MFC r310088, r310090, r310095

r310088:
Put the undocumented df feature of mounting filesystems from device
nodes
under an ifdef. Leave enabled.

Reviewed by: cem
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D8513

r310090:
Mount filesystems without executable permissions since they should never
be used.

Reviewed by: cem
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D8513

r310095:
Use nmount(2) rather than the obsolete mount(2).

Reviewed by: cem
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D8513

MFC r287236:

Use exit() instead of return in main(). The difference in practice
is subtle: C standard requires the language runtime to make return
of int from main() behave like calling exit(), and in FreeBSD we do:

exit(main(argc, argv, env))

In lib/csu/${ARCH}/crt1.c, so the real difference is using exit()
explicitly would use an additional stack frame.

Note however, if there is a on stack pointer is the last reference
of an allocated memory block, returning from the function would,
technically, result in a memory leak because we lost the last
reference to the memory block, and calling exit() from C runtime
could potentionally overwrite that stack frame that used to belong
to the main() function.

In practice, this is normally Okay because eventually the kernel
would tear down the whole address space that belongs to the process
in the _exit(2) system call, but the difference could confuse
compilers (which may want to do stack overflow checks) and static
analyzers.

Replacing return with exit() in main() allows compilers/static
analyzers to correctly omit or generate the right warnings when
they do not treat main() specifically. With the current version
of clang on FreeBSD/amd64, use of exit() would result in slightly
smaller code being generated and eliminated a false positive
warning of memory leak.

MFC r260782:

-h and -H options backwards in manual page.

Hoist all the mergeinfo up to the root in preparation for enforcing merges
to the root only. All MFC's were rerecorded to the root.

Going forward, if an MFC includes mergeinfo, it will need to be made to
the root and committed from the root. Merges with --ignore-ancestry
or diff | patch can go anywhere.

The mergeinfo in HEAD is in a bad state from years of neglect and manual
tampering and this was branched into 10.x. This confuses the coalescing
code and prevents it from doing its job.

Approved by: re (gjb, implicit)

MFC: r257288

Clean up a spurious "." in SEE ALSO.

Approved by: re (glebius)

Copy head (r256279) to stable/10 as part of the 10.0-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Complete revision 243104.

Don't appease clang static analyzer after all and roll back
the free(3) of mntbuf ... again. There's no point in doing
useless extra work when we're about to exit.

See also r240565.

Not reading file history: uqs

bin/df: Fix unitialized use in prtstat

While here:
- use NULL in the context of pointers
- use memset instead of bzero throughout the file
- free memory to appease clang static analyzer

Found by: Coverity Scan (the UNINIT one)

Don't time travel back and use a present date.

Noticed by: mckusick

- Wrap long line;
- Add -, to SYNOPSIS section.

MFC after: 2 weeks

Handle large negative block counts correctly.

MFC after: 2 weeks

Update man pages and clarify a number of options.
Rework block count calculations to work correctly with small "block" sizes.

MFC after: 14 days

Fix section number for pstat and swapinfo cross references.

MFC after: 3 days

Add , (comma) option to print sizes grouped and separated by thousands
using the non-monetary separator returned by localeconv(3), typically
a comma or period.

MFC after: 14 days

Revert 240527:
mntbuf can poit to memory allocated by getmntinfo(3) which can't be freed

PR: bin/171634
Approved by: cperciva (implicit)

Free memory before exiting in order to
help tools understand that we're not leaking it.

PR: bin/171634
Submitted by: Erik Cederstrand <erik@cederstrand.dk>
Approved by: cperciva
MFC after: 3 days

Bump date to today (since I modified it)

PR: bin/165321
Approved by: gjb
MFC after: 3 days

Direct users to swapinfo and pstat from df

PR: bin/165321
Submitted by: jhs
Approved by: bcr
MFC after: 3 days

Minor clarification

MFC after: 3 weeks

Simplify df(1) by factoring out most of the common code:

- In the argc == 0 case, just populate the mount list as before, but
do not calculate widths, update totals or print anything.

- In the argv > 0 case, collect information about the requested file
systems and store it in the mount list, but do not calculate
widths, update totals or print anything.

- In either case, once all the information has been collected,
iterate once through the mount list to calculate widths and totals,
then once more to print everything.

This also fixes two bugs: firstly, column widths were not calculated
correctly if more than one file system was specified on the command
line; and secondly, file systems with MNT_IGNORE were included in the
totals even if -a was not specified.

Noticed by: Paul Schenkeveld
MFC after: 3 weeks

Fix typos - remove duplicate "the".

PR: bin/154928
Submitted by: Eitan Adler <lists at eitanadler.com>
MFC after: 3 days

Staticify internal routines.

Make "human-readable" (-H/-h) output also "humanize" inode counts.
Base 10 is always used for the inode counts as I could not think of any
reason base 2 inode counts would be useful.

