Define _KERNTYPES for things that need it.

add missing DPADD

Accept NAME=label for special so that we can wire down our wedges.

branches: 1.29.52;
Remove special case

move getmntopts(3) to libutil, build and install also it's manpage
bump libutil minor
adjust individual mount_*/Makefile to use libutil getmntopts(3)

libutil (which was added for MNT_GETARGS) is no longer needed.

do not link all mount programs into single image; the total size difference
isn't really that huge now that these are compiled as dynamic

update incorrect comment - getmntopts.c _is_ used by mount(8)

MNT_GETARGS support

Remove extra tab in really an empty line.

if RESCUEDIR is defined, set _PATH_RESCUE to that

Use ${NETBSDSRCDIR}/some/path instead of ${.CURDIR}/../../some/path

use getprogname()

fix nested extern

generated source should include stdlib.h and string.h for prototypes.

also install mount_ufs manpage links
Fixes bin/11412 by Andrew Brown.

Add a 'CLEANFILES+= checkname.c' to remove that file during a make clean.

mount(8) now includes code for all mount_*(8) but mount_portal(8) and
mount_mfs(8); the mount_*(8) are hardlinked to mount (appropriate mount routine
is called depending on program name) - this saves approx. 1.7MB of /sbin
space
mount.c: make all local symbols static

branches: 1.14.6;
resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

define WARNS?=1 for all of sbin/*

don't need getmntopts.c as part of mount(8)

Fix warnings.

remove FFS mounting code from mount(8); there are now _no_ 'internal' mount
types. when using mount(8) with '-a', do _NOT_ remount file systems
that have been mounted once already. (This cannot be 100% precisely
determined (thanks to mfs, union fs, and similar file systems which
don't use a 'real' mounted-from node), and changed options cannot be updated
with mount -a. however, options wouldn't be updated with the old mount -a
anyway, and this solves several annoyances.

convert to new RCS Id conventions; reduce my headache

specify man pages the new way.

Update from 4.4-Lite, with local changes.

there is no librpc any more

Add RCS identifiers.

fixed NFSCLIENT problem.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Fixed a_mask to actually use its argument

Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

branches: 1.8.2;
Remove clause 3 and 4 from TNF licenses

branches: 1.7.12; 1.7.14;
- removes unused ctype.h header
- correct comment in mount_nfs/getnfsargs.c: s/Nead/Need/

From: Slava Semushin <php-coder@altlinux.ru>

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

expose a_num.

Simplify a_uid and a_gid, and fix a_mask to actually use its argument
(fortunately it is always called with 'optarg')
Saves a few bytes from mount_msdos in rescue_tiny.

Add (unsigned char) cast to ctype function

fix copyright (damned copy&paste)

make the code includable into other programs:
* make static all symbols which do not need to be exported
* rename main() to mount_FOO()
* new main() now just calls mount_FOO(), main() is only compiled in if
MOUNT_NOMAIN is not defined
* a_gid(), a_uid() and a_mask() were put into ../mount/fattr.[ch], local
versions removed

Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

branches: 1.5.2;
Remove clause 3 and 4 from TNF licenses

branches: 1.4.20; 1.4.22;
expose a_num.

Simplify a_uid and a_gid, and fix a_mask to actually use its argument
(fortunately it is always called with 'optarg')
Saves a few bytes from mount_msdos in rescue_tiny.

Kill __P(), use ANSI function declarations.

make the code includable into other programs:
* make static all symbols which do not need to be exported
* rename main() to mount_FOO()
* new main() now just calls mount_FOO(), main() is only compiled in if
MOUNT_NOMAIN is not defined
* a_gid(), a_uid() and a_mask() were put into ../mount/fattr.[ch], local
versions removed

move getmntopts(3) to libutil, build and install also it's manpage
bump libutil minor
adjust individual mount_*/Makefile to use libutil getmntopts(3)

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

Whitespace nits

Drop arguments of .Os.

resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

convert to new RCS Id conventions; reduce my headache

branches: 1.1.1;
Update from 4.4-Lite, with local changes.

