bsd.own.mk: rename GCC_NO_* to CC_WNO_*

Rename compiler-warning-disable variables from
GCC_NO_warning
to
CC_WNO_warning
where warning is the full warning name as used by the compiler.

GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH

Using the convention CC_compilerflag, where compilerflag
is based on the full compiler flag name.

introduce some common variables for use in GCC warning disables:

GCC_NO_FORMAT_TRUNCATION -Wno-format-truncation (GCC 7/8)
GCC_NO_STRINGOP_TRUNCATION -Wno-stringop-truncation (GCC 8)
GCC_NO_STRINGOP_OVERFLOW -Wno-stringop-overflow (GCC 8)
GCC_NO_CAST_FUNCTION_TYPE -Wno-cast-function-type (GCC 8)

use these to turn off warnings for most GCC-8 complaints. many
of these are false positives, most of the real bugs are already
commited, or are yet to come.


we plan to introduce versions of (some?) of these that use the
"-Wno-error=" form, which still displays the warnings but does
not make it an error, and all of the above will be re-considered
as either being "fix me" (warning still displayed) or "warning
is wrong."

branches: 1.73.4;
Remove port-acorn26

OK core@

fix broken patch

PR/50729: Izumi Tsutsui: Add "SMALLPROG"-like options to disklabel(8)

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

branches: 1.69.2; 1.69.8;
Add getlabelusesmbr(), as proposed in
http://mail-index.netbsd.org/tech-userlevel/2011/08/25/msg005404.html
This is used by disk tools such as disklabel(8) to dynamically decide is
the undelyling platform uses a disklabel-in-mbr-partition or not
(instead of using a compile-time list of ports).
getlabelusesmbr() reads the sysctl kern.labelusesmbr, takes its value from the
machdep #define LABELUSESMBR.
For evbmips, make LABELUSESMBR 1 if the platform uses pmon
as bootloader, and 0 (the previous value) otherwise.

Define USE_MBR for ofppc.

Disable COMPAT_386BSD_MBRPART. The code is still here if anyone needs it
for some reason. (But I have no idea why that would be -- if you have one
of these really ancient partitions and you're about to run disklabel, you
can easily run fdisk first and change the partition type to NetBSD.)

As it stands, the code will munch FreeBSD installs under some
circumstances, which is really not acceptable behavior.

The code, along with the kernel support that's been disabled by
default for several years, and some related but less dangerous code in
sysinst, should prboably be removed entirely after -6 is branched.

Discussed on tech-kern and tech-userlevel; closes PR 44496.

This is also almost certainly the cause of PR 42521 and PR 38841.

branches: 1.66.2;
System utilities, boot programs and kernel modules are machine
(port) specific not CPU, so use MACHINE not MACHINE_ARCH.

Remove support for NetBSD/playstation2.

Don't use #ifdef __${MACHINE_ARCH}__ to enable machine dependent features.
Instead, use proper macro defined in Makefile per ${MACHINE_ARCH}.

__${MACHINE_ARCH}__ doesn't represent an architecture of tool's target
but an architecture of binaries being compiled, so required features
are not prolery enabled or unintentionally enabled on certain host
and target combinations during src/tools build.

Make dreamcast, evbsh3, and mmeye use -DUSE_MBR.
All of these ports use src/sys/arch/sh3/sh3/disksubr.c
which is MBR aware.

Convert more MACHINE tests to MACHINE_ARCH

branches: 1.61.6; 1.61.8;
The hpcsh kernel does for sure USE_MBR. Make disklabel use it too on hpcsh.
With this, it is possible to bootstrap a MBR partitioned CF card without
partitioning and disklabelling it first on another platform.

branches: 1.60.4; 1.60.10; 1.60.12;
zaurus use MBR partition info.

New NetBSD/landisk port uses MBR too.

Rototil the way disklabel -r reads and writes labels.
In particular the 'read' part plays 'hunt the disklabel' in order to get a
label into a local buffer - from where it can be displayed/edited.
The 'write' part makes a separate scan of the disk looking for places to
write the label.
The main changes are:
- It can no longer write the first 8k of the mbr to the pbr (or v.v.)
- All labels on the disk (that it can find) get updated during a write
- With -A all the labels are displayed (inc. those deleted by -D)
- Addition of -D which will delete (by one's complimenting dk_magic{2}) and
existing labels before writing labels to the expected locations.
- -v gives some verbose output to stderr, -vv more etc
A better basis for processing incorrect endian labels, or labels from other
architectures.

If HOSTPROG is empty, do not pull in libutil, since we do not need it in that
case, and several host platforms do not have libutil. Fixes the build of
these programs as host tools on platforms without libutil.

Rip out the support for writing bootstrap code. installboot(8) contains
support for hp300, hp700 and vax, $MACHINE is never arm32.
Fix usage output to only contain progname() once - bug introduced in rev 1.89

Make disklabel(8) into a host-tool, "nbdisklabel." Move disklabel.c
to main.c to avoid a name collision with lib/libc/gen/disklabel.c
when we build nbdisklabel. Still todo: commit host-tool build
infrastructure to src/tools/disklabel/.

WARNS=2 is the default defined in sbin/Makefile.inc. (thanks wiz)

WARNS=2

Initial import of iyonix port.

The Iyonix is a desktop machine from Castle Technology, based on a 600MHz
XScale[tm] 80321 processor.

* Uses the bootloader from NetBSD/acorn32, which is now 32-bit compatible.
* Currently boots multiuser with a serial console.
* Device support is not yet complete.

With help from abs.

Add hp700 MD options for proper boot code handling.

* Enable SAVEBOOTAREA on i386 and amd64 (by merging the Makefile
controls with the section for the other MBR-using platforms that
already enable this)
* Don't prompt the user to "erase the previous contents of the disk"
when there's no NetBSD MBR partition; SAVEBOOTAREA is sufficient.

These fixes mean that you can create a disklabel (on an i386/amd64) on a disk
that doesn't have a NetBSD MBR partition without trashing the existing MBR.
The previous behaviour was extremely annoying when working with media such
as FAT-formatted CF cards, and didn't really protect people with such from
accidentally trashing part of sector 1 of such disks, and made it extremely
easy to trash sectors 0..15 of those disks instead.

Bring back disklabel -B for now, since hp300 installboot isn't ready
for prime time yet.

Rip out all the '-B bootblock' support, everything should use installboot(8).
(can only affect arm32, hp300 and vax ports, and doesn't affect sysinst).

playstation2 needs MBR support in disklabel(8).

x86_64 -> amd64

Handle MBRs for x86_64.

Enable USE_MBR for all arm ports.

USE_MBR on macppc too

Rename arm26 -> acorn26.

Rather than making FileCore boot block support conditional on __arm32__, make
it conditional on USE_ACORN, and define that if MACHINE is arm26 or acorn32.
This is less than optimal, and maybe we should define it for all ARM systems,
but then disklabel handling across architectures is generally a mess at the
moment.

Preserve the boot area on x68k.

move showinfo() and showpartition() into separate file, for use by
other programs. slightly change argument signature so that globals
aren't depended upon.

branches: 1.38.4;
MBR on cobalt too.

also use mbr partition table on prep.

- use MBR partition table also on arc.
- use defined(USE_MBR) instead of defined(__i386__), like hpcmips and arc.

