Clean up deleted files.

new RCS ID format.

branches: 1.1.2;
Add PCI autoconfiguration support.

move to x86/pci/agp_machdep.c from Blair Sadewitz

branches: 1.4.22; 1.4.52; 1.4.54;
New interrupt code. The basic idea behind it is to hide the differences
in interrupt controllers in struct pic, and try to keep as much
common code as possible. At the lowest (asm) level, this is done
with CPP macros.

The main structure is now struct intrsource, describing an established
interrupt line, of any kind (soft/hard local apic/legacy apic/IO apic).
For quick masking, there may be a maximum of 32 sources per CPU.
Sources can be assigned to any CPU in the MP case, though currently they
all go to the boot CPU.

Merge Bill Sommerfeld's i386 MP branch. This code has some known
caveats, but works quite well in a lot of MP cases, and all
UP cases that I have tested. Parts of this will hopefully be
reworked in the not-too-distant future.

branches: 1.2.2;
add RCSID

branches: 1.1.2; 1.1.4; 1.1.8;
Attach agp gart support @ pchb. Not very clean, but agp support may
be spread over several devices, and the phcb is usually the main one.

Add agp_machdep.c file which implements MD agp functions (currently
just agp_flush_cache).

Remove unnecessary <sys/malloc.h>.

#include <sys/bus.h> instead of <machine/bus.h>.

Move to 2-clause license. Approved by HAYAKAWA Koichi (copyright holder).

branches: 1.10.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.9.52; 1.9.54; 1.9.56;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.7.20; 1.7.22;
merge ktrace-lwp.

de-__P, partially ANSIfy

branches: 1.5.2; 1.5.10; 1.5.12;
Adapt for i386/x86 change.

Put the storage class first.

branches: 1.3.2;
add RCSID

branches: 1.2.2; 1.2.8;
correct the set/get trigger code for ALi M1543 interrupt router.

branches: 1.1.4; 1.1.6;
Add support for ALi M1543 in pcibios.

#include <sys/bus.h> instead of <machine/bus.h>.

branches: 1.8.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.7.52; 1.7.54; 1.7.56;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.5.20; 1.5.22;
merge ktrace-lwp.

de __P, ANSIfy PCIBIOS related files.

branches: 1.3.2;
Adapt for i386/x86 change.

branches: 1.2.2;
add RCSID

branches: 1.1.2; 1.1.4; 1.1.8; 1.1.10;
patch PR port-i386/11114 by MINOURA Makoto.

Remove clause 3 and 4 from TNF licenses

branches: 1.1.2; 1.1.4; 1.1.6; 1.1.8; 1.1.132; 1.1.134; 1.1.136;
patch PR port-i386/11114 by MINOURA Makoto.

Merge thorpej-cfargs2.

branches: 1.51.8;
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: 1.50.44;
fix unused variable warnings

branches: 1.49.2; 1.49.12; 1.49.16;
#include <sys/bus.h> instead of <machine/bus.h>.

Fixes regarding paddr_t/pd_entry_t types in MD x86 code, exposed by PAE:

- NBPD_* macros are set to the types that better match their architecture
(UL for i386 and amd64, ULL for i386 PAE) - will revisit when paddr_t is
set to 64 bits for i386 non-PAE.

- type fixes in printf/printk messages (Use PRIxPADDR when printing paddr_t
values, instead of %lx - paddr_t/pd_entry_t being 64 bits with PAE)

- remove casts that are no more needed now that Xen2 support has been dropped

Some fixes are from jmorse@ patches for PAE.

Compile + tested for i386 GENERIC and XEN3 kernels. Only compile tested for
amd64.

Reviewed by bouyer@.

See also http://mail-index.netbsd.org/tech-kern/2010/02/22/msg007373.html

A pointer typedef entails trading too much flexibility to declare const
and non-const types, and the kernel uses both const and non-const
PMF qualifiers and device suspensors, so change the pmf_qual_t and
device_suspensor_t typedefs from "pointers to const" to non-pointer,
non-const types.

branches: 1.46.2;
Expand PMF_FN_* macros.

Move all copies of ifattr_match() to sys/kern/subr_autoconf.c.

KNF: define struct pareg near the top of the file.

As Jonathan Kollasch points out, elansc_attach() does not actually
use gba after filling it out. Delete it.

Add elansc_shutdown() for turning off the watchdog.

Don't try to unregister the watchdog in elansc_detach() if we are
shutting the system down, because unregistering is designed to fail if
the watchdog is armed.

Delete dead code.

Move elanpar0 and elanpex0 attachment out of the #if NGPIO > 0 block in
elansc_attach().

Let us detach gpio0 at elansc0.

Let us re-attach pci0 at elansc0.

Reduce code duplication between elansc_rescan() and elansc_attach():
call _rescan() from _attach() to attach elanpex0, elanpar0, gpio0, and
pci0.

During shutdown, detach devices in an orderly fashion.

Call the detach routine for every device in the device tree, starting
with the leaves and moving toward the root, expecting that each
(pseudo-)device driver will use the opportunity to gracefully commit
outstandings transactions to the underlying (pseudo-)device and to
relinquish control of the hardware to the system BIOS.

Detaching devices is not suitable for every shutdown: in an emergency,
or if the system state is inconsistent, we should resort to a fast,
simple shutdown that uses only the pmf(9) shutdown hooks and the
(deprecated) shutdownhooks. For now, if the flag RB_NOSYNC is set in
boothowto, opt for the fast, simple shutdown.

Add a device flag, DVF_DETACH_SHUTDOWN, that indicates by its presence
that it is safe to detach a device during shutdown. Introduce macros
CFATTACH_DECL3() and CFATTACH_DECL3_NEW() for creating autoconf
attachments with default device flags. Add DVF_DETACH_SHUTDOWN
to configuration attachments for atabus(4), atw(4) at cardbus(4),
cardbus(4), cardslot(4), com(4) at isa(4), elanpar(4), elanpex(4),
elansc(4), gpio(4), npx(4) at isa(4), nsphyter(4), pci(4), pcib(4),
pcmcia(4), ppb(4), sip(4), wd(4), and wdc(4) at isa(4).

Add a device-detachment "reason" flag, DETACH_SHUTDOWN, that tells the
autoconf code and a device driver that the reason for detachment is
system shutdown.

Add a sysctl, kern.detachall, that tells the system to try to detach
every device at shutdown, regardless of any device's DVF_DETACH_SHUTDOWN
flag. The default for kern.detachall is 0. SET IT TO 1, PLEASE, TO
HELP TEST AND DEBUG DEVICE DETACHMENT AT SHUTDOWN.

This is a work in progress. In future work, I aim to treat
pseudo-devices more thoroughly, and to gracefully tear down a stack of
(pseudo-)disk drivers and filesystems, including cgd(4), vnd(4), and
raid(4) instances at shutdown.

Also commit some changes that are not easily untangled from the rest:

(1) begin to simplify device_t locking: rename struct pmf_private to
device_lock, and incorporate device_lock into struct device.

(2) #include <sys/device.h> in sys/pmf.h in order to get some
definitions that it needs. Stop unnecessarily #including <sys/device.h>
in sys/arch/x86/include/pic.h to keep the amd64, xen, and i386 releases
building.

- rewrite x86 nmi dispatcher so that establish and disesablish are safe
on a running system.
- adapt existing users of the api. (elan)
- adapt tprof_pmi driver to use the api.

branches: 1.37.2;
When elansc(4) is detached, restore BIOS settings on the SC520's
programmable interrupt controller (PIC).

So that I can both detach and *re-attach* elanpar0, elanpex0, and
gpio0, let elansc(4) rescan for them.

branches: 1.35.6;
Use printf_tolog() instead of aprint_error_dev() to report PCI
exceptions and write-protection violations.

While I'm here, add a comment to describe the interrupt configuration.

Add a 'known_mpsafe' argument to intr_establish().

branches: 1.33.2;
Release mutex before unmapping bus space to stop a locking error.

Remove clause 3 and 4 from TNF licenses

branches: 1.31.2; 1.31.4;
Attach elansc(4) at mainbus(4) instead of at pci(4). Attach pci(4)
at elansc(4). Take advantage of a suspend/resume cycle to reconfigure
the SC520's PCI host-bridge bus for higher performance. Update
the manual pages and the NET4501 kernel.

Defer write-protecting the kernel text until after interrupts are
enabled. This is a grotty hack that lets a MULTIPROCESSOR kernel
patch the kernel with x86_patch() without trapping to the debugger.

Complete device_t/softc split.

Define fkb and sfkb once instead of in both elansc_protect_text()
and elansc_protect(). Should stop GCC from complaining about an
unused variable fkb.

Use as many as two Programmable Address Regions (PARs) with 4kB
granularity to protect some of the kernel text that is not protected
by the PAR with 64kB granularity.

Extract shutdown handlers from the detach methods of elanpar(4)
and elanpex(4). Register them using pmf_device_register1(). This
ensures that neither comBIOS nor NetBSD will hang during boot,
after I reboot a Soekris net45xx.

Fix a typo that could lead to a panic when I suspend elanpar(4).

Use PMF_FN_ARGS, PMF_FN_PROTO.

branches: 1.23.2; 1.23.6;
Adapt the code for write-protecting the Interrupt Descriptor Table,
which wasn't such a good idea, to protect physical page 0, instead.

