switch over to unified MAKEDEV

reassing majors for crypto and pf to use the newly defined MI major range

Add a /dev/constty == /dev/console but with minor 1.
Use for tty sessions on 'console' so that output doesn't pinched by anyone
using TIOCCONS to get kernel and system messages.

add isdn telephony dialout devices

Change the way pseudo tty devices are generated in order to avoid
any pipes and running other commands (eg dd).
Measurable speeds up creating pseudo ttys.
(there has to be a better way than editing 51 files! - tedious at only
6 key presses per file)

create /dev/crypto

reserve cdev major # for PF. ok'ed by technical-exec

Remove device nodes for obsolete olms and omms drivers.
Except for i386, the driver never existed for that port anyway...

Add /dev/icp*.

Remove ksyms before mknod ksyms.

fix ugen issue from latest shell changes that treat numeric constants
differently depending on their prefix.

Add ksyms to "std" devices.

Add /dev/power (part of sysmon).

Create two ptys in the init case. You never know, you might need them
(sysinst does).

Add an 'init' target, that creates the device nodes in the case where
MAKEDEV is being called from init(8) to create an MFS /dev. For most
ports, this is the same as 'all'. The only exceptions are i386 and
x86_64.

make ipl consistently "IP Filter"

Miscellaneous fixes from Igor Sobrado, in PRs 20943-20959.

don't need duplicate nsmb entry in 'Special devices'

fix nsmb* to use correct assigned major, hook creation of nsmb devices
into 'all' target

Describe twe device.

Reorder dpti and dpt cases so that 'dptiN' is not interpreted as dpt unit number
'iN'.

add twe control device

Create dpt(4) nodes.

Added verified exec device (veriexec)

actually, don't make /dev/io on i386 neither; the kernel only supports
it as backwards compat with COMPAT_10 option

add comment regarding /dev/io

whitespace consistency cleanup

Include cgd on the list of things to be mknod(8)ed on all arches.

- in local), only run $0.local if it is an existing file
(prevents spurious warning for mfs /dev, amongst other things)
- for mipsco, news68k, newsmips: ensure that "umask 77" is run after $0.local
- for playstation2: add local) target, and use in all) (for consistency)

Add dmoverio.

Add the kttcp device.

create /dev/ttyEstat

Update comment on how to regenerate MAKEDEV(8) after changing the
MAKEDEV script. Not that anybody acts on it, but it's nicer if it's correct.

Add device creation support for systrace.

branches: 1.175.2;
Remove opms(4) and its device, /dev/pms0, from the i386 port, because
it has been obsoleted by pms(4).

Reviewed by fvdl and christos.

Rename ISDN devices, per discussion on tech-kern. The network devices
become ippp (ISDN ppp) and irip (ISDN raw IP). The character device now
are called: /dev/isdn (isdnd <-> kernel communication), /dev/isdnctl (dialing
and other control), /dev/isdntrc* (tracing), /dev/isdnbchan* (raw B channel
access, i.e. for user land PPP) and /dev/isdntel* (telephone devices, i.e.
for answering machines).

Remove trailing semicolons from lines that don't need them.
TABify many lines.

various cleanups inspired by etc.i386/MAKEDEV:

* use "ln -fs foo bar" instead of "rm -f bar; ln -s foo bar"

* remove unnecessary chown root or chgrp wheel directives; older code here
didn't bother doing this, so why be inconsistent

* when making directories, don't use -p and don't redirect stdout to /dev/null

* clean up whitespace

* wrap the main parser code in a function makedev(), and call makedev
instead of "sh $0". (this is a bit faster)

* don't barf when we have fdesc mounted; just skip the devices fdesc provides


(these files should really be automagically generated...)

Use a tab instead of 7 spaces.

remove chown root:wheel . that I added in 1.164; too many warnings with fdesc

Change a 62 to a 16, so that the proper number of slots is skipped.
Fixes PR 15650.

be consistent, and use "mkdir foo 2>/dev/null" instead of "mkdir -p foo"
(to suppress errors if foo already exists)

Use "mkdir -p" instead of "mkdir" to create /dev/altq

remove unnecessary 'chown root.wheel' and be consistent with the rest
of this file...

Make ulpn device nodes (no reset on open).

- ensure . is owned by root:wheel (so new entries inherit the correct group)
- wrap the main parser code in a function makedev(), and call makedev
instead of 'sh $0'. (this is a bit faster)
- remove unnecessary chown root or chgrp wheel directives; older code here
didn't bother doing this, so why be inconsistent
- when making directories, don't use -p and don't redirect stdout to /dev/null

Fix minor mistake made while increasing number of pty devices: where we had
"pty0 pty1 pty2 pty3" before, for a total of 64 device nodes, we now want
just "pty0" for a total of 62 -- the original commit had "pty0 pty1" for a
total of 124, which ate too many inodes and made "ls /dev" a bit messy.

fix another typo in previous commit: the version committed was missing the end of the opty) case.

fix typo in previous commit -- pty device number is 6, not 5!

Add support for up to 930 pty devices (instead of the current 256).

The openpty() routine has been modified to use the extra pty names
before the "traditional" names, so that programs too dumb to use
openpty() will, hopefully, find one of the "traditional" ptys free
even if many others are in use. The modifications to MAKEDEV are
courtesy Andrew Brown, and are pretty clever: the unit numbers used
by the "traditional" names stay the same, to avoid trouble when
upgrading existing systems. The unusual use of "dd" to index an
array in MAKEDEV is because no other simple method seemed feasible
using only the programs on the install media for all ports.

Move 'wscons' to directly after 'std'. Various MAKEDEVs had it pretty much
at the end, and as wscons (actually ttyE0) is required to login on the console,
it is probably better that an out of space MAKEDEV fail on some other device.

Update the various MAKEDEV's to reflect the default of 8 RAIDframe devices.

add nsmb device

Added clockctl

fix a couple of spelling mistakes in the i4b* descriptions

Fix typo in last.

Make radio devices.

