Remove KDIR=, since SYS_INCLUDE=symlinks and KDIR are not supported any more.

branches: 1.2.26;
Install kbdreg.h. The Xserver needs it.

Reorganize installation if include files. Nuke some from atari/include as
they make no sense out of the kernel add some from atari/dev (ioctl's).

TAB/spaces/indents cleanup.

follow fcambus steps by fixing typos for "unknown" in few more error messages.

s/transfered/transferred/

branches: 1.63.4;
Merge support for SATA NCQ (Native Command Queueing) from jdolecek-ncq branch

ATA subsystem was changed to support several outstanding commands, and use
NCQ xfers if supported by both the controller and the disk, including NCQ
error recovery. Set NCQ high priority for BPRIO_TIMECRITICAL xfers
if supported. Added FUA support.

Done some work towards MP-safe, all ATA code tsleep()/wakeup() replaced
by condvars, and switched most code from spl* to mutexes (separate
wd(4) and ata channel lock).

Introduced new option WD_CHAOS_MONKEY to facilitate testing of error
handling, fixed several uncovered issues. Also fixed several problems
with kernel dump to wd(4) disk.

Tested with ahcisata(4), mvsata(4), siisata(4), piixide(4) on amd64,
with and without port multiplier, both disk and ATAPI devices; other
drivers and archs mechanically adjusted and compile-tested. NCQ is
supported for ahcisata(4) and siisata(4) for any controller, for
mvsata(4) only Gen IIe ones for now. Also enabled ATAPI support in
mvsata(4).

Thanks to Matt Thomas for initial ATA infrastructure patch, and
Jonathan A.Kollasch for siisata(4) NCQ changes and general testing.

Also fixes PR kern/43169 (wd(4)); and PR kern/11811, PR kern/47041,
PR kern/51979 (kernel dump)

branches: 1.62.8;
src is too big these days to tolerate superfluous apostrophes. It's
"its", people!

Appease gcc48 -Wunused-but-set-variable warning.

branches: 1.60.2;
- remove unused
- use cpu_{g,s}etmodel() (not committed yet)

branches: 1.59.6; 1.59.10;
Change old-style function defintions to C89 prototypes.

Approved by releng.

branches: 1.58.2; 1.58.6;
Retire varargs.h support. Move machine/stdarg.h logic into MI
sys/stdarg.h and expect compiler to provide proper builtins, defaulting
to the GCC interface. lint still has a special fallback.
Reduce abuse of _BSD_VA_LIST_ by defining __va_list by default and
derive va_list as required by standards.

Split device_t/softc. No crash on TT030.

branches: 1.56.2; 1.56.6;
extern inline -> static inline

Misc KNF and cosmetics.

Always schedule callback functions via add_sicallback(),
which calls softint_schedule(9).

We cannot use BASEPRI() to check recursive calls any longer
since softint(9) has been reorganized to use kernel thread.

branches: 1.53.2; 1.53.4;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Merge local <atari/atari/intr.h> into common <machine/intr.h>.

Ansify function definitions w/o arguments. Generated with sed.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.47.2;
Replace multi-statement macro using GCC extension `({})' with
static inline functions to avoid possible side effects or confusion.

Tested on TT030 by David Ross and on Falcon by Tuomo Makinen on port-atari.

branches: 1.46.2;
Cleanup spelling and syntax in a comment

branches: 1.45.2;
Specify volatile to some pointers to access DMA registers.

No need to specify '&' to get array's address.
No binary changes.

Fix pointer arithmetic botch in rev 1.37.

branches: 1.42.40; 1.42.44; 1.42.46; 1.42.48; 1.42.50;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.41.12; 1.41.14;
Fix a bunch of cast lvalues.

branches: 1.40.4; 1.40.6; 1.40.8; 1.40.10; 1.40.12;
__asm__ -> __asm
__const__ -> const
__inline__ -> inline
__volatile__ -> volatile

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

merge ktrace-lwp.

branches: 1.37.2;
Adapt to compiling with -Wshadow and -Wcast-qual, by adding const
qualification in places and renaming a few local variables.
Also adds a couple uses of __UNVOLATILE() to allow passing volatile
variables to functions wanting e.g. caddr_t.

__KERNEL_RCSID()

branches: 1.35.2;
DMA, not dma nor Dma.

Use PAGE_SIZE rather than NBPG.

branches: 1.33.4;
Spell 'occurred' with two 'r's.

branches: 1.32.2; 1.32.4;
defopt atari scsi options.

Nuke a bunch of cf_unit abuses.

branches: 1.30.2;
Change the way that bit's are cleared in the InterruptPending register of
the 68901. Do this now by _assigning_ ~<bit_to_clear> instead of the
previously used 'andb ~<bit_to_clear>'. The latter caused a rwm-cycle that
caused a race condition to happen when an interrupt arrived between the
Read and Modify-Write.
Anyway, this solved my hanging keyboard problem.

branches: 1.29.6;
Disable SCSI interrupts while polling for the transfer to finish. This
fixes spurious panics while booting my Hades. This bug could also occur
on a TT030.

branches: 1.28.4;
Merge scsipi branch in the mainline. This add support for ATAPI devices
(currently only CD-ROM drives on i386). The sys/dev/scsipi system provides 2
busses to which devices can attach (scsibus and atapibus). This needed to
change some include files and structure names in the low level scsi drivers.

Don't enable the dma-interrupt while polling for dma to end. This sometimes
caused a system panic on the Hades while probing the SCSI devices.

branches: 1.26.2; 1.26.4;
Use m68k cacheops

Make Hades DMA transfers work in polling mode with all interrupts
disabled. This makes dumps to scsi work.

Remove a hardcoded interrupt handler. Now uses intr_establish() for the
SCSI-drq handler on the Hades.

Use the mask register to block the interrups, not the enable register. This
enables the interrupt to be shared (with IDE for example ;-) ).

Fix a rather nasty thinko that made the keyboard useless on the Falcon.
Thanks to Markus Kilbinger for finding it.

Make splbio really splbio. All holes now seem to be plugged. At least
on the Hades & TT030.

backout previous kprintf changes

printf -> kprintf, sprintf -> ksprintf

Add support for Hades platform.