Minor mdoc markup fix to df(1) while here anyway.

MFC after: 3 weeks

Implement -T option as found in Linux df(1). When given, file system type
will be included in output.

Reviewed and tested by: keramida
MFC after: 2 weeks

Significantly reduce the memory leak as noted in BUGS section for
setenv(3) by tracking the size of the memory allocated instead of using
strlen() on the current value.

Convert all calls to POSIX from historic BSD API:
- unsetenv returns an int.
- putenv takes a char * instead of const char *.
- putenv no longer makes a copy of the input string.
- errno is set appropriately for POSIX. Exceptions involve bad environ
variable and internal initialization code. These both set errno to
EFAULT.

Several patches to base utilities to handle the POSIX changes from
Andrey Chernov's previous commit. A few I re-wrote to use setenv()
instead of putenv().

New regression module for tools/regression/environ to test these
functions. It also can be used to test the performance.

Bump __FreeBSD_version to 700050 due to API change.

PR: kern/99826
Approved by: wes
Approved by: re (kensmith)

Back out all POSIXified *env() changes.

Not because I admit they are technically wrong and not because of bug
reports (I receive nothing). But because I surprisingly meets so
strong opposition and resistance so lost any desire to continue that.

Anyone who interested in POSIX can dig out what changes and how
through cvs diffs.

Eliminate error with -W* strict flags and make putenv() calls conforming to
standard in the same way as f.e. gcc internal portable code does.

getblocksize expects pointer to long as a second argument, not
a pointer to u_long.

Fix a bug where the mutual exclusivity of the -l and -t options is not
recognized properly if -l is specified first.

PR: bin/105721
MFC after: 1 week

Fix markup.

Based on The Open Group Base Specifications Issue 6 IEEE Std 1003.1, our
current implementation of df(1) is does not properly format the output under
certain conditions. Right now -kP and -Pk are not the same thing. Further,
when we set the BLOCKSIZE environment variable, we use "1k" instead of "1024",
making the header display incorrectly.

To quote the specification:

"When both the -k and -P options are specified, the following header line
shall be written (in the POSIX locale):

"Filesystem 1024-blocks Used Available Capacity Mounted on\n"

- If -P has been specified, check to make sure that -k has not already been
specified, if so, simply break instead of clobbering the previous blocksize
- Use 1024 instead of 1k to make the header POSIX compliant

Reported by: Andriy Gapon
Discussed with: bde, ru
MFC after: 1 week

Prefer strlcpy to strncpy. In one case, this saves us from re-zeroing
data that's alreday 0. In another, it saves us from zeroing data that
will be overwritten again.

Add the new standard EXIT STATUS section where appropriate.
Sort standard sections in the (documented) preferred order.

/*- or .\"- or #- to begin license clauses.

Scheduled mdoc(7) sweep.

Currently if a mount point is not accessible by the calling user,
invalid information will be printed if the -t flag is specified.

$ df -t ufs
Filesystem 1K-blocks Used Avail Capacity Mounted on
/dev/ad0s1a 495726 139944 316124 31% /
/dev/ad0s1e 253678 6438 226946 3% /tmp
/dev/ad0s1f 56206340 13594248 38115586 26% /usr
/dev/ad0s1d 694126 19812 618784 3% /var
/dev/ad0s1d 694126 19812 618784 3% /var
$

Note that the mount point which is not accessible shows
up as the previous file system that was printed. The reason
for this is that df -t will call statfs(2) on the pathname
supplied by getfsstat(2).

This is done to refresh the file system statistics in the
event that a previous file system had a long delay in
providing its stats.

This change affects the df utility in the following ways:
o Teach df has to deal with statfs(2) failing. If statfs(2)
fails, fall back on the possibly stale stats provided by
the initial call to getfsstat(2).
o Print a warning that the fs stats could possibly be stale
o Modify the man page and document this new behavior
as a bug.

Approved by: bmilekic (mentor)
PR: 68165

Mechanically kill hard sentence breaks.

Convert fsbtoblk() from a macro to a function. The redundant
instances of 64-bit arithmetic were costing 775 bytes, and the
inlining offered no benefit. Moreover, ambiguity as to the argument
types led to the introduction of a bug (see rev 1.56).

Also, remove some casts that are now clearly redundant.

Inspired by: 67467

Use humanize_number(3) to format sizes into a human readable form.

Bump document date for the latest functional change.
Minor markup tweaks.

Fix some style issues in rev 1.58.
Use 64-bit integer math vs. mixed FP & integer.
Add -g to the usage().

Add 'g' to SYNOPSIS.

Fix printing of the "Mounted on" values for 'df -i'.
Fix spacing before "Mounted on" column in general.

Submitted by: bde

Add -c option simular to du(1).

PR: 19635
Submitted by: cyrille.lefevre@laposte.net

Remove clause 3 from the UCB licenses.