Standardize ch* and uk* descriptions.

Make cir and irframe.

Make more USB controller devices. You can easily have 6 these days.

Create wsmuxN and wsmuxctlN together.
Use separate nodes for wsmouse and wskbd.

Allow creation of wsmuxctl devices.

Add `dpti', which is an I2O vendor extension implementing the DPT/Adaptec
control interface. This makes it possible to run the Linux versions of
dptmgr, raidutil, dptelog and other tools under NetBSD.

Create wsfont in the wscons target, so it would be created during a
MAKEDEV all.

Fix a typo.

Make pci device nodes.

Add an entry to make wsmux devices.

Add agp device(s)

add comments for automatic manpage creation
(pointed out by Thomas Klausner)

add vmegen and wsfont devices

Hook in the mly control interface on i386.

Make also ed1 for 'all', create ed0, ed1 for 'floppy'

add ed disk (MCA ESDI disk)

Add ALTQ control devices.

Add a driver for the Mylex DAC960 family (including DEC SWXCR).

Group isdn devices, so "sh MAKEDEV isdn" will create them all.
Add apropriate entries to the archs that already have isdn support
enabled in their cdev array.

Add ISDN devices.

Use highest bit of minor as an extra bit for the partition number to
bump the maximum partition number on NetBSD-i386 to 16. (Approved by fvdl)

Add two more SCSI disks for the floppy. See PR 9912.

Add MAKEDEV support for Cyclom-Y multiport serial boards. Driver
support was already in $MACHINE/$MACHINE/conf.c.

Make iop device files.

Remove watchdog before creating it. Keeps the same behavior as the rest of the creates

lsu -> ld, by popular request.

Make the `watchdog' device node.

ca -> lsu

Get the major no right for uscanner.

Prepare for USB scanners, and add Rio 500 support while messing.

Really, we don't need a 2nd pty0 in the floppy target.

Move to a default of 64 PTYs in "all" targets. Normalize "floppy" and
"minimal" targets to use 16 PTYs.

Fix brain'o in last, pointed out by Johan Danielsson.

lm* -> sysmon

branches: 1.118.2;
pcvt is gone and therefore ttyv[1-7] as well.

Revert to decemal numbering. From talking w/ Jason, we really should
support more than 256 ports. That would mean 3 digits in hex, and I
don't want to write the script for that..

Cyclades-Z tty's are now ttyCZ??, not ttyCZ????. Add support for them
to macppc's MAKEDEV.

Fix typo in last commit.

Handle "bktr" related devices like audio devices:
- Create device nodes with numbered names (e.g. "tuner0").
- Create softlinks with unnumbered names (e.g. "tuner") pointing to device
nodes of unit 0.

branches: 1.113.2;
Add `ses' devices.

MAKEDEV goo for Cyclades-Z multi-port serial.

Add tun2 and tun3 to the all target and create the ugen0 devices in the
usbs target.

add devices for bktr; also add them to all target

Add a comment about urio.

Make urio(4) device node.

pcmcia modems end up at com3 with a GENERIC kernel, so create a tty03 in std

Device entries for ca(4).

add lm0 to 'all' target

Don't use grep to find if we have fdesc mounted. case is faster and
works without having /usr mounted.

Make /dev/usb, a node where USB events can be read.
Also add ttyU to hpcmips.

No need to make audio devices twice in all.

Add `lm*' for LM7[89] hardware monitor

- rename (revert) normal memory disk device name /dev/mdX -> /dev/mdXa
- now raw partitions are used for memory disk ioctls

allow "sh MAKEDEV ttyNN", where NN is 00 to 99.
single-digit "sh MAKEDEV ttyN" works in backward-compatible way.
XXX should bring it to other ports as well?

Correct md* entries.

Back out previous

Final chown user.group -> chown user:group changes.

branches: 1.95.2;
Tidy up comments a bit, to make the manpages generated from the a bit
nicer. Also, add a comment on how to re-generate the manpages if the
MAKEDEV file was changed.

Improve the USB comments.

Add a comment about ttyU*.

Make creation of wsmux devices a little less confusing.

Rename umodem to ttyU. If it walks like a tty and quacks like a tty
it's a tty.

Add umodem device.

Make wsmux devices.

Add "audio" target to create links "audio", "audioctl", "mixer" and
"sound" pointing to the first unit's device nodes.

typo

Make sure fd is a directory before we df it.

Create *wd[23]? device nodes, so that machines with more than two IDE
drives can be installed.
XXX The 1.4 single-floppy install has only 24 free inodes, but 32 are needed
by this. Some more work is needed (e.g., change newfs parameters for the
ram disk, or switch back to a 2-floppy install).

Don't barf when we have fdesc mounted. Just skip the devices fdesc provides

branches: 1.83.2;
build audio[0123] links.

Need pseudo-ttys on INSTALL disk now (due to sysinst).

Add tty2 to 'all' and 'floppy' targets (that's how a PnP or pcmcia modem
will get called with a standart kernel).

Because PCVT only offers 8 virtual consoles by default create only 8
device nodes. Fixes PR port-i386/6876 by Scott Telford.

On platforms that have GENERIC kernels containing 'bpfilter 8' or greater,
create at least eight bpf dev entries. (All except atari).

add "ttyEcfg" (for display-global wscons configuration)

Make ugen devices. None are created by default.

Add wscons to `all' since its absence annoyed perry.

Added RAID devices for everyone!

Add /dev/scsibus* device nodes.

Nuke symlink /dev/ttyv0 -> /dev/vga It isn't needed any more.
(If any linux emulation needs this, it can go in the linux emul
directory in the emulation package.)

/dev/cfs* for Coda.

Make some MIDI devices.

Add MIDI support. The MIDI devices can be accessed as ``raw'' through
the /dev/rmidiN devices, or with a sequencer interface via /dev/music.
So far the only supported MIDI device is the MPU401 port on SoundBlaster
(and only on SB on isapnp, since we do not have locators with multiple
values yet).