move getmntopts(3) to libutil, build and install also it's manpage
bump libutil minor
adjust individual mount_*/Makefile to use libutil getmntopts(3)

Since our gcc doesn't warn about NULL format strings anymore, we can
fix the incorrect err(1, "%s", "") et al.
Closes PR bin/7592 by cgd.

branches: 1.5.10;
resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

Fix warnings.

convert to new RCS Id conventions; reduce my headache

fix quotas

branches: 1.1.1;
Update from 4.4-Lite, with local changes.

move getmntopts(3) to libutil, build and install also it's manpage
bump libutil minor
adjust individual mount_*/Makefile to use libutil getmntopts(3)

PR/16688: Robert Elz: No way exists to use MNT_IGNORE (custom C code excepted)

MNT_GETARGS support

Enable and document the -o softdep option.

branches: 1.8.10;
Add support for a new flag: "nodevmtime". This sets the MNT_NODEVMTIME mount
flag.

Conditionalize the recognition of symbolic link permission by
per fs mount option `symperm'.

branches: 1.6.2;
resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

add support for noatime mount flag

recognize the 'nocoredump' mount option as a standard option (supported by
all file system types).

convert to new RCS Id conventions; reduce my headache

fix quotas

branches: 1.1.1;
Update from 4.4-Lite, with local changes.

PR/59669: rl / Hetzer: document posix1eacls

PR bin/59494

Cause "mount -v [spec] filesystem" to indicate only what was
mounted (and where), to obtain all the rest of the data that
used to be printed, -v must be given twice.

Implemented differently from the patch given in the PR from
greywolf, but the same idea.

While here, avoid the -d option, after indicating what mount
command would have been run, but wasn't, going on to attempt
to report statistics on the never mounted filesystem, and
instead getting those from the intended mount point.

Update the man page to match.

branches: 1.92.2;
new sentence, new line

mount(8): Mention the ability to refer to dk(4) wedges in the manpage

drop reference to smbfs, which was removed in 2020.

while here, s/file-system/file system/

Remove the warning about autoselecting nfs based on : or @.

Ok jmcneill@, mrg@.

branches: 1.88.4;
file systems that are used as what spools?

Sort SEE ALSO.

Merge autofs support from: Tomohiro Kusumi
XXX: Does not work yet

Remove Tn. Remove undefined .Sp. Use tagged list instead of hanging one
with .brs.

Clarify the program naming convention, and use consistently <fstype> instead
of XXX.

expand a little the explanation about discard

describe behaviour of combination of log+async (i.e. same as just log)

branches: 1.82.2;
stress more that discard is experimental, and explain why; also note that
it's not compatible with log

mount appeared in V1
http://minnie.tuhs.org/cgi-bin/utree.pl?file=V1/man/man1/mount.1
Replace contraction.
Bump date.

branches: 1.80.2;
Add reference to mount_chfs, bump date.

Add a reference to secmodel_extensions(9), to make it a little easier to
figure out how to enable mounts for unprivileged users. (Why don't we
just explain vfs.generic.usermount in the mount(8) man page?)

Also add another example of different ways in which the "special"
argument is interpreted by different file systems: mount_tmpfs(8)
ignores it.

Implement experimental support to pass notifications that a file
was deleted from the filesystem to the disk driver, commonly
known as "discard" or "trim".
fs/driver support is in ffs and ata wd for now.
This is what was posted here:
http://mail-index.netbsd.org/tech-kern/2012/02/28/msg012813.html
with minor cleanup, and the global switch replaced by a mount option.

- Improve wording;
- remove superfluous `.Pp' macro (fixes mandoc(1) warning).

From Bug Hunting.

branches: 1.76.2;
Bump date for previous.

Clarify use of the union option for mount, as proposed in PR 45919.
Text proposed by apb.

branches: 1.74.2;
Improve previous; thanks to tsutsui-san for reminding me of the proper
reference. PR 45166

Clarify UFS2 superblock format stuff. Bump date.

New sentence, new line; fix wording; bump date.

