branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

malloc(9) -> kmem(9)

branches: 1.3.8;
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: 1.2.4; 1.2.6; 1.2.10; 1.2.36;
fix eisa_intr_string

Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Merge thorpej-cfargs2.

branches: 1.3.8;
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: 1.2.4;
Mask all interrupts before reading (and clearing) the interrupt request
register in attach.

Picked up by an assert in the qemu code.

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.40;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Use vmem(9) rather than extent(9) to manage IOMMU mappings.

malloc -> kmem

Merge thorpej-cfargs2.

branches: 1.3.8;
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: 1.2.4;
replace EX_NOWAIT with EX_WAITOK in device attach methods.
remove checks for failures that can no longer occur.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Some KNF

Clean up initialization of com_regs structure, in preparation for
some additional changers.

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.34; 1.1.36;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Trailing whitespace

Clean up initialization of com_regs structure, in preparation for
some additional changers.

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.34; 1.1.36;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

hppa: Membar audit in cpu.c.

Don't KASSERT there's an FPU present. QEMU doesn't have one...

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.36; 1.1.40; 1.1.46;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

hppa/dev: Nix trailing whitespace.

No functional change intended.

branches: 1.17.2;
add missing bus_space_*_stream() methods instead of quietly aliasing them
to their potentially byte-swapping regular counterparts

ok skrll@

branches: 1.16.4;
Remove unnecessary include of <sys/malloc.h>.

Use designated initializers for hppa_pci_chipset_tag structs.

Trailing whitespace

explain why we skip over dev 1f func 7 in config space access functions
no functional change

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;
support mmap()

Remove trailing whitespace

KNF

config space accesses to device 15 function 7 cause a machine check exception
on my C200, so weed them out.
Now we can scan the PCI buses without crashing.

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

Mask all interrupts before reading (and clearing) the interrupt request
register in attach.

Picked up by an assert in the qemu code.

branches: 1.4.4;
Re-arrange dino_softc members to improve alignment

branches: 1.3.10; 1.3.18;
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: 1.2.4; 1.2.6; 1.2.10; 1.2.12;
fix eisa_intr_string

Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

KNF

hppa: elroy: support mmap

X works, but has weird colours.

branches: 1.7.2;
add missing bus_space_*_stream() methods instead of quietly aliasing them
to their potentially byte-swapping regular counterparts

ok skrll@

branches: 1.6.4;
Remove unnecessary include of <sys/malloc.h>.

Use designated initializers for hppa_pci_chipset_tag structs.

Merge thorpej-cfargs2.

branches: 1.3.8;
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: 1.2.32;
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: 1.1.4; 1.1.6; 1.1.10; 1.1.12;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.2.4; 1.2.6; 1.2.10;
fix eisa_intr_string

Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Merge thorpej-cfargs2.

branches: 1.2.8;
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: 1.1.4; 1.1.6; 1.1.10; 1.1.50;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

use nglereg.h

have sti_pci_check_rom() return the ROM's size so we can easily unmap it should
we so desire

Revert changes from previous that have nothing to do with STI PCI ROM
grovelling.

Fixes the build.

move the STI PCI ROM grovelling code into sti_pci_machdep.c instead of each
driver having its own copy

drop unnecessary wait and register write when moving the cursor sprite

drop the rectfill workaround here as well

set things up even if we're not the console

gftfb(4): Nix trailing whitespace.

No functional change intended.

more magic number reduction, the way colour map and cursor bitmap acess works
should be obvious now

magic number reduction, consistently use buffer 0

fix tpyo, now we can actually map registers from userland

use HW to draw bitmap fonts

use wrappers for bus_space_*() - makes things shorter, less annoying to read
and reduces unnecessary differences to hyperfb

as suspected the Visualize EG's rectangle fill function works exactly the same
as HCRX's, so apply the same trick to draw less than 32 pixel wide rectangles
and get rid of the workaround

properly return status info in ioctl(WSDISPLAYIO_GVIDEO)

implement ioctl(GCID)

fix tpyo and properly limit what we can mmap()

remove a bunch of #defines that have been moved to stireg.h

STIDEBUG -> GFTFB_DEBUG

branches: 1.14.2;
allow mapping of blitter registers

make gftfb_restore_palette() grab the default colour map from rasops instead
of just writing the driver's map into the hardware ( which may have been
modified by the likes of wsfb )
With this we get a readable console even when exiting X in a less than
graceful manner.