Add '-DSAVEBOOTAREA' to keep MBR partition information
when NetBSD partition starts at sector 0.
Closes PR:port-hpcmips/9146.

add hpcmips support (MBR)

branches: 1.33.2;
On NetBSD/alpha, go out of the way to read the boot area's contents if -r
is specified, so that the boot blocks aren't clobbered when writing labels.

Update a pasto-ed comment.

branches: 1.31.2;
Enable non-cylinder-aligned partition warning on sparc[64] & sun3.
See also PR#6853.

Handle new MBR partition ID on i386. Put all code which decides about
usability of an entry to one place, use same algorithm as kernel (ie,
check signature).

branches: 1.29.2;
Define NUMBOOT=1 for the arm32 port.

CFLAGS->CPPFLAGS

switch to using opendisk(3) from libutil

define WARNS?=1 for all of sbin/*

Fix warnings.

i386 now uses "installboot" for installing boot blocks. Comment out
the CFLAGS+= -DNUMBOOT=2 for now.

Remove -g

Off by one.

Nuke obsolete special case .if for amigas.

Jason says: Use getmaxpartitions(3) to determine the number of partitions
allowed in this architecture.

Add -i (interactive mode to create/edit labels) inspired by the SunOS/Solaris
format/partition command.

don't define NUMBOOT on the alpha, any more. boot block installation
is now done by /usr/mdec/installboot.

Changed to use only one bootblock file on VAX.

Added vax for 2-file boot blocks.

convert to new RCS Id conventions; reduce my headache

NUMBOOT == 2 on the alpha, too.

kill unnecessary .include.

specify man pages the new way.

pull in bsd.own.mk by hand, early, so NOMAN= can be defined there.

Finish 4.4-Lite merge, and convert to use err(3)/warn(3) functions.

disklabel(8) from 4.4 merged with our old disklabel(8) by John Brezak
(brezak@apollo.hp.com). hp300's can now disklabel by themselves.

added -D${MACHINE} for amiga's becuase of differing sizeof (struct disklabel)

Add RCS identifiers.

Use ${COPY}, not -c for install.

Don't build man page if NOMAN set.

make user the cat5 dir exists before trying to install man page there.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

PR/50729: Izumi Tsutsui: Add "SMALLPROG"-like options to disklabel(8)

Fix build.

Fix botch in "make disklabel a MI tool" changes in rev 1.2.

After that chanage, "MAXPARTITIONS" constant is not for the target port.
If host's MAXPARTITIONS is larger than a value of the target label and
target endianness is different from the build host, bswaplabel() could
overwrite data beyond the disklabel and primary boot stored after
LABELSECTOR in images might be corrupted.

This fixes boot failure of sun2 liveimage built by
"build.sh -U -m sun2 release live-image"
on TME.

Should be pulled up to netbsd-7.

branches: 1.2.6;
Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

branches: 1.1.6; 1.1.12;
Add necessary byteswap ops for tools disklabel(8) used with -F option
(treating a target disk as a regular file and suppressing ioctl(2)s)
on reading/writing disklabel in a target file.
This allows cross build enviroment creating bootable disk images
for targets in different endian.
No functional changes to native (non-tools) disklabel(8) command.

Closes PR toolchain/42357.

PR/50729: Izumi Tsutsui: Add "SMALLPROG"-like options to disklabel(8)

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

branches: 1.1.6; 1.1.12;
Add necessary byteswap ops for tools disklabel(8) used with -F option
(treating a target disk as a regular file and suppressing ioctl(2)s)
on reading/writing disklabel in a target file.
This allows cross build enviroment creating bootable disk images
for targets in different endian.
No functional changes to native (non-tools) disklabel(8) command.

Closes PR toolchain/42357.

Change back various occurrences of \*[Le], \*[Ge] (less/greater equal)
and \*(ua (upwards arrow) to literal "<=", ">=" and "^" whenever
appropriate (e.g., in code examples).

Remove workaround for ancient HTML generation code.

revert part of revision 1.28. "e.g." is correct. bad igor!

Grammar fix suggested by textproc/igor.
Syntax fix highlighted by mandoc -Tlint.
Bump date.

branches: 1.27.68;
mark vinum as obsolete.

Sync DK types and FS types with <sys/disklabel.h>.

XXX: More other stuff should be sync'ed?

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

Patches provided by Joel Baker in PR 22308, verified by myself.

xref dkctl(8)

Use
.In header.h
instead of
.Fd #include \*[Lt]header.h\*[Gt]
Much easier to read and write, and supported by groff for ages.
Okayed by ross.

Use getlabel{sector,offset}() instead of LABEL{SECTOR,OFFSET}.

partition, not parition (by Adrian Mrva). file system instead of filesystem.

add FS_APPLEUFS
part of PR #17345

tweak indentation on FS_JFS2 comment

add DTYPE_JFS2 and FS_JFS2 for IBM Journaled File System
this is to match FreeBSD's disklabel.h revision 1.73
FreeBSD change was requested by Hiten Pandya <hiten@uk.FreeBSD.org>

fix DTYPE_LD comment

add DTYPE_LD to match disklabel.h

branches: 1.15.2;
Generate <>& symbolically. I'm avoiding .../dist/... directories for now.

Use .Pp for new paragraphs.

#warn is _not_ the GCC CPP directive that gives a warning, #warning is.
(this code has, apparently, always been wrong, but current GCC is kind
enough to warn about unknown CPP directives.)

s/ ar / are /

branches: 1.11.4;
Introduce new filesystem type FS_CCD so that an operator can mark the
ccd component partition. Note that the ccd driver still allows partitions
of any types as components since an on-disk BSD disklabel isn't available
on some port.

Sync disklabel.5 with disklabel.h and dkio.h more closely.

Note the "RAID" partition type.

Sync with sys/disklabel.h

fix bad .Xr references

Principally, change foo(N) to .Xr foo N
Also, do some other mdoc tagging.

alternate -> alternative, per PR 2643

Update.

convert to new RCS Id conventions; reduce my headache

docs follow code.

specify man pages the new way.

Clean up deleted files.

disklabel(8) from 4.4 merged with our old disklabel(8) by John Brezak
(brezak@apollo.hp.com). hp300's can now disklabel by themselves.

Add RCS indentifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Sort and unify a bit. Add more macros.

Add options to define labelsector and -offset and number of slices.
Make options to chose alternate label position for systems using MBR
more intuitive. -m now selects mode with MBR, -n selects mode without,
independent of the machine defaults.

branches: 1.68.6;
Remove workaround for ancient HTML generation code.

Remove ignored Pp macro, highlighted by mandoc -Tlint.

Document the version disklabel first appeared.
Fix spelling mistakes.
Replace contraction.
Bump date.

Add missing doc flags (B,M,m)

Sort sections. Remove comma in enumeration of two items.

mention that we don't handle more than 2TB disks/partitions.

Shuffle the order of examples to put the interactive option (-i) ahead
of $EDITOR and "disklabel -w -r /dev/rsd0c sd2212 foo"

As pointed out by James Wetterau, we wish to complement, as in the
logical operation, rather than compliment, as in praise, the magic
number of the disk.

There are probably other instances of this in the tree. We should fix
them.

Add a '-l' flag and a 'L' command to interactive mode to list all known
file system types. (Similar to what fdisk does for partition types.)

Closes PR bin/12954.

Use more markup, fix xref, serial comma, other misc. fixes.

Update/rewrite to match reality and in a muchmore succinct format.

Quote Fr so options are sorted properly.