Add mount -o extattr option to enable extended attributs (corrently only
for UFS1).
Remove kernel option for EA backing store autocreation and do it by
default. Add a sysctl so that autocreated attriutr size can be modified.

branches: 1.70.4;
xref rump_smbfs

branches: 1.69.2;
Drop trailing whitespace.

Centralize documentation about exported sysctl variables in the relevant
secmodel's man-page.

Fix typo, and restore line that was lost in previous.

PR kern/26878 FFSv2 + softdep = livelock (no free ram)
PR kern/16942 panic with softdep and quotas
PR kern/19565 panic: softdep_write_inodeblock: indirect pointer #1 mismatch
PR kern/26274 softdep panic: allocdirect_merge: ...
PR kern/26374 Long delay before non-root users can write to softdep partitions
PR kern/28621 1.6.x "vp != NULL" panic in ffs_softdep.c:4653 while unmounting a softdep (+quota) filesystem
PR kern/29513 FFS+Softdep panic with unfsck-able file-corruption
PR kern/31544 The ffs softdep code appears to fail to write dirty bits to disk
PR kern/31981 stopping scsi disk can cause panic (softdep)
PR kern/32116 kernel panic in softdep (assertion failure)
PR kern/32532 softdep_trackbufs deadlock
PR kern/37191 softdep: locking against myself
PR kern/40474 Kernel panic after remounting raid root with softdep

Retire softdep, pass 2. As discussed and later formally announced on the
mailing lists.

branches: 1.65.2;
* Say that type "nfs" is inferred if the path contains a
":" or "@". (Amazingly, this was not documented before).
* Say that the "nfs" inference is deprecated. (It was deprecated in
revision 1.87 of mount.c, dated 2009-01-11.)
* Add "-t nfs" in an example, so as not to rely on the automatic
inference of nfs when the path contains a colon.
* Try to improve the description of how the arguments are interpreted.

New sentence, new line.

Support mount -o rump, which uses the rump server instead of using
the kernel service. E.g. "mount -t efs -o rump /dev/sgidev /mnt"
runs effectively "rump_efs /dev/sgidev /mnt".

branches: 1.62.2;
s/If/It/

Add missing word and bump date for previous.

Mention "-o log".

branches: 1.59.2;
Xref mount_tmpfs(8).

branches: 1.58.10; 1.58.12; 1.58.14;
Explain that nosuid,nodev is also needed for non-privileged mounts.

Mention potential reasons for EPERM.
Requested by Anon Ymous.

branches: 1.56.2; 1.56.4;
Revert the changes I introduced trying to solve tmpfs' NFS export problem.
Requested by yamt@

Add support to allow a file system to not permit being exported over NFS.

Approved by elad@ and wrstuden@

End sentence with a dot.

Add a DIAGNOSTICS section, specifically to address the "Operation not
supported by device" error. Reference options(4). Bump date.

convert "magiclinks" from a per-fs mount option to a system-wide sysctl.
as discussed on tech-kern quite some time ago.

Bump date, use "file system" consistently, no hyphen in "soft
dependencies", and sort options under -o.

Note that softdep and async options can't be used together. From
Dheeraj S in PR#32211, wording and formatting changes by me.

Crosslink mount(8) with mount_udf(8)

document port option; prompted by wiz@ :)

Implement expansion of special "magic" strings in symlinks into
system-specific values. Submitted by Chris Demetriou in Nov 1995 (!)
in PR kern/1781, modified only slighly by me.

This is enabled on a per-mount basis with the MNT_MAGICLINKS mount
flag. It can be enabled at mountroot() time by building the kernel
with the ROOTFS_MAGICLINKS option.

The following magic strings are supported by the implementation:

@machine value of MACHINE for the system
@machine_arch value of MACHINE_ARCH for the system
@hostname the system host name, as set with sethostname()
@domainname the system domain name, as set with setdomainname()
@kernel_ident the kernel config file name
@osrelease the releaes number of the OS
@ostype the name of the OS (always "NetBSD" for NetBSD)

Example usage:

mkdir /arch/i386/bin
mkdir /arch/sparc/bin
ln -s /arch/@machine_arch/bin /bin

branches: 1.46.2;
New sentence, new line, Xref fsck(8), use Dq.

document reload option, description shamelessly stolen off ffs_reload()
comment

Bump date for last.

