Put back all header files defining ioctls

branches: 1.3.2;
Only install wsconsio.h, wsdisplay_usl_io.h, wsksymdef.h and wsksymvar.h
into /usr/include/dev/wscons.

branches: 1.2.8; 1.2.18;
don't install wskbdmap_mfii.h to userland

Rework the way kernel include files are installed. In the new method,
as with user-land programs, include files are installed by each directory
in the tree that has includes to install. (This allows more flexibility
as to what gets installed, makes 'partial installs' easier, and gives us
more options as to which machines' includes get installed at any given
time.) The old SYS_INCLUDES={symlinks,copies} behaviours are _both_
still supported, though at least one bug in the 'symlinks' case is
fixed by this change. Include files can't be build before installation,
so directories that have includes as targets (e.g. dev/pci) have to move
those targets into a different Makefile.

add some more ASCII definitions, move NP->FF for consistency

#define NUL 0 -- isn't that great?

Initial import of cgd's new wscons code.

Merge the [pgoyette-compat] branch

branches: 1.55.2; 1.55.4;
Add WSDISPLAY_MULTICONS option which allows wsdisplay to cooperate with
the console device being overridden. This allows for simultaneous console
output on fb and debug uart.

Add a keymap for converting Linux event codes to wscons keysyms.

wsbellmux.c doesn't need a needs-flag

Nothing needs the NWSBELL flag, so don't bother creating it

New device wsbell - allows for a console beep for non pckbds (usb etc).
Works for platforms without pcppi - can work with spkr at audio and spkr
at pcppi.

To use add the following to your kernel config:
wsbell* at spkr? console?

Ok pgoyette@.

add support for loading fonts in vcons and subsequently resizing screens
- drivers can use this by setting VCONS_LOADFONT and WSSCREEN_RESIZE
- each vcons screen can now have its own font and geometry
- while there, add support for xterm's ESC[18t to report the text buffer's
size

With this tou can:
wsfontload -N foo /usr/share/wscons/fonts/flori.816
wsconsctl -dw font=foo
currently this is limited to drivers that don't use the glyph cache, like genfb

branches: 1.49.2;
Give up making wscons modular for now.

My intent is to make "no" handle dependency right so that you can simplely do:

no wsdisplay*

Normalize wscons devices (except leaving one strange definition).

defflag GLYPHCACHE_DEBUG

branches: 1.46.2;
generic support for caching glyphs in video memory
for speeding up anti-aliased fonts on slow CPUs

branches: 1.45.2; 1.45.6;
dumping ground for generic driver convenience functions
So far it contains only a generic WSDISPLAYIO_GET_EDID implementation that
relies on EDID data being passed to drivers as a device property.

remove VCONS_DRAW_ASYNC
it was always experimental, VCONS_DRAW_INTR is much simpler and Just Works.

add support for command buffering in vcons. still a WIP, same limitations
as async drawing (no ddb, needs interrupts). you can try it with options
VCONS_DRAW_INTR. as with async, there are still occasional glitches.

Add support for asynchronous drawing in vcons.
This is not finished but good enough for others to play with, enable with
options VCONS_DRAW_ASYNC
With this all drawing operations will be posted to a ring buffer instead
of being run directly, and run by a kernel thread. This avoids having to wait
for drawing operations to finish with the kernel lock held ( to a degree at
least ) and scrolling a (slow) framebuffer console should not disrupt other
operations anymore.
Problems:
- we need to switch back to synchronous operations when panicing or entering
ddb, also re-enable async drawing when leaving ddb
- there are still occasional glitches
tested on an SS20 with cg14 and cg6 for dumb and accelerated cases

branches: 1.41.4; 1.41.6; 1.41.8;
add vcons_replay_msgbuf() for use by wsdisplay drivers to replay the message
buffer when attaching so older log messages will show up in the scrollback
buffer
Idea from gimpy

branches: 1.40.26; 1.40.34;
add scrollback support

branches: 1.39.8; 1.39.18; 1.39.22;
Remove (the default) #define XXXDEBUG and add the options to the relevant
files.xxx file.

branches: 1.38.8; 1.38.10;
Remove the getwschar and putwschar accessops from wsdisplay drivers as
requested by uwe@. These were wrong because they were receiving an
emulcookie yet they were accessops (thus having to receive an accesscookie).
Instead, just handle the WSDISPLAYIO_{GET,PUT}WSCHAR ioctls from the
driver's ioctl accessop.

As this reduces the amount of code needed to handle these operations to
two small functions in each driver, remove the WSDISPLAY_CHARFUNCS kernel
option.

Reviewed by, at least, uwe@ and macallan@. No objections in tech-kern@.

branches: 1.37.2; 1.37.4; 1.37.6;
Add support for virtual consoles to igsfb.
Works fine on a Krups, needs testing on other ports.

framework for generic virtual consoles

branches: 1.35.2; 1.35.4; 1.35.6;
merge ktrace-lwp.

branches: 1.34.2;
Make auto repeat of events a compile time option, and have it off by default.
Turn on by WSKBD_EVENT_AUTOREPEAT.

Implement border color customization in wscons(4), only available for vga(4)
at the moment.

This includes the addition of two new wsdisplay ioctls, WSDISPLAY_{G,S}BORDER,
one to get the actual color and one to set it, respectively. Possible colors
match those defined by ANSI (and listed in wsdisplayvar.h).

It also adds two accessops to the underlying graphics device, getborder and
setborder, which mach their ioctl counterparts.

Two kernel options are added: WSDISPLAY_CUSTOM_BORDER, which enables the
ioctls described above (to customize the border color from userland after
boot), and WSDISPLAY_BORDER_COLOR, which sets the color at boot time.
The former is enabled by default on the GENERIC kernel, but not on INSTALL
(among others). The later is always commented out, leaving the usual black
border as a default.

wsconsctl is modified to allow accessing this value easily. For example,
'wsconsctl -d -w border=blue'.

Forgot to define WSDISPLAY_CUSTOM_OUTPUT here.

Implement support to dynamically change wscons console and kernel colors.

Two new ioctls are added to the wsdisplay device, named WSDISPLAY_GMSGATTRS
and WSDISPLAY_SMSGATTRS, used to retrieve the actual values and set them,
respectively (the name, if you are wondering, comes from "message attributes").

A new emulop is added to the underlying display driver (only vga, for now)
which sets the new attribute for the whole screen, without having to clear
it. This is optional, which means that this also works with other drivers
that don't have this new operation.

Five new kernel options have been added, although only documented in
i386 kernels (for now):
- WSDISPLAY_CUSTOM_OUTPUT, which enables the ioctls described above to
change the colors dynamically from userland. This is enabled by default
in the GENERIC kernel (as well as others) but disabled on all INSTALL*
kernels (as this feature is useless there).
- WS_DEFAULT_COLATTR, WS_DEFAULT_MONOATTR, WS_DEFAULT_BG and WS_DEFAULT_FG,
which specify the default colors for the console at boot time. These have
the same meaning as the (already existing) WS_KERNEL_* variables.

wsconsctl is modified to add msg.default.{attrs,bg,fg} and
msg.kernel.{attrs,bg,fg} to the display part, so that colors can be changed
after boot.

Tested on NetBSD/i386 with vga (and vga in mono mode), and on NetBSD/mac68k.
No objections in tech-kern@.

PR/19925: David Ferlier: Add scrolling support to wscons.

Refactor tpcalib and hpc* ports to make tpcalib MI, as discussed on tech-kern.

branches: 1.28.2; 1.28.4;
allow to customize how highlighting and underlining text is substituted
if the display doesn't provide this
submitted by xs@kittenz.org per PR kern/18004

wsdisplay needs wsevent.c. PR 17456

PR/17402: Add wsmoused support by providing get/set char and events.

branches: 1.25.8; 1.25.10;
- convert usage of "defopt" to "defflag" where the relevant option does
not support a value (e.g., it's to be used as "options FOO" instead of
"options FOO=xxx"). options that take a value were converted to
defparam recently.
- minor whitespace & formatting cleanups

- replace "defopt" with "defparam" for options which must take a value,
as config(8) will warn for value-less defparam options
- minor whitespace/formatting cleanup
- consolidate opt_tcp_recvspace.h and opt_tcp_sendspace.h into opt_tcp_space.h

Major rototilling of the wsmux code. No user visible changes (except that
many bugs have been fixed).
Changes:
The wskbd, wsmouse, and wsmux are now "sub-classes" of wsevsrc, which is
a source of ws events. This make the structure of those drivers a little
more uniform.
Many bug fixes involving adding and removing devices from muxes.
When a kernel is configured without wsmux there will now be none (unlike
before where you got a console mux anyway).
The kernel now compiles with all combinations of ws devices present.

branches: 1.22.2;
Stop wscons from conflicting with rcons.

branches: 1.21.2; 1.21.4; 1.21.6;
whitespace police

add definitions for ISO-7 (greek) support

defopt RCONS_4BPP.

branches: 1.18.2; 1.18.8;
Eliminate the need for needs-count for wsmux. This should make
wsdisplays work properly again.

GRR! Kludge around the extremely annoying wsmux configuration lossage
by removing the "| wsdisplay" from the wsmux.c file declaration. This
will cause any kernel which includes wsdisplay but not wsmux explicitly
to fail to link, but at least those of us with multiple wsdisplays on
a single machine can build kernels again.

Add the wsmux pseudo device.

Get rid of the timeout-driven wscons_glue stuff which caused a lot of
headaches.
Now console keyboard and display are connected at autoconfiguration time,
when the last of them is found. Other keyboards / displays remain
unconnected until a new ioctl (WSDISPLAYIO_SETKEYBOARD) is called.