Rip out the support for writing bootstrap code. installboot(8) contains
support for hp300, hp700 and vax, $MACHINE is never arm32.
Fix usage output to only contain progname() once - bug introduced in rev 1.89

Rip out the references to the installation of more than one bootstrap file.
Only arm32 needs the -B stuff at all, all other ports use installboot(8).

Bump date for last. Put the exit EXIT STATUS section in the right place.
New sentence, new line. Correct two typos.

Make it possible to distinguish between exit values that indicate errors,
and those that are warnings. Document the exit statuses in the man page.

Single-letter options do not need a line each; add more
commas; drop trailing whitespace; use Pa where appropriate.

Add flag -F to disklabel. Flag -F indicates that the target of the
disklabel operation is a file, not a disk. With -t disktab and -T
disktype, the user may tell disklabel the "geometry" for the file.

Add Xref to mscdlabel, bump date.

Bring back disklabel -B for now, since hp300 installboot isn't ready
for prime time yet.

Rip out all the '-B bootblock' support, everything should use installboot(8).
(can only affect arm32, hp300 and vax ports, and doesn't affect sysinst).

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

Patches provided by Joel Baker in PR 22308, verified by myself.

xref dkctl(8)

Bump date for last.

Document -I option. From Richard Rauch <rauch@math.rice.edu>
in PR#11890.

.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

Removed "disklabel -I [-r] [-C] disk" from SYNOPSIS.

The code does not permit usage of the -I flag without either -e or -i.

mdoc nits. New sentence, new line. Remove a comment about tahoe.

Fix an argument to -width option of .Bl macro.

Generate <>& symbolically. I'm avoiding .../dist/... directories for now.

The sun2 also requires partitions in integer cylinders.

Note in far more places that -B is only supported on some platforms.

Simplify markup in one place, and sort sections and SEE ALSO.

Drop arguments of .Os.

xref (disklabel.8, fdisk.8, mbrlabel.8) as appropriate

allow -I -i to install a new label interactively.

"boostrap" => "bootstrap"

'disklabel -i' on a disk without a label recommends to use -I. Document
that in the usage.

branches: 1.30.2;
Add a -I flag which, combined with -e, gives the user the opportunity
to edit and create a new label on a virgin disk.

branches: 1.29.2;
s/comprised/composed/, per PR 9586 by Seebs <seebs@ged.plethora.net>

mention that hpcmips and arc use `d' as RAW_PART.

kill some bogus comments about -r: it _does not_ cause the boot area to
be clobbered. since 4.4-Lite(?) disklabel.c has gone to extra effort
to avoid clobbering the boot area when using -r, but the 4.4-Lite manual
pages were apparently not updated to note that!

Update for sparc64.

branches: 1.25.2;
Clean up SYNOPSIS formatting.

Add '-f disktab' option to disklabel, via lib/6623 from
Greg A. Woods <woods@most.weird.com>. (Very) slightly tweaked disklabel.8

Clarify the documentation of the "-B" option, inspired by PR#5262.
Contrary to the PR, the "-B" option does work, but not on all platforms.

Add missing -C flag in SYNOPSIS

Fix typo; the flag for interactive disklabel creation is not -e but -i.
Also, make the description of -C flag into separate paragraph.

additional sun3/sparc warning about the "-r" option.

add a warning about sun3 and sparc systems requiring partitions in integer cylinders or the boot ROMs get upset, per PR#3267. Is that it?

- Use .Nm "" instead of .Nm disklabel in SYNOPSIS.
- Do a trick to remove whitespace before close paren from generated output.
- Delete space at the end of line.

all but the first .Nm in SYNOPSIS need to be spelt out. [bin/4256]

- cleanup use of .Nm
- xref installboot(8)
- deprecate register

s/except/accept/, from John F. Woods in PR bin/3802

Fix broken uses of Dd. Both the mdoc and mdoc.samples pages agree:
.Dd is supposed to be invoked like:
.Dd month day, year
e.g. ".Dd January 25, 1989", rather than:
.Dd "month day, year"
which is what these pages did.

Add -i (interactive mode to create/edit labels) inspired by the SunOS/Solaris
format/partition command.

PR/1551: clarify usage in manual page
PR/2452: add -t option to print label in disktab(5) format.
Misc cleanups...

Allow display and editing of partition size & offset fields in
<cyl/track/sector> format. This format is always excepted on input
(`-e' and `-R'); to display it use `-C'.

PR487. From John Kohl. (more examples and some corrections)

convert to new RCS Id conventions; reduce my headache

Finish 4.4-Lite merge, and convert to use err(3)/warn(3) functions.

disklabel(8) from 4.4 merged with our old disklabel(8) by John Brezak
(brezak@apollo.hp.com). hp300's can now disklabel by themselves.

Fix spelling errors

Update to -mandoc macros

Add RCS indentifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Make disklabel(8) into a host-tool, "nbdisklabel." Move disklabel.c
to main.c to avoid a name collision with lib/libc/gen/disklabel.c
when we build nbdisklabel. Still todo: commit host-tool build
infrastructure to src/tools/disklabel/.

Translate an error message from C to english.

Before: disklabel: warning, partition g: offset % cylinder-size != 0
After: disklabel: warning, partition g: not starting on cylinder boundary

Warn overlapping partitions (except the one marked as FS_UNUSED).
Address PR#10777.

A variable referenced isn't `write only'.
Wrap long line and fix some indent while here.

Fix some interesting disklabel lossage:

If the input to 'disklabel' (non -i) referenced a partition two or more
larger than that permitted by npartitions it would error out with 'bad
partition name' (and the option given to re-edit if -e). If a partition one
larger was used it would be dropped from the disklabel with a warning and the
label written anyway! The (off by one) check for a bad partition name was
against npartitions anyway, rather than MAXPARTITIONS. npartitions was just
read from the disklabel and could have been set to an arbitrary value.

Given 'disklabel -i' sets npartitions automatically, make the normal case
check partition names against MAXPARTITIONS, and if necessary increase
npartitions to the size needed to hold the last partition used.

While here remove two 'write only' uses of a 'part' variable.

Make it possible to distinguish between exit values that indicate errors,
and those that are warnings. Document the exit statuses in the man page.

branches: 1.130.2;
Add flag -F to disklabel. Flag -F indicates that the target of the
disklabel operation is a file, not a disk. With -t disktab and -T
disktype, the user may tell disklabel the "geometry" for the file.

Fix remaining issues:
- use getulong everywhere for proper range checking
- fix error checking in getulong

Fix breakage of last commit (disklabel -R just plain didn't work!)
Convert two large nasty #defines into much smaller (still nasty) #defines
that call proper functions.
Fixes PR 24765

use strtoul() instead of atoi(). have proper range check for each fields.

* Enable SAVEBOOTAREA on i386 and amd64 (by merging the Makefile
controls with the section for the other MBR-using platforms that
already enable this)
* Don't prompt the user to "erase the previous contents of the disk"
when there's no NetBSD MBR partition; SAVEBOOTAREA is sufficient.

These fixes mean that you can create a disklabel (on an i386/amd64) on a disk
that doesn't have a NetBSD MBR partition without trashing the existing MBR.
The previous behaviour was extremely annoying when working with media such
as FAT-formatted CF cards, and didn't really protect people with such from
accidentally trashing part of sector 1 of such disks, and made it extremely
easy to trash sectors 0..15 of those disks instead.