Revert to polled-DMA on Falcon.

During a rather lengthy debug session with Markus Kilbinger (Thanks again!),
we managed to kill the following SCSI-driver bugs for the Falcon:
- The Byte_Count_zero bit of the falcon DMA controller does not always
tell the truth! This caused the SCSI-driver to choke on devices that
disconnected in the middle of a DMA-transfer (mostly removables).
- Printing debug info about the 5380 on the Falcon is *only* permitted
when DMA is not active.
- Some functions forgot to remove possibly pending sofware interrupts
- Some debug options didn't allow debugging a single target while they
could easily be made to do so. This is fixed.

Pull down from release. Sorry, should have been the other way around....

branches: 1.14.4;
Prototype + -Wall fixes.

- Fixup the ipending function for the Falcon. This seemed to be the source
of a lot of instability problems on the Falcon. I also enabled DMA on
interrupt basis for the Falcon.
- Try to handle targets that request too much data more sanely.
- Some fixups for strict prototypes + -Wall

Thanks to Markus Kilbinger for providing the debugging support.

Yet more prototyping and -Wall fixes.

Prototyping.

New device attachment scheme:

- split softc size and match/attach out from cfdriver into
a new struct cfattach.

- new "attach" directive for files.*. May specify the name of
the cfattach structure, so that devices may be easily attached
to parents with different autoconfiguration semantics.

Just prototyping.

A bit of reshuffling. Also some stricter prototyping.

Oops, should have been updated in conjunction with ncr5380.c .

Fix #includes.

The changes frpm Paul Goyette with respect to the driver trace require
DBG_PID to be defined to an interger value....make it so.

Get in sync with the mac68k version to incorporate Allen Briggs' changes
- emit MSG_ABORT/MSG_REJECT at the right times:
- make MIN_PHYS machine dependent.

- Handle devices that do not support identify messages (Allen Briggs)
- Add pdma_ready() hook (Allen Briggs)
- Increase timeout to 1000 (Matthias Pfaller)

- Add 'restbyte' handling for TT.
- Don't _ever_ do DMA for less than 512 bytes on the Falcon
- Fix bug in autosense-handling. Now asks for the correct number of bytes.
Now it won't read ghost bytes on the tape anymore.
- Add missing braces as suggested by Matthias Pfaller
- Make it possible to debug requests on a specified number of targets
- Add debug option to show only transaction with error code != 0

New version of NCR5380-SCSI driver. This version is far more generic than
the previous version. The machine dependent part contains the configuration
for both the TT and FALCON. The configuration files have been changed to
select SCSI-support for either TT, Falcon or both.
The configuration file for the Atari-Falcon will only get _really_ usefull
when the Falcon video part is finished.

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."

Remove unnecessary NULL-initialization of TODR handle fields.

Initialize the TODR cookie. Not used now, but will be in the future.

Use aprint_*(9) for device attach messages.

Make local devsw functions static.

No visible regression on TT030.

hardclock_ticks -> getticks()

For struct timecounter, use C99 initializers.
Compile tested. No functional changes intended.

Misc cosmetics and cleanup.

branches: 1.59.28;
Add d_discard to all struct cdevsw instances I could find.

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

branches: 1.58.2;
kill sprintf

- remove unused
- use cpu_{g,s}etmodel() (not committed yet)

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.

branches: 1.55.2;
use sizeof(buffer) - 1, not sizeof(buffer - 1)

branches: 1.54.2;
Fix buffer overrun (off by one). PR port-atari/42717

branches: 1.53.2;
Split device_t/softc. No crash on TT030.

branches: 1.52.2;
Remove clause 3 (UCB advertising clause) from the University of Utah
copyright. Confirmed by Mike Hibler, mike at cs.utah.edu - thanks!
Also, merge UCB and Utah copyright texts back into one, as they
originally were.

Extra verification by snj@.

branches: 1.51.2; 1.51.4;
Include "ioconf.h" to declare struct cfdriver foo_cd.

branches: 1.50.2; 1.50.4;
Use proper pointer types and names rather than casts for readability.

KNF, ANSIfy, etc.

Explicitly initialize the MFP Timer-B for delay(9) from atari_hwinit()
rather than using if(!atari_realconfig) in clockmatch().
(I doubt the latter one has actually been called..)

Add a workaround for annoying
"WARNING: negative runtime; monotonic clock has gone backwards"
message. Partially taken from hp300.

Ansify function definitions w/o arguments. Generated with sed.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.42.4;
Move tc_init(&clk_timecounter); to stop the "timecounter:" message
appearing in the middle of the clock0 attach message. Tested by
Tuomo Makinen

branches: 1.41.2; 1.41.4; 1.41.6;
Use device_private() and device_lookup_privat() to get softc.

branches: 1.40.6; 1.40.8; 1.40.10; 1.40.12; 1.40.14;
Convert Atari to generic TODR and timecounter.

Simplify delay based on x86 version.

branches: 1.38.20; 1.38.26; 1.38.32;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.37.24; 1.37.26; 1.37.34;
Remove leading __ from __(const|inline|signed|volatile) -- it is obsolete.

merge ktrace-lwp.

branches: 1.35.2;
Adapt to compiling with -Wshadow and -Wcast-qual, by adding const
qualification in places and renaming a few local variables.
Also adds a couple uses of __UNVOLATILE() to allow passing volatile
variables to functions wanting e.g. caddr_t.

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

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

__KERNEL_RCSID()

branches: 1.32.2;
Merge the nathanw_sa branch.

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

Use CFATTACH_DECL().

Declare all cfattach structures const.

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.

branches: 1.27.2; 1.27.6; 1.27.14;
Various typos in comments (neccessary, sceme, choise, ...).

branches: 1.26.2;
Symmetric has one s and two m's.

Add sanity checking on register B of the mc146818.

branches: 1.24.2;
Fix y2001 problem in the initial read of the RTC.

branches: 1.23.2;
Mark 'n' explicitely as output of the 'mulul' instruction.