For some reason the drawing engine occasionally scribbles past the right
boundary when filling rectangles, especially annoying when we draw whitespaces
As a workaround we draw all rectangles less than 50 pixels wide by drawing
a 50 pixel rectangle into off-screen memory to the right of the visible fb and
then copy the portion we want. Keeps track of the colour and size of the
off-screen rectangle so we can avoid redrawing it whenever possible.

fix brainfart - only update fbi_fbsize, not the size of the visible fb...

in gftfb_ioctl():
- identify ourselves as WSDISPLAY_TYPE_STI
- return full fb geometry in WSDISPLAYIO_GET_FBINFO

if the framebuffer is wider than the visible area, use the full width for the
glyphcache

support WSDISPLAYIO_SVIDEO so X can turn the monitor off
so far I only know how to turn off video output, not sync(s). Better than
nothing though.

add hardware cursor support, mostly for X

avoid one more instance of unnecessary blitter stalling

keep track of hw settings for blitter, fill of fb access to avoid unnecessary
register writes
while there, remove some accidentially left in debug goop

play the same lazy cursor (un)draw trick as cgsix and friends
visible speedup in things like systat

enable glyph cache, anti-aliased fonts etc.

blitter time!
everything except drawing characters i now done by hardware
next step - glyphcache

crude beginning of a native driver for PCI Visualize EG cards
so far it supports:
- colour
- virtual consoles
todo:
- mmap
- hardware acceleration
- STI refactoring

use BINC writes to draw characters instead of abusing the rectangle primitive
small but measurable speedup, fewer register writes, and we demonstrate how
to use BINC

use de-obfuscated register names where possible
NFCI

use nglereg.h

branches: 1.24.4;
hppa/dev: Nix trailing whitespace.

No functional change intended.

some cleanup:
- remove code to initialize a DAC that HCRX doesn't have
- decode / explain some more magic numbers
add more comments describing what we do and why

remove glyphcache and debug goop
we're not likely to be able to use any off-screen memory on this hardware, so
don't pretend we will.

now that we can draw rectangles in exactly the size we want without abusing
the pixel mask, we no longer need to draw characters in two passes - just
disable transparency, set fore- and background, and go.
Only clear the character cell when drawing blanks.

finally figured out how to automatically mask off pixels at the right when
drawing rectangles so we can drop the workaround

set things up even if we're not the console

fix tpyo, now we can actually map registers from userland

simplify putchar() a bit, use 16bit accesses for fonts that are more than 8
pixels wide

remove sc_putchar - don't need it anymore

now that we know how, draw characters by hardware

get rid of the workarounds to draw rectangles less than 32 pixels wide
turns out the hardware applies a bit mask to all rectangle fills, and we
can turn the background transparent. With that we just mask off the pixels
we don't need in the less than 32 pixels wide case.
Next step - figure out how to abuse that to draw characters
I bet Visualize EG works exactly the same way in that regard.

properly return status info in ioctl(WSDISPLAYIO_GVIDEO)

implement ioctl(GCID)

hyperfb(4): Nix trailing whitespace.

No functional change intended.

magic number reduction, NFC

do cursor position updated the way NGLE does - poke reg 28 and wait_fifo
instead of waiting for the whole thing to go idle

fix tpyo and properly limit what we can mmap()

hyperfb(4): Nix trailing whitespace, sprinkle KNF.

No functional change intended -- just whitespace (and comments).

use the blitter to draw small rectangles as well by abusing set/clear ops
and the plane mask

hand X a 24bit framebuffer if the hardware supports it

remove a bunch of #defines that have been moved to stireg.h

add cursor sprite support

we have a blitter - use it

first shot at a native driver for HCRX cards
this will give you a fully working wsdisplay in 8bit colour living in the
overlay planes, X/wsfb will also work.
No acceleration or cursor sprite support yet.

s/requset/request/ in comments.

branches: 1.5.12;
Merge thorpej-cfargs2.

branches: 1.4.8;
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: 1.3.4;
Mask all interrupts before reading (and clearing) the interrupt request
register in attach.

Picked up by an assert in the qemu code.

branches: 1.2.4;
Trailing whitespace

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.36;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Rename min/max -> uimin/uimax for better honesty.

These functions are defined on unsigned int. The generic name
min/max should not silently truncate to 32 bits on 64-bit systems.
This is purely a name change -- no functional change intended.

HOWEVER! Some subsystems have

#define min(a, b) ((a) < (b) ? (a) : (b))
#define max(a, b) ((a) > (b) ? (a) : (b))

even though our standard name for that is MIN/MAX. Although these
may invite multiple evaluation bugs, these do _not_ cause integer
truncation.

To avoid `fixing' these cases, I first changed the name in libkern,
and then compile-tested every file where min/max occurred in order to
confirm that it failed -- and thus confirm that nothing shadowed
min/max -- before changing it.