Use correct sector number when the netbsd partition is in the extended
partition chain (and not the first extended partition).

Homogenize usage messages: make the 'usage' word all lowercase, as this seems
to be the most common practice in our tree.

fix spelling, s/extented/extended/

Commit changes proposed on tech-kern Thu, 6 Nov 2003
- factor out disksubr.c between sun3, sparc and sparc64. Keep the sun3
groveling code to find a NetBSD disklabel in the first sector (so that it
can find a label at the old sun3 LABELOFFSET) as a fallback is not
label at LABELOFFSET, or sun label is present.
- Fix the sun3 LABELOFFSET (was 64, but the kernel wrote the NetBSD label at
128)
- Make next68k disksubr.c always write a next-compatible disklabel.
- remove #ifdef __sparc__ hack from disklabel(8), and change it to issue
a DIOCWDINFO after writing the disklabel to the raw partition in the
-r/-I case (so that the kernel can convert the label if needed).

Bring back disklabel -B for now, since hp300 installboot isn't ready
for prime time yet.

Rip out all the '-B bootblock' support, everything should use installboot(8).
(can only affect arm32, hp300 and vax ports, and doesn't affect sysinst).

Overhaul MBR handling (part 1):

<sys/bootblock.h>:
* Added definitions for the Master Boot Record (MBR) used by
a variety of systems (primarily i386), including the format
of the BIOS Parameter Block (BPB).
This information was cribbed from a variety of sources
including <sys/disklabel_mbr.h> which this is a superset of.

As part of this, some data structure elements and #defines
were renamed to be more "namespace friendly" and consistent
with other bootblocks and MBR documentation.
Update all uses of the old names to the new names.

<sys/disklabel_mbr.h>:
* Deprecated in favor of <sys/bootblock.h> (the latter is more
"host tool" friendly).

amd64 & i386:
* Renamed /usr/mdec/bootxx_dosfs to /usr/mdec/bootxx_msdos, to
be consistent with the naming convention of the msdosfs tools.

* Removed /usr/mdec/bootxx_ufs, as it's equivalent to bootxx_ffsv1
and it's confusing to have two functionally equivalent bootblocks,
especially given that "ufs" has multiple meanings (it could be
a synonym for "ffs", or the group of ffs/lfs/ext2fs file systems).

* Rework pbr.S (the first sector of bootxx_*):
+ Ensure that BPB (bytes 11..89) and the partition table
(bytes 446..509) do not contain code.
+ Add support for booting from FAT partitions if BOOT_FROM_FAT
is defined. (Only set for bootxx_msdos).
+ Remove "dummy" partition 3; if people want to installboot(8)
these to the start of the disk they can use fdisk(8) to
create a real MBR partition table...
+ Compile with TERSE_ERROR so it fits because of the above.
Whilst this is less user friendly, I feel it's important
to have a valid partition table and BPB in the MBR/PBR.

* Renamed /usr/mdec/biosboot to /usr/mdec/boot, to be consistent
with other platforms.

* Enable SUPPORT_DOSFS in /usr/mdec/boot (stage2), so that
we can boot off FAT partitions.

* Crank version of /usr/mdec/boot to 3.1, and fix some of the other
entries in the version file.

installboot(8) (i386):
* Read the existing MBR of the filesystem and retain the BIOS
Parameter Block (BPB) in bytes 11..89 and the MBR partition
table in bytes 446..509. (Previously installboot(8) would
trash those two sections of the MBR.)

mbrlabel(8):
* Use sys/lib/libkern/xlat_mbr_fstype.c instead of homegrown code
to map the MBR partition type to the NetBSD disklabel type.


Test built "make release" for i386, and new bootblocks verified to work
(even off FAT!).

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

Patches provided by Joel Baker in PR 22308, verified by myself.

Read correct sector when following extended partition chain.
Require extended partition chain to be through increasing sector numbers
in order to abort loops.

use bounded copy (a few unbounded copy remains - need work)

Allow for the netbsd partition being in the extended partition list.

Add support for UFS2. UFS2 is an enhanced FFS, adding support for
64 bit block pointers, extended attribute storage, and a few
other things.

This commit does not yet include the code to manipulate the extended
storage (for e.g. ACLs), this will be done later.

Originally written by Kirk McKusick and Network Associates Laboratories for
FreeBSD.

Rename `sigset' locals to avoid symbol shadowing warning.

writable, not writeable.

Use getlabel{sector,offset}() instead of LABEL{SECTOR,OFFSET}.

Handle MBRs for x86_64.

While cleaning up other places that invoked $EDITOR, noticed that
this one had all the infrastructure of fork/exec/wait, like the
others, but called system instead of exec creating an extra PID
and associated memory usage during the edit.

Fix bug causing segmentation fault when no fs type specified

add FS_APPLEUFS
part of PR #17345

need disklabel_mbr.h for USE_MBR build

display / read the p_cdsession field in the FS_ISO9660 case

Alpha boot block stuff has moved to <sys/bootblock.h>

Make this build again after the alpha/disklabel.h and dev/dec/dec_boot.h
changes.

When accessing the MBR magic as a 16-bit value, consider its on-disk
layout would be little-endian, too.

Remove `const' attribute from writelabel prototype and function to make it
compile again on Alpha.

Rather than making FileCore boot block support conditional on __arm32__, make
it conditional on USE_ACORN, and define that if MACHINE is arm26 or acorn32.
This is less than optimal, and maybe we should define it for all ARM systems,
but then disklabel handling across architectures is generally a mess at the
moment.

minor WARNS=2 stuff

Fix big-endian vs. MBR parameters.

Don't error out if the mbr cannot be read.
This makes "disklabel cd0" working also on platforms with MBR support.

convert to use getprogname()

When -I is specified, we mean to initialize the label if it's not there
already. So, don't fail if there appears to be a corrupt label or
no 'fake' label; get the 'default' label (which is generated
from DIOCGDEFLABEL) instead.

Test read(2) failure by read(..., sizeof(n)) != sizeof(n) instead of
< sizeof(n) to avoid unwanted type promotion.

Kill whitespace at the end of line.

move showinfo() and showpartition() into separate file, for use by
other programs. slightly change argument signature so that globals
aren't depended upon.

- convert to KNF ANSI style guide
- reorder arguments of runcmd(), getnum(), defnum() to be consistent
with other functions (pass struct disklabel * first)

Change "cpg" to "cpg/sgs" in the printed partition table header, in an
attempt to avoid confusion. (PR #9230)

Format string cleanups by Bill Sommerfeld.

allow -I -i to install a new label interactively.

* use strcasecmp() instead of strcmp() so that filesystem type and disk type
searchs (amongst others) are case insensitive.
* in interactive mode (-i), when editing entries display supported disk types
and filesystem types when given `?' (when ``[?]'' appears in the prompt
this feature is supported for the question).
* support `m' as a suffix equivalent to `M'
* in interactive mode, be a bit more sensible about handling errors and EOF
* implement dumpnames(), which takes a char ** and size, and displays
as per ls -F (sorted, listed vertically) but indented by one tab
* don't assume d_typename and d_packname are NUL terminated
* fix up some comments and some warning messages (bad cut & pastos :)
* deprecate deffstypename() and getfstypename()
* be consistent when using sizeof()

'disklabel -i' on a disk without a label recommends to use -I. Document
that in the usage.

warnx?/errx? audit. don't pass variable/function return value alone.
use with "%s". from openbsd.

branches: 1.84.2;
Add a -I flag which, combined with -e, gives the user the opportunity
to edit and create a new label on a virgin disk.

add system(3) for spawning the editor - this is handy when EDITOR is set to
something like 'editor arg1 arg2 ...'

branches: 1.82.2;
Provide some information when confirming "Write outside MBR partition".