Attach two new devices to the AMD Elan SC520 System Controller,
elansc(4).

elanpex(4) is for PCI exception reporting. I've already found some
kernel bugs by reading the exceptions reported. Beware that it
will spam the console a lot while the kernel and pcictl(8) probe
non-existing addresses in PCI configuration space.

elanpar(4) protects the kernel text from writes by the CPU and by
PCI bus masters. As you might guess, this is not compatible with
setting breakpoints using a debugger; detach the device using
'drvctl -d elanpar0' before you try to set breakpoints. In the
future, I hope to extend elanpar(4) to provide general-purpose RAM
write-protection.

Use aprint_*_dev(), device_xname(), and device_t. Join a line.

Get rid of sc_suspended, it is redundant because the device_t
contains the same state.

Miscellaneous changes designed to improve elansc suspend/detachment:

1 Synchronize access to the softc and hardware with a mutex.
2 Add the sc_suspended flag, and observe it in the watchdog routines.
3 Re-order operations in elansc_attach() and elansc_detach() to
protect against access while the device is not fully configured.
4 Cancel detachment if sysmon_wdog_unregister() is cancelled by a
signal.
5 Add an empty childdetached method so that I can detach gpio0 at
elansc0. (XXX cannot individually re-attach gpio0.)

Add a detach routine.

Add a suspend/resume handler. Do not suspend if the watchdog timer
is active. Reinitialize the watchdog timer on resume. If the
device is not powered, neither tickle the watchdog nor set the
watchdog's mode, but return EBUSY, instead.

branches: 1.16.28; 1.16.34; 1.16.38; 1.16.42;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.14.14; 1.14.16;
Use aprint_*().

branches: 1.13.2; 1.13.4; 1.13.6;
u_intN_t -> uintN_t

merge ktrace-lwp.

Move a little more under #if NGPIO > 0

-put gpio stuff inside #if NGPIO > 0
-kill gba_name in attach arg, specify interface attribute

Add support for the gpio(4) framework to the elansc(4) system controller.

From Alexander Yurchenko (grange@openbsd), via OpenBSD.
Approved by jmcneill.

branches: 1.8.16;
netbsd.org -> NetBSD.org

All "netbsd.org" is now gone from src/sys/arch.

Fix uninitialized variable warnings.

Fix uninitialized variable warnings

branches: 1.5.2;
Use PAGE_SIZE rather than NBPG.

Tidy up CFATTACH_DECL() formatting.

Merge Bill Sommerfeld's i386 MP branch. This code has some known
caveats, but works quite well in a lot of MP cases, and all
UP cases that I have tested. Parts of this will hopefully be
reworked in the not-too-distant future.

Declare all cfattach structures const.

branches: 1.1.2; 1.1.4; 1.1.6; 1.1.8;
Add a driver for the AMD Elan SC520 System Controller. The "elansc"
driver attaches where "pchb" would normally attach (it matches at a
higher match priority). The "elansc" driver currently provides support
for the watchdog timer built-in the SC520.

Thanks to Jasper Wallace for laying the ground-work for this (most
notably by providing a work-around for a watchdog-related bug in the
SC520).

Remove clause 3 and 4 from TNF licenses

branches: 1.9.6; 1.9.8; 1.9.10;
Define some "special" values of register MMCR_WPVMAP.

Add definitions for several more registers on the AMD Elan SC520
System Controller.

branches: 1.7.2; 1.7.4; 1.7.8;
Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

branches: 1.6.2;
Add offset for Host Bridge Target Interrupt Control register.

branches: 1.5.2; 1.5.6; 1.5.10;
Add definitions for the AMD Elan 520 Host Bridge Master Interrupt
Control register.

branches: 1.4.2; 1.4.6; 1.4.8;
Define some more registers on the AMD Elan SC520, according to the
datasheet.

branches: 1.3.20;
merge ktrace-lwp.

Add support for the gpio(4) framework to the elansc(4) system controller.

From Alexander Yurchenko (grange@openbsd), via OpenBSD.
Approved by jmcneill.

branches: 1.1.2; 1.1.4; 1.1.6; 1.1.8; 1.1.14; 1.1.30;
Add a driver for the AMD Elan SC520 System Controller. The "elansc"
driver attaches where "pchb" would normally attach (it matches at a
higher match priority). The "elansc" driver currently provides support
for the watchdog timer built-in the SC520.

Thanks to Jasper Wallace for laying the ground-work for this (most
notably by providing a work-around for a watchdog-related bug in the
SC520).

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: 1.13.16;
Stop potential misuse of vendor names and USB vendor IDs in root hub
device and string descriptors.

Firstly: Few vendors have identical PCI-SIG vendor IDs and USB-IF vendor
IDs. As such, using the PCI vendor ID as a USB vendor ID may trample
on whomever is allocated that USB vendor ID.

Secondly: The vendor of the host controller hardware implementation has
little to nothing to do with our usbroothub implementation. Thus we
should not potentially associate any problems therewith to such third
party.

This change will result in root hubs being identified by USB Vendor ID
0x0000. Root hub vendor string will now be "NetBSD" (or, specifically:
ostype). Product ID (0x0000) and product strings remain unchanged.

branches: 1.12.16;
Merge nick-nhusb

- API / infrastructure changes to support memory management changes.
- Memory management improvements and bug fixes.
- HCDs should now be MP safe
- conversion to KERNHIST based debug
- FS/LS isoc support on ehci(4).
- conversion to kmem(9)
- Some USB 3 support - mostly from Takahiro HAYASHI (t-hash).
- interrupt transfers now get proper DMA operations
- general bug fixes
- kern/48308
- uhub status notification improvements
- umass(4) probe fix (applied to HEAD already)
- ohci(4) short transfer fix

branches: 1.11.2;
adjust for the new findvendor

branches: 1.10.4; 1.10.8;
make pci_intr_string and eisa_intr_string take a buffer and a length
instead of relying in local static storage.

branches: 1.9.2; 1.9.12; 1.9.16;
#include <sys/bus.h> instead of <machine/bus.h>.

Revert earlier changes to callers of pci_find{vendor,product} since
the original routine entrypoint name has been restored.

Requested by dyoung@ and mrg@

Update all callers of the pci_find{vendor,product} routines to now call
these routines through their global pointers.

Cleanup interrupt establish error messages. Do not mix
aprint_error/aprint_normal/printf calls for a single line.

struct device * -> device_t, use device_private()
No functional changes intended.

branches: 1.4.10;
ryo shimizu: gscsehci doesn't work.

Correctly initialize softc. Fix from PR port-i386/39109.

branches: 1.3.2; 1.3.4;
Remove clause 3 and 4 from TNF licenses

branches: 1.2.2; 1.2.4;
split device/softc for USB host controllers and the usb (control)
device,
this is hairy stuff, and I've only tested with uhci/ehci at pci,
please test the rest and report problems

branches: 1.1.2; 1.1.4; 1.1.16; 1.1.36;
Add gcscehci(4), an EHCI glue driver for the AMD CS5536 companion
controller. This driver enables USB 2.0 high speed support on the
Soekris net5501 (comBIOS ver. 1.31m).

KNF. No functional change.

branches: 1.15.6;
Use the universal pciide_detach() for detach function of all remaining
pciide-core drivers, except pciide at pnpbios which isn't actually PCI.

branches: 1.14.2; 1.14.4;
Apply back changes that were reverted on Jul 24 and Jul 26 (general ata/wdc
cleanup and SATA PMP support), now that I'm back to fix the fallouts.

Revert, with intention of restoring in a less invasive way, the SATA Port
Multiplier code.

ok christos@

Revert dsl@'s changes of Sun, 15 Jul 2012 10:55:35 +0000 and
Sun, 15 Jul 2012 10:56:50 +0000, excepting the kernel version bump.
First step in reverting regressions to ata(4) subsystem during the addition of
port multiplier support.

Some namespace protection (and add greppablity).
Prefix the DRIVE_ and DRIVET_ constants from atavar.h with ATA_.
Don't use an enum for drive_type - you don't know how big it will be.
Move driver_type to avoid implicit structure padding (esp on arm).
This change is purely lexical and mechanical.

Update to 6.99.9 - this wasn't done when the SATA PMP changes
were made - I'm sure they warranted a bump.

Add sata Port MultiPlier (PMP) support to the ata bus layer,
as described in
http://mail-index.netbsd.org/tech-kern/2012/06/23/msg013442.html
PMP support in integrated to the atabus layer.
struct ata_channel's ch_drive[] is not dynamically allocated, and ch_ndrive
(renamed to ch_ndrives) closely reflects the size of the ch_drive[] array.
Add helper functions atabus_alloc_drives() and atabus_free_drives()
to manage ch_drive[]/ch_ndrives.
Add wdc_maxdrives to struct wdc_softc so that bus front-end can specify
how much drive they really support (master/slave or single).
ata_reset_drive() callback gains a uint32_t *sigp argument which,
when not NULL, will contain the signature of the device being reset.
While there, some cosmetic changes:
- added a drive_type enum to ata_drive_datas, and stop encoding the
probed drive type in drive_flags (we were out of drive flags anyway).
- rename DRIVE_ATAPIST to DRIVE_ATAPIDSCW to better reflect what this
really is
- remove ata_channel->ata_drives, it's redundant with the pointer in
ata_drive_datas
- factor out the interpretation of SATA signatures in sata_interpet_sig()

propagate these changes to the ATA HBA drivers, and add support for PMP
to ahcisata(4) and siisata(4).