Sort out the inclusion of wscons source files: the files are only
compiled if the corresponding devices are configured.
(Files calling wscons functions have to check whether these are present.)

branches: 1.13.4;
defopt WSCONS_SUPPORT_PCVTFONTS

defopt RCONS_2BPP and RCONS_16BPP
add support for 16bpp framebuffers as found on color nexten.

defopt WSDISPLAY_DEFAULTSCREENS

make the timeout for process/vt synchronization "config" settable and
ddb/gdb patchable

Improvements to the VT100 emulation:
-display DEC special graphics and DEC technical characters as far as
possible
-implement the font switching controls (need documentation!)
-behave well if double-width characters are requested
-simplify the state machine: store CSI command modifiers in variables
instead of dedicating own states to each of them

First cut on a VT100 emulation.
It should be able to parse escape sequences up to VT300, but not everything
is implemented. Most notably, there is no font handling - all displayable
characters are handed to the graphics driver. To solve this, a serious
interface change to the graphics driver is needed (Unicode?).

prepare for VT100 emulation

add options for new compatibility stuff, add wsdisplay_compat_usl.c

Simple screen attribute handling for wscons.
The graphics device driver passes a "default attribute" for normal text
output to the wscons framework. If the emulation module needs more
attributes (for different "renditions") it can allocate them via a
callback.
For now, only the "sun" emulation makes use of it.

defopt the WSEMUL options.

some attributes are now global

Initial import of MI keyboard translation for dev/wscons.

Initial import of cgd's new wscons code.

branches: 1.2.2;
Instead of a Linux keymap, convert Linux event codes to USB scan codes and
use the ukbd keymaps instead.

Add a keymap for converting Linux event codes to wscons keysyms.

branches: 1.2.2;
Instead of a Linux keymap, convert Linux event codes to USB scan codes and
use the ukbd keymaps instead.

Add a keymap for converting Linux event codes to wscons keysyms.

Fix spello in a comment

branches: 1.6.28;
Fixed wrong cast and invalid array access in the calibration framework, as
documented in PR kern/45872. The AA() macro accessed sample coordinates as
long integers, whereas they are really stored as signed integers.

Fixes calibration on my Wetab device.

Comment out a variable only used in #if 0 code

branches: 1.4.64; 1.4.88; 1.4.94; 1.4.98; 1.4.104;
s/REGENTS/AUTHOR/

branches: 1.3.20; 1.3.22;
merge ktrace-lwp.

branches: 1.2.2;
Constify. Drivers pass const default calibration data to
WSMOUSEIO_SCALIBCOORDS using __UNCONST, so make sure we don't try to
modify it.

branches: 1.1.2;
Refactor tpcalib and hpc* ports to make tpcalib MI, as discussed on tech-kern.

Do not overflow int64_t on HP Spectre x360 13-inch ae019TU

The machine has pen touch panel with (18344, 10544) -> (3840, 2160)
conversion.

branches: 1.12.14;
don't forget to store bounds rectangle in raw mode

branches: 1.11.86; 1.11.106; 1.11.110; 1.11.114;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.10.4;
__unused removal on arguments; approved by core.

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

s/REGENTS/AUTHOR/

Remove (the default) #define XXXDEBUG and add the options to the relevant
files.xxx file.

branches: 1.6.20; 1.6.22;
merge ktrace-lwp.

branches: 1.5.2;
Constify. Drivers pass const default calibration data to
WSMOUSEIO_SCALIBCOORDS using __UNCONST, so make sure we don't try to
modify it.

Fix misleading indentation.

nuke trailing whitespace

branches: 1.2.2; 1.2.6; 1.2.8;
Add calibration support to uep driver.
Untested, still need a userland utility to calibrate with.

Refactor tpcalib and hpc* ports to make tpcalib MI, as discussed on tech-kern.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.3.4;
s/REGENTS/AUTHOR/

branches: 1.2.20; 1.2.22;
merge ktrace-lwp.

branches: 1.1.2; 1.1.14;
Refactor tpcalib and hpc* ports to make tpcalib MI, as discussed on tech-kern.

retire our private definitions for the scan1/3/5/7/9 DEC graphics
symbols, use the unicode definitions instead (which apparently didn't
exist when I wrote that)

branches: 1.3.118; 1.3.140;
Revert previous; the _e003U definition is used elsewhere.

On a VT100, the "scanline 5" character is also used for box drawing, so
represent it as U+2500 (BOX DRAWINGS LIGHT HORIZONTAL). This makes NetHack
DECgraphics mode work properly on a wscons console using an IBM-encoded font.

branches: 1.1.20; 1.1.22;
complete the font tables for the DEC special fonts; assign some codes
from the private Unicode area (0xe000 ff.) to stuff I didn't find in
the Unicode database

branches: 1.1.2; 1.1.6;
Allow big parts of the vt100 emulation code (what is in
wsemul_vt100_subr.c) to be used by alternative terminal emulators
(which are not in-tree yet but can be loaded as LKMs).
For this, split out that part of the state structure which is used
by the sharable code and include that in the original vt100 state.
This is only a query-replace and sed(1) job for now, it makes sense
to rearrange things a bit so that even more code can be changed --
will do so later.

For device modules that provide both auto-config and /dev/xxx
interfaces, make sure that initialization and destruction
follow the proper sequence. This is triggered by the recent
changes to the devsw stuff; per riastradh@ the required call
sequence is:

devsw_attach()
config_init_component() or config_cf*_attach()
...
config_fini_component() or config_cf*_detach()
devsw_detach()

While here, add a few missing calls to some of the detach
routines.

Testing of these changes has been limited to:
1. compile without build break
2. no related test failures from atf
3. modload/modunload work as well as
before.

No functional device testing done, since I don't have any
of these devices. Let me know of any damage I might cause
here!

XXX Some of the modules affected by this commit are already
XXX broken; see kern/56772. This commit does not break
any additional modules (as far as I know).

Explicitly include generated ioconf.h for struct cfdrivers.

branches: 1.12.10;
Add missing SETBELL() for WSKBDIO_COMPLEXBELL.
wskbd_bell_data structure has value fields and operation bitmask field.
"xset b * * 0" doesn't touch period field (and doesn't initialized it)
and doesn't set DOPERIOD flag (this means that period field should not
be used). SETBELL() handles this situation appropliately.
This happened during copying wsbell.c from wskbd.c.
Fix PR xsrc/54245.

Fix the bell period.

White space and indent fix.

branches: 1.9.2; 1.9.6;
Remove redundant includes of malloc.h

branches: 1.8.4;
Implement missing ioctls for the default bell.

Change sc_dying from u_char to bool.

Remove dying ftom vb_args and use sc_dying instead.

Move duplicate definition of SETBELL macro into wsbelldata.h.

Revert previous - it's no longer needed since data structure sizes no
longer depend on which wscons child devices exist

For now, #include wsdisplay.h to get the definition of NWSDISPLAY

The symbol is currently used to conditionally include members of a
data structure, so without this include there could be different
structure definitions, with different sizes, and data corruption is
likely.

XXX For cleaner modularization, all of the needs-flag entries in
XXX wscons should be removed.

Add module infrastructure for wsbell(4)

New device wsbell - allows for a console beep for non pckbds (usb etc).
Works for platforms without pcppi - can work with spkr at audio and spkr
at pcppi.

To use add the following to your kernel config:
wsbell* at spkr? console?

Ok pgoyette@.

branches: 1.2.4; 1.2.6;
Move duplicate definition of SETBELL macro into wsbelldata.h.

Seperate the definitions for the console beep. This code will be shared
with a new device.

Ok pgoyette@.

branches: 1.1.4; 1.1.6;
New device wsbell - allows for a console beep for non pckbds (usb etc).
Works for platforms without pcppi - can work with spkr at audio and spkr
at pcppi.

To use add the following to your kernel config:
wsbell* at spkr? console?

Ok pgoyette@.

branches: 1.1.4; 1.1.6;
New device wsbell - allows for a console beep for non pckbds (usb etc).
Works for platforms without pcppi - can work with spkr at audio and spkr
at pcppi.

To use add the following to your kernel config:
wsbell* at spkr? console?

Ok pgoyette@.

branches: 1.3.4; 1.3.6;
Change sc_dying from u_char to bool.

Remove dying ftom vb_args and use sc_dying instead.

New device wsbell - allows for a console beep for non pckbds (usb etc).
Works for platforms without pcppi - can work with spkr at audio and spkr
at pcppi.

To use add the following to your kernel config:
wsbell* at spkr? console?

Ok pgoyette@.

If you're going to use device_t, may as well include sys/device.h; unb0rks
the build for sparc{,64}.

Use aprint API for output and convert to device_t/softc separation.

branches: 1.17.8; 1.17.22; 1.17.24; 1.17.28;
KNF: No variable names in the prototype.

branches: 1.16.32; 1.16.62; 1.16.64;
Make wsdisplay_set_console_kbd() a little less twisted.

Major rototilling of the wsmux code. No user visible changes (except that
many bugs have been fixed).
Changes:
The wskbd, wsmouse, and wsmux are now "sub-classes" of wsevsrc, which is
a source of ws events. This make the structure of those drivers a little
more uniform.
Many bug fixes involving adding and removing devices from muxes.
When a kernel is configured without wsmux there will now be none (unlike
before where you got a console mux anyway).
The kernel now compiles with all combinations of ws devices present.

branches: 1.14.2;
ANSIfy.

branches: 1.13.2; 1.13.4;
- Add ioctl commands to wsdisplay which set/get some LCD status,
backlight, brightness and contrast.
- Add 9 command key symbols which make wskbd to call wsdisplay's
iocontrol.
- Add special key map variant KB_MACHDEP. This variant does not
have any real keymap but it's just a placeholder for machine dependent
maps which will be overwrite by machine dependent keyboard driver.

Add cnbell() support.

Done something half sensible if a USB console keyboard is unplugged instead
of panic()ing.
XXX Untested. I will test it the next time I can borrow an iMac.

branches: 1.10.2; 1.10.8;
Add the wsmux pseudo device.

Get rid of the timeout-driven wscons_glue stuff which caused a lot of
headaches.
Now console keyboard and display are connected at autoconfiguration time,
when the last of them is found. Other keyboards / displays remain
unconnected until a new ioctl (WSDISPLAYIO_SETKEYBOARD) is called.

branches: 1.8.4;
remove wsdisplay_has_emulator(), this is not meaningful with virtual screens,
and it did not work with external (serial) consoles what can cause that
a display has no screen and thus no "focus"

extend the interface for reset-like commands from the keyboard

-actually reset the terminal emulator on close()
-add a callback to reset the terminal emulator by keyboard command

Split the wskbd_attach arguments into console and autoconfiguration parts
(again); put some constant data into shared structures. This makes it
more look like the mouse part, and it helps USB integration.
Implement wskbd_enable() to enable/disable the keyboard.