Use `wsmouse' and `wskbd' rather than `wms' and `wkbd', and make some
wskbds in the `wscons' meta-target.

Nuke ums and ukbd (USB mouse and keyboard), add wms and wkbd (wscons dito).

Add some comments about USB and make the USB devices in `all'.

Add USB support. Supported so far:
* UHCI and OHCI host controllers on PCI
* Hubs
* HID devices withe special drivers for mouse and keyboard
* Printers

change wscons tty names to ttyE* to avoid clashes with the pty namespace

add nwscons tty nodes

Introduce a couple of variables to make this clearer.

Create dty (dialout) device nodes for the com driver.

Make some more disk devices; people may have more than 2 wd disks
or 4 sd disks these days.

Fixes PR #3858: /dev/fd uniformly updated in MAKEDEV to make directory
(and file nodes) owned by root.wheel, and /dev/fd directory mode 755. Now
it matches mtree (and the rest of the system's permissions/owner layout).

add appropriate entry to make "random" as documented in rnd(4).
add "random" entry to "all".

branches: 1.58.2;
* st* and wt* have perms 660 instead of 640
* add enss* to ss*
* invoke MAKEDEV.local as $0.local
* reset umask to 077 after running MAKEDEV.local
* warn on unknown devices

Create /dev/audioctl and allow creation of multiple audio devices.

Finally finish fixing pr-230 by assigning minor+8 and minor+16 device
names for the lower density minor dev numbers. My shell code would
even work for a wt1 device, but I would seriously pity anyone trying
to get two of these things to work on one machine. Heck, no one should
even try to use ONE wt device.

Make MAKEDEV and MAKEDEV.local mutually consistant. Reported in
pr #2596 by Tatoku Ogaito, but my fix is different from his.
This is purely an aesthetic change.

Make all tty??/pty?? pairs that can be opened by openpty().
Remove use of awk for generating names.
Make all consistent as possible.
Closes PR 3420.

Make all MAKEDEVs make fd/* similar. No need for awk or expr.
Inspired by PR 3420. Some MAKEDEVS had the fd separated out
of std and fd added to std.

Satellite receiver driver for ISA cards from PlanetConnect by Jason and
Herb for Canada Connect Corp.

Added a new device entry for ipfilter: /dev/ipauth

Add ipnat and ipstate at the appropriate minor numbers off the ipfilter
device.

Use a shell loop to build fd/*, rather than an awk script.

Unroll the pty creation loop.

add 'ipl' device node entry, for ip-filter.

Device rename: `rd' => `md'

change rd* target to produce rd?[abc] and rrd?[abc] devices.
add "ramdisk" target which is basically floppy + rd0, needed for
new install system.

Correct char device major number for `rd'.

Document `ch'.

Add `ss' and `uk'. Sync `all'. General cleanup. (No more expr!)

add Advanced Power Management driver
Includes bootstrap support from FreeBSD (Tatsumi HOSOKAWA and <ukai>)

forgot about joy1

Added joystick support (from Matthieu Herrb); also RCS/CVS ID police.

Collapse some cases together.

Allow more than one wt, and fix up permissions. Unclutter st.

Add no rewind device nodes for the wt driver

Add entries for the ccd.

etc/etc.i386/MAKEDEV create mixer device. PR# port-i386/1208.

add tun to all

Remove silly unit number limitations on several items.

Change vga to ttyv0, with a symlink for X.

Fix minor numbers for pms.

branches: 1.29.2;
Fix typo.

add a 'floppy' target, with minimal devices useful for install.
make /dev/fd 'fd', seperate from 'std', because we need 64 more
device nodes on floppies like i need a hole in my head.

make all device nodes for vnd's, but only make 2 vnd's...

add pty1 to all, kill the ttyv0 weird window system hack

create device nodes for tun driver

Update for new lpt.

Update for new mouse drivers.

clean up:
(1) have all actually make all of the various devies (like vga!)
(2) re-org a bit
(3) mark some XXX's
(4) delete some old cruft

having symbolic links in /dev is a bad idea (for ttyname()?)

Forgot to add `rm' command in previous change.

Fix vnd minor number calculation.

;;

make /dev/audio and /dev/sound

Make device special files for vnd

add /usr/sbin to PATH for chown, as suggested by brad@fcr.com

add mcd: Mitsumi CD driver by Holger Veit and Brian Moore
and do minor cleanup

don't make /dev/acct

Add /dev/io, used to get I/O privileges. This has been in the kernel for a
while.

fix tty?? (serial device) creation

Add pms, allow for multiple mice, and add mice to listing at top of file.

Added mms and lms busmice, plus mouse-* symlinker.

add support for terry lambert's loadable kernel modules.
needs a bit of cleanup, but overall: SLICK!

have makedev create /dev/zero

fix broken patchkit patch for pty stuff...

add a ;; to make sh parse it, and add acct to the list of devs to remove
for "std"

cleaned up, added support for various devices. hacked, from patch 135

make /dev/acct as a standard device

after 0.2.2 "stable" patches applied

branches: 1.1.1;
Initial revision

MAKEDEV: New `virtio' target for all virtio-related devices.

We can use this to add more as they come up so we don't have to keep
tweaking every etc.${MACHINE}/MAKEDEV.conf every time. Currently
this makes nodes for viocon(4) and vio9p(4).

(Not all GENERIC kernels enable vio9p(4) but I suspect the ones that
don't really should; for those that are disabled because PAGE_SIZE !=
4096, I expect it's just a matter of some additional indexing to make
vio9p(4) work with other page sizes.)

PR kern/59211: vio9p(4): missing from various GENERICs and MAKEDEVs

branches: 1.36.2;
MAKEDEV: Tidy some entries on x86 and Arm.

While here, reduce some diffs arising from ordering and formatting
between these architectures.

Only difference between evbarm and aarch64 now is /dev/vchiq. Not
sure offhand if it makes sense on aarch64 or only 32-bit Arm.