Thanks to:
- Protocase (http://www.protocase.com/) which provided a system
with lots of controllers, SATA PMP and drive slots
- Conservation Genomics Laboratory, Department of Biology, New Mexico State
University for hosting the above system
- Brook Milligan, who set up remote access and has been very responsive
when SATA cable move was needed

branches: 1.9.4;
Neither pci_dma64_available(), pci_probe_device(), pci_mapreg_map(9),
pci_find_rom(), pci_intr_map(9), pci_enumerate_bus(), nor the match
predicate passed to pciide_compat_intr_establish() should ever modify
their pci_attach_args argument, so make their pci_attach_args arguments
const and deal with the fallout throughout the kernel.

For the most part, these changes add a 'const' where there was no
'const' before, however, some drivers and MD code used to modify
pci_attach_args. Now those drivers either copy their pci_attach_args
and modify the copy, or refrain from modifying pci_attach_args:

Xen: according to Manuel Bouyer, writing to pci_attach_args in
pci_intr_map() was a leftover from Xen 2. Probably a bug. I
stopped writing it. I have not tested this change.

siside(4): sis_hostbr_match() needlessly wrote to pci_attach_args.
Probably a bug. I use a temporary variable. I have not tested this
change.

slide(4): sl82c105_chip_map() overwrote the caller's pci_attach_args.
Probably a bug. Use a local pci_attach_args. I have not tested
this change.

viaide(4): via_sata_chip_map() and via_sata_chip_map_new() overwrote the
caller's pci_attach_args. Probably a bug. Make a local copy of the
caller's pci_attach_args and modify the copy. I have not tested
this change.

While I'm here, make pci_mapreg_submap() static.

With these changes in place, I have tested the compilation of these
kernels:

alpha GENERIC
amd64 GENERIC XEN3_DOM0
arc GENERIC
atari HADES MILAN-PCIIDE
bebox GENERIC
cats GENERIC
cobalt GENERIC
evbarm-eb NSLU2
evbarm-el ADI_BRH ARMADILLO9 CP3100 GEMINI GEMINI_MASTER GEMINI_SLAVE GUMSTIX
HDL_G IMX31LITE INTEGRATOR IQ31244 IQ80310 IQ80321 IXDP425 IXM1200
KUROBOX_PRO LUBBOCK MARVELL_NAS NAPPI SHEEVAPLUG SMDK2800 TEAMASA_NPWR
TEAMASA_NPWR_FC TS7200 TWINTAIL ZAO425
evbmips-el AP30 DBAU1500 DBAU1550 MALTA MERAKI MTX-1 OMSAL400 RB153 WGT624V3
evbmips64-el XLSATX
evbppc EV64260 MPC8536DS MPC8548CDS OPENBLOCKS200 OPENBLOCKS266
OPENBLOCKS266_OPT P2020RDB PMPPC RB800 WALNUT
hp700 GENERIC
i386 ALL XEN3_DOM0 XEN3_DOMU
ibmnws GENERIC
macppc GENERIC
mvmeppc GENERIC
netwinder GENERIC
ofppc GENERIC
prep GENERIC
sandpoint GENERIC
sgimips GENERIC32_IP2x
sparc GENERIC_SUN4U KRUPS
sparc64 GENERIC

As of Sun Apr 3 15:26:26 CDT 2011, I could not compile these kernels
with or without my patches in place:

### evbmips-el GDIUM

nbmake: nbmake: don't know how to make /home/dyoung/pristine-nbsd/src/sys/arch/mips/mips/softintr.c. Stop

### evbarm-el MPCSA_GENERIC
src/sys/arch/evbarm/conf/MPCSA_GENERIC:318: ds1672rtc*: unknown device `ds1672rtc'

### ia64 GENERIC

/tmp/genassym.28085/assym.c: In function 'f111':
/tmp/genassym.28085/assym.c:67: error: invalid application of 'sizeof' to incomplete type 'struct pcb'
/tmp/genassym.28085/assym.c:76: error: dereferencing pointer to incomplete type

### sgimips GENERIC32_IP3x

crmfb.o: In function `crmfb_attach':
crmfb.c:(.text+0x2304): undefined reference to `ddc_read_edid'
crmfb.c:(.text+0x2304): relocation truncated to fit: R_MIPS_26 against `ddc_read_edid'
crmfb.c:(.text+0x234c): undefined reference to `edid_parse'
crmfb.c:(.text+0x234c): relocation truncated to fit: R_MIPS_26 against `edid_parse'
crmfb.c:(.text+0x2354): undefined reference to `edid_print'
crmfb.c:(.text+0x2354): relocation truncated to fit: R_MIPS_26 against `edid_print'

branches: 1.8.2;
Pave the way for detachment of pciide(4)-family ATA controllers.

branches: 1.7.18; 1.7.24; 1.7.26;
Split device_t and softc for ATA devices, as well as wd(4). Other
cosmetic changes where appropriate.

branches: 1.6.14; 1.6.18;
Use a two clause license for all the code I contributed.

The envsys code will be changed later.

branches: 1.5.2; 1.5.4;
- Always use PIO Format 1, which allows independent control of command
and data (PIO Format 0 selects the speed of the slowest drive).
- Set correctly Ultra DMA and Multi-word DMA modes.
- Update comments to be more correct after reading again the datasheet.

branches: 1.4.2; 1.4.4; 1.4.8; 1.4.12; 1.4.14; 1.4.16; 1.4.18;
Set the PIO mode to the correct drive not for drive1 always.

Improve a comment.

Fix a typo and remove unneeded assignments.

Import i386/gcscide(4). A driver for the IDE Controller of the
AMD CS5535 Companion device found in the decTOP.

gcscide0 at pci0 dev 15 function 2
gcscide0: National Semiconductor/AMD CS5535 IDE Controller (rev. 0x00)

Supports Ultra DMA mode 4, Pio Mode 4 and MDMA mode 2.

"Go for it" jmcneill@.

Split gcscpcib into MI part, and MD pci attachement which is also in
charge of attaching the MD pcib device.
Will be used by the upcoming evbmips loongson support.

Make this build without gpio(4).

struct cfdata * -> cfdata_t

branches: 1.6.8;
Fix previous.

Explicitly add struct pcib_softc to the softc since the pcib functions
we call expect this.

branches: 1.4.2; 1.4.4; 1.4.6; 1.4.12; 1.4.14; 1.4.18; 1.4.20; 1.4.22; 1.4.24;
Remove the unused mutex from softc.

emo-style(9).

Fix up some comments, from Yojiro UO in private email.

Add gcscpcib(4): a driver for the AMD CS5535 and CS5536 Companion Device
that acts as PCI-ISA bridge and supports a Timecounter, Watchdog Timer
and GPIO.

Adapted by Yojiro UO and minor tweaks by me from OpenBSD. Tested on
CS5535 and CS5536. This closes PR kern/37577.

s/attachement/attachment/ and s/detachement/detachment/.

Add a flags argument to gcscpcib_attach().
Add a flag argument which cause the waychdog to not be attached.
Use it to disable the watchdog on loongson-based platforms, where
the CS5536's reset output is not wired to the CPU reset line :(

Split gcscpcib into MI part, and MD pci attachement which is also in
charge of attaching the MD pcib device.
Will be used by the upcoming evbmips loongson support.

Split gcscpcib into MI part, and MD pci attachement which is also in
charge of attaching the MD pcib device.
Will be used by the upcoming evbmips loongson support.

branches: 1.1.2; 1.1.4; 1.1.6; 1.1.14;
Add gcscpcib(4): a driver for the AMD CS5535 and CS5536 Companion Device
that acts as PCI-ISA bridge and supports a Timecounter, Watchdog Timer
and GPIO.

Adapted by Yojiro UO and minor tweaks by me from OpenBSD. Tested on
CS5535 and CS5536. This closes PR kern/37577.

Merge thorpej-cfargs2.

branches: 1.16.8;
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: 1.15.68;
#include <sys/bus.h> instead of <machine/bus.h>.

Drop 3rd and 4th clauses from David Young's license.
Reviewed and approved by dyoung@ (copyright holder).

branches: 1.13.12;
Use aprint_naive, cleanup error printing.

branches: 1.12.2;
device_t/softc split and other related cosmetic changes.

Remove clause 3 and 4 from TNF licenses

branches: 1.10.2; 1.10.4;
use aprint_*_dev and device_xname
OK joerg

branches: 1.9.6;
Support detachment of geodegcb(4) and its children, geodecntr(4)
and geodewdog(4). Support detachment of gscpcib0 and its child,
gpio0.

branches: 1.8.28; 1.8.34; 1.8.42;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.6.4; 1.6.8; 1.6.10;
convert to timecounters (from branch simonb-timecounters)

branches: 1.5.4; 1.5.6;
Move my bit-twiddling macros to libkern.h from my drivers, where
I had duplicated them. Improve the macros' names. Simplify their
implementation.

A brief description of each macro is below.

BIT(n): Return a bitmask with bit m set, where the least
significant bit is bit 0.

BITS(m, n): Return a bitmask with bits m through n, inclusive,
set. It does not matter whether m>n or m<=n.
The least significant bit is bit 0.

A "bitfield" is a span of consecutive bits defined by a
bitmask, where 1s select the bits in the bitfield. SHIFTIN,
SHIFTOUT, and SHIFTOUT_MASK help read and write bitfields
from device registers.

SHIFTIN(v, mask): Left-shift bits `v' into the bitfield
defined by `mask', and return them. No
side-effects.

SHIFTOUT(v, mask): Extract and return the bitfield selected
by `mask' from `v', right-shifting the
bits so that the rightmost selected bit
is at bit 0. No side-effects.

SHIFTOUT_MASK(mask): Right-shift the bits in `mask' so that
the rightmost non-zero bit is at bit
0. This is useful for finding the
greatest unsigned value that a bitfield
can hold. No side-effects. Note that
SHIFTOUT_MASK(m) = SHIFTOUT(m, m).