Switch keyboard to "event mode" on open.
Nuke wskbd_set_translation(). It doesn't fit into the "screen" model.

Move the translation of keypad and function keys into the realm of the
terminal emulation. For this, change the interface to wsdisplay_kbdinput()
to take a "keysym_t" as argument. From there, the code is handed to the
appropriate emulation module via a new entry point: "translate".
Nuke the ioctls dealing with global assignment of character sequences
to keypad/function keys.
The "sun" emulation works much better now!

add "sleeping ok" argument to wsdisplay_switch()

Initial import of cgd's new wscons code.

Get rid of the timeout-driven wscons_glue stuff which caused a lot of
headaches.
Now console keyboard and display are connected at autoconfiguration time,
when the last of them is found. Other keyboards / displays remain
unconnected until a new ioctl (WSDISPLAYIO_SETKEYBOARD) is called.

branches: 1.5.4;
remove wsdisplay_has_emulator(), this is not meaningful with virtual screens,
and it did not work with external (serial) consoles what can cause that
a display has no screen and thus no "focus"

Insert lots of ``#if NWSMOUSE > 0'' and ``#if NWSDISPLAY > 0'' to make it
possible to have a wsmouse and/or wskbd without having a wsdisplay.

Move the translation of keypad and function keys into the realm of the
terminal emulation. For this, change the interface to wsdisplay_kbdinput()
to take a "keysym_t" as argument. From there, the code is handed to the
appropriate emulation module via a new entry point: "translate".
Nuke the ioctls dealing with global assignment of character sequences
to keypad/function keys.
The "sun" emulation works much better now!

Make it possible to build a kernel with a wscons display device but not
a wscons keyboard device.

Initial import of cgd's new wscons code.

fix various typos in comments.

KNF: No variable names in the prototype.

branches: 1.9.20; 1.9.22;
merge ktrace-lwp.

branches: 1.8.16;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.7.6;
alloc_attr -> allocattr

Approved by Matthias Drochner.

branches: 1.6.10;
ANSIfy.

branches: 1.5.20; 1.5.22;
change the "mapchar" interface to return a "quality" or "match level"
arguments which allows to find approximations for characters which are
not present in a font

add dummy character mapping function

Change the calling interface for text output (to the graphics driver)
to take a single character at a time, where the character is an "int" now.
The old interface (took a string) was never called with more than 1
char to print, and the "int" allows us to handle charsets cleanly.

Simple screen attribute handling for wscons.
The graphics device driver passes a "default attribute" for normal text
output to the wscons framework. If the emulation module needs more
attributes (for different "renditions") it can allocate them via a
callback.
For now, only the "sun" emulation makes use of it.

Initial import of cgd's new wscons code.

Complete initializers so those files compile.

branches: 1.3.20;
merge ktrace-lwp.

branches: 1.2.16;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.1.48;
Initial import of cgd's new wscons code.

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: 1.7.28; 1.7.30;
Consistently ifdef a variable and all uses

branches: 1.6.110; 1.6.120; 1.6.126;
merge ktrace-lwp.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.4.16;
add/cleanup RCSIDs

ANSIfy.

branches: 1.2.26; 1.2.28;
Simple screen attribute handling for wscons.
The graphics device driver passes a "default attribute" for normal text
output to the wscons framework. If the emulation module needs more
attributes (for different "renditions") it can allocate them via a
callback.
For now, only the "sun" emulation makes use of it.

Initial import of cgd's new wscons code.

merge ktrace-lwp.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.9.6;
alloc_attr -> allocattr

Approved by Matthias Drochner.

branches: 1.8.8;
add/cleanup RCSIDs

ANSIfy.

branches: 1.6.6; 1.6.8;
Remove register declarations.

branches: 1.5.8;
change the "mapchar" interface to return a "quality" or "match level"
arguments which allows to find approximations for characters which are
not present in a font

add dummy character mapping function

Change the calling interface for text output (to the graphics driver)
to take a single character at a time, where the character is an "int" now.
The old interface (took a string) was never called with more than 1
char to print, and the "int" allows us to handle charsets cleanly.

Simple screen attribute handling for wscons.
The graphics device driver passes a "default attribute" for normal text
output to the wscons framework. If the emulation module needs more
attributes (for different "renditions") it can allocate them via a
callback.
For now, only the "sun" emulation makes use of it.

Initial import of cgd's new wscons code.

PR 58681: add a port of the OpenBSD viogpu(4) driver written by
Joshua Stein <jcs@openbsd.org>.
Port by George Matsumura with help from Jiaxun Yang.

Add WSDISPLAYIO_GFONT ioctl to query the installed font name.

Add type for videocore6 display.

branches: 1.127.2;
wscons: Add HOLLYWOOD display and YUY2 pixel format types

branches: 1.126.4;
wsmouse: add support for "precision scrolling" events and (GET|SET)PARAMS

WSCONS_EVENT_HSCROLL and WSCONS_EVENT_VSCROLL are two new wscons event
types that allow scrolling with a higher precision ("smoothness") than an
emulated scroll wheel, and are useful for touch input drivers.

WSMOUSEIO_GETPARAMS and WSMOUSEIO_SETPARAMS are two new ioctls that allow
the speed and direction of precision scrolling to be configured.

both features were originally implemented in OpenBSD.

add an ioctl() to get a list of fonts currently available via wsfont

wscons(4): WSKBD_TYPE_PC_AT is automatically translated to
WSKBD_TYPE_PC_XT in modern kernels

Added drivers for Hyper-V Synthetic Keyboard and Video device.

allocate wsdisplay type 66 for ssdfb(4)

branches: 1.121.4;
Instead of a Linux keymap, convert Linux event codes to USB scan codes and
use the ukbd keymaps instead.

add WSKBD_TYPE_TCAKP for Texas Instruments TCA8418

New device wsbell - allows for a console beep for non pckbds (usb etc).
Works for platforms without pcppi - can work with spkr at audio and spkr
at pcppi.

To use add the following to your kernel config:
wsbell* at spkr? console?

Ok pgoyette@.

branches: 1.118.2;
Add type for ARM PrimeCell PL11x

add fbinfo flag for wildcat split framebuffer workaround

branches: 1.116.2;
Needed <sys/time.h> that it got in the past by side-effect loading
of <sys/stat.h> from <fcntl.h>. Go figure.

branches: 1.115.2;
add WSDISPLAY_TYPE_PLATINUM, for platinumfb

use optimized refresh with VCONS_DRAW_INTR even when in polling mode.

add WSDISPLAY_TYPE_TEGRA

add WSDISPLAY_TYPE_MESON

add MGX

tab police

branches: 1.109.2;
add WSDISPLAY_TYPE_ALLWINNER

branches: 1.108.2; 1.108.10;
Add WS{DISPLAY,KBD}_TYPE_* for epoc32.

add ioctl(WSDISPLAYIO_GET_FBINFO)

add WSDISPLAY_TYPE_OMAP3

add WSDISPLAY_TYPE_VC4 and WSDISPLAYIO_BUS_SOC

branches: 1.104.2;
Support i.MX51's LCD framebuffer on Netwalker.
from Kenichi Hashimoto.

branches: 1.103.2;
ass WSDISPLAY_TYPE_VALKYRIE

Add the Usermode vnc framebuffer and (vnc) remote keyboard types.

Amiga wscons/Xorg support, first phase.
Implemented wscons for CV64 and CV64/3D. Other graphics cards drivers are
prepared for it, but will not be attempted before Xorg is not running.
The wscons support is disabled by default. A GENERIC kernel should behave
like always. Use WSCONS to try out a kernel with wscons support.
Done by rkujawa@ and phx@.

branches: 1.100.2; 1.100.6;
Add and return WSKBD_TYPE_LUNA and WSDISPLAY_TYPE_LUNA on WS*_GTYPE ioctls.

XXX: should we still sync <sys/dev/wsconsio.h> and src/sbin/wsconsctl/util.c?

make the EDID buffer pointer void *

make sizes unsigned

add two new ioctl()s:
WSDISPLAYIO_GET_EDID to retrieve EDID data from display drivers
WSDISPLAYIO_SET_POLLING to control polling with VCONS_DRAW_INTR

add support for the interesting parts of ISO-2 and KOI8-R fonts
to the vga(4) driver

branches: 1.95.2;
Switch NetBSD/hp300 to wscons with rasops. Simply ported from OpenBSD/hp300.

- Only A1416 Kathmandu (topcat) framebuffer on 425t is tested, but
all other variants (TigerShark, Hyperion, DaVinci, GatorBox, Renaissance)
should also work if they are working on OpenBSD/hp300.
- sti(4) and SGC bus support are not pulled because I don't have 425e
and I can't confirm that 362 and 382 actually have SGC bus.
(I'll commit a DIO based dumb driver for 362 and 382 framebuffers later)
- Xorg server with wsfb driver will also be integrated soon
once after keycode with NoSymbol problem is addressed.
(We have to re-think what code should be used on WSDISPLAY_COMPAT_RAWKBD)
- MI HIL keyboard and mouse drivers are working fine though
cngetc via hilkbd has some problem (still we can input commands).
- No old HP-UX like HIL ioctl compatibility (we removed COMPAT_HPUX anyway).
grfinfo(8) and hilinfo(8) will be removed shortly.

Demonstrated on NetBSD booth at Open Source Conference 2011 Kagawa.

Implement new WSDISPLAYIO_GET_BUSID ioctl.
It returns the bus id and allows userland (like Xorg) to create mapping
of ttyE? and bus id. For now only PCI is implemented.

First discussed with macallan@ then public on tech-kern@ and tech-x11@

branches: 1.93.2; 1.93.4;
clarify how backlight control ioctl()s are supposed to work
TODO: adapt r128fb and radeonfb

branches: 1.92.2; 1.92.4;
add WSDISPLAY_TYPE_XVR1000

Reserve WSDISPLAY_TYPE for DisplayLink DL-1x0/DL-1x5 USB LCDs and
USB-VGA adaptors (which may require a new ioctl for DUMBFB ops). From
FUKAUMI Naoki.