Recognize 'from_mount' as the device, which is especially useful when
putting root on NFS. (From my PR, 20305.)

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

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

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

removable with only two es, from Adrian Mrva.

Sort SEE ALSO.

PR/16688: Robert Elz: No way exists to use MNT_IGNORE (custom C code excepted)

Use .Fl t instead of \fI-t\fP.

New sentence, new line. By Robert Elz with minimal fixes.

MNT_GETARGS support

Improve mount_mfs examples. Inspired by [misc/16193] from
Brian A. Seklecki <lavalamp@hurt.spiritual-machines.org>

Drop empty lines and an unnecessary .Pp, sort sections.

Prominently provide explanation of who is allowed to mount file
systems.

Drop arguments of .Os.

If device and mountpoint are given without -t, and without a : in the
device, check the disklabel for filesystem type. Fall back to ffs as ever.

Xr for mount_overlay(8)

Make sure .Dl has the line to be shown as argument (on same line).

Add example text to answer PR 10232.

branches: 1.26.2;
Enable and document the -o softdep option.

branches: 1.25.8;
xref mount_ntfs(8) in SEE ALSO

branches: 1.24.2;
Add support for a new flag: "nodevmtime". This sets the MNT_NODEVMTIME mount
flag.

add mount_filecore to "see also"-list

Word the comments about `async' a bit more strongly.

Document noasync and nosync.
Clarify the meaning of sync.
Run all the -o keywords through the Cm (keyword emphasis) macro.

fix bad .Xr references

Add a clarification how the missing data are looked up if either
mount source or target are omitted. Adresses PR 3240.
XXX This is not the complete truth. Things change with the -u
option. Should be fixed in "mount.c".

Conditionalize the recognition of symbolic link permission by
per fs mount option `symperm'.

branches: 1.17.2;
Fix .Nm usage.

resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

* getopt returns -1 not EOF
* use .Nm correctly

add support for MNT_NOATIME; from Lennart Augustsson in PR bin/3304.

document the 'nocoredump' mount option, which sets the MNT_NOCOREDUMP
mount flag, disabling core dumps on the file system.

fdesc file system does not allow files created in it. pr#941

update docs to match the code.

remove FFS mounting code from mount(8); there are now _no_ 'internal' mount
types. when using mount(8) with '-a', do _NOT_ remount file systems
that have been mounted once already. (This cannot be 100% precisely
determined (thanks to mfs, union fs, and similar file systems which
don't use a 'real' mounted-from node), and changed options cannot be updated
with mount -a. however, options wouldn't be updated with the old mount -a
anyway, and this solves several annoyances.

Remove xref to mount_lofs and add xref to mount_ados and mount_msdos.

convert to new RCS Id conventions; reduce my headache

Update from 4.4-Lite, with local changes.

update for new (actually old... *sigh*) mount flags

Add RCS indentifiers.

incorporate changes from 0-9-base to 0-9-ALPHA

branches: 1.3.2;
changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

PR bin/59494

Cause "mount -v [spec] filesystem" to indicate only what was
mounted (and where), to obtain all the rest of the data that
used to be printed, -v must be given twice.

Implemented differently from the patch given in the PR from
greywolf, but the same idea.

While here, avoid the -d option, after indicating what mount
command would have been run, but wasn't, going on to attempt
to report statistics on the never mounted filesystem, and
instead getting those from the intended mount point.

Update the man page to match.

branches: 1.107.4;
remove impossible test.

PR/56529: RVP: mount should try ffs when DIOCGWEDGEINFO returns an empty
dkw_ptype.

Add some { } around an if body that is followed by an "else {".

Remove the warning about autoselecting nfs based on : or @.