Examples:

/*
* Register definitions taken from the RFMD RF3000 manual.
*/
#define RF3000_GAINCTL 0x11 /* TX variable gain control */
#define RF3000_GAINCTL_TXVGC_MASK BITS(7, 2)
#define RF3000_GAINCTL_SCRAMBLER BIT(1)

/*
* Shift the transmit power into the transmit-power field of the
* gain-control register and write it to the baseband processor.
*/
atw_rf3000_write(sc, RF3000_GAINCTL,
SHIFTIN(txpower, RF3000_GAINCTL_TXVGC_MASK));


/*
* Register definitions taken from the ADMtek ADM8211 manual.
*
*/
#define ATW_RXSTAT_OWN BIT(31) /* 1: NIC may fill descriptor */
/* ... */
#define ATW_RXSTAT_DA1 BIT(17) /* DA bit 1, admin'd address */
#define ATW_RXSTAT_DA0 BIT(16) /* DA bit 0, group address */
#define ATW_RXSTAT_RXDR_MASK BITS(15,12) /* RX data rate */
#define ATW_RXSTAT_FL_MASK BITS(11,0) /* RX frame length, last
* descriptor only
*/

/* Extract the frame length from the Rx descriptor's
* status field.
*/
len = SHIFTOUT(rxstat, ATW_RXSTAT_FL_MASK);

branches: 1.4.2; 1.4.4;
Use aprint_*().

branches: 1.3.2; 1.3.4; 1.3.6;
merge ktrace-lwp.

branches: 1.2.6;
Suppress false 'WARNING: LAST RESET DUE TO WATCHDOG EXPIRATION!'
indications by clearing the WDOVF status flag for real. The way
to do that is by writing 1, not 0.

Add a driver for the watchdog timer on the AMD Geode SC1100.

#include <sys/bus.h> instead of <machine/bus.h>.

branches: 1.8.12;
Mhz -> MHz in aprint, use aprint_naive.

branches: 1.7.2;
device_t/softc split and other related cosmetic changes.

Remove clause 3 and 4 from TNF licenses

branches: 1.5.6; 1.5.8; 1.5.10;
Support detachment of geodegcb(4) and its children, geodecntr(4)
and geodewdog(4). Support detachment of gscpcib0 and its child,
gpio0.

branches: 1.4.28; 1.4.34; 1.4.42;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.2.4; 1.2.6; 1.2.10; 1.2.12; 1.2.14;
convert to timecounters (from branch simonb-timecounters)

branches: 1.1.2; 1.1.4;
file geodecntr.c was initially added on branch simonb-timecounters.

Drop 3rd and 4th clauses from David Young's license.
Reviewed and approved by dyoung@ (copyright holder).

branches: 1.8.60; 1.8.74;
Per discussion on tech-kern and tech-userlevel, move the bit-twiddling
macros, __BIT, __BITS, SHIFTIN, SHIFTOUT, and __arraycount() from
lib/libkern/libkern.h to sys/cdefs.h. Add a __-prefix to SHIFTIN
and SHIFTOUT, and add a manual page for the bit-twiddling macros,
bits(3).

Make the __BIT and __BITS macros "widthless," as best I can, by
changing their type to uintmax_t from uint32_t. XXX The manual
page lags this change by a bit.

Define __PRIxBIT and __PRIxBITS printf(3) format strings.

branches: 1.7.4;
convert to timecounters (from branch simonb-timecounters)

branches: 1.6.4; 1.6.6;
Change macro names to avoid collisions:

BIT -> __BIT
BITS -> __BITS

Move my bit-twiddling macros to libkern.h from my drivers, where
I had duplicated them. Improve the macros' names. Simplify their
implementation.

A brief description of each macro is below.

BIT(n): Return a bitmask with bit m set, where the least
significant bit is bit 0.

BITS(m, n): Return a bitmask with bits m through n, inclusive,
set. It does not matter whether m>n or m<=n.
The least significant bit is bit 0.

A "bitfield" is a span of consecutive bits defined by a
bitmask, where 1s select the bits in the bitfield. SHIFTIN,
SHIFTOUT, and SHIFTOUT_MASK help read and write bitfields
from device registers.

SHIFTIN(v, mask): Left-shift bits `v' into the bitfield
defined by `mask', and return them. No
side-effects.

SHIFTOUT(v, mask): Extract and return the bitfield selected
by `mask' from `v', right-shifting the
bits so that the rightmost selected bit
is at bit 0. No side-effects.

SHIFTOUT_MASK(mask): Right-shift the bits in `mask' so that
the rightmost non-zero bit is at bit
0. This is useful for finding the
greatest unsigned value that a bitfield
can hold. No side-effects. Note that
SHIFTOUT_MASK(m) = SHIFTOUT(m, m).

Examples:

/*
* Register definitions taken from the RFMD RF3000 manual.
*/
#define RF3000_GAINCTL 0x11 /* TX variable gain control */
#define RF3000_GAINCTL_TXVGC_MASK BITS(7, 2)
#define RF3000_GAINCTL_SCRAMBLER BIT(1)

/*
* Shift the transmit power into the transmit-power field of the
* gain-control register and write it to the baseband processor.
*/
atw_rf3000_write(sc, RF3000_GAINCTL,
SHIFTIN(txpower, RF3000_GAINCTL_TXVGC_MASK));


/*
* Register definitions taken from the ADMtek ADM8211 manual.
*
*/
#define ATW_RXSTAT_OWN BIT(31) /* 1: NIC may fill descriptor */
/* ... */
#define ATW_RXSTAT_DA1 BIT(17) /* DA bit 1, admin'd address */
#define ATW_RXSTAT_DA0 BIT(16) /* DA bit 0, group address */
#define ATW_RXSTAT_RXDR_MASK BITS(15,12) /* RX data rate */
#define ATW_RXSTAT_FL_MASK BITS(11,0) /* RX frame length, last
* descriptor only
*/

/* Extract the frame length from the Rx descriptor's
* status field.
*/
len = SHIFTOUT(rxstat, ATW_RXSTAT_FL_MASK);

branches: 1.4.4; 1.4.6; 1.4.8; 1.4.10;
u_intN_t -> uintN_t

merge ktrace-lwp.

branches: 1.2.6;
Cosmetic: s/writing writing/writing/ in comments.

Add a driver for the watchdog timer on the AMD Geode SC1100.

update copyright to 2 clause version of the NetBSD Foundation

branches: 1.3.14;
device_t/softc split and other related cosmetic changes.

branches: 1.2.4; 1.2.6; 1.2.12; 1.2.66; 1.2.68; 1.2.70;
convert to timecounters (from branch simonb-timecounters)

branches: 1.1.2; 1.1.4;
file geodevar.h was initially added on branch simonb-timecounters.

KNF. No functional change.

branches: 1.11.12; 1.11.30;
#include <sys/bus.h> instead of <machine/bus.h>.

Drop 3rd and 4th clauses from David Young's license.
Reviewed and approved by dyoung@ (copyright holder).

branches: 1.9.14;
device_t/softc split and other related cosmetic changes.

Remove clause 3 and 4 from TNF licenses

branches: 1.7.6; 1.7.8; 1.7.10;
Support detachment of geodegcb(4) and its children, geodecntr(4)
and geodewdog(4). Support detachment of gscpcib0 and its child,
gpio0.

Cosmetic: wrap some lines. Join some lines. struct device * ->
device_t. Use device_t accessor functions.

branches: 1.5.28; 1.5.34; 1.5.38; 1.5.42;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.3.2; 1.3.4; 1.3.6;
Per discussion on tech-kern and tech-userlevel, move the bit-twiddling
macros, __BIT, __BITS, SHIFTIN, SHIFTOUT, and __arraycount() from
lib/libkern/libkern.h to sys/cdefs.h. Add a __-prefix to SHIFTIN
and SHIFTOUT, and add a manual page for the bit-twiddling macros,
bits(3).

Make the __BIT and __BITS macros "widthless," as best I can, by
changing their type to uintmax_t from uint32_t. XXX The manual
page lags this change by a bit.

Define __PRIxBIT and __PRIxBITS printf(3) format strings.

branches: 1.2.4; 1.2.6;
convert to timecounters (from branch simonb-timecounters)

branches: 1.1.2; 1.1.4;
file geodewdg.c was initially added on branch simonb-timecounters.

opencrypto: Make freesession callback return void.

No functional change intended: all drivers already return zero
unconditionally.

glxsb(4): Prune dead branches. Assert session id validity.

glxsb(4): Return zero, not error, if we've issued crypto_done.

glxsb(4): Remove rijndael dependency.

This doesn't actually seem to depend on it in any way.

XXX Compile-tested only.

glxsb(4): Don't use prev msg's last block as IV for next msg in CBC.

This violates the security contract of the CBC construction, which
requires that the IV be unpredictable in advance; an adaptive adversary
can exploit this to verify plaintext guesses.

XXX Compile-tested only.

- Use aprint*() more in xxx_attach().
- Add missing aprint_naive("\n").
- Remove extra spaces and tabs.
- KNF.

Need <sys/rndsource.h>, not <sys/cprng.h>, for rnd source stuff.

branches: 1.12.4;
Merge tls-earlyentropy branch into HEAD.

branches: 1.11.2;
PR/48701: Timo Buhrmester: spurious messages while booting silently (gpio,
glxsb); use aprint_normal()
>Category: kern
>Synopsis: spurious messages while booting silently (gpio, glxsb)
>Confidential: no
>Severity: non-critical
>Priority: medium
>Responsible: kern-bug-people
>State: open
>Class: sw-bug
>Submitter-Id: net

branches: 1.10.8; 1.10.12;
First step of random number subsystem rework described in
<20111022023242.BA26F14A158@mail.netbsd.org>. This change includes
the following:

An initial cleanup and minor reorganization of the entropy pool
code in sys/dev/rnd.c and sys/dev/rndpool.c. Several bugs are
fixed. Some effort is made to accumulate entropy more quickly at
boot time.