G/c duplicate definition of WSCONS_EVENT_ASCII.

branches: 1.89.2;
provide wscons_event compatibility with 5.0.

branches: 1.88.22; 1.88.26; 1.88.34;
add type for Fujitsu AG-10e

branches: 1.87.4; 1.87.6;
add WSDISPLAY_TYPE_PXALCD for PXA2x0 LCD controller.

add WSDISPLAY_TYPE_CRIME for the SGI O2's graphics hardware

branches: 1.85.4;
add WSDISPLAY_TYPE_GENFB

branches: 1.84.2; 1.84.6; 1.84.8;
add #defines for ADB and pseudo mice

Add type for sgimips/light(4).

Add WSKBD_TYPE_ZAURUS (Sharp Zaurus keyboard).

Welcome to evbppc/virtex -- port to Xilinx Virtex series FPGA's with embedded
ibm405d5 core.

OK by Simon Burge

branches: 1.80.4; 1.80.6;
Initial import of bluetooth stack on behalf of Iain Hibbert. (plunky@,
NetBSD Foundation Membership still pending.) This stack was written by
Iain under sponsorship from Itronix Inc.

The stack includes support for rfcomm networking (networking via your
bluetooth enabled cell phone), hid devices (keyboards/mice), and headsets.

Drivers for both PCMCIA and USB bluetooth controllers are included.

branches: 1.79.2; 1.79.8; 1.79.10;
Bunch of changes to wscons in preparation for splash screen support:
* Add WSDISPLAY_TYPE_VESA for vesafb. While here, fix a typo in a comment.
* Add WSDISPLAYIO_SSPLASH and WSDISPLAYIO_SPROGRESS ioctls. The former
toggles the splash screen on and off, and the latter updates the progress
animation.
* Prevent more than one hw driver from claiming to be the console.
* In vcons, keep two pointers to the screen's vcons_data. This lets us
override the original (ie with null emulops during boot), and restore
them later on.

Add support to automatically repeat mouse button events in wsmouse(4) and
change wsconsctl(4) so that this is configurable.

This is specially useful for mice that provide page up/down buttons instead
of a real wheel and that do not send events repeatedly from the hardware.
(E.g.: Logitech Marble Mouse.)

No objections in tech-kern@.

branches: 1.77.2; 1.77.4; 1.77.6;
Add WSKBD_TYPE_EWS4800 for EWS4800 machines.

Some devices provide more than three (X, Y, and Z) "directions". So add
a W "coordinate" that can be used for these.
This changes the type of wsmouse_input(). To avoid changing a lot of drivers
a compatibilty #define is provided. Maybe changing all drivers would have
been better?

branches: 1.75.6;
TAB/space cosmetics.

branches: 1.74.2;
Make it possible for a console driver to deliver raw ASCII characters
to wskbd. This is for special cases only, and may even be considered a
hack, but it is cheap and very local.

branches: 1.73.2;
nuke trailing whitespace

Support wsdisplay(4) attachments of hd44780 LCD controllers

Add new wskbd type MATRIXKP

branches: 1.70.4; 1.70.6;
Import STI wscons(4) driver from OpenBSD and add new driver for PS/2
keyboard / mouse pots of LASI.

Implement border color customization in wscons(4), only available for vga(4)
at the moment.

This includes the addition of two new wsdisplay ioctls, WSDISPLAY_{G,S}BORDER,
one to get the actual color and one to set it, respectively. Possible colors
match those defined by ANSI (and listed in wsdisplayvar.h).

It also adds two accessops to the underlying graphics device, getborder and
setborder, which mach their ioctl counterparts.

Two kernel options are added: WSDISPLAY_CUSTOM_BORDER, which enables the
ioctls described above (to customize the border color from userland after
boot), and WSDISPLAY_BORDER_COLOR, which sets the color at boot time.
The former is enabled by default on the GENERIC kernel, but not on INSTALL
(among others). The later is always commented out, leaving the usual black
border as a default.

wsconsctl is modified to allow accessing this value easily. For example,
'wsconsctl -d -w border=blue'.

Implement support to dynamically change wscons console and kernel colors.

Two new ioctls are added to the wsdisplay device, named WSDISPLAY_GMSGATTRS
and WSDISPLAY_SMSGATTRS, used to retrieve the actual values and set them,
respectively (the name, if you are wondering, comes from "message attributes").

A new emulop is added to the underlying display driver (only vga, for now)
which sets the new attribute for the whole screen, without having to clear
it. This is optional, which means that this also works with other drivers
that don't have this new operation.

Five new kernel options have been added, although only documented in
i386 kernels (for now):
- WSDISPLAY_CUSTOM_OUTPUT, which enables the ioctls described above to
change the colors dynamically from userland. This is enabled by default
in the GENERIC kernel (as well as others) but disabled on all INSTALL*
kernels (as this feature is useless there).
- WS_DEFAULT_COLATTR, WS_DEFAULT_MONOATTR, WS_DEFAULT_BG and WS_DEFAULT_FG,
which specify the default colors for the console at boot time. These have
the same meaning as the (already existing) WS_KERNEL_* variables.

wsconsctl is modified to add msg.default.{attrs,bg,fg} and
msg.kernel.{attrs,bg,fg} to the display part, so that colors can be changed
after boot.

Tested on NetBSD/i386 with vga (and vga in mono mode), and on NetBSD/mac68k.
No objections in tech-kern@.

Add WSDISPLAYIO_[GS]MODE type _DUMBFB - mapped fb (no registers)
Add WSDISPLAYIO_LINEBYTES ioctl - # bytes/row

Used by ffb driver and XFree wsfb driver module - From OpenBSD

Add SGI types for mouse and keyboard.

make the scroll lines ioctl struct take u_int's again. The wsconsctl
command does not deal with shorts well.

White space changes only;
1. put value defines under the fields that they refer too, all the time
not just in some of the ioctls.
2. use #define<tab>value consistently.

Re-group scrolling ioctl's and provide new numbers for them! (thanks to
John Heasley for catching that).

PR/19925: David Ferlier: Add scrolling support to wscons.

branches: 1.61.2;
Add sun display types.

Add console type for SGI GR2 family.

Add display type for SGI Newport.

branches: 1.58.2;
Add sun keyboards and display.

undo constification of 'data' for wsdisplay font - unfortunately we need
to be able to adjust the font to display adapter needs very soon (in consinit()),
so it's not possible to do font copy

make 'name' and 'data' of struct wsdisplay_font const, mark data arrays
in font sources const

Define Atari wskeyboard type.

Added new ioctl command, WSMOUSEIO_GETID to tell touch panel identifier
for tpctl(8).

Add WSKBD_TYPE_MAPLE and WSMOUSE_TYPE_MAPLE for Dreamcast Maple bus devices.

Fix some small unclear documentation and cleanup a bit...

PR/17402: Add wsmoused support by providing get/set char and events.

branches: 1.50.2;
Rename WSDISPLAYIO_USEFONT to WSDISPLAYIO_SFONT.

Approved by Matthias Drochner <drochner@netbsd.org>.

Add WSMOUSE_TYPE_MAXINE.

added amiga display/mouse/keyboard defines

Simply entries added in previous.

Add identifiers for HP devices.

Fix typo in comment.

Major rototilling of the wsmux code. No user visible changes (except that
many bugs have been fixed).
Changes:
The wskbd, wsmouse, and wsmux are now "sub-classes" of wsevsrc, which is
a source of ws events. This make the structure of those drivers a little
more uniform.
Many bug fixes involving adding and removing devices from muxes.
When a kernel is configured without wsmux there will now be none (unlike
before where you got a console mux anyway).
The kernel now compiles with all combinations of ws devices present.

branches: 1.43.2;
Add a WSKBD_TYPE_ADB for Macs.

Get/set keyclick volume.

branches: 1.41.2;
add encoding for ISO-2

introduce hard limit on maximum font size (WSDISPLAY_MAXFONTSZ - 512KB)
and keymap size (WSKBDIO_MAXMAPLEN - 64KB)

branches: 1.39.2;
Make Sony's fonts ISO8859-1 compliant.

Add RiscPC (arm32) keyboard. It basicly returns AT scancodes but since it has to
use a seperate keyboard driver (pckbdc not available) i cant guarantee that it is
100% compatible with AT scan codes

branches: 1.37.2;
Added a fontencoding for the Sony fonts.

Add a display type for the Dreamcast PowerVR.

These ioctl commands should be '_IOW' to set value.
WSMOUSEIO_SRES
WSMOUSEIO_SSCALE
WSMOUSEIO_SRATE

- Add ioctl commands to wsdisplay which set/get some LCD status,
backlight, brightness and contrast.
- Add 9 command key symbols which make wskbd to call wsdisplay's
iocontrol.
- Add special key map variant KB_MACHDEP. This variant does not
have any real keymap but it's just a placeholder for machine dependent
maps which will be overwrite by machine dependent keyboard driver.

add definitions for ISO-7 (greek) support

Change type `kbd_t' from `u_int_16_t' to `u_int32_t'.
This eliminates bogus casts from `int *' to `u_int_16_t*'.

branches: 1.31.2;
add EGA display type

Add VAX display types along with the SPARCbook P9100.

branches: 1.29.2;
Add Archimedes mouse/kbd/video

HPC/PsPC keyboard types.

Add a new frame buffer identifier for the hpcmips.

WSMOUSE_SCALIBCOORDS had not worked because ioctrl code was wrong.

add the NeXT mouse type

Absolute pointing device support.
- Wsmouse_input() get new argument 'flag', which indicates whether x/y/z are
relative or absolute.
- Wsmouse get new io controls, WSMOUSEIO_SCALIBCOORDS and
WSMOUSEIO_GCALIBCOORDS.

Don't define WSDISPLAY_FONTORDER_L2R as 0. The function wsfont_lock uses
the value 0 specially.

Got too excited with cut and paste in previous. Noted by itojun.

Add two new members to 'struct wsdisplay_font': 'bitorder' and 'byteorder'.

branches: 1.20.2;
Add one new dispaly type for DEC TC TX 8/24bpp framebuffer. Currently acts
as an 8bpp pseudo color device.

Add the wsmux pseudo device.

Get rid of the timeout-driven wscons_glue stuff which caused a lot of
headaches.
Now console keyboard and display are connected at autoconfiguration time,
when the last of them is found. Other keyboards / displays remain
unconnected until a new ioctl (WSDISPLAYIO_SETKEYBOARD) is called.