Fix some bugs in clkread(). These bugs caused strange time leaps.

branches: 1.21.2;
Fix y2k buglet in /dev/rtc reported by Julian Coleman. Note that the input
and output format of /dev/rtc is now changed.

branches: 1.20.2;
Change the way that bit's are cleared in the InterruptPending register of
the 68901. Do this now by _assigning_ ~<bit_to_clear> instead of the
previously used 'andb ~<bit_to_clear>'. The latter caused a rwm-cycle that
caused a race condition to happen when an interrupt arrived between the
Read and Modify-Write.
Anyway, this solved my hanging keyboard problem.

branches: 1.19.8;
Update for changes to config.

Use generic date conversion module.

branches: 1.17.6;
Get rid of BROKEN_INDIRECT_CONFIG

Tweak the clockframe structure a bit so we are able to save a few cycles
at interrupt time. (From Gordon Ross).

Move up the initialisations for the delay routine in the initialisation
process. It's now also available to the different device probes.

Implement an rtc-device. Credits for the idea goes to Wolfgang Solfrank.
Because there is no way yet to tell init(1) to execute commands before
single user-mode is entered, the inittodr() function is still partly
functional. It retrieves the value of the RTC and sets the time as if
the RTC was running at UTC. The resettodr() function is a No-Op. The
RTC can only be changed by a write to /dev/rtc.
The TIMEZONE and DST options are no longer needed and removed from the
config files.

backout previous kprintf changes

printf -> kprintf, sprintf -> ksprintf

Corrections for removed <sys/cpu.h> and some more prototypes.

New device attachment scheme:

- split softc size and match/attach out from cfdriver into
a new struct cfattach.

- new "attach" directive for files.*. May specify the name of
the cfattach structure, so that devices may be easily attached
to parents with different autoconfiguration semantics.

First bunch of changes due to stricter prototype checking.

Some extra range-tests on the RTC values.

Add an nvram-driver. This driver makes it possible to read/write the
configuration info stored in the nvram on the mc146818.

Remove forgotten #endif

Add kernel option for separate {stat,prof}clock

PROF -> GPROF

Use <dev/ic/mc146818.h>. Change systemclock settings. Usable values for
HZ are now 48/64/96. This reduces the interrupt overhead, because we
don't need the extra division by 4 in the interrupt handler.

Shift system clocks from TT-mfp to ST-mfp and remove some dead-code.

branches: 1.1.1;
Initial revision

Final bit of PR 41200: headers that declare ioctls should include sys/ioccom.h.
This covers most if not all of the MD headers.

XXX: a lot of the ioctl definitions in some of these files are cutpasted.

branches: 1.4.14; 1.4.32;
Remove clause 3 (UCB advertising clause) from the University of Utah
copyright. Confirmed by Mike Hibler, mike at cs.utah.edu - thanks!
Also, merge UCB and Utah copyright texts back into one, as they
originally were.

Extra verification by snj@.

branches: 1.3.100; 1.3.106; 1.3.108;
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: 1.1.1;
Initial revision

PR/49207: Kamil Rytarowski: Add sys/clock.h with generic time macros
(derived from clock_subr.h). Keep clock_subr.h with the kernel structures
and functions to reduce diffs, and have clock.h only include standalone
constants and macros.

branches: 1.6.22;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

branches: 1.5.8; 1.5.130;
The register spacing for the Milan is different.

branches: 1.4.42;
Add an nvram-driver. This driver makes it possible to read/write the
configuration info stored in the nvram on the mc146818.

note that most of dev/ic's contents have changed names

Use <dev/ic/mc146818.h>. Change systemclock settings. Usable values for
HZ are now 48/64/96. This reduces the interrupt overhead, because we
don't need the extra division by 4 in the interrupt handler.

branches: 1.1.1;
Initial revision

branches: 1.1.2;
Explicitly initialize the MFP Timer-B for delay(9) from atari_hwinit()
rather than using if(!atari_realconfig) in clockmatch().
(I doubt the latter one has actually been called..)

TAB/spaces/indents cleanup.

Merge support for SATA NCQ (Native Command Queueing) from jdolecek-ncq branch

ATA subsystem was changed to support several outstanding commands, and use
NCQ xfers if supported by both the controller and the disk, including NCQ
error recovery. Set NCQ high priority for BPRIO_TIMECRITICAL xfers
if supported. Added FUA support.

Done some work towards MP-safe, all ATA code tsleep()/wakeup() replaced
by condvars, and switched most code from spl* to mutexes (separate
wd(4) and ata channel lock).

Introduced new option WD_CHAOS_MONKEY to facilitate testing of error
handling, fixed several uncovered issues. Also fixed several problems
with kernel dump to wd(4) disk.

Tested with ahcisata(4), mvsata(4), siisata(4), piixide(4) on amd64,
with and without port multiplier, both disk and ATAPI devices; other
drivers and archs mechanically adjusted and compile-tested. NCQ is
supported for ahcisata(4) and siisata(4) for any controller, for
mvsata(4) only Gen IIe ones for now. Also enabled ATAPI support in
mvsata(4).

Thanks to Matt Thomas for initial ATA infrastructure patch, and
Jonathan A.Kollasch for siisata(4) NCQ changes and general testing.

Also fixes PR kern/43169 (wd(4)); and PR kern/11811, PR kern/47041,
PR kern/51979 (kernel dump)

branches: 1.27.18; 1.27.48;
- more cosmetics
- use queue(3) macro properly

KNF and cosmetics.

Always schedule callback functions via add_sicallback(),
which calls softint_schedule(9).

We cannot use BASEPRI() to check recursive calls any longer
since softint(9) has been reorganized to use kernel thread.

branches: 1.24.2; 1.24.4;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Merge local <atari/atari/intr.h> into common <machine/intr.h>.

Ansify function definitions w/o arguments. Generated with sed.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.18.44; 1.18.52; 1.18.54; 1.18.58;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.17.26;
merge ktrace-lwp.

branches: 1.16.16;
__KERNEL_RCSID()

branches: 1.15.6;
remove trailing \n in panic(). approved perry.

branches: 1.14.40; 1.14.44;
Add option to sleep until ST_DMA becomes available.

branches: 1.13.2;
Fix return value of the interrupt function.

The ST-DMA driver now uses intr_establish.

backout previous kprintf changes

printf -> kprintf, sprintf -> ksprintf

Add missing spl0() call.