- use MBR partition table also on arc.
- use defined(USE_MBR) instead of defined(__i386__), like hpcmips and arc.

disklabel now understands the p_sgs partition field (shift to compute
segment size from block size).

newfs_lfs now reads the disklabel to find segment, block, and fragment
sizes. Because reading this info from the wrong fs type could result in
very poor fs layout (e.g. ffs has "16" where the segshift would go,
resulting in 512-*megabyte* segments for 8K blocks), newfs_lfs refuses
to create a filesystem on a partition not labeled "4.4LFS".

Man pages for newfs_lfs updated to reflect this change.

change "millisecond" to "microsecond" for headswitch and track seek
comments, to resolve PR 8189. This makes the documentation for these
fields consistent. Not that it matters - a grep of syssrc shows that
these fields aren't actually used for anything and should probably be
GC'd.

allow changing partition info from inside the interactive handler.

Fix typo. ('_i386__' -> '__i386__')
Pointed by D'Arcy J.M. Cain, thanks.

add hpcmips support (MBR)

branches: 1.75.2;
don't unbalance { } pairs with #ifdefs or ctags(1) gets confused

Try $TMPDIR before resorting to _PATH_TMP.

Make BSDLFS a EXT2FS-like filesystem, that is, dont print or parse the cpg
field.
According to disklabel.h, its LFS semantics are "segment shift" (log2(segment
size)), but in the code it is used nowhere, and there are even plans to
allow non-poweroftwo segment sizes, so it won't ever work.
While at this, simplify the disktab-like output routine... here, currently,
BSDFFS, BSDLFS, EX2FS and ADOS do the same, so don't duplicate the code.

Print bsize and fsize in disktab-format output also for LFS and ADOS.

Print/parse block size, fragment size and cpg for FS_BSDLFS and FS_ADOS.
(cpg has different semantics for LFS and ADOS, but is in use.)
XXX The disktab reading/disktab format output writing needs still to be done.

On NetBSD/alpha, go out of the way to read the boot area's contents if -r
is specified, so that the boot blocks aren't clobbered when writing labels.

Add a 'chain' command in interactive mode that allows you to vary the
size of the partitions, while the starting point gets adjusted automagically.
This chaining works for partitions that are not 'unused'.

STRICT_ALIGNMENT only applies to the start of partitions, not length.

Fix disklabel -t to work with disks that don't end on a cylinder. Before it
wasn't adding a "su" entry, so when the disktab was read, the sectors per
unit was initialized to "nc"*"sc" which was wrong.

Fixes PR/7446 reported by Matthias Buelow <mkb@altair.mayn.de>.

Allow partitions to extend to the end of the disk, even when
STRICT_CYLINDER_ALIGNMENT defined. This allows existing valid disklabels
to work (including those generated by sysinst).

If checklabel() failed, the label has not been written, so exit(1).

instead of coding an alpha boot block checksum routine here, use
the macro now provided by disklabel.h

branches: 1.63.2;
Use <sys/disklabel_mbr.h>

Enable non-cylinder-aligned partition warning on sparc[64] & sun3.
See also PR#6853.

Add '-f disktab' option to disklabel, via lib/6623 from
Greg A. Woods <woods@most.weird.com>. (Very) slightly tweaked disklabel.8

Adjust for DKTYPENAME changes.

bzero->memset, bcopy->memcpy, bcmp->memcmp

-fix some lint warnings
-fix overflow conditions (PR bin/5534, Zdenek Salvet <salvet@ics.muni.cz>)
(+ one more: can VAX SMD drives be >4G?)
-fix output of "*" at odd end cylinder number (not odd size)
-break some lines >80 cols

Shift column headers into the right position.

const poisoning.

Fix from Chris Demetriou to modify the way in which arm32 filecore
checksums are calculated paying attention to the fact that the way
the checksum works a sectors filled with a single byte value will
always checksum correctly.

when reading the MBR, seek to DOSBBSECTOR * DEV_BSIZE rather than just
DOSBBSECTOR. DOSBBSECTOR is 0 so it doesn't really matter, but the former
is what was really intended.

in makebootarea() (if NUMBOOT is nonzero), before reading the contents
of the old boot area so that 'disklabel -r -w' won't clobber anything,
be sure to seek to the offset of the boot area (like readlabel() does).
Otherwise, if the machine has code which looks up machine-dependent label
bits, the offset read from will be incorrect, and disklabel will end up
secretly replacing the boot area with other data (which probably doesn't
look like a boot block). The only port this currently affects is
NetBSD/arm32, because it's the only one of the three ports that use
NUMBOOT > 0 that has MD label grovelling code.

factor out some more common bits into 'confirm', and make it a bit more
bullet-proof. also, make the arm32 #ifdefs a bit more consistent with
the i386 #ifdefs (without actually changing the behaviour of the code).

use warnx when warning about an old BSD partition ID (i386 MBR code),
since err/warn seem to be used consistently throughout the code, rather
than fprintf.

KNF arm32-specific label-grovelling code. Also, clean up error messages
(errx instead of err, kill bogus newlines, make error messages look more
"normal").

Handle new MBR partition ID on i386. Put all code which decides about
usability of an entry to one place, use same algorithm as kernel (ie,
check signature).

Fix usage; -i and -C flags are alway effective regardless of NUMBOOT.

branches: 1.47.2;
On the sparc, do not write to the raw disk at all if `-r' is on.

Added support for filecore partitions on arm32 platforms.

The number of fs types is FSMAXTYPES not DKMAXTYPES.
Correct read of formatted file (used with -e or -R) so that fstype with
spaces will work. Also enlarge the fstype field by 2 chars so that
"Linux Ext2" will not be truncated.

Backout prototype of opendisk.

Add a prototype for opendisk().

switch to using opendisk(3) from libutil

fix sccsid conflict

Define boot1 if NUMBOOT > 1.
Noticed in pr#3850, but fixed differently.

Fix warnings.

use mkstemp() instead of mktemp()

Add -i (interactive mode to create/edit labels) inspired by the SunOS/Solaris
format/partition command.

Add missing '}', from David Carrel <carrel@ipsec.com>, PR #2873.

PR/1551: clarify usage in manual page
PR/2452: add -t option to print label in disktab(5) format.
Misc cleanups...

Fix disklabel to deal with some syntax errors a little better.

Mostly, when someone specifies a partition entry like:

c: 1295850 0
or
g: 1295850 0 4.2BSD

disklabel would dump core. Now it generates a warning like it should
in the first case and complains about too few fields in the second.

Perhaps I should just default to unused in the first case, but this prevents
core dumps. (pr bin/2659)

Don't display cylinder comments if the number of cylinders is zero. This
closes pr bin/2683.

On the sparc on an extra DIOCWDINFO after writing to the raw device.

Allow display and editing of partition size & offset fields in
<cyl/track/sector> format. This format is always accepted on input
(`-e' and `-R'); to display it use `-C'.