PR port-amd64/58093: /dev/efi missing on x86
PR port-arm/58100: /dev/ttyVI* missing on aarch64
PR port-arm/58101: /dev/nvme* missing on aarch64
PR port-arm/58102: /dev/raid* missing on aarch64

branches: 1.35.2;
viocon(4): New virtio tty driver imported from OpenBSD.

viocon* at virtio?

/dev/ttyVI??

Tested under qemu with:

qemu-system-aarch64 ... \
-device virtio-serial \
-chardev socket,path=/tmp/ttyVI00,server=on,wait=off,id=ttyVI00 \
-device virtconsole,chardev=ttyVI00,name=org.NetBSD.dev.ttyVI00 \
...

I updated MAKEDEV.conf to create /dev/ttyVI?? on all ports where it
looks likely to work based on:
(a) having pci or a non-pci virtio attachment,
(b) `qemu-system-$ARCH -M ?' mentioned something resembling the port,
and
(c) `qemu-system-$ARCH -device virtio-serial' launched without
complaining about the virtio-serial device.

(Criterion (c) excluded sparc and sparc64.)

smbios: Add character device for accessing SMBIOS tables

The /dev/smbios character device gives an aperture into physical memory
that allows read-only access to the SMBIOS header and tables.

acpi: add character device for accessing ACPI tables

The /dev/acpi character device gives an aperture into physical memory
that allows only read access to known ACPI tables: RSDP, XSDT/RSDT, and
the root tables. Adapt acpidump(8) to use this interface by default,
falling back to the old /dev/mem method if it is not available or if
ACPIDUMP_USE_DEVMEM=1 is set in the environment. The user visible benefit
of this change is that "options INSECURE" is no longer required to
dump ACPI tables.

Create 8 /dev/ldX by default, as with other disk drivers and ports

remove nsmb from MD list of devices to create for 'all'

branches: 1.30.2;
Create up to pci19.

We probably don't need more than one pci file, but the code needs to be
changed for it, which might introduce problems, and we're just before a
branch.

Not needed for evbarm because it uses devpubd by default.

Stopgap fix for PR xsrc/54388.

Remove the satlink driver. It was disabled everywhere, had no man page and
no use either. Spotted by thorpej in PR/21345, ok christos.

Add tprof

Remove ISDN from the kernel. It has remained unmaintained for a long time,
is of poor quality, and is now an obstacle to MP-ification. It was removed
ten years ago from FreeBSD for the same reason.

This retires a big user of the mbuf API, and will ease maintenance of the
kernel.

branches: 1.26.2; 1.26.4;
Add NVMe devices.

branches: 1.25.24;
add more sd and pci devices.
remove duplicate ch0.

Add a pass-through ioctl for mfi(4), allowing userland to send raw commands
to the controller. This is compatible with the linux and FreeBSD
implementations.
Add the needed conversion for mfi ioctls in COMPAT_LINUX
Allocate a character major number, and create /dev/mfi0 by default
on amd64 and i386.
This allows (along with a hand-created /emul/linux/proc/devices file)
to run the MegaCLI linux binary provided by LSI.

branches: 1.23.2;
Add 'bpf' to all ramdisk targets (*), or floppy targets if ramdisk calls that.
Sort bpf to come directly after 'std' where relevant.
* (apart from etc.evbppc who's ramdisk target *only* builds md0,
which just doesn't make sense).
Should allow a fighting chance for dhcpcd to get an IP address.

give wedges to everyone

branches: 1.21.4; 1.21.6;
Import the Flash and NAND subsytem code contributed by the University
of Szeged, Hungary.

The commit includes:
- Flash layer, which gives a common API to access flash devices
- NAND controller subsystem for the flash layer
- An example OMAP driver which is used on BeagleBoard or alike ARM boards

branches: 1.20.4;
Now that the installation floppies are including tmpfs as well, stop
distinguishing MAKEDEV init and MAKEDEV all for x86.

Create 8 dk instances by default.

Can't type. xkbd -> xbd.

Create 4 Xen disks by default and all three Xen devices to make it
usable out-of-the-box.

branches: 1.16.2;
Move "makedev cfs" for the vcoda device from MI MAKEDEV.tmpl to
MD MAKEDEV.conf. Not all ports have vcoda in their majors.foo files
and we shouldn't (re)assign major numbers on all ports before 5.0.

Should close PR port-sgimips/38962.

branches: 1.15.12; 1.15.16;
Create /dev/bio

Increase the number of raid(4) device node sets to four and the number
of wd(4) device node sets to eight. This fixes another problem
reported in PR port-amd64/35532

Create /dev/cmos.

Fix MAKEDEV on shark: Move ses(4) devices from MI code (MAKEDEV.tmpl) to
those platforms that actually have the driver.

See also:
http://mail-index.netbsd.org/tech-userlevel/2007/01/08/0001.html

Fixes PR 26511 properly.

OK'd by wrstuden@

Enable /dev/io for COMPAT_FREEBSD.

branches: 1.10.4;
Bluetooth fixes by Iain Hibbert:
Change the way in which bluetooth devices attach to system. The
new way is for devices to attach directly to a btdevN device via
its own control file /dev/btdevN.
- bthub(4) is replaced by btdev(4).
- /dev/bthubctl is replaced by /dev/btdevN.
- configuration now uses proplib(3) property lists.
- btcontrol(8) updated to use new API, and now uses private
- XML config file /var/db/btdev.xml.

Add a /dev/amr* control file for amr(4) devices, which allows sending raw
commands to the controller.
Add a amrctl(8) control tool, which for now only allows to get status
from the adapter (status of adapter, logical volumes and and individual
drives).
From FreeBSD, with some adjustements by Andrew Doran and me.

Add new xen device xsd_kva (for Xen3 dom0 operations)

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.

Add xencons to the default list of devices. Fix port-xen/29887 by Juan RP.

branches: 1.5.2;
Create the xbd1* devices as well as xbd0*, this is useful to install
domUs via CDROM or unmounted FS in the sysinst menu.