Ok jmcneill@, mrg@.

branches: 1.103.4;
Make command line arguments -r and -w actually override defaults from fstab,
previously -w would only cancel a preceeding -r.

No longer strip -o rw.

try to be more clear about what is not found.

change warning message

previously attempting to use mount -t ext2 like myself would result in
the warning "mount: mount_ext2 not found for /mnt", which (if you're me)
can be misunderstood as "/mnt is not an ext2 filesystem"...

change it to "mount: mount_ext2 not found"

branches: 1.100.2;
gcc 4.9 produces vfork clobberation warnings for "name" and "buflen".
Since there's no performance reason to use vfork here, just replace it
with fork and be done with warnings once and for all.

No code change: corrected spelling in comment, removed trailing white spaces.

use return instead of exit.

branches: 1.97.2;
don't print the NULL, but the error

return exit code from mount_XXXX again

use getfsspecname

Accept NAME=label for special so that we can wire down our wedges.

branches: 1.93.2;
Use __dead

In case we are union-mounting /, don't force MNT_UPDATE. This makes
the following work:

mount -t tmpfs -o union tmpfs /

(some caveats are implied, such as if you "mkdir /usr" you're
screwed, but then again you'll get there with "rm -rf /usr" even
without union -- we supply rope)

per discussion with zafer, use case for jibbed

Remove puffs| from vfstype before making comparison to determine if
fs is mounted.

Fixes slightly-miscategorized kern/37626.

If getargs and vfs has PUFFS_TYPEPREFIX, use mount_puffs to query args.

disable automatic mountd reloading for now because it makes the service
temporary unavailable. PR/41331 from FUKAUMI Naoki.

branches: 1.88.2;
Support mount -o rump, which uses the rump server instead of using
the kernel service. E.g. "mount -t efs -o rump /dev/sgidev /mnt"
runs effectively "rump_efs /dev/sgidev /mnt".

Warn that autoselecting nfs based on : or @ in the device path will
be removed in a future release.

branches: 1.86.2;
Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

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

branches: 1.84.12; 1.84.16;
cast sizeof() to int for printf widths.

kill MFSTYPENAME

sprinkle volatile.

since yamt prohibits any flag with MNT_GETARGS, don't set MNT_UPDATE for
/ when we do getargs.

Coverity CID 1688: Fix memory leak.

Check for allocation failures in malloc, calloc, realloc, asprintf, and
vasprintf and try to handle them.

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

Fix core dump when doing
mount fileserver:/somepath
by not passing a NULL pointer to getfsfile(3).
(Bug was introduced in rev 1.74 as part of fix for PR 28644.)

branches: 1.76.2;
Kill __P(), use ANSI function declarations.

Move local variable declaration to variable declaration section at
top of function, and at least away from after active code within
a block. Fixes build problem with gcc 2 (for vax).

PR #28644: Use realpath to make mount more forgiving of non-canonical
mount paths entered on the command line.

Try the DIOCGWEDGEINFO ioctl first. If that succeeds (i.e. the block
device is a wedge), use the partition type string from the dkwedge_info
structure to get the file system type.

Handle RESCUEDIR

Print fsid on verbose printing. It is useful when debugging nfs.

Use PRIu64 to printf uint64_t. Compiles again on sparc64.

Replace the statfs() family of system calls with statvfs().
Retain binary compatibility.

Recognize 'from_mount' as the device, which is especially useful when
putting root on NFS. (From my PR, 20305.)

realloc pedant

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

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

do not link all mount programs into single image; the total size difference
isn't really that huge now that these are compiled as dynamic

Cosmetic changes.

Properly terminate the output from mount -v.

make sure options is not NULL before we look in it. Thanks Charles.

MNT_GETARGS support

if _PATH_RESCUE is defined, try that before _PATH_SBIN & _PATH_USRSBIN

When attempting to guess the filesystem type from the disklabel, don't
consider a out-of-range partition letter at the end of the special
device node to be a fatal error; just return NULL and let the caller
fall back to FFS.

This fixes the "mount -u /kern/rootdev /" done by the script installer.