A generic interface, "rndsink", is added, for stream generators to
request that they be re-keyed with good quality entropy from the pool
as soon as it is available.

The arc4random()/arc4randbytes() implementation in libkern is
adjusted to use the rndsink interface for rekeying, which helps
address the problem of low-quality keys at boot time.

An implementation of the FIPS 140-2 statistical tests for random
number generator quality is provided (libkern/rngtest.c). This
is based on Greg Rose's implementation from Qualcomm.

A new random stream generator, nist_ctr_drbg, is provided. It is
based on an implementation of the NIST SP800-90 CTR_DRBG by
Henric Jungheim. This generator users AES in a modified counter
mode to generate a backtracking-resistant random stream.

An abstraction layer, "cprng", is provided for in-kernel consumers
of randomness. The arc4random/arc4randbytes API is deprecated for
in-kernel use. It is replaced by "cprng_strong". The current
cprng_fast implementation wraps the existing arc4random
implementation. The current cprng_strong implementation wraps the
new CTR_DRBG implementation. Both interfaces are rekeyed from
the entropy pool automatically at intervals justifiable from best
current cryptographic practice.

In some quick tests, cprng_fast() is about the same speed as
the old arc4randbytes(), and cprng_strong() is about 20% faster
than rnd_extract_data(). Performance is expected to improve.

The AES code in src/crypto/rijndael is no longer an optional
kernel component, as it is required by cprng_strong, which is
not an optional kernel component.

The entropy pool output is subjected to the rngtest tests at
startup time; if it fails, the system will reboot. There is
approximately a 3/10000 chance of a false positive from these
tests. Entropy pool _input_ from hardware random numbers is
subjected to the rngtest tests at attach time, as well as the
FIPS continuous-output test, to detect bad or stuck hardware
RNGs; if any are detected, they are detached, but the system
continues to run.

A problem with rndctl(8) is fixed -- datastructures with
pointers in arrays are no longer passed to userspace (this
was not a security problem, but rather a major issue for
compat32). A new kernel will require a new rndctl.

The sysctl kern.arandom() and kern.urandom() nodes are hooked
up to the new generators, but the /dev/*random pseudodevices
are not, yet.

Manual pages for the new kernel interfaces are forthcoming.

branches: 1.9.12;
KNF, same object code generated.

bcopy -> memcpy

bzero -> memset

branches: 1.6.8; 1.6.14;
device_t/softc split and other related cosmetic changes.

branches: 1.5.2; 1.5.4;
use aprint_*_dev and device_xname
OK joerg

branches: 1.4.6;
Start detangling lock.h from intr.h. This is likely to cause short term
breakage, but the mess of dependencies has been regularly breaking the
build recently anyhow.

use __KERNEL_RCSID()

branches: 1.2.2; 1.2.4; 1.2.14; 1.2.16; 1.2.18; 1.2.24; 1.2.26; 1.2.28; 1.2.30;
Merge some of the less invasive changes from the vmlocking branch:

- kthread, callout, devsw API changes
- select()/poll() improvements
- miscellaneous MT safety improvements

branches: 1.1.2;
Import driver for the AMD Geode LX AES Security Block, from OpenBSD. The
glxsb(4) driver provides random numbers and AES acceleration.

Merge thorpej-cfargs2.

branches: 1.19.8;
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: 1.18.66;
Pickup NGPIO.

branches: 1.17.2;
Don't attach GPIO framework if NGPIO = 0

#include <sys/bus.h> instead of <machine/bus.h>.

Move all copies of ifattr_match() to sys/kern/subr_autoconf.c.

Fix the gscpcib(4) detachment routine: add a call to pcibdetach().

Write child-detachment and rescan routines to let us detach and
re-attach gpio0 and isa0 at gscpcib0:

# drvctl -d isa0
# drvctl -d gpio0
# drvctl -r -a isabus gscpcib0
# drvctl -r -a gpiobus gscpcib0

Add struct pcib_softc to the gscpcib's softc, it is needed since we call
pcibattach() later. This fixes the driver and lets the GPIO pins attach.
While here, remove the empty gscpcib_childdetach function. Fixes GPIO
on my Soekris net4801 where I can now control the error LED...

"please commit" joerg@

struct cfdata * -> cfdata_t

branches: 1.11.14;
device_t/softc split and other related cosmetic changes.

branches: 1.10.6; 1.10.8; 1.10.10;
Support detachment of geodegcb(4) and its children, geodecntr(4)
and geodewdog(4). Support detachment of gscpcib0 and its child,
gpio0.

use __KERNEL_RCSID()

branches: 1.8.28; 1.8.34; 1.8.38; 1.8.40; 1.8.42;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

Nuke checks for SMALL_KERNEL define; this is not a NetBSD option.

branches: 1.5.18; 1.5.22; 1.5.24;
u_intN_t -> uintN_t

Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

branches: 1.2.6;
kill gba_name in attach arg, specify interface attribute

Add NS Geode PCI-ISA bridge GPIO support.

branches: 1.2.18;
merge ktrace-lwp.

branches: 1.1.6;
Add NS Geode PCI-ISA bridge GPIO support.

Share Intel hardware random number generator support between amd64 and
i386 port. This will benefit EM64T systems using Intel i9xx chipsets.

branches: 1.3.2; 1.3.4; 1.3.6;
merge ktrace-lwp.

branches: 1.2.16;
Update some URLs.

branches: 1.1.2; 1.1.4; 1.1.28;
Add support for sampling the random number generator on
the 810, 815, 820, and 840 chipsets. From OpenBSD, modified
for NetBSD by me.

Some changes for the ichlpcib driver:

- Moved to x86/pci, so that EM64T systems running NetBSD/amd64 can use it.
- Added support for the TCO on ICH6 or newer chipsets, adapted from
FreeBSD.
- Added timecounter support for the power management timer, adapted from
OpenBSD.
- Plus some misc/cosmetic changes.

Thanks to yukonbob on irc@freenode for testing the TCO part on ICH4-M.
Tested by me with ICH7 too.

branches: 1.19.8; 1.19.16; 1.19.22; 1.19.26;
match ICH-7M / -DH
(The watchdog doesn't reset the box for me, but it is present according
to the datasheet.)

__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

Add "name" parameter to powerhook_establish, to aid debugging. No objections
on tech-kern@

branches: 1.15.4; 1.15.6;
cube@ says that acpi_pci_fixup() only writes in devices interrupt line
registers, so restore the PCI-ISA bridge IRQ routing here again.

Don't try to fixup PCI interrupt routing here; we're going to use ACPI to
do this for us instead. This prevents us from having to write chipset
specific support -- yay!

Add powerhook to restore IRQ routing on resume.

branches: 1.12.2; 1.12.8; 1.12.10;
Use aprint_*().

branches: 1.11.2; 1.11.4; 1.11.6;
u_intN_t -> uintN_t

Attach to ICH6, ICH6-M and ICH7 Intel I/O Controller Hub.
Tested by seb with ICH6-M.

branches: 1.9.2;
Change the rest of the sysctl subsystem to use const consistently.
The __UNCONST macro is now used only where necessary and the RW macros
are gone. Most of the changes here are consumers of the
sysctl_createv(9) interface that now takes a pair of const pointers
which used not to be.

de-__P, partially ANSIfy

branches: 1.7.2; 1.7.4;
On my laptop, the SMI handler for the TCO interrupt appearently
re-tickles the watchdog, keeping the system from rebooting.
So enable the SMI for TCO only if the hardware reset is not available,
as a last resort.

branches: 1.6.2;
implement ICH-based Intel SpeedStep(TM), based on both the linux and
freebsd drivers for the same. if found and supported, export a
"machdep.speedstep_state" sysctl that can be set to "0" (low) or "1"
(high).

tested on a dell inspiron 8500 (supported, working) and a dual P4
system (supports ichlpcib not speedstep, comes up properly disabled.)

add one more PCI ID (ICH4-M).

Fix grammar in error message.

Fix typo.

Stop the timer explicitly before initialization.
Apparently certain BIOS requires this. Noted by yamt.

Add Intel ICHn PCI-LPC bridge driver.
It is a pcib, but with sysmon watchdog support.

Reduce the massive code duplication regarding joy(4). Split it into
MI and MD parts, and make ISA/ISAPNP/PCI joy(4) attachments MI.

branches: 1.2.2;
add RCSID

branches: 1.1.8; 1.1.10; 1.1.12;
Added a PCI frontend for the legacy ISA joystick driver.
Some PCI soundcards don't seem to use the generic gameport function with
interface 0x10 used here, but have either an own BAR dedicated to this
(i.e. Sonic Vibes or ESS Solo-1) or specify their own device (see
PCI_PRODUCT_CREATIVELABS_SBJOY in sys/dev/pci/pcidevs.h).
Probably these use a similar simple sheme and adding a frontend for them would
be trivial, but I don't own any of these cards, so I didn't.

Clean up deleted files.

Incorporate recent changes from Wolfgang. Recognize the 815.

Remove many of the `static's, to facilitate debugging.

Numerous changes. Many bugs fixed, better autoconfig, a few new features.