Pull down from release. Sorry, should have been the other way around....

branches: 1.7.4;
Prototypes.

First bunch of changes due to stricter prototype checking.

The Falcon needs a delay after writing the dma_data register (Thomas Gerner)

branches: 1.4.2;
- Remove race condition from DMA interrupt handler
- Add function to get current DMA address for Falcon SCSI.

Complete rewrite, made necessary because Falcon SCSI needs to share
NCR5380 and DMA access with floppy.

- Remove register access functions from dma.c -> fd.c and inline them
- Add write protect checking (mistereously disapeared during port)
- Make fdclose() return a sensible value
- Prefix 'dma-defines' with DMA_

branches: 1.1.1;
Initial revision

Driver "kqfilter" entry points return an error code, so if an invalid
filter is requested, return EINVAL rather than 1.

Change the kqueue filterops::f_isfd field to filterops::f_flags, and
define a flag FILTEROP_ISFD that has the meaning of the prior f_isfd.
Field and flag name aligned with OpenBSD.

This does not constitute a functional or ABI change, as the field location
and size, and the value placed in that field, are the same as the previous
code, but we're bumping __NetBSD_Version__ so 3rd-party module source code
can adapt, as needed.

NetBSD 9.99.89

- malloc(9) -> kmem(9)
- Use sel{record,remove}_knote().

branches: 1.14.16;
Use C99 initializer for filterops

Mostly done with spatch with touchups for indentation

@@
expression a;
identifier b,c,d;
identifier p;
@@
const struct filterops p =
- { a, b, c, d
+ {
+ .f_isfd = a,
+ .f_attach = b,
+ .f_detach = c,
+ .f_event = d,
};

branches: 1.13.22;
Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

branches: 1.12.4; 1.12.12; 1.12.18;
Welcome to 4.99.55:

- Add a lot of missing selinit() and seldestroy() calls.

- Merge selwakeup() and selnotify() calls into a single selnotify().

- Add an additional 'events' argument to selnotify() call. It will
indicate which event (POLL_IN, POLL_OUT, etc) happen. If unknown,
zero may be used.

Note: please pass appropriate value of 'events' where possible.
Proposed on: <tech-kern>

branches: 1.11.20; 1.11.36; 1.11.40;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.10.26;
merge ktrace-lwp.

branches: 1.9.16;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

__KERNEL_RCSID()

branches: 1.7.6;
si_ -> sel_

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

branches: 1.5.42; 1.5.46;
Merge the netbsd-1-2 branch into the mainline.

Catchup with -current:
- Extra argument to boot()
- select -> poll

Add missing spl(x). (Jason Thorpe)

branches: 1.2.6;
Mouse driver added.

branches: 1.1.1;
Initial revision

Change proc_lock -> &proc_lock.

Make sure to acquire proc_lock mutex on calling psignal(9) in EV_WAKEUP().

Without this, Xserver could cause panic on the DIAGNOSTIC kernel as x68k:
http://mail-index.netbsd.org/source-changes/2012/08/15/msg036632.html

branches: 1.8.12; 1.8.22; 1.8.26;
Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.7.4; 1.7.12; 1.7.18;
Welcome to 4.99.55:

- Add a lot of missing selinit() and seldestroy() calls.

- Merge selwakeup() and selnotify() calls into a single selnotify().

- Add an additional 'events' argument to selnotify() call. It will
indicate which event (POLL_IN, POLL_OUT, etc) happen. If unknown,
zero may be used.

Note: please pass appropriate value of 'events' where possible.
Proposed on: <tech-kern>

branches: 1.6.20; 1.6.36; 1.6.40;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.5.26;
merge ktrace-lwp.

branches: 1.4.16;
Move UCB-licensed code from 4-clause to 3-clause licence.

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

branches: 1.3.6;
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

branches: 1.2.42; 1.2.46;
Catchup with -current:
- Extra argument to boot()
- select -> poll

branches: 1.1.1;
Initial revision

fix typos in comments, mainly s/harware/hardware/.

branches: 1.98.2;
Remove unnecessary <sys/malloc.h> include.

atari/fdc(4): fix printf format specifiers for FLP_DEBUG build.

Fix typo in comment.

TAB/spaces/indents cleanup.

Use aprint_*(9) for device attach messages.

Make local devsw functions static.

No visible regression on TT030.

s/wich/which in comments.

fix various typos, mainly s/trasfering/transferring/ and s/theese/these/.

s/simular/similar/

Merge thorpej-cfargs2.

branches: 1.88.8;
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: 1.87.14;
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: 1.86.2; 1.86.4;
Typos.

Merge support for SATA NCQ (Native Command Queueing) from jdolecek-ncq branch

ATA subsystem was changed to support several outstanding commands, and use
NCQ xfers if supported by both the controller and the disk, including NCQ
error recovery. Set NCQ high priority for BPRIO_TIMECRITICAL xfers
if supported. Added FUA support.

Done some work towards MP-safe, all ATA code tsleep()/wakeup() replaced
by condvars, and switched most code from spl* to mutexes (separate
wd(4) and ata channel lock).

Introduced new option WD_CHAOS_MONKEY to facilitate testing of error
handling, fixed several uncovered issues. Also fixed several problems
with kernel dump to wd(4) disk.

Tested with ahcisata(4), mvsata(4), siisata(4), piixide(4) on amd64,
with and without port multiplier, both disk and ATAPI devices; other
drivers and archs mechanically adjusted and compile-tested. NCQ is
supported for ahcisata(4) and siisata(4) for any controller, for
mvsata(4) only Gen IIe ones for now. Also enabled ATAPI support in
mvsata(4).

Thanks to Matt Thomas for initial ATA infrastructure patch, and
Jonathan A.Kollasch for siisata(4) NCQ changes and general testing.

Also fixes PR kern/43169 (wd(4)); and PR kern/11811, PR kern/47041,
PR kern/51979 (kernel dump)

branches: 1.84.6;
Use C99-style initializers for struct dkdriver.