Integrate Xen to the i386 build process:
- Add xen devices to MAKEDEV
- Add Xen kernels to list of kernel to build
- Add INSTALL_XENU to the install kernels
- introduce the xbd disk devices to sysinst.

This will add 3 kernels to the i386 release:
XEN0 for use as a Xen domain0 kernel
XENU for use on a non-privileged domain
INSTALL_XENU to install NetBSD on a non-privileged domain virtual disk.

Only make bpf, not bpfN

determine the platform number of disk partitions by looking
at MAXPARTITIONS/OLDMAXPARTITIONS in kernel sources, so that it wouldn't
need to be specified separately in MAKEDEV.conf

change platform MAKEDEV.conf to contain only MD targets and nothing else;
simplify the parsing in MAKEDEV.awk accordingly

MD part of unified MAKEDEV
this contains information about disk partitions used by platform,
and MD MAKEDEV targets, such as 'init', MD part of 'all',
as well as any other MD-specific targets not covered by MAKEDEV.tmpl

Distribute the MICROVM kernel on amd64 and i386.

See sys/arch/{amd64,i386}/conf/MICROVM for usage instructions.

branches: 1.70.4;
x86: Release the ramdisks too.

This way we will get cgdroot.fs (and zfsroot.fs too) in the release.

PR misc/57534

branches: 1.69.2; 1.69.10;
Retire the non-PAE-i386-PV configuration files. Keep only PAE-i386-PV.
Non-PAE has been dropped years ago by Xen.

The content of XEN3_* is merged into XEN3PAE_*, with "options PAE" set.

branches: 1.68.8; 1.68.14; 1.68.16;
remove vga@isa and pcdisplay@isa from i386 GENERIC, and create a new
LEGACY kernel that includes them instead. now radeon@pci is able to
properly claim wsdisplay0 on i386 systems, and radeondrmkms has a good
chance of working.

this "fixes" PR#49290.

branches: 1.67.12;
Disable GENERIC_USERMODE kernel auto-build to avoid problems with the current
autobuild-system.

Move i386 and amd64 usermode configurations to their respective directories
and make the usermode kernels buildalbe under build.sh.

The resulting kernels are build and packaged correctly as are the associated
modules.

branches: 1.65.4;
Reinstate INSTALL kernel for embedded ramdisk installs.

Reasons being:
- INSTALL is GENERIC with an embedded ramdisk, and as such, can benefit from
features included within.
- INSTALL_FLOPPY has its own config(5) file, and is tailored for "small"
floppy images; it misses features/drivers that could be needed to boot
in a decent environment for recent x86 machines (like ACPI)
- makes it closer to floppies distrib available for amd64

While here, comment out INSTALL_FLOPPY and bootfloppy-big image build. NetBSD
does not use the 3.6MiB image for El Torito cdroms anymore.

Remove the FLOPPYMAX limit; i386 needs 4 floppies now. Modify boot.cfg and
release/contents to reflect reality.

See http://mail-index.netbsd.org/port-i386/2011/02/08/msg002307.html

No comments, no objections.

branches: 1.64.2;
Build the "MONOLITHIC" kernel configuration by default and include
in the release.

Using the modularise "GENERIC" kernel on an existing NetBSD 5.0 system
is difficult and error-prone. The "MONOLITHIC" kernel provides an
easy way to test a new kernel or to upgrade an existing system.

Revert previous per comment from ad@ in PR 41668, which is no longer orphaned.

Add MONOLITHIC kernel to i386 release binaries. PR#41668

remove Xen2 from the list of kernels to build for a release.
ok bouyer@

Add XEN3PAE_DOM0 to the list of kernels to build for a release.

branches: 1.59.2; 1.59.4; 1.59.8;
As part of make release copy the miniroot.kmod into installation/miniroot/
and create sums for it.

This means all the pieces to do network installation are easily accessible
following a release build (and will also come out of the build servers).

netbsd-GENERIC is in binary/kernel/
pxeboot_ia32.bin is in installation/misc/
miniroot.kmod is in installation/miniroot/

branches: 1.58.2;
Put the GENERIC kernel onto i386 installation CDs and load a kernel module
from the boot loader to provide the miniroot.

For PR kern/38537:

- Make MULTIPROCESSOR mandatory on i386.

Installation changes:

- Update installation section of release notes to match reality.
- Rename INSTALL to INSTALL_FLOPPY, retire INSTALL_LARGE.
- Build INSTALL kernel from GENERIC, like on amd64.
- Update boot menu to allow disabling ACPI and/or SMP.
- Remove GENERIC.NOACPI from the installed kernel list.
- TODO: install default boot.cfg in etc.tgz.
- TODO: possibly enable PCI fixup stuff at runtime if ACPI is disabled.

Build changes:

- No longer build ALL, it's for verification, is slow to build and the
build process is already crippled by the number of kernels built.
- No longer build GENERIC.NOACPI.

branches: 1.56.2;
Convert all the etc/etc.*/Makefile.inc to use RELEASEDIR/RELEASEMACHINEDIR
just like etc/Makefile does.

branches: 1.55.2; 1.55.4; 1.55.6;
build XEN3PAE kernels as part of release.

Make SMP the default on i386. For the time being the noacpi kernel does
not have it enabled.

-GENERIC_DIAGNOSTIC

Remove GENERIC_LAPTOP and INSTALL_LAPTOP. To compensate, switch
bootfloppy-big to actually use INSTALL_LARGE (with ACPI) and 5 floppies.

branches: 1.51.4; 1.51.6;
Add a comment reminding to update the release notes if the list of
kernels change.

Add GENERIC.NOACPI kernel that offers - surprise! - no ACPI
(in contrast to GENERIC).