Add ioctl definitions for setting mouse resolution, scale factor and sample
rate. Add two new display types (DEC PX, PXG). Replace instances of '???' in
comments about DEC hardware with meaningful representations.

branches: 1.16.4;
"NEXT" is a keyboard type, not a keymap variation.

wscons based console video & keyboard support for the NeXT
from Matt DeBergalis

forgot to commit last week: define wsmouse type constants for busmice

change the way fonts are handled: now we have a display-global ioctl
to download fonts and a per-screen call to select a font
allows easier sharing of ressources in the display driver
change the parameters to the load_font driver vector to pass all
available font information

add ioctl()s for screen allocation/deallocation

Add USB keyboard type.

Add a wscons display type `unknown' that the generic VGA driver can
return until someone fixes it for real.

Add a special wsmouse type for USB mice.

Define a new keyboard eveny type "all keys up". This allows to reduce
the amount of statekeeping in drivers for LK201-like keyboards, and it
might be useful to resync after resets or overflows.
(reused the unused WSCONS_EVENT_KEY_OTHER event ID)
Implement the necessary decoding bits.

add definition for pointing device's absolute z coordinates

define a z axis for pointing devices which support this

Fix ioctl definitions for WSKBDIO_{SET,GET}LEDS.
from Takuya Koumoto <takuya-k@is.aist-nara.ac.jp>

Move the translation of keypad and function keys into the realm of the
terminal emulation. For this, change the interface to wsdisplay_kbdinput()
to take a "keysym_t" as argument. From there, the code is handed to the
appropriate emulation module via a new entry point: "translate".
Nuke the ioctls dealing with global assignment of character sequences
to keypad/function keys.
The "sun" emulation works much better now!

implement a way to get raw keyboard data (for X)

Initial import of MI keyboard translation for dev/wscons.

Initial import of cgd's new wscons code.

avoid a dangling pointer error by making sure all exits don't leave
a pointer that won't be actually used. found by GCC 14.

nouveau: Kick out genfb on firmware framebuffer before initializing.

PR kern/53126

branches: 1.165.4;
wsdisplay(4): KNF: return foo, not return (foo)

No functional change intended.

wsdisplay(4): KASSERT(a && b) -> KASSERT(a); KASSERT(b)

Better diagnostics this way. No other functional change. Omit
needless #ifdef DIAGNOSTIC while here; the compiler can optimize the
dead code away.

libkern: Make KASSERT verify expression is valid if !DIAGNOSTIC.

This way it is no longer necessary to mark variables __diagused if
they are used in KASSERT conditions.

Fix fallout from this by removing now-unnecessary and `#ifdef
DIAGNOSTIC'.

Don't do the same for KDASSERT if !DEBUG -- unlike KASSERT and
DIAGNOSTIC, variables needed by KDASSERT and DEBUG are likely to be
expensive to compute (and potentially difficult for a compiler to
prove flushable), so we don't want to require them under !DEBUG.

Explicitly include generated ioconf.h for struct cfdrivers.

Use C99 designated initializers for struct consdev.

branches: 1.160.8;
localify

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.

Halt the old console on both cnattach and preattach

Call cn_halt for old console in wsdisplay_preattach too

Enable us to enter DDB from serial console when WSDISPLAY_MULTICONS
is enabled. This is necessary for machines on which framebuffer
console cannot be disabled at run time.

opt_wsdisplay_compat.h is generated for all ports including those
that do not support wscons.

OK Jared

Call cn_halt on the old console device

be consistent in checking cn_getc return value.

- add or adjust /* FALLTHROUGH */ where appropriate
- add __unreachable() after functions that can return but won't in
this case, and thus can't be marked __dead easily

Suspend multicons output to the old console device while replaying
the vcons msgbuf.

Revert previous as it does not completely solve the problem of detaching a wsdisplay device

Remove KASSERT in wsdisplay_cndetach. The caller likely doesn't know the
state of the console, so simply do cleanup if a previous console has been
attached.

Add hw.wsdisplay.multicons sysctl that can be used to disable mirroring wsdisplay output to serial port.

WSDISPLAYIO_SFONT: Do not attempt to print to the console in the middle of
resizing when DEBUG is defined.

Introduce wsdisplay_isconsole(struct wsdisplay_softc *), which
returns sc->sc_isconsole

fix wsdisplay/wskbd and cngetc() return value issues:
- return -1 for error / no character
- allow 0 as a character

cngetc() itself considers any value >= 0 valid, and this matches
the usage in other drivers that return -1 for "not data", vs
polling until something arrives.

removes ^G from spewing out the askname/etc prompts when no key
has been pressed (these come from cngetsn(), ignoring unknown
non printable characters.)

from @skrll with help from @mlelstv.

branches: 1.145.2; 1.145.4;
ifdef speeding ticket.

Add WSDISPLAY_MULTICONS option which allows wsdisplay to cooperate with
the console device being overridden. This allows for simultaneous console
output on fb and debug uart.

Use M_WAITOK if we aren't going to check for NULL before dereferencing
Same function already uses M_WAITOK unconditionally, so this is a safe change

While here, drop null check of M_WAITOK. it shouldn't fail.

add support for loading fonts in vcons and subsequently resizing screens
- drivers can use this by setting VCONS_LOADFONT and WSSCREEN_RESIZE
- each vcons screen can now have its own font and geometry
- while there, add support for xterm's ESC[18t to report the text buffer's
size

With this tou can:
wsfontload -N foo /usr/share/wscons/fonts/flori.816
wsconsctl -dw font=foo
currently this is limited to drivers that don't use the glyph cache, like genfb

Double size of WSDISPLAYIO_SFONT name buffer to 32 characters.

We've got a font known as "DejaVu Sans Mono", and that is 17 bytes
by the time it's NUL terminated.

branches: 1.140.2; 1.140.4;
KNF. Remove extra spaces. No functional change.

to garnish, dust with _KERNEL_OPT

branches: 1.138.4;
Add d_discard to all struct cdevsw instances I could find.

All have been set to "nodiscard"; some should get a real implementation.

branches: 1.137.2;
Change (mostly mechanically) every cdevsw/bdevsw I can find to use
designated initializers.

I have not built every extant kernel so I have probably broken at
least one build; however I've also found and fixed some wrong
cdevsw/bdevsw entries so even if so I think we come out ahead.

add function to detach wscons from a console. This allows to
switch the console to a different driver.

branches: 1.135.6; 1.135.10;
delete virtual screens on detach - this allows to hot-unplug
a udl@usb monitor without crash

branches: 1.134.4; 1.134.8;
Rename ttymalloc() to tty_alloc(), and ttyfree() to tty_free() for
consistency. Remove some unnecessary malloc.h inclusions as well.

branches: 1.133.2;
add just as much "detach" code to keep the box from panicking on
removal of a USB "udl" device
(needs more work, but I got the hardware for a quick test only)

branches: 1.132.2;
various aprint_* fixes.

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: 1.130.2;
give up some silly macros which were only used ~once, and reduce
dependency on "locators.h"
makes the code more LKM frienly

add hooks to allow terminal emulations to be installed by LKMs
(these are not available in early bootstrap, so this is not an
option for the system's default emulation)

-rename the "KS_GROUP_Ascii" key symbol category to "Plain" because
this is what it meant all the time -- it includes characters >127
-delegate translation of all key symbols >127 to the terminal emulation
layer -- formerly everything non-special was passed to the tty subsystem
directly, involving truncation to a "char", which effectively meant
some ISO-1 assumption

Expand PMF_FN_* macros.

Avoid deadlock in tty code if a terminal emulation responds to
type/status/etc inquiries. (PR kern/37915)
This is clearly a design problem in tty, but we need a cheap fix now.
The problem is that ttyinput() tries to pull a spinlock which
is already held on calls to t_oproc.
The workaround is based on the fact that within wscons code, the
wsdisplay_emulinput() function is only called directly from
wsdisplaystart(). So we can be sure that the tty lock is held,
and use an inofficial entry point in ttc.c which avoids the locking.
These ate certainly more assumptions than needed by the fix
proposed in the PR, but it doesn't affect (and slow down) other
tty drivers.

- reduce the number of #ifdefs.
- build compat glues if MODULAR.

provide wscons_event compatibility with 5.0.

In wsdisplay_kbdholdscreen(), just return if sc->sc_focus is NULL.
There is apparently no guarantee sc->sc_focus will be non-NULL,
as proved by a crash I experienced earlier today (admittedly on 4.0.1).

branches: 1.122.2; 1.122.4; 1.122.6;
use device_lookup_private to get softc
use device_lookup to get device_t

use device_lookup_private to get sc
ok cube

branches: 1.120.2; 1.120.4; 1.120.6; 1.120.8;
Split device_t and softc for wskbd(4), wsmouse(4) and that creepy wsmux(4).

-add a function to vga_post which can be called from DDB to get a
usable VGA console ("call ddb_vgapost")
-allow to switch from/to screens occupied by an X server if the graphics
device is console and in polling mode (ie DDB)
This together allows to get a DDB session on a VGA console if the
system crashed while X11 was running.
As long as the protocol to tell X servers about virtual screen switches
is as primitive as it is, it is unsafe to restart an X session afterwards.
So this is basically for crash analysis.

Use PMF_FN_ARGS, PMF_FN_PROTO.

branches: 1.117.2; 1.117.6;
-Don't detach/attach the X server in the pmf suspend/resume functions
which are called somewhere in the middle of system suspend. Since the
X server accesses hardware directly it is outside our control whether
the devices it accesses are already/still suspended or not, so the only
way is to detach it before any device suspends and re-attach after
everything is awake again.
-For that, export a function ("wsdisplay_handlex" for now) which is
to be called from central suspend/resume code.
-The right way to handle the (normally impossible) case that the X
server is not detached on suspend is to return an error which should
abort the suspend process. pmf doesn't yet handle errors of device
suspend handlers, so as a temporary measure try to suspend anyway,
to get at least a text console.
-Improve error handling of X server attach/detach and maintain a flag
which tracks whether the X server is really active.

make the mutex securing a flag which controls screen switched adaptive
for sanity -- it is not yet in a code path which is called from
keyboard interrupt handlers

On suspend/resume, just detach/attach the X server (or whatever might
be using the "syncops" protocol) rather than doing a full switch
to virtual screen 1. The latter doesn't have the expected result if
the X server is running on screen 1, and is too much policy for inside
the kernel anyway.
Introduce a mutex+condvar to synchronize these operations, to replace
the spltty()/tsleep(). (for now only used by the new functions,
should be extended to the normal vt switching code)

Switch to console on suspend and shutdown, switch back to where the
system was before on resume. This is the second attempt and got more
complicated due to the async nature of VT_ACTIVE. After the initial
switch, check that we ended up on the first screen and if not, wait for
the switch to happen.

branches: 1.113.2;
Merge jmcneill-pm branch.

branches: 1.112.2; 1.112.4;
- Factor out too many copies of the same bit of tty code.
- Fix another tty signalling/wakeup problem.

branches: 1.111.2;
Calling device_private once is enough. Reported by macallan@

Initialise the callbacks for tty.t_rstrt_ch in ttymalloc
as all drivers but Sun/SPARC's kd.c use the same arguments.
Separate callout_reset into callout_schedule and the initial
callout_setfunc using that.