We have three sets of DTYPE_ constants in the kernel:
altq Drop Type
disklabel Disk Type
file Descriptor Type
(not to mention constants that contain the string DTYPE).
Let's make them two, by changing the disklabel one to be DisK TYPE since the
other disklabel constants seem to do that. Not many userland programs use
these constants (and the ones that they do are mostly in ifdefs). They will
be fixed shortly.

declare error

make more drivers use disk_ioctl, and add a dev parameter to it so that
we can merge the "easy" disklabel ioctls to it. Ultimately all this will
go do dk_ioctl once all the drivers have been converted.

branches: 1.80.2;
src is too big these days to tolerate superfluous apostrophes. It's
"its", people!

Add d_discard to all struct cdevsw instances I could find.

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

Add d_discard to all struct bdevsw instances I could find.

I've set them all to nodiscard. Some of them (wd, dk, vnd, ld,
raidframe, maybe cgd) should be implemented for real.

branches: 1.77.2;
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.

branches: 1.76.8; 1.76.18; 1.76.22;
Misc KNF.

Misc KNF.

- split device_t/softc
- use device_lookup_private() directly rather than via local macro

extern inline -> static inline

branches: 1.72.2; 1.72.4;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Read AD_CFG_SWITCH via volatile pointer so that
the default density is detected correctly.

Remove extra whitespace added by a dumb tool which replaced bcopy with memcpy.

bcopy -> memcpy

bzero -> memset

Ansify function definitions w/o arguments. Generated with sed.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.63.2;
g/c BUFQ_FOO() macros and use bufq_foo() directly.

branches: 1.62.4; 1.62.6; 1.62.12;
Use device_private() and device_lookup_privat() to get softc.

branches: 1.61.16; 1.61.18; 1.61.20; 1.61.22; 1.61.24;
Merge the ppcoea-renovation branch to HEAD.

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

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

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

Fix typo: in disk_init() call: fd -> sc.

branches: 1.59.2;
Merge disk init changes from the vmlocking branch. These seperate init /
destroy of 'struct disk' from attach / detach.

branches: 1.58.4; 1.58.6; 1.58.8;
It's not a good idea for device drivers to modify b_flags, as they don't
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.

branches: 1.57.2; 1.57.4;
Merge some of the less invasive changes from the vmlocking branch:

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

branches: 1.56.2; 1.56.4; 1.56.10;
Make io_data (char *) rather than adding an extra cast
since it's used only in this file to calculate buffer address.

More fixes after the caddr_t removal.
Mostly cast to char* for pointer arithmetic,
but also one missing indirection, and one "void *v, x;" fix.

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

branches: 1.53.26;
__asm__ -> __asm
__const__ -> const
__inline__ -> inline
__volatile__ -> volatile

merge ktrace-lwp.

- change the way to specify a bufq strategy. (by string rather than by number)
- rather than embedding bufq_state in driver softc,
have a pointer to the former.
- move bufq related functions from kern/subr_disk.c to kern/subr_bufq.c.
- rename method to strategy for consistency.
- move some definitions which don't need to be exposed to the rest of kernel
from sys/bufq.h to sys/bufq_impl.h.
(is it better to move it to kern/ or somewhere?)
- fix some obvious breakage in dev/qbus/ts.c. (not tested)

branches: 1.50.2;
Adapt to compiling with -Wshadow and -Wcast-qual, by adding const
qualification in places and renaming a few local variables.
Also adds a couple uses of __UNVOLATILE() to allow passing volatile
variables to functions wanting e.g. caddr_t.

move buffer queue related stuffs from buf.h to their own header, bufq.h.

Spell controller with two ls. Inspired by miod@openbsd.

__KERNEL_RCSID()

branches: 1.46.2;
DMA, not dma nor Dma.

Use aprint_normal() for cfprint routines.

implement separate read/write disk statistics:
- disk_unbusy() gets a new parameter to tell the IO direction.
- struct disk_sysctl gets 4 new members for read/write bytes/transfers.
when processing hw.diskstats, add the read&write bytes/transfers for
the old combined stats to attempt to keep backwards compatibility.

unfortunately, due to multiple bugs, this will cause new kernels and old
vmstat/iostat/systat programs to fail. however, the next time this is
change it will not fail again.

this is just the kernel portion.

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

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.

Convert to new device buffer queue interface.

Approved by: Leo Weppelman <leo@netbsd.org>

branches: 1.37.2; 1.37.6; 1.37.14;
issueing -> issuing

branches: 1.36.8;
Nuke a bunch of cf_unit abuses.

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.

Fix doubled 'the's in comments.

Grabage collect a 'struct buf *' left over from recent disksort changes.

Fix a bug in disksort_*() which caused non-optimal ordering when multiple
active partitions were on a single spindle. Add a b_rawblkno member to
struct buf which contains the non-partition-relative block number to sort
by.

Update for sys/buf.h/disksort_*() changes.

Determine the default floppy type on the TT/Falcon by reading the config
switches. (Thanks to Michael Schwingen for the missing bits of information!).

branches: 1.29.2;
Change the way that bit's are cleared in the InterruptPending register of
the 68901. Do this now by _assigning_ ~<bit_to_clear> instead of the
previously used 'andb ~<bit_to_clear>'. The latter caused a rwm-cycle that
caused a race condition to happen when an interrupt arrived between the
Read and Modify-Write.
Anyway, this solved my hanging keyboard problem.

branches: 1.28.8;
Declare 'fd_cd'.

Update for changes to config.

Implement DIOCGDEFLABEL.

branches: 1.25.10;
Oops, this one was missed in the BROKEN_INDIRECT_CONFIG cruisade.

- Make it's behaviour on minor numbers identical to that of the Hades
floppy driver.
- Fixup format types in the debug-printf's
- Don't print empty lines on unconfigured drives.
- When a drive is detected, show it's default format.

---- Warning: minor numbers are shifted! ---

backout previous kprintf changes

printf -> kprintf, sprintf -> ksprintf

change cfprint_t type definition to take a const char *, rather than
a char *, because that's what was really intended, and because
if the print function modifies the string, various things could become
unhappy (so the string should _not_ be modified).

Yet more prototyping and -Wall fixes.

Prototyping.

Take out direct access to the YM2149. The chip definitions are moved
from video.h to ym2149reg.h.

Another bit of prototyping...

New device attachment scheme:

- split softc size and match/attach out from cfdriver into
a new struct cfattach.