Make a wrapper match function to check the bus and device numbers, rather
than insisting that every driver do it.

Always use direct configuration.

new RCS ID format.

Default to decent performance.

Merge changes from Wolfgang Stanglmeier's 941012 release.

branches: 1.2.2;
Update from Wolfgang Stanglemeier.

Add NCR 53c8XX driver, originally from Wolfgang Stanglmeier.

Clean up deleted files.

Numerous changes. Many bugs fixed, better autoconfig, a few new features.

new RCS ID format.

branches: 1.2.2;
Update from Wolfgang Stanglemeier.

Add NCR 53c8XX driver, originally from Wolfgang Stanglmeier.

Clean up deleted files.

Convert port, IRQ, and DRQ numbers to ints.

new RCS ID format.

branches: 1.3.2;
Update from Wolfgang Stanglemeier.

Make this work.

branches: 1.1.2;
Add NCR 53c8XX driver, originally from Wolfgang Stanglmeier.

#include <sys/bus.h> instead of <machine/bus.h>.

branches: 1.9.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.8.52; 1.8.54; 1.8.56;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.6.20; 1.6.22;
merge ktrace-lwp.

de __P, ANSIfy PCIBIOS related files.

branches: 1.4.2;
Adapt for i386/x86 change.

branches: 1.3.2;
add RCSID

branches: 1.2.4; 1.2.6; 1.2.8;
use I386_PCI_INTERRUPT_LINE_NO_CONNECTION instead of magic number.

branches: 1.1.2; 1.1.8;
Interrupt router drivers from UCHIYAMA Yasushi's PCI BIOS patches,
cleaned up and reworked a bit be me.

branches: 1.1.2; 1.1.10;
Interrupt router drivers from UCHIYAMA Yasushi's PCI BIOS patches,
cleaned up and reworked a bit be me.

#include <sys/bus.h> instead of <machine/bus.h>.

branches: 1.9.14;
Remove clause 3 and 4 from TNF licenses

branches: 1.8.52; 1.8.54; 1.8.56;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.6.20; 1.6.22;
merge ktrace-lwp.

branches: 1.5.12;
de __P, ANSIfy PCIBIOS related files.

branches: 1.4.2;
Adapt for i386/x86 change.

branches: 1.3.2;
add RCSID

branches: 1.2.4; 1.2.6; 1.2.8;
- deal with FIRESTAR_PIR_SELECT_PIRQ case on set/get_trigger(),
to prevent a panic on a Panasonic Let's Note AL-N2T516J5.
- add several debug printf which can be enabled by FIRESTARDEBUG.
by UCHIYAMA Yasushi <uch@netbsd.org>

- use I386_PCI_INTERRUPT_LINE_NO_CONNECTION instead of magic number.

branches: 1.1.2; 1.1.8;
Interrupt router drivers from UCHIYAMA Yasushi's PCI BIOS patches,
cleaned up and reworked a bit be me.

branches: 1.2.2; 1.2.4;
- deal with FIRESTAR_PIR_SELECT_PIRQ case on set/get_trigger(),
to prevent a panic on a Panasonic Let's Note AL-N2T516J5.
- add several debug printf which can be enabled by FIRESTARDEBUG.
by UCHIYAMA Yasushi <uch@netbsd.org>

- use I386_PCI_INTERRUPT_LINE_NO_CONNECTION instead of magic number.

branches: 1.1.2; 1.1.8;
Interrupt router drivers from UCHIYAMA Yasushi's PCI BIOS patches,
cleaned up and reworked a bit be me.

struct device * -> device_t, use device_private()
No functional changes intended.

branches: 1.14.8;
device/softc split

branches: 1.13.2; 1.13.4; 1.13.6;
Remove clause 3 and 4 from TNF licenses

branches: 1.12.2; 1.12.4;
use aprint_*_dev and device_xname
OK joerg

branches: 1.11.52;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.9.14; 1.9.16;
Use aprint_*().

branches: 1.8.2; 1.8.4; 1.8.6;
merge ktrace-lwp.

branches: 1.7.6;
de-__P, partially ANSIfy

branches: 1.6.4; 1.6.6;
pass string length (= boundary info) to pci_devinfo so that we do not run over
the end of memory region

add __KERNEL_RCSID()

branches: 1.4.2; 1.4.8;
* Add "struct device" to the softc.
* Add missing notice.

Problems pointed out by Jonathan Stone.

branches: 1.3.2;
Tidy up CFATTACH_DECL() formatting.

Merge Bill Sommerfeld's i386 MP branch. This code has some known
caveats, but works quite well in a lot of MP cases, and all
UP cases that I have tested. Parts of this will hopefully be
reworked in the not-too-distant future.

branches: 1.1.2;
file p64h2apic.c was initially added on branch sommerfeld_i386mp_1.

Use {,e}isabus_attach().

Merge thorpej-cfargs2.

branches: 1.25.8;
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: 1.24.68;
#include <sys/bus.h> instead of <machine/bus.h>.

On x86, change the bus_space_tag_t to a pointer to a struct
bus_space_tag. For now, bus_space_tag's only member is
bst_type, the type of space, which is either X86_BUS_SPACE_IO
or X86_BUS_SPACE_MEM. In the future, new bus_space_tag members
will refer to override-functions installed by a new function,
bus_space_tag_create(9).

Add pointers to constant struct bus_space_tag, x86_bus_space_io and
x86_bus_space_mem. Use them to replace most uses of X86_BUS_SPACE_IO
and X86_BUS_SPACE_MEM.

Add an x86-specific bus_space_is_equal(9) implementation that compares
the two tags' bst_type.

branches: 1.22.14; 1.22.20; 1.22.22;
device_t/softc split and other related cosmetic changes.

Remove clause 3 and 4 from TNF licenses

branches: 1.20.2; 1.20.4;
use aprint_*_dev and device_xname
OK joerg

branches: 1.19.52;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.17.14; 1.17.16;
Use aprint_*().

branches: 1.16.2; 1.16.4; 1.16.6;
merge ktrace-lwp.

branches: 1.15.6;
de-__P, partially ANSIfy

branches: 1.14.4; 1.14.6;
Phase out the use of a string as first "attach args" member to control
which bustype should be attached with a specific call to config_found()
(from a "mainbus" or a bus bridge).
Do it for isa/eisa/mca and pci/agp for now. These buses all attach to
an mi interface attribute "isabus", "eisabus" etc., and the autoconf
framework now allows to specify an interface attribute on config_found()
and config_search(), which limits the search of matching config data
to these which attach to that specific attribute.
So we basically have to call config_found_ia(..., "foobus", ...) where
such a bus is attached.
As a consequence, where a "mainbus" or alike also attaches other
devices (eg CPUs) which do not attach to a specific attribute yet,
we need at least pass an attribute name (different from "foobus") so
that the foo bus is not found at these places. This made some minor
changes necessary which are not obviously related to the mentioned buses.

pass string length (= boundary info) to pci_devinfo so that we do not run over
the end of memory region

branches: 1.12.2;
Adapt for i386/x86 change.

match anything which claims to be PCI-EISA bridge
the explicit check for Intel PCI-EISA bridge was left in, just in case
it wouldn't identify itself as PCI-EISA bridge

fixes PR kern/9589 by Johan Danielsson

Use aprint_normal() for cfprint routines.

Tidy up CFATTACH_DECL() formatting.

Merge Bill Sommerfeld's i386 MP branch. This code has some known
caveats, but works quite well in a lot of MP cases, and all
UP cases that I have tested. Parts of this will hopefully be
reworked in the not-too-distant future.

Declare all cfattach structures const.

branches: 1.6.2;
add RCSID

branches: 1.5.26; 1.5.28;
bzero->memset

Use config_defer().

Nuke __BROKEN_INDIRECT_CONFIG.

Use __BROKEN_INDIRECT_CONFIG where appropriate.

Add an autoconfig node for PCI-EISA bridges. Configures EISA and ISA
bridges via callback.

- Share pchb(4) between i386 and amd64; one copy is enough for both.
- Move some of the x86 PCI devices into x86/pci/files.pci.
- Add more x86 stuff into x86/conf/files.x86.

ok joerg.

Merge the ppcoea-renovation branch to HEAD.

This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.

TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.

NOTES:
pmppc was removed as an arch, and moved to a evbppc target.

branches: 1.65.2; 1.65.6;
Add agp support for i965Q.

Based on a patch for FreeBSD by Eric Anholt.
OKed on tech-kern some months ago.

branches: 1.64.6; 1.64.14; 1.64.18; 1.64.22;
Fix description of the 82443BX workaround to use "stepping" instead
of "revision".

According to http://www.intel.com/design/chipsets/specupdt/290639.htm,
the BIOS bug workaround for the i82443BX chipset's DRAM leadoff timing
parameter is not needed for revisions >= C0, so avoid tweaking that
parameter in that case.

Earlier, this would trigger NMIs on fully memory-populated Compaq
1850R systems, where the BIOS appears to program and require a non-
standard value for this parameter.

Thanks to Chris Ross for the diagnosis and the fix!

branches: 1.62.2;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.60.4; 1.60.6;
Add support for 82945GM and (untested) 82945P AGP. FreeBSD and
Linux both treat the 945 series exactly the same as the 915 series
for AGP, so just match a few extra device IDs.

branches: 1.59.2;
Use aprint_*().