For the benefit of modern hardware, enable ACPI by default in INSTALL and
GENERIC kernels. If ACPI is an issue on your hardware, 'boot -c' and
'disable acpi' should be a workaround. ACPI-enabled kernels works fine
on pre-acpi hardware.
for amd64:
- add ACPI to INSTALL and GENERIC, remove the *_ACPI config files.
- get rid of the bootfloppy-big.fs boot image, and got to a 3-floppy boot
image
for i386:
- introduce INSTALL_LARGE which has ACPI and some devices with big firmware
- move some devices from INSTALL to INSTALL_LARGE
- Boot floppies still use INSTALL, and bootfloppy-big.fs is still there
(for thoses who want to build el-torito floppy emulation boot CD) and use
INSTALL.
For both, drop the 'iso-image' code in etc/ to make the iso bootable, we'll
use something else to build bootable CDs.

Install pxeboot_ia32.bin as part of a "make release" on i386 and amd64.

branches: 1.47.2;
Stop building the INSTALL_SMALL, INSTALL_TINY, INSTALL_PS2, GENERIC_TINY and
GENERIC_PS2TINY kernels.
These kernels are for i386 (and small i486) systems that have almost all
long since met their makers.
Add ALL (to help avoid bitrot).

Put the kernels one per line for easier commenting out

Add a kernel config file for Xen3 Dom0, built as part or i386 release.

Rename XEN kernel config files to be more consistent:

INSTALL_XEN3_U -> INSTALL_XEN3_DOMU
INSTALL_XENU -> INSTALL_XEN2_DOMU
XEN0 -> XEN2_DOM0
XEN3_U -> XEN3_DOMU
XENU -> XEN2_DOMU

OK'd by bouyer@

Add & enable boot-CDs (with ramdisk kernels only, no install sets),
built as part of a regular release. For more information, see
http://mail-index.netbsd.org/tech-install/2006/05/11/0001.html

OK: core@

Build XEN3_U and INSTALL_XEN3_U kernels as part of release process.

branches: 1.41.2;
Integrate Xen to the i386 build process:
- Add xen devices to MAKEDEV
- Add Xen kernels to list of kernel to build
- Add INSTALL_XENU to the install kernels
- introduce the xbd disk devices to sysinst.

This will add 3 kernels to the i386 release:
XEN0 for use as a Xen domain0 kernel
XENU for use on a non-privileged domain
INSTALL_XENU to install NetBSD on a non-privileged domain virtual disk.

Add new kernel configuration for ACPI based SMP systems which can also be
used to utilize Hyper-Threading.

Allow the user to specify which file system image to use as the
boot image of the CD-ROM via the CDROM_BOOT_IMAGE variable. The
user may want to create a bootable image with bootlap-big.fs which
uses the INSTALL_LAPTOP kernel configuration instead of the default
boot-big.fs image which uses the INSTALL kernel configuration.

branches: 1.38.2;
- Enable OBJ support for etc/; a scratch dir is needed for iso-image.
- Add ${.CURDIR} to various rules.
- Remove the ${KERNOBJDIR} (sys/arch/${MACHINE}/compile) objdir creation
from here.
- Rework iso-image to copy temporary files (such as kernels or boot files)
into a temporary directory instead of stomping over ${RELEASEDIR}, and
use mkisofs -graft-points to merge the temporary directory and
${RELEASEDIR}/${MACHINE}.
NOTE: mkisofs 2.0 is highly recommended for functioning -graft-points
- Add some more comments

Install release files under "${RELEASEDIR}/${MACHINE}/..." instead of
"${RELEASEDIR}/...".

${RELEASEDIR} is never cleaned , and ${RELEASEDIR}/${MACHINE} is only
cleaned if UPDATE is not defined.

- add snap_post, which makes "clean depend all release" in ../distrib/
for all platforms.
- remove now unnecessary replicated code in the various snap_md_post targets
- remove snap_md_pre; nothing uses it any more
- use ${NETBSDSRCDIR} and ${KERNSRCDIR} as appropriate

add GENERIC.MP to KERNEL_SETS on architectures with MP support

- consistency cleanup
- ensure INSTALLATION_DIRS contains the appropriate extra directories
for a given platform

Do sums of installation/* correctly.

Go into the new stand/misc directory to grab the host-OS prebuilt tools
for installation/misc.

Add a MAKESUMS variable which invokes makesums with CKSUM in its environment.
Use this whenever we want to invoke makesums.

Change the way kernel sets are specified, and make them
more consistent. To quote the comment in etc/Makefile
that describes how it's done:

# This target builds the kernels specified by each port. A port may
# specify the following kernels:
#
# KERNEL_SETS The list of kernels that will be
# packaged into sets, named
# kern-${kernel}.tgz. These kernels
# are also placed in the binary/kernels
# area of the release package as
# netbsd-${kernel}.gz.
#
# EXTRA_KERNELS Additional kernels to place in the
# binary/kernels area of the release
# package as netbsd-${kernel}.gz, but
# which are not placed into sets. This
# allows a port to provide e.g. a netbootable
# installation kernel containing a ramdisk.
#
# BUILD_KERNELS Additional kernels to build which are
# not placed into sets nor into the
# binary/kernels area of the release
# package. These are typically kernels
# that are built for inclusion only in
# installation disk/CD-ROM/tape images.
#

build GENERIC_PS2TINY and make kern-ps2tiny kern set

add INSTALL_PS2 to BUILD_KERNELS

Add GENERIC_DIAGNOSTIC to the list of kernels to be distributed.

Enable building of additional kernel sets.

Add INSTALL_LAPTOP to kernels to be built.

Add standalone "iso-image" target that can be run with DESTDIR and RELEASEDIR
set in src/etc after a "make release" in both "src" and "xsrc", to create a
ISO-image of the release in $RELEASEDIR/installation/cdrom.

Hook for architecture dependent pre/post-processing in etc/etc.*/Makefile.inc
are available as 'iso-image_md_post' and 'iso-image_md_pre', see
etc/etc.i386/Makefile.inc as an example. Might be useful for setting up
bootable CDs on alpha, sparc, ...

Reviewed by Todd Whitesel and Thomas Klausner.

install dosboot.com in the release(7) hierarchy.