XXX this is still gross, and breaks things like
"mount /my/strange/dev/path/b /mnt". Perhaps it should stat the node
and use the minor number as an index instead?

Use __MNT_FLAGS from <sys/mount.h> instead of rolling our own list. Also
when -v is used, print even the silent flags.

Sync getopt() / man page with actual getopt options.

Add x_mount, which does not include any MOUNT_PROGS.
Installer does not require all vfs progs and all binaries
are crunched after all.

Factor out some function declarations into a header file.

When failed to read disklabel to deduce filesystem type,
- don't warn. It's just too verbose when we know there is
no disklabel and want to use the default filesystem type.
- close the file descriptor so that further mount success.

Cosmetic changes.

mount(8) now includes code for all mount_*(8) but mount_portal(8) and
mount_mfs(8); the mount_*(8) are hardlinked to mount (appropriate mount routine
is called depending on program name) - this saves approx. 1.7MB of /sbin
space
mount.c: make all local symbols static

When determining the filesystem type automatically, only try the raw device
for the disklabel if the given device fails with EBUSY. Also make disklabel
errors non fatal (just fall back to ffs as per pre-autofilesystem behaviour)
Based on further discussion with Launey Thomas <ljt@alum.mit.edu>

Format string cleanups by Bill Sommerfeld.

When opening a partition to automatically determine the filesystem type,
always use the raw partition in case it is already mounted and this is
an 'update' mount. Patch from Launey Thomas <ljt@alum.mit.edu>

If device and mountpoint are given without -t, and without a : in the
device, check the disklabel for filesystem type. Fall back to ffs as ever.

branches: 1.47.4;
If we didn't print an open paren, no need to print a closing paren
(by backing out part of previous commit).

Print softdep information. Also print sync+async writes in verbose mode.

branches: 1.45.2; 1.45.4; 1.45.8;
Add support for a new flag: "nodevmtime". This sets the MNT_NODEVMTIME mount
flag.

from Erik Bertelsen <erik@mediator.uni-c.dk>
{ put } { in } { lots } { of } { these } { to } { shut } { up } { egcs }

const poisoning.

Rewrite options and fstype handling for upgrade mounts.

Add MNT_SYMPERM to optnames[].

lint

Take out last change (wait for mount_mfs completion).
I changed it in mount_mfs instead.

Fix a problem with asynchronous processing of mfs mounts: Calls to
statfs() or getmountinfo() did not tell about the freshly mounted
ramdisk.
XXX Imo, "mount_mfs" should not return before the mount is done.
This is the only place where this can be done cleanly. But this would
require a substancial restructuring of "newfs". For now, wait in "mount"
until the filesystem appears.

Fix a misbehaviour introduced in rev. 1.34 - mountfs() returned in
"verbose" mode after printing the exec args.
This invalidates most of my comment in the previous commit. (Not all,
there are still bogosities with mount_mfs.)

When checking whether a filesystem is already mounted (for -a), get and
check all mounts (getmntinfo), not only the top one (statfs).
Otherwise we might miss lower level mounts on the same mountpoint.
XXX "mount" behaves differently with the "verbose" flag (-v) in some
cases, probably due to asynchronous calls to fs dependant mount programs.

Import modified mangle() from mount(8).
Also rewrite catopt() with a more sane interface.

PR/2974: VaX#n8: fix the rest of the problems that were not addressed by 4384.
- vfork exit -> _exit
- debug and verbose consistency with fsck.

PR/4384: Brian Grayson: Pass the real argv[0] to the exec'ed child.
Also while I am here, remove bogus XXX comment about mount_mfs.

branches: 1.32.2;
resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

* getopt returns -1 not EOF
* use .Nm correctly

When we mount -u and there is no vfstype specified, deduce the vfstype from
statfs.

Fix warnings.

More accurate failure message for unknown filesystem types, as
suggested in PR#1274.

NULL => 0 (Arne Juul; PR#3629)

add support for MNT_NOATIME; from Lennart Augustsson in PR bin/3304.

recognize MNT_NOCOREDUMP and print the 'nocoredump' flag for file
systems which have MNT_NOCOREDUMP set.