Share Intel hardware random number generator support between amd64 and
i386 port. This will benefit EM64T systems using Intel i9xx chipsets.

branches: 1.57.2; 1.57.4;
PR/32550: Michael Eriksson: No agp(4) support for i915 integrated graphics

branches: 1.56.2;
merge ktrace-lwp.

branches: 1.55.4; 1.55.6;
de-__P, partially ANSIfy

branches: 1.54.4; 1.54.6;
Phase out the use of a string as first "attach args" member to control
which bustype should be attached with a specific call to config_found()
(from a "mainbus" or a bus bridge).
Do it for isa/eisa/mca and pci/agp for now. These buses all attach to
an mi interface attribute "isabus", "eisabus" etc., and the autoconf
framework now allows to specify an interface attribute on config_found()
and config_search(), which limits the search of matching config data
to these which attach to that specific attribute.
So we basically have to call config_found_ia(..., "foobus", ...) where
such a bus is attached.
As a consequence, where a "mainbus" or alike also attaches other
devices (eg CPUs) which do not attach to a specific attribute yet,
we need at least pass an attribute name (different from "foobus") so
that the foo bus is not found at these places. This made some minor
changes necessary which are not obviously related to the mentioned buses.

Attempt to handle unknown ServerWorks chips, although I wouldn't bet on it
working cleanly.

Update Serverworks chip names, and add a few more.

Be more careful about what ServerWorks devices we attempt to map PCI buses
behind.
This should address PR 21458 and PR 25670.

pass string length (= boundary info) to pci_devinfo so that we do not run over
the end of memory region

branches: 1.49.2;
add yet another ServerWorks variant
from PR port-i386/24570 by Ray Phillips

XXX: gcc

Fix uninitialized variable warnings

Add missing blank line.
Don't treat pointers as boolean values.

Add support for i865G integrated graphics support to agp(4). Patches
contributed by Michael D. Allen in PR port-i386/22600.

Add support for i855 agp. From FreeBSD.

Approved by: Frank van der Linden <fvdl@netbsd.org>

branches: 1.43.2;
Handle 64bit DMA addresses on PCI for platforms that can (currently only
enabled on amd64). Add a dmat64 field to various PCI attach structures,
and pass it down where needed. Implement a simple new function called
pci_dma64_available(pa) to test if 64bit DMA addresses may be used.
This returns 1 iff _PCI_HAVE_DMA64 is defined in <machine/pci_machdep.h>,
and there is more than 4G of memory.

From PR port-i386/20464, by Yoshihisa Nakagawa: Match i845G AGP bridge.

One more ServerWorks host bridge, seen on Dell PowerEdge server
fix provided in PR port-i386/20333 by Sverre Froyen

Remove duplicate assignment.

Add a couple more ServerWorks host bridges that explicitly need IO space
enabled.

Use aprint_normal() for cfprint routines.

Tidy up CFATTACH_DECL() formatting.

Merge Bill Sommerfeld's i386 MP branch. This code has some known
caveats, but works quite well in a lot of MP cases, and all
UP cases that I have tested. Parts of this will hopefully be
reworked in the not-too-distant future.

Declare all cfattach structures const.

Add support for auxillary busses on the Intel 82452 PCI-Host
bridge. PR 17353, from MOCHIDA Shuji.

branches: 1.33.2; 1.33.4;
* Add "pcitag_t *pba_bridgetag" to pci_attach_args. This is set to
NULL for root PCI busses. For busses behind a bridge, it points to
a persistent copy of the bridge's pcitag_t. This can be very useful
for machine-dependent PCI bus enumeration code.
* Implement a machine-dependent pci_enumerate_bus() for sparc64 which
uses OFW device nodes to enumerate the bus. When a PCI bus that is
behind a bridge is attached, pci_attach_hook() allocates a new PCI
chipset tag for the new bus and sets it's "curnode" to the OFW node
of the bridge. This is used as a starting point when enumerating
that bus. Root busses get the OFW node of the host bridge (psycho).
* Garbage-collect "ofpci" and "ofppb" from the sparc64 port.

Also correct the attach flags to enable IO space on the Intel 450NX.
OK'ed by fvdl.

Add the i830 bridge.

Also correct the attach flags to enable IO space on the ServerWorks
CNB20LE.

add RCSID

branches: 1.28.2;
-fix botched switch/case nesting which made AGP on i810 in GFX mode fail
-remove the check for i810's internal graphics completely: we'll attach
AGP whether in GFX or AGP mode anyway, and the SMRAM register test
was of questionable value (should have masked with 0xc0, but even then
the builtin graphics appeared enabled although I used an external
PCI card)

Clean up the AGP match/attach code somewhat.

Apparently some ServerWorks Host-PCI bridges only get their memory space
enabled, even though IO space does work. A few drivers (like ahc)
will only work reliably with IO space, so check for this condition
and correct it.

wrap decl in #if NAGP > 0

Attach agp gart support @ pchb. Not very clean, but agp support may
be spread over several devices, and the phcb is usually the main one.

Add agp_machdep.c file which implements MD agp functions (currently
just agp_flush_cache).

branches: 1.23.4; 1.23.6; 1.23.8;
- Pequr -> Serverworks. It pays to do a cvs update first...
- Pay attention only to the low byte of config reg 0x44 on Serverworks chips,
as Linux does.
- Compress duplicatated code.

Fix NRND == 0 case (unused variable).

Add support for sampling the random number generator on
the 810, 815, 820, and 840 chipsets. From OpenBSD, modified
for NetBSD by me.

Support the second PCI bus on SeverWorks chipsets. From OpenBSD.

BIOS BUG WORKAROUND! The 82443BX datasheet indicates that the only legal
setting for the "Idle/Pipeline DRAM Leadoff Timing (IPLDT)" parameter
(bits 9:8) is 01. Unfortunately, some BIOSs do not set these bits properly.

Based on a hint from OpenBSD.

branches: 1.18.4;
-attach auxilary PCI buses at the 450NX chipset,
tested by hubertf in a small configuration (1 PXB - 2 PCI buses)
-being here, remove the per-VID/DID matching code which was commented
out for a while

branches: 1.17.12;
Match pchb by class/subclass, like pcib.
Add a note about this strange Pequr host bridge in Compaq machines.

Add the Intel 82443BX PCI host bus adapter. From Soren S. Jorvang, PR #6250.

Recognise the ALI M1541 in order to support motherboards with the
ALI Aladdin V chipset.

Nuke __BROKEN_INDIRECT_CONFIG.

Use __BROKEN_INDIRECT_CONFIG where appropriate.

Update for changes to config.

add yet another PCI brige -- from pr 4780 from Nathan J. Williams

Switch off CPU-PCI write posting for the Saturn chipset; it is known
to cause problems with PCI busmasters.

Fix tipo inherited from old version of TNF copyright template.

adapt to changed device names

Recognize some more chipsets.

branches: 1.6.4;
Pull thorpej-bus-dma branch into mainline.

branches: 1.5.2;
Remove comment about bus dma tag. I've been convinced otherwise.

Add a comment about why the bus dma tag is not forwarded from parent
to child if an auxillary PCI bus is attached to the bridge. Also,
remove an extra ';'.

add support for dual PCI bus Intel host/PCI bridge. Thanks to Chris Demetriou for his assistance in educating me about the vagaries of PCI and its configuration model.

add support for ACC Micro host bridges

Implement a simple generic PCI-Host bridge driver. At the moment, this
driver is a place-holder, which will nicely print version information
about your PCI chipset (try with "options PCIVERBOSE"). Eventually,
this can be used to enable/disable features/bugs of individual PCI
chipsets.

Share Intel hardware random number generator support between amd64 and
i386 port. This will benefit EM64T systems using Intel i9xx chipsets.

Add hardware RNG support for Intel i925X chipset.

Add hardware RNG support for Intel i945G/P and i955X chipsets.

branches: 1.16.2; 1.16.4;
PR/32550: Michael Eriksson: No agp(4) support for i915 integrated graphics

branches: 1.15.2;
u_intN_t -> uintN_t

merge ktrace-lwp.

branches: 1.13.2; 1.13.6; 1.13.14; 1.13.16; 1.13.22; 1.13.26;
Add hardware random number generator support for Intel 865 and
875P chipsets.

branches: 1.12.2; 1.12.8;
Add missing calls to bus_space_unmap(9) when the hardware RNG is not
detected, cannot be activated or doesn't work.

Remove senceless benchmark loop for i8xx hardware random generator
attach code. The throughput was neither calculated nor displayed.

Add hardware random generator support for Intel i845, i850 and i860
chipsets.

branches: 1.9.10; 1.9.12;
add RCSID

Mix random data directly into the pool and increase entropy instead of
estimating entropy with polling based timing.

branches: 1.7.4; 1.7.6;
The `speed' statistic from the RNG was brought to you by the letter `B' and
the letter `S'. I don't think anyone will miss it.

branches: 1.6.4;
- Zero is a valid random data. Don't give up to attach.
- 4 cycle is enough to get 4 octet of data.

branches: 1.5.4;
Fix speeling in cooment.

Print a newline after the random number speed message.

Change the RNG callout -- don't spin until data is available. If
none is available, just wait until the next clock tick.

fix busy-wait logic against random number register. (missing semicolon)

Add support for sampling the random number generator on
the 810, 815, 820, and 840 chipsets. From OpenBSD, modified
for NetBSD by me.

Share Intel hardware random number generator support between amd64 and
i386 port. This will benefit EM64T systems using Intel i9xx chipsets.

branches: 1.2.2; 1.2.4; 1.2.6;
u_intN_t -> uintN_t

branches: 1.1.2; 1.1.4; 1.1.44;
Add support for sampling the random number generator on
the 810, 815, 820, and 840 chipsets. From OpenBSD, modified
for NetBSD by me.

Clean up deleted files.

