware.
- Add PCI bus reconfiguration support to p5pb (PCI_NETBSD_CONFIGURE), allows using G-REX without boot to AmigaOS.
- Bring p5pb registers desription up to date with my knowledge.
- Add a lot of debugging code to p5pb (enabled by P5PB_DEBUG kernel option).
- Fix PCI_NETBSD_CONFIGURE in mppb (now it really works).
- Fix quirk handling in zbus.
- Update GENERIC.in and regen the kernel configs.
- Misc cleanup.

Add cv3dpb, a driver for CyberVision64/3D PCI bridge, which was split from p5pb.

Changes to Phase5 hardware support:
- Rework p5pb driver - simplify, cleanup, make more flexible.
- Add p5membar driver, which handles PCI resources autoconfigured by the firmware.
- Introduce intermediate p5bus layer, between zbus and CSPPC/BPPC on-board devices (p5pb, cbiiisc, bppcsc).
- Add experimental G-REX support to p5pb (first slot support only).
- Split CV64/3D PCI bridge support into separate cv3dpb driver (to be committed later).

Approved by phx.

branches: 1.3.2; 1.3.6;
Added experimental support for CV64/3D PCI bridge to p5pb(4). Approved by phx.

Minor changes to Amiga PCI MD code:
- Add bus enumeration and resource assignment (PCI_NETBSD_CONFIGURE) to mppb(4) - no more need to run AmigaOS to configure the bus.
- Add PCI configuration address register to p5pb(4).
- Add PCI I/O space support to p5pb(4).
- Move common code from p5pb(4) and mppb(4) to new pci_machdep.c file.
- Fix style.

Add basic PCI support for amiga port.
Add missing bus_space(9) methods needed for MI PCI.
Add p5pb(4) - Phase5 PCI bridge driver (and the man page).
Add quirks table to zbus(4) - needed for p5pb.
Change approved by phx.

Clean up p5pb console glue a bit.

Tons of enchancements to MD PCI code:
- The G-REX 4000 and G-REX 1200 are now really supported and working (still only 1st slot but the rest are a matter of time).
- Add bridge identification to p5pb, simplifies the driver and make it more user-friendly.
- Retire P5PB_GENFB and P5PB_OLD_FIRMWARE options, these are now handled automatically, depending on used hardware and firmware.
- Add PCI bus reconfiguration support to p5pb (PCI_NETBSD_CONFIGURE), allows using G-REX without boot to AmigaOS.
- Bring p5pb registers desription up to date with my knowledge.
- Add a lot of debugging code to p5pb (enabled by P5PB_DEBUG kernel option).
- Fix PCI_NETBSD_CONFIGURE in mppb (now it really works).
- Fix quirk handling in zbus.
- Update GENERIC.in and regen the kernel configs.
- Misc cleanup.

Changes to Phase5 hardware support:
- Rework p5pb driver - simplify, cleanup, make more flexible.
- Add p5membar driver, which handles PCI resources autoconfigured by the firmware.
- Introduce intermediate p5bus layer, between zbus and CSPPC/BPPC on-board devices (p5pb, cbiiisc, bppcsc).
- Add experimental G-REX support to p5pb (first slot support only).
- Split CV64/3D PCI bridge support into separate cv3dpb driver (to be committed later).

Approved by phx.

branches: 1.1.2; 1.1.6;
Added experimental support for CV64/3D PCI bridge to p5pb(4). Approved by phx.

Remove unnecessary <sys/malloc.h> include.

<sys/extent.h> not needed here.

make pci_intr_string and eisa_intr_string take a buffer and a length
instead of relying in local static storage.

kill sprintf
fix unused variables

branches: 1.1.2; 1.1.12; 1.1.16;
Minor changes to Amiga PCI MD code:
- Add bus enumeration and resource assignment (PCI_NETBSD_CONFIGURE) to mppb(4) - no more need to run AmigaOS to configure the bus.
- Add PCI configuration address register to p5pb(4).
- Add PCI I/O space support to p5pb(4).
- Move common code from p5pb(4) and mppb(4) to new pci_machdep.c file.
- Fix style.