I have left a handful of bootloaders that are too annoying to
compile-test, and some dead code:

cobalt ews4800mips hp300 hppa ia64 luna68k vax
acorn32/if_ie.c (not included in any kernels)
macppc/if_gm.c (superseded by gem(4))

It should be easy to fix the fallout once identified -- this way of
doing things fails safe, and the goal here, after all, is to _avoid_
silent integer truncations, not introduce them.

Maybe one day we can reintroduce min/max as type-generic things that
never silently truncate. But we should avoid doing that for a while,
so that existing code has a chance to be detected by the compiler for
conversion to uimin/uimax without changing the semantics until we can
properly audit it all. (Who knows, maybe in some cases integer
truncation is actually intended!)

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.34; 1.1.36;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Remove unnecessary include of <sys/malloc.h>.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Use {,e}isabus_attach().

Merge thorpej-cfargs2.

branches: 1.6.8;
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: 1.5.14;
Trailing whitespace

branches: 1.4.4;
Reorder to avoid null deref before null test

branches: 1.3.4; 1.3.6; 1.3.10;
fix eisa_intr_string

kill sprintf

Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Change todr_chip_handle::cookie -> todr_chip_handle::todr_dev, and
make it a device_t. Upcoming functional changes will require the
device_t associated with a TODR device.

Change todr_chip_handle::bus_cookie -> todr_chip_handle::todr_devaux.
Nothing was using the old field, but I decided to keep it around just
in cause something needs it in the future.

And with these largely mechanical yet semantically meaningful changes,
thus spake the Oracle: "Welcome to NetBSD 11.99.2."

Put the TODR handle in the pdc_softc.

KNF

branches: 1.4.36;
Trailing whitespace

branches: 1.3.4; 1.3.30;
Fix typo in previous. Eesh.

Add d_discard to a cdevsw I apparently missed.

branches: 1.1.2; 1.1.4; 1.1.6;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Merge thorpej-cfargs2.

branches: 1.2.8;
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: 1.1.4; 1.1.6; 1.1.10; 1.1.50;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Trailing whitespace

branches: 1.2.10;
remove checks for failure after memory allocation calls that cannot fail:

kmem_alloc() with KM_SLEEP
kmem_zalloc() with KM_SLEEP
percpu_alloc()
pserialize_create()
psref_class_create()

all of these paths include an assertion that the allocation has not failed,
so callers should not assert that again.

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.12;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Merge thorpej-cfargs2.

branches: 1.4.2;
Fix all remaining typos, mainly in comments but also in few definitions and log messages, reported by me in PR kern/54889.
Also fixed some additional typos in comments, found on review of same files or typos.

branches: 1.3.2;
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: 1.2.4; 1.2.6; 1.2.10; 1.2.50;
make pci_intr_string and eisa_intr_string take a buffer and a length
instead of relying in local static storage.

Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

have sti_pci_check_rom() return the ROM's size so we can easily unmap it should
we so desire

break long lines

move the STI PCI ROM grovelling code into sti_pci_machdep.c instead of each
driver having its own copy

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

have sti_pci_check_rom() return the ROM's size so we can easily unmap it should
we so desire

break long lines

move the STI PCI ROM grovelling code into sti_pci_machdep.c instead of each
driver having its own copy

hppa/dev: Nix trailing whitespace.

No functional change intended.

branches: 1.3.16;
Fix silent freeze on probing sti(4) framebuffer on 712/60. PR/52162

Ok'ed by skrll@.
Should be pulled up to netbsd-9 and netbsd-8.

branches: 1.2.4; 1.2.16;
Trailing whitespace

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.28; 1.1.36;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

more FX5/10 support:
- setup byte swapping for 8bit framebuffer access
- expose 8bit overlay on WSDISPLAYIO_MODE_MAPPED on FX5/10
with this we can run X with wsfb in 8bit with correct colours and hw cursor

don't dump registers in the middle of setting things up...

humble beginnings of FX5/10 support. This will:
- provide a working, accelerated console at least on my FX5
- ROPs and alpha blending don't work, so we use a putchar() based cursor()
and disable alpha fonts
- X11 can use a hw cursor and 24bit framebuffer, but since we don't know how
attributes work (yet) this will give us the blue channel piped through the
8bit palette. With a linear ramp that's at least somewhat usable.
Next steps:
- provide an 8bit fb with palette so we can run X in colour
- find the FX5's equivalent of the VISFX_FATTR register so we can either force
24bit colour or use proper WIDs

move the STI PCI ROM grovelling code into sti_pci_machdep.c instead of each
driver having its own copy

branches: 1.34.4;
remove code to stash the console font in video memory - this was an experiment
based on guesswork which probably won't work if the card runs in anything
other than 1280x1024, and wasn't significantly faster than drawing characters
line by line using BINC writes.

no need to set read mode when doing host-to-vram alpha blending
while there, remove some commented out debug goop

s/zeeo/zero/ in comment.