Recognize all mount flags exported from the kernel (but only print the
ones that we used to print). Note unrecognized flags(s) when printing
mount list.

Call mount_ffs instead of mount_ufs; PR #1379

fix a goof in my previous code, that i thought of while reading pr 749.
add a -A flag, which does the same thing as '-a', but doesn't check for
remounts.

remove FFS mounting code from mount(8); there are now _no_ 'internal' mount
types. when using mount(8) with '-a', do _NOT_ remount file systems
that have been mounted once already. (This cannot be 100% precisely
determined (thanks to mfs, union fs, and similar file systems which
don't use a 'real' mounted-from node), and changed options cannot be updated
with mount -a. however, options wouldn't be updated with the old mount -a
anyway, and this solves several annoyances.

Give an argument to the warning message for when realpath() fails.

specify precision, not minimum field width. pointed out by John Kohl

don't assume f_fstypename is larger than MFSNAMELEN or is nul-terminated

sys_siglist[] -> strsignal()

convert to new RCS Id conventions; reduce my headache

Change type list code to match umount(8).

compile against new headers

branches: 1.13.2;
Display type of mount; suggested by Thomas Eberhardt.

Update from 4.4-Lite, with local changes.

allow external mount types, even if LETS_GET_SMALL

oops, forgot one

fs type names, and much clean

update for new (actually old... *sigh*) mount flags

should bzero automatic sockaddr's before use

Changes to VFS readdir semantics
NFS changes for better cookie support
ISOFS changes for better Rockridge support and support for generation numbers

Recognize `noauto' (and `auto', just for symmetry) in fstab.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

remove FFS mounting code from mount(8); there are now _no_ 'internal' mount
types. when using mount(8) with '-a', do _NOT_ remount file systems
that have been mounted once already. (This cannot be 100% precisely
determined (thanks to mfs, union fs, and similar file systems which
don't use a 'real' mounted-from node), and changed options cannot be updated
with mount -a. however, options wouldn't be updated with the old mount -a
anyway, and this solves several annoyances.

convert to new RCS Id conventions; reduce my headache

branches: 1.1.1;
Update from 4.4-Lite, with local changes.

branches: 1.1.2; 1.1.4;
Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

pathadj() is required to succeed. If it cannot, simply issuing a
warning and continuing is not good enough. Change the warning to
an error (and hence immediate exit) instead, that's all we can do.

Problem pointed out by tlaronde@polynum.com

XXX - pullup -10 (others?)

Refactor remaining mount_* commands to use the common pathadj()
function for resolving paths.

Make pathadj() no longer warn about symlinks. Symlinks in /dev are
regularly used in several places like LVM . The warning was also
only visible when calling a mount_* command directly as mount(8)
itself would resolve the path witout warning before passing it to
a mount_* command.

"when in doubt, let the kernel decide"

branches: 1.1.2; 1.1.4; 1.1.20;
Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

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

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

convert to new RCS Id conventions; reduce my headache

Update from 4.4-Lite, with local changes.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

branches: 1.6.20; 1.6.24;
Don't abuse writing to const strings. We can pass const strings to this,
and we write to them. Allocate memory instead.

Kill __P(), use ANSI function declarations.

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

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

Factor out some function declarations into a header file.

resolve conflicts from lite-2 import, with two major modifications
that require all of /sbin/mount_* and /sbin/umount to be modified:
* incorporate vfslist.c, adding checkvfsname() and makevfslist()
* getmntopts() now takes 4 args (adding int *altflagp)

branches: 1.1.1;
Initial revision

Refactor mount utilities to provide a mount_fs_parseargs() routine.
Use this routine both in mount_fs and rump_fs to provide equivalent
command line parameters and therefore usage interchangeability.
While doing this, combine some common mountgoop to mountprog.h

branches: 1.4.20; 1.4.24;
Don't abuse writing to const strings. We can pass const strings to this,
and we write to them. Allocate memory instead.

un-__P

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

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

Factor out some function declarations into a header file.