Have port-i386 use installation/floppy, for conformance with release(7).

branches: 1.21.2;
Fix this so that release notes are build properly upon `make release'.

branches: 1.20.2;
Build INSTALL_SMALL.

Revert last two changes. As pointed out by enami tsugutomo, snap_md_post
is the wrong place to do the splits, as the tar files are not yet made!

Don't include ${RELEASDIR} in INSTALLATION_DIRS. Error pointed out by
Urban Boquist in PR install/9689.

Build the split sets in place on "make release".

Sum the floppy directory, too.

Clean up a bit.

Put a copy of GENERIC in the kernel directory, too.

Provide some more automation in the "make release" process. Also build
and install:
o INSTALL and INSTALL_TINY kernels (for boot floppy images)
o GENERIC_TINY kernel
o Boot floppy images
Ideas borrowed from etc/etc.sun3/ setup.

branches: 1.12.2; 1.12.6;
Update all this crap so that `make snapshot' makes snapshots in release(7)
format, rather than in some unknown format. Also clean up the code a bit,
remove some code duplication, and add documentation to the Makefile.

Use "${MAKE}" instead of "make" for cross compilation.

These commits do two things:

1. Provide a consistent and extensible framework for compiling
kernels for the "make snapshot" target.

2. Add OBJDIR support for the "make snapshot" target, in the form
of three new make variables: KERNOBJDIR, KERNSRCDIR, and KERNCONFDIR.
Default values are in the Makefile.inc, and for OBJDIR they should
be defined in /etc/mk.conf as

KERNSRCDIR=${BSDSRCDIR}/sys
KERNOBJDIR=${BSDOBJDIR}/sys/arch/${MACHINE}/compile

The next step is to make a new target in the src/etc/Makefile for "release"
(or something like that) to make release(7) format files.

GENERIC{AHA,OTHER} have been merged into GENERIC; fixes PR port-i386/3274.

restore snap_md target (sigh...)

Remove obsolete snapshot functionality (which is now in
/usr/src/distrib) -- Makefile.inc is now a no-op. I chose not to
remove it completely because port specific stuff might show up someday.

Fix the snap_md target to use the correct kernel names; GENERICAHA and
GENERICBT were replaced with GENERICADP and GENERICOTHER some time ago.
XXX The kc_* targets here are still broken, but aren't currently used
XXX anyhow. This change allows one to at least do a "make snapshot".

don't make floppies until it works

add support for making i386 floppies. more needs to go on the kernel
floppies, but this is a damned good start.

yes, i'm a moron

update for new kernel name

add 'snapshot' target, including machine-dependent targets

removed obsolete garbage associated with obsolete snapshot
functionality. Snapshots are built with tools in src/distrib

Use "${BINOWN}" and "${BINGRP}" instead of "bin"

now using sh builtin 'echo'... so kill 'echo' entry in FLOPPYBINS

branches: 1.3.2;
more support for building floppiee easily. This method will become
unwieldy, eventually, but until then, it's fine.

don't need -f for umount, now that alias problems kludged around

add support for making i386 floppies. more needs to go on the kernel
floppies, but this is a damned good start.

Undo loading pciverbose module by default.
See item 14 in src/doc/TODO.modules.

Make use of the pcidevs database by default.

Express more confidence in our ability to boot on hardware with ACPI
enabled and don't warn our users that it might not and already suggest
workarounds.

The ability to disable ACPI and SMP is still there, by dropping to
the boot prompt.

branches: 1.5.4;
Instead of explicitly specifying a kernel name, ommit to allow the bootloader to iterate
through the list of kernel names it is configured to try. This way there is
fallback if /netbsd is not present. netbsd is the first name to be tried anyway.
Issue brought up on tech-kern@ by Patrick Welche <prlw1 AT cam ac uk> where a system
without /netbsd hung on boot.
Suggestion by rudolf <netbsd AT eq cz>
https://mail-index.netbsd.org/tech-kern/2018/02/16/msg023122.html

bootmenu doesn't parse "prompt" as part of a command list, it must
stand alone.

Use random seed file (saved entropy) from default location by default
in i386 and amd64 boot.cfg.

branches: 1.2.8;
add clear=1

branches: 1.1.4; 1.1.6;
Give i386 and amd64 a default boot.cfg.

use single minimal default disktab file for most archs, replacing
MD disktab where possible (everything but vax)
the MD disktab was often either empty, or contained entries even for disks
which support geometry autodetection (SCSI), and/or non-interesting/long
obsolete entries

the old records are still available in example disktab in
/usr/share/examples/disktab/disktab in case anyone would need them

Fix some white space niggles.

Back out last change because we now have a special disktab for the
FFS regression test.

Add a comment to the "floppy288" entry that it is needed for the
FFS regression test.

Add an entry for 2.88MB floppies, even though we can't write them
directly yet. Useful for El Torito bootable CDs, though.

Add ta=4.2BSD to both floppy3 and floppy5 entries. (If you are
disklabeling your floppy, you are most likely putting on a 4.2BSD
file system. This is in support of the distrib/i386/floppies
building via vnds.)

RCS ID police

don't forget backslashes, and spelling. from dgy@mcs.com

sc = ns*nt, not nc*nt.

got rid of sf entry for cp3100, and added new cp3100 layout, for testing
on boat anchor

give 1.2M floppies an "a" partition

branches: 1.1.1;
Initial revision

Move the sample fstab files to /usr/share/examples/fstab/. Duplicates
or very similar configs removed.

fix mfs examples

Note that you now must have swap entries to satisfy swapctl(8)

New and improved fstab samples, now with examples of procfs, kernfs
and cd9660 file systems, an mfs mounted /tmp, an example swap
partition, correct fsck scheduling, and even comments pointing users
at man pages for more information. I'm still not sure we need these
sample files at all, but at least they are now useful to the new user.

Move the sample fstab files to /usr/share/examples/fstab/. Duplicates
or very similar configs removed.

fix mfs examples

Uncomment out the CD-ROM entry and remove outdated comment for the
reason why it was commented out.