Include the total number of sectors in the text representation of
the label; calculating based on the disk geometry is not always
accurate (notably, for SCSI disks). Also be a bit more careful
about printing out fixed-width types.

sys/errno.h -> errno.h
sys/signal.h -> signal.h

Make sure to reset signal mask on errors.

make sure that the alpha boot block is checksummed in a way more
acceptable to charles; #ifdef it, and do it explicitly here, rather
than doing a WDINFO ioctl for the -r case.

Don't do a DIOCWDINFO after already writing the label.

Use POSIX signals.

re-order the DIOCWLABEL in the rflag case, so it actually works right.

convert to new RCS Id conventions; reduce my headache

do the DIOCWDINFO in writelabel() even if -r is specified.
On most machines it's redundant, but it gives others the chance to
take special actions, e.g. updating a checksum.

Fix typos, and minor cosmetic changes.

Change the magic question to an absolute, flat lie that should hopefully make
people think twice before toying with it. B-P

more cleanups from Jim Jegers, passed over by me.

Clean up this festering pile of rotten code a bit. To wit:
* Add prototypes.
* Remove a strange and apparently pointless bit of code propagated
from 386BSD.
* Collapse NOWRITE and WRITEABLE into one.
* Use the same algorithm as the kernel for finding the label.
* Redo some of the parsing more sanely.

Eliminate uses of some obsolete functions.

Finish 4.4-Lite merge, and convert to use err(3)/warn(3) functions.

Fix i386 disklabel bug

branches: 1.14.2;
usage message with less surprise

disklabel(8) from 4.4 merged with our old disklabel(8) by John Brezak
(brezak@apollo.hp.com). hp300's can now disklabel by themselves.

fix include file

don't core so easily

386BSD -> NetBSD and/or i386

We said we're going to default to `n', now do it!

Check boundary of *c* partition against the DOS partition table, not the *a*
partition, as we might want the latter to not necessarily be at the beginning
of the NetBSD section.

merge in changes from netbsd-0-9-ALPHA2

Add RCS identifiers.

branches: 1.5.2;
Cleanup for GCC 2.

changed "Id" to "Header" for rcsids

added rcs ids to all files

after 0.2.2 "stable" patches applied

branches: 1.1.1;
Initial revision

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

branches: 1.13.6; 1.13.12;
Add necessary byteswap ops for tools disklabel(8) used with -F option
(treating a target disk as a regular file and suppressing ioctl(2)s)
on reading/writing disklabel in a target file.
This allows cross build enviroment creating bootable disk images
for targets in different endian.
No functional changes to native (non-tools) disklabel(8) command.

Closes PR toolchain/42357.

- u_short -> uint16_t
- KNF a bit

Make disklabel(8) into a host-tool, "nbdisklabel." Move disklabel.c
to main.c to avoid a name collision with lib/libc/gen/disklabel.c
when we build nbdisklabel. Still todo: commit host-tool build
infrastructure to src/tools/disklabel/.

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

Patches provided by Joel Baker in PR 22308, verified by myself.

- convert to KNF ANSI style guide
- reorder arguments of runcmd(), getnum(), defnum() to be consistent
with other functions (pass struct disklabel * first)

branches: 1.8.12;
- cleanup use of .Nm
- xref installboot(8)
- deprecate register

Fix warnings.

convert to new RCS Id conventions; reduce my headache

Finish 4.4-Lite merge, and convert to use err(3)/warn(3) functions.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

branches: 1.5.6; 1.5.12;
Add necessary byteswap ops for tools disklabel(8) used with -F option
(treating a target disk as a regular file and suppressing ioctl(2)s)
on reading/writing disklabel in a target file.
This allows cross build enviroment creating bootable disk images
for targets in different endian.
No functional changes to native (non-tools) disklabel(8) command.

Closes PR toolchain/42357.

- u_short -> uint16_t
- KNF a bit

- convert to KNF ANSI style guide
- reorder arguments of runcmd(), getnum(), defnum() to be consistent
with other functions (pass struct disklabel * first)

branches: 1.2.10;
RCS Id Police.

Fix warnings.

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

move dk_ioctl to a header file for the benefit of x-building.

branches: 1.11.6; 1.11.12;
Remove 3rd and 4th clauses in christos' license. OK christos.

Add a '-l' flag and a 'L' command to interactive mode to list all known
file system types. (Similar to what fdisk does for partition types.)

Closes PR bin/12954.

Rototil the way disklabel -r reads and writes labels.
In particular the 'read' part plays 'hunt the disklabel' in order to get a
label into a local buffer - from where it can be displayed/edited.
The 'write' part makes a separate scan of the disk looking for places to
write the label.
The main changes are:
- It can no longer write the first 8k of the mbr to the pbr (or v.v.)
- All labels on the disk (that it can find) get updated during a write
- With -A all the labels are displayed (inc. those deleted by -D)
- Addition of -D which will delete (by one's complimenting dk_magic{2}) and
existing labels before writing labels to the expected locations.
- -v gives some verbose output to stderr, -vv more etc
A better basis for processing incorrect endian labels, or labels from other
architectures.

Remove `const' attribute from writelabel prototype and function to make it
compile again on Alpha.

- KNF printlabel.c
- add a function to print only one partition's info.
- print the partition information if it was modified in interactive mode.
- improve on the chaining code. [still assumes that partition offsets increase
monotonically]. We could check for overlap too.

move showinfo() and showpartition() into separate file, for use by
other programs. slightly change argument signature so that globals
aren't depended upon.

- convert to KNF ANSI style guide
- reorder arguments of runcmd(), getnum(), defnum() to be consistent
with other functions (pass struct disklabel * first)

branches: 1.4.4;
allow changing partition info from inside the interactive handler.

branches: 1.3.4;
Add a 'chain' command in interactive mode that allows you to vary the
size of the partitions, while the starting point gets adjusted automagically.
This chaining works for partitions that are not 'unused'.

Fix warnings.

Add -i (interactive mode to create/edit labels) inspired by the SunOS/Solaris
format/partition command.

disklabel(8): convert malloc(x * y) to reallocarr

PR/50729: Izumi Tsutsui: Add "SMALLPROG"-like options to disklabel(8)

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

move dk_ioctl to a header file for the benefit of x-building.

- simplify getinput.
- add adjust command.

branches: 1.35.6; 1.35.12;
Change printf formats to match the data type of the values being
printed. There's now a lot of PRIu16 and PRIu32, some PRIu8, some
SCNu32, and a few cases where %u and %d were reversed. Multiplication
of 32-bit and 8-bit values is cast to uint64_t and printed with PRIu64.

Inspired by a report from Patrick Welche on current-users.

Change getnum() to use intmax_t instead of int, so it doesn't overflow
between 2^31 and 2^32. Adjust call sites accordingly. PR 43354.

- Display "Enter '?' for help" when starting interactive mode
- Use %.40g rather than %g when printing sectors and MB for existing
partition size/offset.
Changes [1.93802e+06c, 1953525105s, 953870M]:
to: [1938021c, 1953525105s, 953869.6875M]:

Remove 3rd and 4th clauses in christos' license. OK christos.

fix sign-compare issues

branches: 1.30.26; 1.30.28; 1.30.32;
Add a '-l' flag and a 'L' command to interactive mode to list all known
file system types. (Similar to what fdisk does for partition types.)

Closes PR bin/12954.

Avoid indexing beyond the end of a malloced array.
Rework loop to be less confusing.
Coverty 519 made me read this code, but this isn't the 'bug' it found.