Use aprint API for output and convert to device_t/softc separation.

branches: 1.108.2; 1.108.14; 1.108.16; 1.108.20;
It was probably not a good idea to put that scroll stuff to accessops.
But now that it is there, keep it from crashing the box if there is
no valid focus screen.

Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.106.2; 1.106.4; 1.106.8;
__unused removal on arguments; approved by core.

add wsdisplay_preattach() to attach an early console that can be overridden
by wsdisplay_cnattach() for instance when a hardware-specific display driver
attaches.
as discussed on tech-kern

now that we have -Wno-unused-parameter, back out all the tremendously ugly
code to gratuitously access said parameters.

More -Wunused fallout. sprinkle __unused when possible; otherwise, use the
do { if (&x) {} } while (/* CONSTCOND */ 0);
construct as suggested by uwe in <20061012224845.GA9449@snark.ptc.spbu.ru>.

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

Adapt MD code to KAUTH_DEVICE_TTY_OPEN, batch #2 from Matt Fleming, thanks!

Also, add forgotten splx() calls in some places.

branches: 1.100.4; 1.100.6;
- Use the LWP cached credentials where sane.
- Minor cosmetic changes.

add a missing splx() in an error path

branches: 1.98.4;
integrate kauth.

Remove the getwschar and putwschar accessops from wsdisplay drivers as
requested by uwe@. These were wrong because they were receiving an
emulcookie yet they were accessops (thus having to receive an accesscookie).
Instead, just handle the WSDISPLAYIO_{GET,PUT}WSCHAR ioctls from the
driver's ioctl accessop.

As this reduces the amount of code needed to handle these operations to
two small functions in each driver, remove the WSDISPLAY_CHARFUNCS kernel
option.

Reviewed by, at least, uwe@ and macallan@. No objections in tech-kern@.

pass the right cookie to ioctl() and mmap() instead of struct wsscreen *

correct typo

Add an extra cookie to the ioctl and mmap wsdisplay accessops that points
to the screen on which they are being called. The driver cannot guess
this by itself but it is needed to implement, at least, the getwschar and
putwschar functions in the correct place. There are no functional changes
yet.

Tested on i386 (vga, vga_raster, machfb, vesafb), macppc and sparc64.
Suggested and reviewed by macallan@.

Use the existing ioctl path to call WSDISPLAYIO_GBORDER/WSDISPLAYIO_SBORDER
instead of a separate accessops entry. There is no need to introduce bloat
for the majority of drivers which don't implement this feature.
This should also resolve PR kern/33186 by Valeriy E. Ushakov.

Use device_cfdata().

Use device_unit().

branches: 1.90.2; 1.90.4;
cleanup more SET/CLR/ISSET lossage

branches: 1.89.2;
Bunch of changes to wscons in preparation for splash screen support:
* Add WSDISPLAY_TYPE_VESA for vesafb. While here, fix a typo in a comment.
* Add WSDISPLAYIO_SSPLASH and WSDISPLAYIO_SPROGRESS ioctls. The former
toggles the splash screen on and off, and the latter updates the progress
animation.
* Prevent more than one hw driver from claiming to be the console.
* In vcons, keep two pointers to the screen's vcons_data. This lets us
override the original (ie with null emulops during boot), and restore
them later on.

wsevent cleanup:

- Add a wsevent_inject function that atomically adds a set of events to an
event queue and change all code that directly messed with a queue to use it.
- Replace the WSEVENT_WAKEUP macro with a regular function.
- Make WSEVENT_QSIZE, PWSEVENT and splwsevent private definitions to
wsevent.c, instead of exposing them in the header file.
- Make the wsevent_init function take a process to attach to the queue,
instead of leaving this task to the caller (which always did it).

Reviewed in tech-kern@.

branches: 1.87.2; 1.87.4; 1.87.6;
merge ktrace-lwp.

Call nanotime() directly, instead of doing the
microtime()/TIMEVAL_TO_TIMESPEC() dance.

Change the driver open function's conditional for overriding exclusive tty
use from checking the proc's uid to suser(9), and account for the use of
privileges. Noted by David Holland in PR kern/31126.

branches: 1.84.2;
PR-30566: Poll must not return <sys/errno.h> values.
Start with those places I can easily test.

- sprinkle const.
- avoid variable shadowing.

nuke trailing whitespace

branches: 1.81.4; 1.81.6;
Implement border color customization in wscons(4), only available for vga(4)
at the moment.

This includes the addition of two new wsdisplay ioctls, WSDISPLAY_{G,S}BORDER,
one to get the actual color and one to set it, respectively. Possible colors
match those defined by ANSI (and listed in wsdisplayvar.h).

It also adds two accessops to the underlying graphics device, getborder and
setborder, which mach their ioctl counterparts.

Two kernel options are added: WSDISPLAY_CUSTOM_BORDER, which enables the
ioctls described above (to customize the border color from userland after
boot), and WSDISPLAY_BORDER_COLOR, which sets the color at boot time.
The former is enabled by default on the GENERIC kernel, but not on INSTALL
(among others). The later is always commented out, leaving the usual black
border as a default.

wsconsctl is modified to allow accessing this value easily. For example,
'wsconsctl -d -w border=blue'.

Implement support to dynamically change wscons console and kernel colors.

Two new ioctls are added to the wsdisplay device, named WSDISPLAY_GMSGATTRS
and WSDISPLAY_SMSGATTRS, used to retrieve the actual values and set them,
respectively (the name, if you are wondering, comes from "message attributes").

A new emulop is added to the underlying display driver (only vga, for now)
which sets the new attribute for the whole screen, without having to clear
it. This is optional, which means that this also works with other drivers
that don't have this new operation.

Five new kernel options have been added, although only documented in
i386 kernels (for now):
- WSDISPLAY_CUSTOM_OUTPUT, which enables the ioctls described above to
change the colors dynamically from userland. This is enabled by default
in the GENERIC kernel (as well as others) but disabled on all INSTALL*
kernels (as this feature is useless there).
- WS_DEFAULT_COLATTR, WS_DEFAULT_MONOATTR, WS_DEFAULT_BG and WS_DEFAULT_FG,
which specify the default colors for the console at boot time. These have
the same meaning as the (already existing) WS_KERNEL_* variables.

wsconsctl is modified to add msg.default.{attrs,bg,fg} and
msg.kernel.{attrs,bg,fg} to the display part, so that colors can be changed
after boot.

Tested on NetBSD/i386 with vga (and vga in mono mode), and on NetBSD/mac68k.
No objections in tech-kern@.

Add WSDISPLAYIO_[GS]MODE type _DUMBFB - mapped fb (no registers)
Add WSDISPLAYIO_LINEBYTES ioctl - # bytes/row

Used by ffb driver and XFree wsfb driver module - From OpenBSD

return ENODEV if functions are not compiled in the kernel.

PR/19925: David Ferlier: Add scrolling support to wscons.

branches: 1.76.2;
Do much more sanity checks when handling entries in the sc_scr array
of struct wsdisplay_softc. Unused entries are NULL, and there were
a lot of places where we did not checked that the caller did not
requested an operation on a NULL entry.

While we are there, one bit of KNFification, and make return code more
consistent by always returning the same code (EINVAL) when a screen
number higer than the maximum is requested.

branches: 1.75.2;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

Pass lwp pointers throughtout the kernel, as required, so that the lwpid can
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V

Add cn_halt and cn_flush entries to consdevs. (needed for dma-only console
devices).

Use u_int variables for some variables that are passed to
functions that expect u_int parameters.
Remove a semi-colon after the trailing brace of a function.

Use aprint_normal() in cfprint routines.

merge kqueue branch into -current

kqueue provides a stateful and efficient event notification framework
currently supported events include socket, file, directory, fifo,
pipe, tty and device changes, and monitoring of processes and signals

kqueue is supported by all writable filesystems in NetBSD tree
(with exception of Coda) and all device drivers supporting poll(2)

based on work done by Jonathan Lemon for FreeBSD
initial NetBSD port done by Luke Mewburn and Jason Thorpe

Add trailing ; to CFATTACH_DECL.

Use CFATTACH_DECL().

Declare all cfattach structures const.

remove trailing \n in panic(). approved perry.

Merge the gehenna-devsw branch into the trunk.

This merge changes the device switch tables from static array to
dynamically generated by config(8).

- All device switches is defined as a constant structure in device drivers.

- The new grammer ``device-major'' is introduced to ``files''.

device-major <prefix> char <num> [block <num>] [<rules>]

- All device major numbers must be listed up in port dependent majors.<arch>
by using this grammer.

- Added the new naming convention.
The name of the device switch must be <prefix>_[bc]devsw for auto-generation
of device switch tables.

- The backward compatibility of loading block/character device
switch by LKM framework is broken. This is necessary to convert
from block/character device major to device name in runtime and vice versa.

- The restriction to assign device major by LKM is completely removed.
We don't need to reserve LKM entries for dynamic loading of device switch.

- In compile time, device major numbers list is packed into the kernel and
the LKM framework will refer it to assign device major number dynamically.

PR/17402: Add wsmoused support by providing get/set char and events.

branches: 1.63.2;
Rename WSDISPLAYIO_USEFONT to WSDISPLAYIO_SFONT.

Approved by Matthias Drochner <drochner@netbsd.org>.