Trailing whitespace

use hardware to draw anti-aliased characters

some minor updates:
- clear image planes before they become visible when switching to fb mode
- adjust FIFO slots, now that we knw a bit more about how they work

hppa/summitfb(4): Nix trailing whitespace again.

(ab)use the image buffer as glyph cache, since on this thing we can set source
and destination buffer independently

Use KM_SLEEP in attach path code.

reshuffle summitfb_setup():
- remove magic number inits that are no longer magic
- annotate the remaining inits if we know the register names
- zero the attribute plane
while there, cull excessive calls to summitfb_setup()

clean up aroound sti_fetchfonts(), mke sure we call wsfont_init() nefore
trying to add s font to the system
while there, allow mmap()in the entire fb aperture, no sense in restricting it

Oops. Turn off debug.

KNF

Remove some unnecessary casts and parenthesis

hppa/summitfb(4): Nix trailing whitespace.

No functional change intended.

now that we can do ROPs, use them and enable all the lazy cursor update stuff

some updates:
- setup attributes for RGB8 in the image planes and 8I in the overlay
- explicitly select LUT 0 for the overlay so we don't depend on what STI sets
anymore - now this should work on all FX cards
- expose 24bit framebuffer to X
- make sure we switch overlay transparency according to WSDISPLAY_SMODE
- remove unnecessary register writes when accessing the colour map
- no need to call summitfb_setup_fb() in summitfb_do_cursot() - that was for
HCRX and friends
- support screen blanking

set the right cursor sprite colours, now that we know how

avoid console output between sti_init() and wsdisplay_cnattach()

do not zero the cursor position register on each colour map access, instead
do it when initializing fb access
-> no more disappearing cursor sprite

remove declaration of sti_unitoroman[] since it lives in stivar.h now

some udates:
- fetch the ROM font, convert it to ISO and hand it to wsfont so we can mimic
firmware output should we feel like it
- set read mode to match write mode whenever we touch the latter

hppa/summitfb(4): Nix trailing whitespace again.

add support for alpha fonts and glyph cache, disabled by default since there
is guesswork involved
also, store bitmap fonts in video memory, mostly to show how to change colours
around

hppa/summitfb(4): Nix trailing whitespace.

No functional change intended.

reshuffle unknown register inits, move the ones identified

hppa/summitfb(4): Nix trailing whitespace.

Plus some other minor whitespace/comment cleanups.

some cleanup & optimization:
- don't write the pixel mask register for operations that don't use it
- track write mode, only wait if we change it
- always sync when switching in and out of fb mode ( for X... )
- drop sc_hwmode
... about 10% speedup and no more glitches when starting X

make sure we apply the pixel mask when drawing characters so we don't scribble
outside the character cell

now that we know how, use the blitter to:
- draw characters
- scroll
- rectangle fills
no ROP support yet, so the cursor is drawn using putchar
while there get rid of some unused cargo-culted goop from gftfb
MUCH faster now

CV: ----------------------------------------------------------------------

use summitreg.h

add hardware cursor support
while there remove debug goop, fix ioctl(GCID)

hppa/summitfb(4): Fix SUMMITFB_DEBUG build.

hppa/summitfb(4): KNF

No functional change intended.

beginnings of a driver for Visualize FX graphics cards
So far it does:
- work on my FX4
- use the 8bit overlay, in R3G3B2
- all drawing by software, clear screen by hw
- X works in 8bit with wsfb

Use vmem(9) rather than extent(9) to manage IOMMU mappings.

malloc -> kmem

Merge thorpej-cfargs2.

branches: 1.4.8;
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: 1.3.4;
replace EX_NOWAIT with EX_WAITOK in device attach methods.
remove checks for failures that can no longer occur.

Trailing whitespace

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.36;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

fix recently committed typos by msaitoh in few more places, as well as few more.
mainly s/contigous/contiguous/ and s/miliseconds/milliseconds/ in comments.

few more typos in the word "register", mainly s/resister/register/.
In one comment I assume that it was meant to be pull-up resistor.

src is too big these days to tolerate superfluous apostrophes. It's
"its", people!

branches: 1.1.4; 1.1.6; 1.1.10;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

s/requset/request/ in comments.

branches: 1.4.12;
Merge thorpej-cfargs2.

branches: 1.3.8;
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: 1.2.4;
Mask all interrupts before reading (and clearing) the interrupt request
register in attach.

Picked up by an assert in the qemu code.

branches: 1.1.4; 1.1.6; 1.1.10; 1.1.40;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.