Gracefully handle out of memory condition.

Rototil the way disklabel -r reads and writes labels.
In particular the 'read' part plays 'hunt the disklabel' in order to get a
label into a local buffer - from where it can be displayed/edited.
The 'write' part makes a separate scan of the disk looking for places to
write the label.
The main changes are:
- It can no longer write the first 8k of the mbr to the pbr (or v.v.)
- All labels on the disk (that it can find) get updated during a write
- With -A all the labels are displayed (inc. those deleted by -D)
- Addition of -D which will delete (by one's complimenting dk_magic{2}) and
existing labels before writing labels to the expected locations.
- -v gives some verbose output to stderr, -vv more etc
A better basis for processing incorrect endian labels, or labels from other
architectures.

Move WARNS=3 to the Makefile.inc, and add a little const to the remaining
programs that did not compile before.

Make disklabel(8) into a host-tool, "nbdisklabel." Move disklabel.c
to main.c to avoid a name collision with lib/libc/gen/disklabel.c
when we build nbdisklabel. Still todo: commit host-tool build
infrastructure to src/tools/disklabel/.

Recognize more units and be more helpful when we enter bad units.

Add missing ' in prompt text.
Pointed out by Allen Briggs.

Add the ability to define a partition as starting after another partition.
If chaining is on, display all changed partitions (not just this one).

OK'ed by Christos.

Handle partition offset and size as unsigned entities.

disallow creation of partitions which start beyond (or extend beyond)
the end of the disk.

branches: 1.19.2;
minor WARNS=2 stuff

- KNF printlabel.c
- add a function to print only one partition's info.
- print the partition information if it was modified in interactive mode.
- improve on the chaining code. [still assumes that partition offsets increase
monotonically]. We could check for overlap too.

move showinfo() and showpartition() into separate file, for use by
other programs. slightly change argument signature so that globals
aren't depended upon.

- convert to KNF ANSI style guide
- reorder arguments of runcmd(), getnum(), defnum() to be consistent
with other functions (pass struct disklabel * first)

in cmd_name(), put the default prompt into a temporary buffer rather
than overwrite the existing d_packname. noted by enami@.

* use strcasecmp() instead of strcmp() so that filesystem type and disk type
searchs (amongst others) are case insensitive.
* in interactive mode (-i), when editing entries display supported disk types
and filesystem types when given `?' (when ``[?]'' appears in the prompt
this feature is supported for the question).
* support `m' as a suffix equivalent to `M'
* in interactive mode, be a bit more sensible about handling errors and EOF
* implement dumpnames(), which takes a char ** and size, and displays
as per ls -F (sorted, listed vertically) but indented by one tab
* don't assume d_typename and d_packname are NUL terminated
* fix up some comments and some warning messages (bad cut & pastos :)
* deprecate deffstypename() and getfstypename()
* be consistent when using sizeof()

branches: 1.13.4;
Use '$' instead of '-' to select remainder of disk interactively

When setting size of partitions interactively allow '-' for remainder of disk

allow changing partition info from inside the interactive handler.

branches: 1.10.4;
Accept 'Y' for label disk as well as 'y'

Add a 'chain' command in interactive mode that allows you to vary the
size of the partitions, while the starting point gets adjusted automagically.
This chaining works for partitions that are not 'unused'.

branches: 1.8.2;
Adjust for DKTYPENAME changes.

The number of fs types is FSMAXTYPES not DKMAXTYPES.
Correct read of formatted file (used with -e or -R) so that fstype with
spaces will work. Also enlarge the fstype field by 2 chars so that
"Linux Ext2" will not be truncated.

Don't pass return value of writelabel() to strerror(), since
the value is not a errno but just 0 or 1.

Fix warnings.

Off by one.

fix reversed test for maxpartitions.

Jason says: Use getmaxpartitions(3) to determine the number of partitions
allowed in this architecture.

Add -i (interactive mode to create/edit labels) inspired by the SunOS/Solaris
format/partition command.

Add an entry of (belated) virt68k. Required to build live-image.

XXX: should be added to https://www.netbsd.org/developers/new-port.html

branches: 1.58.2;
Fix an old bug in NATIVELABEL_ONLY case in PR/50729 by me. (sigh)

'disklabel -r -w' writes a disklabel at a wrong sector in
NATIVELABEL_ONLY && !LABELUSESMBR && LABELSECTOR != 0 case
if the target disk doesn't have a valid disklabel, due to
incorrect LABEL_OFFSET value.

Found and investigated on NetBSD/hp300 bootable CD tests.
Maybe this affects ports that use distrib/utils/x_disklabel
but have no MBR support, i.e. only NetBSD/hp300 10.0 and
NetBSD/ews4800mips 9.0 and later.

Should be pulled up to netbsd-10 and netbsd-9.

branches: 1.57.2;
disklabel(8): convert malloc(x * y) to reallocarr

first check, then copy

branches: 1.55.2;
s/occurence/occurrence/

s/parition/partition/

Fix a trivial wrong comment. x68k's arch is not m68010.

branches: 1.52.2;
Fix NATIVELABEL_ONLY build.

Add options to define labelsector and -offset and number of slices.
Make options to chose alternate label position for systems using MBR
more intuitive. -m now selects mode with MBR, -n selects mode without,
independent of the machine defaults.

branches: 1.50.2;
Avoid misaligned access in disklabel(8) in find_label()

Introduce a new helper variable tlp and use it for memory access.

Detected with MKSANITIZER/UBSan

A patch by <christos>

Add initial support for ARMv8 (AARCH64) (by nisimura@ and ryo@)

- sys/arch/evbarm64 is gone and integrated into sys/arch/evbarm. (by skrll@)
- add support fdt. evbarm/conf/GENERIC64 fdt (bcm2837,sunxi,tegra) based generic 64bit kernel config. (by skrll@, jmcneill@)

branches: 1.48.2;
Remove port-acorn26

OK core@

use warnx() rather than warn() in a case where errno is not relevant.

branches: 1.46.2; 1.46.4;
PR/50729: Izumi Tsutsui: Add "SMALLPROG"-like options to disklabel(8)

fix mistake in previous

make table smaller in size.

Fix the 3 programs that use DTYPE_ (disklabel disk types) constants from the
kernel. Two of them are inside ifdefs.

Add OpenRISC 1000 & UCB RISC-V platform support.

Fix typo in "dreamcast" port name.

Changes to existing files to enable building AARCH64 userland.
evbarm64-el
This is clang only. While gcc4.8 supports aarch64, no netbsd support has
been written for aarch64 with gcc4.8.

Print uint32_t field as such.

branches: 1.38.2;
Rename NetBSD/hp700 to NetBSD/hppa.

Unfortunately our VCS isn't very helpful here.

Add m68000/coldfire.
Add evbcf.

A number of new earm MACHINE_ARCH have shown up recently; make sure
we can deduce their endianness.

%td is for ptrdiff_t not for off_t

CVE 1020933: Prevent integer overflow by using wider type

CVE 1020935: Prevent overflow

Add an arch_endian entry for x86_64.

Make sure to initialize byteorder if native.

Fix tpyos.

Make disklabel a MI tool. It will use MACHINE/MACHINE_ARCH to determine
the disklabel params as well as allowing command-line options of -M <machine>
and -B {le,be} to specify MACHINE and byteorder to be used.

move dk_ioctl to a header file for the benefit of x-building.

branches: 1.27.2;
PR bin/45744
from Julian Fagir
Removing options that have had implementations removed.