- new "attach" directive for files.*. May specify the name of
the cfattach structure, so that devices may be easily attached
to parents with different autoconfiguration semantics.

First bunch of changes due to stricter prototype checking.

Fix #includes.

New generic disk framework. Highlights:

- New metrics handling. Metrics are now kept in the new
`struct disk'. Busy time is now stored as a timeval, and
transfer count in bytes.

- Storage for disklabels is now dynamically allocated, so that
the size of the disk structure is not machine-dependent.

- Several new functions for attaching and detaching disks, and
handling metrics calculation.

Old-style instrumentation is still supported in drivers that did it before.
However, old-style instrumentation is being deprecated, and will go away
once the userland utilities are updated for the new framework.

For usage and architectural details, see the forthcoming disk(9) manual
page.

Make sure that the motor will be turned off when a floppy is inserted.

Use general bound-check function.

branches: 1.10.2;
minphys() functions really should return void.

fix minphys so that it actually works properly

- Change character interface as suggested by Charles
- HD floppies really have 2 sides ;-)
- Cleanup drive/density selection

Change of function params for Falcon SCSI.

Fix bug on handling HD-floppy's.

Small changes in dma locking code.

- Remove register access functions from dma.c -> fd.c and inline them
- Add write protect checking (mistereously disapeared during port)
- Make fdclose() return a sensible value
- Prefix 'dma-defines' with DMA_

Make it consistent with conf.c .

Fdclose --> fdclose

branches: 1.1.1;
Initial revision

Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

branches: 1.2.172;
- Remove register access functions from dma.c -> fd.c and inline them
- Add write protect checking (mistereously disapeared during port)
- Make fdclose() return a sensible value
- Prefix 'dma-defines' with DMA_

branches: 1.1.1;
Initial revision

Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

branches: 1.1.1;
Initial revision

s/ for for / for / in comment.

TAB/spaces/indents cleanup.

Correct flipped glyph of 'N' in the "New Line" in the DEC graphics chars.

merge ktrace-lwp.

__KERNEL_RCSID()

branches: 1.2.66;
Make them compliant to iso-8859-1 (Thomas Gerner).

branches: 1.1.1;
Initial revision

s/ for for / for / in comment.

TAB/spaces/indents cleanup.

merge ktrace-lwp.

__KERNEL_RCSID()

branches: 1.2.66;
Make them compliant to iso-8859-1 (Thomas Gerner).

branches: 1.1.1;
Initial revision

Remove unnecessary <sys/malloc.h> include.

TAB/spaces/indents cleanup.

Use aprint_*(9) for device attach messages.

Make local devsw functions static.

No visible regression on TT030.

Merge thorpej-cfargs2.

branches: 1.52.6;
The Amiga and Atari ports abuse some autoconfiguration internals as part
of their early console bring-up, so we need to expose some of the new
internals to them and adapt the call sites.

branches: 1.51.2;
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: 1.50.12;
Make local functions static.

branches: 1.49.28;
Add d_discard to all struct cdevsw instances I could find.

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

branches: 1.48.2;
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.

branches: 1.47.2; 1.47.12; 1.47.16;
dependant -> dependent

A few missed device_t, cfdata_t cleanups

- use CFATTACH_DECL_NEW() (no softc)
- use device_t, cfdata_t etc.

branches: 1.44.2;
Remove clause 3 (UCB advertising clause) from the University of Utah
copyright. Confirmed by Mike Hibler, mike at cs.utah.edu - thanks!
Also, merge UCB and Utah copyright texts back into one, as they
originally were.

Extra verification by snj@.

branches: 1.43.2; 1.43.4;
Misc KNF.

Include "ioconf.h" to declare struct cfdriver foo_cd.

branches: 1.41.2; 1.41.4;
Remove extra whitespace added by a dumb tool which replaced bcopy with memcpy.

bcopy -> memcpy

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.36.44; 1.36.52; 1.36.58;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.35.26;
merge ktrace-lwp.

branches: 1.34.16;
Licence cleanup (suggested by wiz).

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

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

__KERNEL_RCSID()

branches: 1.31.6;
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

replace the somewhat strange poll routine with standard nopoll() (a.k.a
seltrue())

Use CFATTACH_DECL().

Declare all cfattach structures const.

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.

branches: 1.26.4;
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.

branches: 1.25.2; 1.25.4; 1.25.8;
remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

remove/move more mach vm header files:

<vm/pglist.h> -> <uvm/uvm_pglist.h>
<vm/vm_inherit.h> -> <uvm/uvm_inherit.h>
<vm/vm_kern.h> -> into <uvm/uvm_extern.h>
<vm/vm_object.h> -> nothing
<vm/vm_pager.h> -> into <uvm/uvm_pager.h>

also includes a bunch of <vm/vm_page.h> include removals (due to redudancy
with <vm/vm.h>), and a scattering of other similar headers.

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: 1.22.8; 1.22.18;
Changed grf mmap that it is closer to XFree. VGA memory on et4000/et6000
is now at offset 0xa0000.

Update for changes to config.

- Provide the reasonable size that can/may be mmap-ed. This might differ
from the current frame buffer size.
- The et-driver now allows for a 16Kb io-region (was 8Kb) that can be
mapped. The XFree server uses this.

Fix a kernel-dump that occurred when accessing a grf-device that was configured
in the config file but not found by the autoconfig code.

branches: 1.18.6;
Allow the control registers to be mapped too.

Get rid of BROKEN_INDIRECT_CONFIG

backout previous kprintf changes

printf -> kprintf, sprintf -> ksprintf

Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Catchup with -current:
- Extra argument to boot()
- select -> poll

Prepare for (soon to come) et4000 PCI console.

change cfprint_t type definition to take a const char *, rather than
a char *, because that's what was really intended, and because
if the print function modifies the string, various things could become
unhappy (so the string should _not_ be modified).

Yet more prototyping and -Wall fixes.

Fix typo in neew config changes.

New device attachment scheme:

- split softc size and match/attach out from cfdriver into
a new struct cfattach.

- new "attach" directive for files.*. May specify the name of
the cfattach structure, so that devices may be easily attached
to parents with different autoconfiguration semantics.