Make a wrapper match function to check the bus and device numbers, rather
than insisting that every driver do it.

Rename pciprobe() to pcimatch(), and move it to pci_machdep.c.

Always use direct configuration.

new RCS ID format.

branches: 1.2.2;
Update some comments.

branches: 1.1.2;
Add PCI autoconfiguration support.

Add support for PCI_BUS_FIXUP and PCI_ADDR_FIXUP on amd64.

branches: 1.19.52; 1.19.54; 1.19.56; 1.19.58;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.17.20; 1.17.22;
u_intN_t -> uintN_t

merge ktrace-lwp.

branches: 1.15.2;
It turns out that pci_addr_fixup() doesn't require PCIBIOS either. Decouple,
run it from mainbus_attach().

PR/30387: Nicolas Joly: -current i386 GENERIC_LAPTOP kernel compilation failure

de __P, ANSIfy PCIBIOS related files.

branches: 1.12.2;
Adapt for i386/x86 change.

branches: 1.11.2;
add RCSID

- Correct a value of subend.
Pointed out by enami tsugutomo <enami@but-b.or.jp>.

branches: 1.9.2;
added context parameter to pciaddr_resource_{reserve,allocate}
and to pci_device_foreach().
added new function pci_device_foreach_min().

- fixup only #0 bus --> fixup maxbus.
- Don't pciaddr_do_resource_allocate if device is AGP
to avoid conflict.

branches: 1.7.4; 1.7.6;
Fix a problem uncovered by rev 1.5 of pcibios.c:

Avoid interpreting the upper 32 bits of 64-bit BARs as a 32-bit BAR.
Otherwise, the code would assume that the value 0 was incorrect and either:
(a) [on bus 0] "fix up" the address to some nonzero value, thus placing
the decoded address range outside of 32-bit address space, or
(b) [elsewhere] completely disable the device.

The fact that this behaviour depends on the bus number of the device is
already XXX'd.

XXX: This will need revisiting if and when we ever want to handle a PCI bus
XXX: with more than 32 bits of address space on an i386.

The onboard Adaptec 7890 on my Dell Precision Workstation 410 works again.

fix oversight introduced in previous my commmitment (revision 1.4),
pointed out by Michael Shalayeff <mickey@openbsd.org>.

reserve AGP space to avoid resource conflict.

use PCIBIOS_PRINTV() defined in pcibios.h, instead of homegrown DPRINTF().

branches: 1.3.2;
don't destroy PCI bus space extent for rbus_machdep.c

branches: 1.2.2;
Skip fixup phase when system BIOS setting were perfect.
don't write 0 to command register. (it is harmful).

PCI I/O address fixup routine.
When BIOS don't asign PCI I/O address space to device,
allocate it without conflict and write to PCI configuration header.

Add support for PCI_BUS_FIXUP and PCI_ADDR_FIXUP on amd64.

branches: 1.6.74; 1.6.76; 1.6.78; 1.6.80;
u_intN_t -> uintN_t

merge ktrace-lwp.

branches: 1.4.12;
de __P, ANSIfy PCIBIOS related files.

branches: 1.3.4; 1.3.24;
added context parameter to pciaddr_resource_{reserve,allocate}
and to pci_device_foreach().
added new function pci_device_foreach_min().

branches: 1.2.6; 1.2.8;
don't destroy PCI bus space extent for rbus_machdep.c

branches: 1.1.4;
PCI I/O address fixup routine.
When BIOS don't asign PCI I/O address space to device,
allocate it without conflict and write to PCI configuration header.

Add support for PCI_BUS_FIXUP and PCI_ADDR_FIXUP on amd64.

branches: 1.8.52; 1.8.54; 1.8.56; 1.8.58;
__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.6.20; 1.6.22;
merge ktrace-lwp.

Remove unneeded dependancy on PCIBIOS.

branches: 1.4.2; 1.4.18;
Reset configuration of bridge order before enumerating bridge devices.
This is required when two or more bridges on the same bus, and some of
them are incompletely initialized by BIOS.
Reported by NAKAGAWA Yoshihisa <y-nakaga@nwsl.mesh.ad.jp>

add RCSID

branches: 1.2.2;
when fixing up buses, keep track of the parent of each bus,
and the pcitag to access the primary side of that bridge.

branches: 1.1.2; 1.1.10; 1.1.12;
Code from UCHIYAMA Yasushi's PCI BIOS patches to renumber the PCI
busses.

Add support for PCI_BUS_FIXUP and PCI_ADDR_FIXUP on amd64.

branches: 1.4.74; 1.4.76; 1.4.78; 1.4.80;
merge ktrace-lwp.

de __P, ANSIfy PCIBIOS related files.

branches: 1.2.4; 1.2.24;
when fixing up buses, keep track of the parent of each bus,
and the pcitag to access the primary side of that bridge.

branches: 1.1.2; 1.1.10; 1.1.12;
Code from UCHIYAMA Yasushi's PCI BIOS patches to renumber the PCI
busses.

This is no longer needed.

New bus.h implementation/interface:
- No more distinction between i/o-mapped and memory-mapped
devices. It's all "bus space" now, and space tags
differentiate the space with finer grain than the
bus chipset tag.
- Add memory barrier methods.
- Implement space alloc/free methods.
- Implement region read/write methods (like memcpy to/from
bus space).
This interface provides a better abstraction for dealing with
machine-independent chipset drivers.

backout previous kprintf changes

printf -> kprintf, sprintf -> ksprintf

modify these to provide a new, better-specified PCI interface
(soon to be documented on mailing lists; eventually in section 9 manual
pages), most importantly:
(1) support interrupt pin swizzling on non-i386 systems with
PCI-PCI bridges (per PPB spec; done, but meaningless, on i386).
(2) provide pci_{io,mem}_find(), to determine what I/O or memory
space is described by a given PCI configuration space
mapping register.
(3) provide pci_intr_map(), pci_intr_string(), and
pci_intr_{,dis}establish() to manipulate and print info about
PCI interrupts.
(4) deprecate the pci_map_* functions, and provide them only
as compatibility interfaces (in pci_compat.c) which will
eventually go away, implemented as wrappers around
the functions described above.
(5) make pci functions take as an argument a machine-dependent
cookie, to allow more flexibility in implementation.

fix typos in word "successfully", mainly s/succesfully/successfully/.

in many device attach paths, allocate memory with M_WAITOK instead of M_NOWAIT
and remove code to handle failures that can no longer happen.

branches: 1.50.20;
__USE(id) in the !PCIBIOSVERBOSE case.
Fixes PR 49181.

branches: 1.49.12; 1.49.28;
#include <sys/bus.h> instead of <machine/bus.h>.

Rename to use PCI_PRODUCT_INTEL_82801DBM_LPC

branches: 1.47.14; 1.47.22; 1.47.32;
Remove clause 3 and 4 from TNF licenses

branches: 1.46.44; 1.46.46; 1.46.48;
For consistency use PCIBIOS_PRINTV instead of printf even under
#ifdef PCIBIOSVERBOSE.

__unused removal on arguments; approved by core.

- sprinkle __unused on function decls.
- fix a couple of unused bugs
- no more -Wno-unused for i386

branches: 1.43.2; 1.43.4;
add missing initializers.

Add an 'uninit' callback to PCI ICU drivers. This allows (for example) the
PIIX support to release its mapping on the edge/level control registers.

Now that these are guaranteed to be unmapped, capture and restore the
registers in piixpcib(4)'s powerhook. The same will need to be done on a
per-chipset basis.

Concerns were raised about calling pci_intr_fixup on resume WRT hotplug
devices, so this has been removed.

Ok cube@.

Allow pci_intr_fixup() to be called more than once.

Fix AMD 766 PMC entry.

Match the following products:

* SIS 962 and SIS 963
* AMD 766 and AMD 768
* ALI M1533

... syncing the code with openbsd.

branches: 1.38.2;
Add another fixup entry Arthur Dimitrelis

branches: 1.37.2; 1.37.6; 1.37.8;
add some more Intel LPC controllers

branches: 1.36.2; 1.36.4;
PCI Interrupt router support for VIA VT823[1357] Southbridges.
From OpenBSD.

branches: 1.35.2;
u_intN_t -> uintN_t

Add Intel 82801FBM ICH6M LPC Interface Bridge. There seem to be more
pci products missing, but I can only test this one.

merge ktrace-lwp.

branches: 1.32.2;
Rework the configuration scheme for PCI fixups:

* bus enumeration fixups are not PCIBIOS-specific, interrupt fixups are done
by both PCIBIOS and ACPI. The redundancy is very redundant. Therefore,
rename PCIBIOS_*_FIXUP to PCI_*_FIXUP, use PCI_INTR_FIXUP in place of
ACPI_PCI_FIXUP, and change code refences to match.

* move the fixup defines from opt_pcibios.h to opt_pcifixup.h to reflect
the above.

* fix up the PCI bus numbering in mainbus_attach(), right after we detect the
configuration mode. This probably renders the fixup in pcibios.c
redundant -- but it should be harmless.

These changes make cardbus work in ACPI-only machines, when PCI_BUS_FIXUP
and PCI_INTR_FIXUP are defined.

de-__P, partially ANSIfy

branches: 1.30.4; 1.30.6;
Deal with edge versus level interrupts properly. From Masanori Kanaoka.

de __P, ANSIfy PCIBIOS related files.

Support for PIRQ[E-H], found in recent intel south bridges
(ICH2 and later), which fixes PR/23700.
The changes are from Hiroyuki Bessho and Masanori Kanaoka in PR/23700
with a little modification of interrupt router lookup from mine.