Convert ioctl code to use EPASSTHROUGH instead of -1 or ENOTTY for
indicating an unhandled "command". ERESTART is -1, which can lead to
confusion. ERESTART has been moved to -3 and EPASSTHROUGH has been
placed at -4. No ioctl code should now return -1 anywhere. The
ioctl() system call is now properly restartable.

Added wsmux_set_display() to fix the problem that wskbd wan't connected
to any display if keyboard was attached before display and both aren't console.

add/cleanup RCSIDs

Make wsdisplay_set_console_kbd() a little less twisted.

Put in some tests to see if the input source is NULL before using it.
Suggested by Havard Eidnes <he@netbsd.org>.

Cosmetics.

Major rototilling of the wsmux code. No user visible changes (except that
many bugs have been fixed).
Changes:
The wskbd, wsmouse, and wsmux are now "sub-classes" of wsevsrc, which is
a source of ws events. This make the structure of those drivers a little
more uniform.
Many bug fixes involving adding and removing devices from muxes.
When a kernel is configured without wsmux there will now be none (unlike
before where you got a console mux anyway).
The kernel now compiles with all combinations of ws devices present.

branches: 1.55.2;
Don't use wsmux_getmux() if we NWSMUX == 0. Fixes PR kern/14252.
This is only a stop gap measure until the real wsmux changed get in.

ANSIfy.

Two changes to the wsmux code:
* Allow the wsmux used by wsdisplay for the keyboard(s) to be explicitely
specified with the kbdmux locator.
* Allow keyboards and mice that have a mux to be opened in the regular way.
These changes should be totally backwards compatible.

branches: 1.52.2;
introduce hard limit on maximum font size (WSDISPLAY_MAXFONTSZ - 512KB)
and keymap size (WSKBDIO_MAXMAPLEN - 64KB)

branches: 1.51.2;
do this console device lookup thing the right way this time
(This "dev" argument to the console functions is nonsense - nothing
should depend on it.)

Add `l_poll' to `struct linesw' and provide an xxxpoll() entry point
in each tty driver to indirect through it.

This allows tty line-disciplines to handle poll(2) system calls.

branches: 1.49.2;
Redo previous; we need to notify kbd driver.

Avoid null pointer dereference so that boot -d works again.

Use device_lookup.

for calling sc->sc_acessops->pollc,
use wsdisplay_cd.cd_devs[unit] in accessing wsdisplay_softc
instead of casting pointer.
XXX: i compiled it in i386, hpcmips, alpha.

Fix at least alpha kernel builds. You can't pun a pointer to a dev_t on
alpha. This is also a bad idea. I really don't know the ws code at all
or I would just fix the whole issue.

notify console polling mode or not to lower level framebuffer driver.
notify EMUL mode or not to lower level framebuffer driver.
these may use framebuffer level hardware acceleration/software optimization.

In wsdisplay_switchtoconsole(), ensure that the switch happens
synchronously. Suggested by drochner.

Adapt to the new line discipline scheme.

- Add ioctl commands to wsdisplay which set/get some LCD status,
backlight, brightness and contrast.
- Add 9 command key symbols which make wskbd to call wsdisplay's
iocontrol.
- Add special key map variant KB_MACHDEP. This variant does not
have any real keymap but it's just a placeholder for machine dependent
maps which will be overwrite by machine dependent keyboard driver.

* rename vga_stdscreen* -> vga_25lscreen*
* in vga_init(), set the screen type to WSCONS_DEFAULT_TYPE, which defaults
to "80x25".
XXX: the code currently makes no attempt to ensure that a font
with the appropriate width & height is available, effectively
limiting this default to either "80x25" or "80x24" at this
time.
* make wsdisplay_screentype_pick() non static, so that vga_init() can use it

Introduce WSDISPLAY_NULLSCREEN to allow low level device to detach control
process. Hpcmips frame buffer driver(hpcfb) call wsdisplay_switch with
WSDISPLAY_NULLSCREEN to notify user process of suspend/resume event.

Change the kernel mmap interface so that the offset to map is an
"off_t" and the return value is a "paddr_t" to allow mappings
at offsets past 2^31 bytes. Somewhat inspired by FreeBSD, which
only changed the offset to a "vm_offset_t".

Includes updates for the i386, pc532 and sh3 mmmmap from Jason Thorpe.

branches: 1.37.4;
Remove register declarations.

New callout mechanism with two major improvements over the old
timeout()/untimeout() API:
- Clients supply callout handle storage, thus eliminating problems of
resource allocation.
- Insertion and removal of callouts is constant time, important as
this facility is used quite a lot in the kernel.

The old timeout()/untimeout() API has been removed from the kernel.

Add cnbell() support.

replace embedded _rcsid[] string by __KERNEL_RCSID(), remove _copyright[]

-allow the "show_screen" call to the display driver to complete
asynchronously, in the same style like the process attach/detach functions
-intercept the "cnpollc" call which originally went directly to the
keyboard driver and keep track whether the console is in "polling" state
(DDB!)
-pass a NULL callback to the screen switcher and the process attach/detach
functions if the console is "polling", to tell them that asynchronous
completion is forbidden

Done something half sensible if a USB console keyboard is unplugged instead
of panic()ing.
XXX Untested. I will test it the next time I can borrow an iMac.

branches: 1.31.2;
Split wsdisplay_usl_ioctl() into two functions -- one which is screen-dependent
and one which isn't. The latter is now used for ttyEcfg, enabling the
VT-switching ioctls to work on it. (This allows Linux X servers to work when
/emul/linux/dev/tty0 is linked to /dev/ttyEcfg.)

branches: 1.30.2;
add wsdisplay_switchtoconsole() to allow other parts of kernel to force
the wscons to switch to console; code taken from wsdisplay_shutdownhook(),
which is now just a wrapper around wsdisplay_switchtoconsole()

branches: 1.29.2;
At shutdown, switch to screen 0 on console display so that we see some
notification from the kernel.

introduce new variable wsdisplay_defaultscreens (initialized to
WSDISPLAY_DEFAULTSCREENS or 0 if it's not defined) and use it instead
of WSDISPLAY_DEFAULTSCREENS as approptiate, so that number of screens
added on bootup is patchable

move printing of info about added screen into separate routine and
print just one message for all screens added in wsdisplay_common_attach(),
such as:
wsdisplay0: screen 1-7 added (80x25, vt100 emulation)

while here, do minor const poisoning

Reviewed by: drochner
Tested by: jdolecek

defopt COMPAT_14

Add the wsmux pseudo device.

add some #ifdefs to make it link w/o wskbd
(especially for vax)

Get rid of the timeout-driven wscons_glue stuff which caused a lot of
headaches.
Now console keyboard and display are connected at autoconfiguration time,
when the last of them is found. Other keyboards / displays remain
unconnected until a new ioctl (WSDISPLAYIO_SETKEYBOARD) is called.

Sort out the inclusion of wscons source files: the files are only
compiled if the corresponding devices are configured.
(Files calling wscons functions have to check whether these are present.)

branches: 1.22.2; 1.22.4;
If no screen got the "focus" (which happens with serial console or if
the current screen is deleted), automatically give focus to the next
screen created. Saves the need to switch to a usable screen manually.

Don't initialize any virtual screens (except the system console) at
autoconfiguration time; this can be done better from rc.local.
(For INSTALL kernels, WSDISPLAY_DEFAULTSCREENS can be defined which
sets up that many screens with default parameters.)

Don't bother copying output queue into automatic buffer; instead, pass
contiguous chunk of outq to emulation routine. Fix based on patch
from M. Drochner, modified to call output routine once with each
contiguous chunk when the ring buffer wraps around.

Change WSDISPLAYBURST from OBUFSIZ-1 to OBUFSIZ*4, for a ~15x
improvement in the "cat /usr/share/misc/termcap" benchmark on 486/75

ttwrite queues data for us in OBUFSIZ chunks. If we only consume
OBUFSIZ-1 bytes per call to wsdisplaystart(), we're *guaranteed* to
fall behind when given maximum-sized reads (which will defer
processing the leftovers until ttrstrt gets called, after a timeout)

Note that OBUFSIZ*4 may be overkill; OBUFSIZ*3 may be sufficient to
get maximal performance; OBUFSIZ*2 resulted in only about a 2.5x
performance improvement, but it's 2am and I've done enough tweaking
for the night.

improve error handling in 2 cases:
1. If the current screen becomes invalid (ie no focus anymore), always
set the keyboard to translating mode. Otherwise, we could get stuck
because the command keystokes don't come through.
2. Catch errors in attaching to a process (X server) - For this,
implement a callback mechanism similar to the detach case. Add an
argument to report an errno via callback.

remove wsdisplay_has_emulator(), this is not meaningful with virtual screens,
and it did not work with external (serial) consoles what can cause that
a display has no screen and thus no "focus"

-extend the interface for reset-like commands from the keyboard,
add handling of a "close all" command
-split out code common with wsdisplay_delscreen() into
wsdisplay_closescreen()
-add ttymodem call to signal a HUP to the application if the screen
is closed
-add flag to force a "clear screen" on close (XXX should actually be
controlled from userland)
-remove checks for existence of "reset" entries in the emulations - this
is required now

-actually reset the terminal emulator on close()
-add a callback to reset the terminal emulator by keyboard command

change the way fonts are handled: now we have a display-global ioctl
to download fonts and a per-screen call to select a font
allows easier sharing of ressources in the display driver
change the parameters to the load_font driver vector to pass all
available font information

rename "struct wsdisplay_conf" -> "struct wsscreen_internal" to reflect
its real meaning
support allocation and deallocation of virtual screens at runtime
implement a "control device" (minor number = 255) and ioctl()s for screen
allocation/deallocation
minor fixes to consistency checks / error handling

make sure programs cannot write to the screen while it is in graphics mode

Split the wskbd_attach arguments into console and autoconfiguration parts
(again); put some constant data into shared structures. This makes it
more look like the mouse part, and it helps USB integration.
Implement wskbd_enable() to enable/disable the keyboard.

branches: 1.10.2;
Insert lots of ``#if NWSMOUSE > 0'' and ``#if NWSDISPLAY > 0'' to make it
possible to have a wsmouse and/or wskbd without having a wsdisplay.

Switch keyboard to "event mode" on open.
Nuke wskbd_set_translation(). It doesn't fit into the "screen" model.