First bunch of changes due to stricter prototype checking.

Sync grf with underlying view on open.

Fix grfmmap()

Remove the pointer to view_t from the grf-structure, use viewview(dev)
to obtain the pointer instead. This removes a kernel-crash that happened
when something was written to the console while the console was being
resized.

Minor fixes.

map --> mmap

branches: 1.1.1;
Initial revision

TAB/spaces/indents cleanup.

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

branches: 1.1.78; 1.1.116;
Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Remove unnecessary <sys/malloc.h> include.

TAB/spaces/indents cleanup.

Change old-style function defintions to C89 prototypes.

Approved by releng.

branches: 1.17.12; 1.17.16;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

bzero -> memset

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.13.78; 1.13.86; 1.13.92;
merge ktrace-lwp.

__KERNEL_RCSID()

branches: 1.11.14;
possible has two s.

branches: 1.10.42; 1.10.46;
Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Prepare for (soon to come) et4000 PCI console

When having a perfect match on the resolution, try to match the screen
depth as best we can.

Make sure the prototypes and inline definitions are included.

Right align color values in the bytes.

Add Falcon video support. Thanks to Thomas Gerner.

- Add grf_get_best_mode()
- Only update 'real' colormap when actually displaying
- Fix TT-HIGH video mode

Better way to select a display mode?

Make ite ioctl work and fixup the colormap stuff.

branches: 1.1.1;
Initial revision

TAB/spaces/indents cleanup.

fix various typos in comments.

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: 1.34.58; 1.34.60;
Misc KNF.

branches: 1.33.2; 1.33.4;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Remove extra whitespace added by a dumb tool which replaced bcopy with memcpy.

bcopy -> memcpy

Ansify function definitions w/o arguments. Generated with sed.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.27.44; 1.27.52; 1.27.58;
- remove volatile from *regkva member in struct grfabs_et_priv
and use temporary variables on register accesses
- also remove volatile from *memkva member in struct grfabs_et_priv
because it isn't used to refer memory
XXX: I'm not sure if bm->plane should be volatile or not
but I don't think previous code treats it as volatile anyway.
(though I'm not sure how compiler handled "volatile caddr_t")

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

branches: 1.25.14;
Do not disable access to tseng private registers

branches: 1.24.4; 1.24.6; 1.24.8; 1.24.10; 1.24.12;
merge ktrace-lwp.

branches: 1.23.16;
Uppercase CPU, plural is CPUs.

__KERNEL_RCSID()

branches: 1.21.2;
Config option for et4000 boards with 2MB memory.

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

possible has two s.

branches: 1.18.4; 1.18.8;
remove include of <vm/vm.h>. <vm/vm.h> -> <uvm/uvm_extern.h>

remove/move more mach vm header files:

<vm/pglist.h> -> <uvm/uvm_pglist.h>
<vm/vm_inherit.h> -> <uvm/uvm_inherit.h>
<vm/vm_kern.h> -> into <uvm/uvm_extern.h>
<vm/vm_object.h> -> nothing
<vm/vm_pager.h> -> into <uvm/uvm_pager.h>

also includes a bunch of <vm/vm_page.h> include removals (due to redudancy
with <vm/vm.h>), and a scattering of other similar headers.

Now what are we doing with the return value of splx()? (probably a pasto)

Remove redundant test.

Init the preset_row_scan register. This fixes a sometimes partly scrolled-up
console.

branches: 1.13.8;
Nuke the pieces of code obsoleted by .../pci/pci_tseng.c

More work on Tseng support. Now both my et6000 _and_ the et6100 work. I just
whish that programming MDRAM latency values came out of the BlackMagic(tm)
domain :-(

Changed grf mmap that it is closer to XFree. VGA memory on et4000/et6000
is now at offset 0xa0000.

Checkpoint my work on et6000 support.

Let et4000 work with other RAMDAC than STG1703

Use NOWAIT when malloc()-ing a buffer for switching screens. It might be
running from an interrupt handler.

branches: 1.7.8;
- Provide the reasonable size that can/may be mmap-ed. This might differ
from the current frame buffer size.
- The et-driver now allows for a 16Kb io-region (was 8Kb) that can be
mapped. The XFree server uses this.

Record physical addresses in the bitmap structure instead of VA's. The
latter breaks the mmap-functions.

Shift screen a bit to the left.

backout previous kprintf changes

*** empty log message ***

printf -> kprintf, sprintf -> ksprintf

Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

expand the attr array to hold all the bytes we store in it.
not that the code actually does anything with the extra bytes,
but at least it builds with gcc 4.5.

Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.1.168; 1.1.176; 1.1.182;
- add a iteet_ioctl function so iteconfig(1) can be used to select the
screen geometry.
- Re-arrange et_loadmode() and et_inittextmode() to write into shadow
register set instead of writing to the card-registers directly.
- provide et_hwsave()/et_hwrest() functions to transfer the card-registers
to and from the shadow set.

fix a/an grammar in obvious cases

fix various typos in comments and log messages, mainly s/hight/high/
s/hight/height/ and removing double m in time(ing).

branches: 1.29.6;
TAB/spaces/indents cleanup.

fix multiplei repetitive typos in comments, messages and documentation. mainly because copy paste code big amount of files are affected.

Misc KNF.

branches: 1.26.2; 1.26.4;
Drop 3rd and 4th clause. Approved by leo@ and thomas@ (copyright holders).

Remove extra whitespace added by a dumb tool which replaced bcopy with memcpy.

bcopy -> memcpy

bzero -> memset

Ansify function definitions w/o arguments. Generated with sed.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.18.78; 1.18.86; 1.18.92;
merge ktrace-lwp.

__KERNEL_RCSID()

branches: 1.16.2;
Use PAGE_SIZE rather than NBPG.

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

possible has two s.

branches: 1.13.8; 1.13.12;
Fix doubled 'the's in comments.

branches: 1.12.8;
Changed grf mmap that it is closer to XFree. VGA memory on et4000/et6000
is now at offset 0xa0000.

Fix a bug in the colormap handling. (Thomas Gerner)

branches: 1.10.8;
s/atari_round_page/m68k_round_page/

Fix the 'shifting screens' bug on the Falcon. This solution is not perfect
since it gives a bit of flicker while switching frame buffers. From
Thomas Gerner.