Note that you now must have swap entries to satisfy swapctl(8)

New and improved fstab samples, now with examples of procfs, kernfs
and cd9660 file systems, an mfs mounted /tmp, an example swap
partition, correct fsck scheduling, and even comments pointing users
at man pages for more information. I'm still not sure we need these
sample files at all, but at least they are now useful to the new user.

ufs -> ffs; PR #1381

branches: 1.1.1;
Initial revision

i386 doesn't need ld.so.conf at this time.

branches: 1.2.2; 1.2.40; 1.2.42; 1.2.44;
support fully qualified machdep names.

Prototype file that selects i387 if the fpu is present.

Arch-specific rc.conf files are not really autonomous rc.conf
files, they are appended to the end of etc/defaults/rc.conf.

So rename them to rc.conf.append for clarity, as suggested by mrg@. Adapt
Makefile accordingly.

Modify etc/defaults/Makefile so that architectures can specify an additional
rc.conf file. This one should reside under etc/etc.${MACHINE}/, and will
get automatically appended to etc/defaults/rc.conf at build time if present.

This is used by i386 and amd64 to append a small MD rc.conf(5) configuration
at the end of the defaults/rc.conf file, so that powerd(8) can be started
by default when we are running in a Xen environment. This is needed to support
save/restore functions for domains.

From all the alternatives proposed to fix that issue (from /etc/rc.conf
parsing in postinstall to etc/defaults/rc.conf arch-hooks) I believe
this one will appease everyone because it:
- does not touch etc/defaults/rc.conf template file,
- patches it at build time for MD hooks only when required,
- does not need to parse/modify a user-specified file like /etc/rc.conf (which
is a complex, error-prone operation),
- only enables powerd(8) by default when conditions are met (Xen environment)
while still allowing root to shoot himself in the foot if he wants to
override this manually in /etc/rc.conf.

See also http://mail-index.netbsd.org/tech-userlevel/2011/07/25/msg005246.html

default powerd=YES if acpi is present

branches: 1.1.2;
Arch-specific rc.conf files are not really autonomous rc.conf
files, they are appended to the end of etc/defaults/rc.conf.

So rename them to rc.conf.append for clarity, as suggested by mrg@. Adapt
Makefile accordingly.

chown user.group -> chown user:group

according to cvs log of basesrc/etc/etc.sun3/ttyaction and
archive of source-changes, it was backed out with MAKEDEV
because of portability issue.
but i think ttyaction doesn't have that issue as it's never
executed on non-NetBSD environment.

branches: 1.1.4;
Add a ttyaction file for port-i386 that properly sets the ownership
on the console device when you log in using wscons on the console.

Why haven't we had defaults for this all this time?

Switch default console tty from /dev/console to /dev/constty

With this switch processes (such as xconsole) can open /dev/console
without breaking login on the text or serial console. This can be
trivially triggered by enabling xdm in rc.conf and hitting
Ctrl+Alt+F1 or equivalent once booted.

The changes:
- Add entry for /dev/console or /dev/constty if missing
- If a port's had /dev/console 'on' switch it off and enable /dev/constty
- If a port did not have /dev/console 'on', leave /dev/constty off

Some ports had /dev/console off and /dev/ttyE0 enabled, presumably to
avoid just this issue. It may make sense to adjust these also (but not
in this pass)

As discussed on current-users

branches: 1.20.34; 1.20.36;
Per discussion on tech-userlevel, finally fix PR 10127:
move all ttyE* entries that use "vt100" emulation to wsvt25 term type.
The terminfo vt220 entry lacked (correctly) a delete key entry, which
was a regression against the netbsd-5 termcap entry. On the other hand,
only a very small number of foreign systems lacks support for wsvt25
nowadays.

branches: 1.19.18; 1.19.20;
Add a constty entry, marked 'off', with otherwise the same settings
as the console. Now when coming up in single user mode the tset in
/.profile can set the terminal type correctly.
Note: I believe constty should be 'on' and console 'off', but thats a
separate discussion

branches: 1.18.26;
remove ad-hok list of pty's from port tty files, and auto-generate it so
that it is consistent, and in the proper order.

Revert 'console' back to /dev/console (from /dev/constty)

Change 'console' login to use /dev/constty so that its output isn't
stolen by code that requests system messages using TIOCCONS.

Back out last change as suggested by Hubert Feyrer in private e-mail:
The "ttys" file is modified by "sysinst" during installation.

Enable for "getty" processes because because the i386 port used wscons(4)
default these days and the default "wsconf.conf" creates four screens.

Added ttyp[g-zA-Z] (bin/15594).

Do what we should have been doing long ago; use /dev/console by default as the
boot console tty, so that the console is on the same place that the kernel
dmesg output goes, whether wscons or serial. Users who want to use tty00
or ttyE0 explicitly can change /etc/ttys themselves.

Also use "vt100" as terminal type for /dev/console (for a reasonable baseline
that also works with wscons; pccons is long since no longer default).

This addresses PR install/13249, i386-specific, but may be appropriate to
apply to all ports.

This should be the final part of moving to 64 PTYs by default. Added
entries for tty{r,s}{0,1,2,3,4,5,6,7,8,9,a,b,c,d,e,f} to /etc/ttys.

Thanks to Matthew Green for pointing out the need.

branches: 1.10.8;
turn off ttyE[123] by default -- default installs don't configure them.

branches: 1.9.2;
wscons is default now - replace the "ttyv0" getty by a number of
"ttyEn" lines.
Thanks to "Johnny C. Lam" <lamj@stat.cmu.edu> for pointing out.

RCS ID police

Change vga to ttyv0, with a symlink for X.

branches: 1.6.2;
reality

changes from 0.9-ALPHA2 to 0.9-BETA

branches: 1.4.2;
Remove a spurious `unknown' that was screwing people over.

changes for init...

changed `console' to `vga'

branches: 1.1.1;
Initial revision