Move the translation of keypad and function keys into the realm of the
terminal emulation. For this, change the interface to wsdisplay_kbdinput()
to take a "keysym_t" as argument. From there, the code is handed to the
appropriate emulation module via a new entry point: "translate".
Nuke the ioctls dealing with global assignment of character sequences
to keypad/function keys.
The "sun" emulation works much better now!

access the "raw keyboard" flag inside spltty() to make sure there is
no screen switch in between

Add calls to process / VT synchronization functions (as provided in
wsdisplay_compat_usl.c).
add "sleeping ok" argument to wsdisplay_switch()
implement a way to get raw keyboard data (for X)
some cleanup

Simple screen attribute handling for wscons.
The graphics device driver passes a "default attribute" for normal text
output to the wscons framework. If the emulation module needs more
attributes (for different "renditions") it can allocate them via a
callback.
For now, only the "sun" emulation makes use of it.

Fix calculation of console minor device number.
Being here, remove BROKEN_INDIRECT_CONFIG.

Initial import of MI keyboard translation for dev/wscons.

let all 8 bits go through

Initial import of cgd's new wscons code.

wsdisplay(4): Use C99 initializer for usl_syncops.

Avoid sketchy function pointer cast while here.

branches: 1.53.6;
Move proc_lock into the data segment. It was dynamically allocated because
at the time we had mutex_obj_alloc() but not __cacheline_aligned.

- Use kmem_intr_{alloc,free}, because on time out we need to free from an
interrupt context.
- Remove parens from returns and useless casts from void.

branches: 1.51.2; 1.51.4;
use sizeof(*var) instead of sizeof(type)

use kmem, don't check for allocation failure with KM_SLEEP

to garnish, dust with _KERNEL_OPT

branches: 1.48.14;
fix a bunch of _IO() ioctl()s to actually work on LP64/BE
( the data pointer is abused to pass an integer, we get a pointer to it so
we need to deref it to something the same size as a pointer to get the
correct value )
now the PCVT compat code ( which is used for VT switching in and out of X )
works on sparc64

branches: 1.47.8; 1.47.18;
Remove pfind() and pgfind(), fix locking in various broken uses of these.
Rename real routines to proc_find() and pgrp_find(), remove PFIND_* flags
and have consistent behaviour. Provide proc_find_raw() for special cases.
Fix memory leak in sysctl_proc_corename().

COMPAT_LINUX: rework ptrace() locking, minimise differences between
different versions per-arch.

Note: while this change adds some formal cosmetics for COMPAT_DARWIN and
COMPAT_IRIX - locking there is utterly broken (for ages).

Fixes PR/43176.

branches: 1.46.2; 1.46.4;
Don't call usl_sync_check_sig from an interrupt context. Call it only if waitok.
Stack trace: mutex_vector_enter <- usl_sync_check_sig <- usl_detachproc <-
wsdisplay_switch <- wskbd_translate <- wskbd_input <-
pckbd_input <- pckbcintr <- intr_biglock_wrapper <-
Xintr_ioapic_edge1
Reported by Anon Ymous

branches: 1.45.2;
Network protocol interrupts can now block on locks, so merge the globals
proclist_mutex and proclist_lock into a single adaptive mutex (proc_lock).
Implications:

- Inspecting process state requires thread context, so signals can no longer
be sent from a hardware interrupt handler. Signal activity must be
deferred to a soft interrupt or kthread.

- As the proc state locking is simplified, it's now safe to take exit()
and wait() out from under kernel_lock.

- The system spends less time at IPL_SCHED, and there is less lock activity.

branches: 1.44.6; 1.44.8;
restructure things a bit so that proclist_mutex is held between the
p_find() and the psignal() which is its purpose after all

branches: 1.43.2; 1.43.8;
Use aprint API for output and convert to device_t/softc separation.

branches: 1.42.2;
Use callout_setfunc/callout_schedule instead of callout_reset.
Use mstohz for the calculations.

branches: 1.41.6; 1.41.8; 1.41.10;
Merge some of the less invasive changes from the vmlocking branch:

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

branches: 1.40.2; 1.40.4;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.39.2;
Merge newlock2 to head.

Adapt to recent machdep scope changes.

Pointed out by Kouichirou Hiratsuka, thanks!

branches: 1.37.2;
__unused removal on arguments; approved by core.

COMPAT_10 implies COMPAT_11, so there's no need to test both.
Also collapse two tightly-nested #ifs into one.

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

Use kauth(9) instead of securelevel.

okay cube@

branches: 1.33.4; 1.33.6;
- Use the LWP cached credentials where sane.
- Minor cosmetic changes.

integrate kauth.

Don't use intarg uninitialized in wsdisplay_usl_ioctl2. Fixes CID 2507

branches: 1.30.4; 1.30.6; 1.30.8; 1.30.10; 1.30.12;
merge ktrace-lwp.

back out the changes to KD* ioctl()s for now, seem to break more than they
do good.

- remove a leftover debug printf()
- fix the other cases of _IO() type ioctl() abuse ( mainly keyboard related
stuff )

Make the VT_* ioctl()s work in LP64/big endian platforms, like sparc64.
The problem is that these ioctl()s are declared as _IO() and expect to pass an
integer as argument, instead of a pointer. When dereferencing the argument
pointer in the ioctl() handler as an int we get the upper 32bit of the value so
we simply dereference it as long. Other _IO() ioctl()s may need similar fixes.

Tested on sparc64, sparc and macppc.

branches: 1.26.10;
PR kern/28837: fix undef typo.

use bounded string ops

remove license clauses 3 and 4 from my cpoyright notices

branches: 1.23.2;
Back out the lwp/ktrace changes. They contained a lot of colateral damage,
and need to be examined and discussed more.

Pass lwp pointers throughtout the kernel, as required, so that the lwpid can
be inserted into ktrace records. The general change has been to replace
"struct proc *" with "struct lwp *" in various function prototypes, pass
the lwp through and use l_proc to get the process pointer when needed.

Bump the kernel rev up to 1.6V

Merge the nathanw_sa branch.

Silence strict-alias warning.

Remove breaks after returns, unreachable returns and returns after
returns(!).

Convert ioctl code to use EPASSTHROUGH instead of -1 or ENOTTY for
indicating an unhandled "command". ERESTART is -1, which can lead to
confusion. ERESTART has been moved to -3 and EPASSTHROUGH has been
placed at -4. No ioctl code should now return -1 anywhere. The
ioctl() system call is now properly restartable.

add/cleanup RCSIDs

ANSIfy.

branches: 1.15.2; 1.15.4;
KNF - variable declarations.

Introduce WSDISPLAY_NULLSCREEN to allow low level device to detach control
process. Hpcmips frame buffer driver(hpcfb) call wsdisplay_switch with
WSDISPLAY_NULLSCREEN to notify user process of suspend/resume event.

In wsdisplay_usl_ioctl2(), use a u_long for the request to be passed off
to wsdisplay_internal_ioctl() instead of an int.
With this, the pcvt compat ioctl's work on an Alpha and probably other
64 bit archs.

branches: 1.12.4;
New callout mechanism with two major improvements over the old
timeout()/untimeout() API:
- Clients supply callout handle storage, thus eliminating problems of
resource allocation.
- Insertion and removal of callouts is constant time, important as
this facility is used quite a lot in the kernel.

The old timeout()/untimeout() API has been removed from the kernel.

-allow the "show_screen" call to the display driver to complete
asynchronously, in the same style like the process attach/detach functions
-intercept the "cnpollc" call which originally went directly to the
keyboard driver and keep track whether the console is in "polling" state
(DDB!)
-pass a NULL callback to the screen switcher and the process attach/detach
functions if the console is "polling", to tell them that asynchronous
completion is forbidden

branches: 1.10.2;
Split wsdisplay_usl_ioctl() into two functions -- one which is screen-dependent
and one which isn't. The latter is now used for ttyEcfg, enabling the
VT-switching ioctls to work on it. (This allows Linux X servers to work when
/emul/linux/dev/tty0 is linked to /dev/ttyEcfg.)

branches: 1.9.2; 1.9.4;
Don't let bd.which be unitialized... Thanks Dave!

PR/7665: Dave Sainty: X server bell pitch setting doesn't reflect reality
Applied with a little more error checking.

branches: 1.7.2; 1.7.4; 1.7.6;
make the timeout for process/vt synchronization "config" settable and
ddb/gdb patchable

improve error handling in 2 cases:
1. If the current screen becomes invalid (ie no focus anymore), always
set the keyboard to translating mode. Otherwise, we could get stuck
because the command keystokes don't come through.
2. Catch errors in attaching to a process (X server) - For this,
implement a callback mechanism similar to the detach case. Add an
argument to report an errno via callback.

* defopt COMPAT_{09,10,11,12,13} and COMPAT_NOMID (from unsaved buffers).
TODO: revisit interaction between native compat and emul compat usage.

defopt COMPAT_FREEBSD

change initialization order to avoid a race condition if a switch is
performed at the same time

clean up if a controlling process exits while a switch is in progress,
initialize the flag used for keeping track of this

Implement the VT access and keyboard control ioctl()s needed for
compatibility to PCVT, syscons and the Linux console.

allow placement of the glyph cache anywhere in VRAM
mostly for weird HP graphics hardware that has usable off-screen memory to the
right of the visible screen instead of below it

branches: 1.13.2;
remove check for unused bits in attr2idx() - this really wasn't helpful

allow drivers to specify horizontal alignment of glyph cache cells
for things like SX which have alignment restrictions

branches: 1.11.30;
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: 1.10.2; 1.10.4;
use sizeof(*var) instead of sizeof(type)

make glyph caches reconfigurable

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.

to garnish, dust with _KERNEL_OPT

branches: 1.6.6;
try not to crash when calling glyphcache_wipe() on an uninitialized cache
first step to fix PR 48492

branches: 1.5.2;
glyphcache_init(): bail if we don't have enough off-screen memory to be useful

allow caching of glyphs with attributes other than the default
while there, also add support for underlined characters

branches: 1.3.2;
deal with negative cache sizes

no need to convert cell number to coordinates every time we draw a glyph
from cache - just stick the coordinates directly into the map

branches: 1.1.2; 1.1.4;
generic support for caching glyphs in video memory
for speeding up anti-aliased fonts on slow CPUs