-b: removed in -r1.4
-s: removed in -r1.2

branches: 1.26.2;
Add getlabelusesmbr(), as proposed in
http://mail-index.netbsd.org/tech-userlevel/2011/08/25/msg005404.html
This is used by disk tools such as disklabel(8) to dynamically decide is
the undelyling platform uses a disklabel-in-mbr-partition or not
(instead of using a compile-time list of ports).
getlabelusesmbr() reads the sysctl kern.labelusesmbr, takes its value from the
machdep #define LABELUSESMBR.
For evbmips, make LABELUSESMBR 1 if the platform uses pmon
as bootloader, and 0 (the previous value) otherwise.

Use __dead

Change printf formats to match the data type of the values being
printed. There's now a lot of PRIu16 and PRIu32, some PRIu8, some
SCNu32, and a few cases where %u and %d were reversed. Multiplication
of 32-bit and 8-bit values is cast to uint64_t and printed with PRIu64.

Inspired by a report from Patrick Welche on current-users.

match printf formats and types from Patrick Welche

Add necessary byteswap ops for tools disklabel(8) used with -F option
(treating a target disk as a regular file and suppressing ioctl(2)s)
on reading/writing disklabel in a target file.
This allows cross build enviroment creating bootable disk images
for targets in different endian.
No functional changes to native (non-tools) disklabel(8) command.

Closes PR toolchain/42357.

Don't use #ifdef __${MACHINE_ARCH}__ to enable machine dependent features.
Instead, use proper macro defined in Makefile per ${MACHINE_ARCH}.

__${MACHINE_ARCH}__ doesn't represent an architecture of tool's target
but an architecture of binaries being compiled, so required features
are not prolery enabled or unintentionally enabled on certain host
and target combinations during src/tools build.

fix sign-compare issues in vax-specific code.

branches: 1.19.4;
Remove the \n and tabs from the __COPYRIGHT() strings.
(Tweak some to use a consistent format.)

branches: 1.18.2;
Remove clause 3 and 4 from TNF licenses

branches: 1.17.12; 1.17.14;
Let the operator use the DISKLABELSECTOR and DISKLABELOFFSET
environment variables to override the sector number, and the
bytes-offset in the sector, where disklabel(8) writes a disklabel.

If disklabel(8) was compiled as a host tool, then the defaults
still come from the port's LABELOFFSET and LABELSECTOR #defines.
Otherwise, the defaults still come from the sysctls, kern.labelsector
and kern.labeloffset.

This change helps me create a bootable CompactFlash image with both
a Master Boot Record (MBR) and a BSD disklabel for the RouterBOARD
153. The RB 153 is an evbmips board with a CompactFlash slot. It
searches the MBR on a CompactFlash card for a partition containing
an ELF kernel. When disklabel(8) is built for evbmips, it will
ordinarily overwrite an MBR with the BSD disklabel.

Remove defunct B from getopt string.

include <signal.h> where signal(3) is used

Remove more duplicate #includes,
from Slava Semushin <slava.semushin@gmail.com>

Add a '-l' flag and a 'L' command to interactive mode to list all known
file system types. (Similar to what fdisk does for partition types.)

Closes PR bin/12954.

GCC4 fixes from Mike Pumford

PR/33695: Arnaud Lacombe: Add compatibility syntax for FreeBSD interoperability.

fix gcc warning.

Coverty 521: fclose() temporary file used when editing label.

Fix acorn builds

Fix alpha build - pointed out by Kurt Schreiner on port-alpha

Rototil the way disklabel -r reads and writes labels.
In particular the 'read' part plays 'hunt the disklabel' in order to get a
label into a local buffer - from where it can be displayed/edited.
The 'write' part makes a separate scan of the disk looking for places to
write the label.
The main changes are:
- It can no longer write the first 8k of the mbr to the pbr (or v.v.)
- All labels on the disk (that it can find) get updated during a write
- With -A all the labels are displayed (inc. those deleted by -D)
- Addition of -D which will delete (by one's complimenting dk_magic{2}) and
existing labels before writing labels to the expected locations.
- -v gives some verbose output to stderr, -vv more etc
A better basis for processing incorrect endian labels, or labels from other
architectures.

When building as host tool don't include "../../include/util.h".

getlabel{offset,sector} and opendisk functions declared in that file
are not used in this case anyway, and <util.h> pulls in unwatned
includes (e.g. <utmpx.h> doesn't exist on FreeBSD 4.*).

XXX: We can probably undo the #if !HAVE_NBTOOL_CONFIG_H kludge in
<util.h> now.

Rip out the support for writing bootstrap code. installboot(8) contains
support for hp300, hp700 and vax, $MACHINE is never arm32.
Fix usage output to only contain progname() once - bug introduced in rev 1.89

Remove the never changed 'bootxx' variable

Remove all the code that was under 'NUMBOOT > 1' nothing has set that for
a long time.

Make disklabel(8) into a host-tool, "nbdisklabel." Move disklabel.c
to main.c to avoid a name collision with lib/libc/gen/disklabel.c
when we build nbdisklabel. Still todo: commit host-tool build
infrastructure to src/tools/disklabel/.

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

Patches provided by Joel Baker in PR 22308, verified by myself.

Try $TMPDIR before resorting to _PATH_TMP.

convert to new RCS Id conventions; reduce my headache

Finish 4.4-Lite merge, and convert to use err(3)/warn(3) functions.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Change printf formats to match the data type of the values being
printed. There's now a lot of PRIu16 and PRIu32, some PRIu8, some
SCNu32, and a few cases where %u and %d were reversed. Multiplication
of 32-bit and 8-bit values is cast to uint64_t and printed with PRIu64.

Inspired by a report from Patrick Welche on current-users.

Remove local definitions of PRIxxxx macros in favour of the
centralised definitions from tools/compat/compat_defs.h, which
are included via nbtool_config.h.

Oops, brain fart. PRIu32 should be "u", not "d".

Define PRIu32 in host environments that don't have it defined.
Define it to "d", as we are unlikely to support hosts where int is
smaller than 32 bits anyway :).

Make disklabel(8) into a host-tool, "nbdisklabel." Move disklabel.c
to main.c to avoid a name collision with lib/libc/gen/disklabel.c
when we build nbdisklabel. Still todo: commit host-tool build
infrastructure to src/tools/disklabel/.

Print everything unsigned - helps a bit for disks with [2^31..2^32) sectors.

If a partition is adjusted so that it falls off the end of the disk, display
a '+' character (aka the '*').

OK'ed by Christos.

Partition size and offset inside the disklabel are unsigned. Print
them as unsigned so that we don't print negative values when we're
halfway through.

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

Patches provided by Joel Baker in PR 22308, verified by myself.

Make output fit into 80 columns when a partition doesn't end on a cylinder
boundary.

Line up the cylinder numbers in the printlabel output a bit better.

add FS_APPLEUFS
part of PR #17345

display / read the p_cdsession field in the FS_ISO9660 case

minor WARNS=2 stuff

- KNF printlabel.c
- add a function to print only one partition's info.
- print the partition information if it was modified in interactive mode.
- improve on the chaining code. [still assumes that partition offsets increase
monotonically]. We could check for overlap too.

branches: 1.1.2;
move showinfo() and showpartition() into separate file, for use by
other programs. slightly change argument signature so that globals
aren't depended upon.