- Provide the reasonable size that can/may be mmap-ed. This might differ
from the current frame buffer size.
- The et-driver now allows for a 16Kb io-region (was 8Kb) that can be
mapped. The XFree server uses this.

Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Prepare for (soon to come) et4000 PCI console.

Don't use an absolute I/O adress.

Prototype + -Wall fixes.

Take the type of the attached monitor into account when selecting a
display-mode. Includes hack to make things work on non-VGA monitors.

Allow selection of Falcon (videl) video modes other than the mode selected
on startup (Thomas Gerner).

Add Falcon video support. Thanks to Thomas Gerner.

TAB/spaces/indents cleanup.

Drop 3rd and 4th clauses. Approved by thomas@ (copyright holder).

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.3.168; 1.3.176; 1.3.182;
Prepare for (soon to come) et4000 PCI console.

Take the type of the attached monitor into account when selecting a
display-mode. Includes hack to make things work on non-VGA monitors.

Allow selection of Falcon (videl) video modes other than the mode selected
on startup (Thomas Gerner).

TAB/spaces/indents cleanup.

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.13.78; 1.13.86; 1.13.92;
merge ktrace-lwp.

Adapt to compiling with -Wshadow and -Wcast-qual, by adding const
qualification in places and renaming a few local variables.
Also adds a couple uses of __UNVOLATILE() to allow passing volatile
variables to functions wanting e.g. caddr_t.

branches: 1.11.42;
Changed grf mmap that it is closer to XFree. VGA memory on et4000/et6000
is now at offset 0xa0000.

Add member to bitmap struct that gives the size of mappable memory.

*** empty log message ***

Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Prepare for (soon to come) et4000 PCI console.

Take the type of the attached monitor into account when selecting a
display-mode. Includes hack to make things work on non-VGA monitors.

Right align color values in the bytes.

Add Falcon video support. Thanks to Thomas Gerner.

Add prototype for grf_get_best_mode()

Make ite ioctl work and fixup the colormap stuff.

branches: 1.1.1;
Initial revision

TAB/spaces/indents cleanup.

fix multiplei repetitive typos in comments, messages and documentation. mainly because copy paste code big amount of files are affected.

Misc KNF.

branches: 1.22.2; 1.22.4;
Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Remove extra whitespace added by a dumb tool which replaced bcopy with memcpy.

bcopy -> memcpy

bzero -> memset

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.15.78; 1.15.86; 1.15.92;
merge ktrace-lwp.

__KERNEL_RCSID()

branches: 1.13.2;
DMA, not dma nor Dma.

Use PAGE_SIZE rather than NBPG.

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

possible has two s.

branches: 1.9.22; 1.9.26;
Changed grf mmap that it is closer to XFree. VGA memory on et4000/et6000
is now at offset 0xa0000.

s/atari_round_page/m68k_round_page/

- Provide the reasonable size that can/may be mmap-ed. This might differ
from the current frame buffer size.
- The et-driver now allows for a 16Kb io-region (was 8Kb) that can be
mapped. The XFree server uses this.

Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Prepare for (soon to come) et4000 PCI console.

Yet more prototyping and -Wall fixes.

Prototyping.

First bunch of changes due to stricter prototype checking.

Add Falcon video support. Thanks to Thomas Gerner.

Remove 3rd and 4th clause on Leo Weppelman's license. OK leo@.

Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.1.168; 1.1.176; 1.1.182;
Prepare for (soon to come) et4000 PCI console

TAB/spaces/indents cleanup.

s/changable/changeable/ in comments.

fix typos, mainly in words minimum and maximum, but also few others.

Final bit of PR 41200: headers that declare ioctls should include sys/ioccom.h.
This covers most if not all of the MD headers.

XXX: a lot of the ioctl definitions in some of these files are cutpasted.

branches: 1.9.14; 1.9.32;
Remove clause 3 (UCB advertising clause) from the University of Utah
copyright. Confirmed by Mike Hibler, mike at cs.utah.edu - thanks!
Also, merge UCB and Utah copyright texts back into one, as they
originally were.

Extra verification by snj@.

branches: 1.8.66; 1.8.72; 1.8.74;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.7.4;
s/independant/independent/, from Zafer.

branches: 1.6.20; 1.6.22;
merge ktrace-lwp.

branches: 1.5.12;
Add some mulitple include protection.
Include kbdmap.h from iteioctl.h so that kdump compiles, otherwise
the size of struct kbdmap as used in various ioctl defines is unknown.

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

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

branches: 1.3.42;
Changed grf mmap that it is closer to XFree. VGA memory on et4000/et6000
is now at offset 0xa0000.

Prepare for (soon to come) et4000 PCI console.

branches: 1.1.1;
Initial revision

TAB/spaces/indents cleanup.

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

branches: 1.13.2; 1.13.12;
Split device_t/softc. Also rename some softc variables. Tested on TT030.

XXX: old ugly config_console() hack should go away...

branches: 1.12.2;
Remove clause 3 (UCB advertising clause) from the University of Utah
copyright. Confirmed by Mike Hibler, mike at cs.utah.edu - thanks!
Also, merge UCB and Utah copyright texts back into one, as they
originally were.

Extra verification by snj@.

branches: 1.11.4; 1.11.6; 1.11.8;
Remove all the __P() from sys (excluding sys/dist)
Diff checked with grep and MK1 eyeball.
i386 and amd64 GENERIC and sys still build.

branches: 1.10.44; 1.10.52; 1.10.58;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.9.26;
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;
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.

branches: 1.6.42; 1.6.46; 1.6.54;
Checkpointing my et4000 work. Note that the et-console will not be
functional until the minimal pci-support is checked in.

Catchup with -current:
- Extra argument to boot()
- select -> poll

Prepare for (soon to come) et4000 PCI console.

First bunch of changes due to stricter prototype checking.

Remove the pointer to view_t from the grf-structure, use viewview(dev)
to obtain the pointer instead. This removes a kernel-crash that happened
when something was written to the console while the console was being
resized.

branches: 1.1.1;
Initial revision

TAB/spaces/indents cleanup.

Use aprint_*(9) for device attach messages.