OK'ed by: imp, core

And a bandaid so that the output of "available space" is correct when
using -m and -g switches and "available space" is negative (i.e. when
the file system is already using the root-reserved minimum free space).

Obtained from: Stefan Farfeleder <stefan@fafoe.narf.at>
PR: bin/62536
Submitted by: Peter van Dijk <peter@dataloss.nl>
Approved by: grog (mentor), bde

Make NULL a (void*)0 whereever possible, and fix the warnings(-Werror)
that this provokes. "Wherever possible" means "In the kernel OR NOT
C++" (implying C).

There are places where (void *) pointers are not valid, such as for
function pointers, but in the special case of (void *)0, agreement
settles on it being OK.

Most of the fixes were NULL where an integer zero was needed; many
of the fixes were NULL where ascii <nul> ('\0') was needed, and a
few were just "other".

Tested on: i386 sparc64

Use int rather than size_t storage for printf field widths to avoid
many casts.

Reviewed by: bde

Fix format strings (intmax_t is %jd, not %qd)

Reviewed by: tjr

Update the statfs structure with 64-bit fields to allow
accurate reporting of multi-terabyte filesystem sizes.

You should build and boot a new kernel BEFORE doing a `make world'
as the new kernel will know about binaries using the old statfs
structure, but an old kernel will not know about the new system
calls that support the new statfs structure. Running an old kernel
after a `make world' will cause programs such as `df' that do a
statfs system call to fail with a bad system call.

Reviewed by: Bruce Evans <bde@zeta.org.au>
Reviewed by: Tim Robbins <tjr@freebsd.org>
Reviewed by: Julian Elischer <julian@elischer.org>
Reviewed by: the hoards of <arch@freebsd.org>
Sponsored by: DARPA & NAI Labs.

Be a little bit more correct WRT counting numbers vs. integer numbers.

Get this area compiling with the highest WARNS= that it works with.
Obsolete WFORMAT= junk also removed where possible.

OK'ed by: obrien
Tested on: sparc64, alpha, i386

Catch up with revs 1.49-1.50 of df.c: don't mention -t in the BUGS section
since it has been fixed.

Fixed exit code in previous commit. "var++" to set a flag to nonzero
is a style bug at best. When the variable isn't a flag, it potentially
overflows after a large number of settings. Here the number of settings
is limited by ARG_MAX, but the variable is the exit code so it became
bogus after the second setting and effectively overflowed to 0 after
approx. 128 settings.

Fixed some style bugs involving comments in and near previous commit.

Clarification of previous commit message: df -t didn't give undefined
behaviour, and the behaviour used to conform perfectly with the man
page, since the buggy behaviour is documented in the BUGS section. -t
just worked when no files or file systems were specified, and was just
ignored if a file or file system was specified.

From the df man page:
-t Only print out statistics for filesystems of the specified types.

Make the behavior of df(1) conform to its man page (behavior is otherwise
undefined).

Submitted by: Rob Braun <bbraun@apple.com>
Obtained from: Apple

df(1) and ls(1) print units in 'four or fewer' not 'three or less'.

PR: 35523
Submitted by: Tomas Svensson <tsn@gbdev.net>

WARNS cleaning for the Alpha. On alpha, size_t is a long, and it
solicits a warning when used for the '*' in printf("%*d"). Cast
to u_int for universal use.

Fix a shedload of warnings, some memory leaks and clean up WARNS
and lint. This is now WARNS=9, std=c99 clean on i386.

Quiet warnings about copyright[].

Back out rev 1.44; getbsize(3)'s original interface has been restored.

Approved by: markm

Fix for changed getbsize arg type.

Fix 'SYNOPSIS' and 'usage'

s/filesystem/file system/ as discussed on -developers

- Introduce a new struct xvfsconf, the userland version of struct vfsconf.
- Make getvfsbyname() take a struct xvfsconf *.
- Convert several consumers of getvfsbyname() to use struct xvfsconf.
- Correct the getvfsbyname.3 manpage.
- Create a new vfs.conflist sysctl to dump all the struct xvfsconf in the
kernel, and rewrite getvfsbyname() to use this instead of the weird
existing API.
- Convert some {set,get,end}vfsent() consumers to use the new vfs.conflist
sysctl.
- Convert a vfsload() call in nfsiod.c to kldload() and remove the useless
vfsisloadable() and endvfsent() calls.
- Add a warning printf() in vfs_sysctl() to tell people they are using
an old userland.

After these changes, it's possible to modify struct vfsconf without
breaking the binary compatibility. Please note that these changes don't
break this compatibility either.

When bp will have updated mount_smbfs(8) with the patch I sent him, there
will be no more consumers of the {set,get,end}vfsent(), vfsisloadable()
and vfsload() API, and I will promptly delete it.

Consistently use __FBSDID

Unconditionally update the maximum field width statistics when we
refetch the filesystem information in MNT_WAIT mode. This avoids
incorrect column alignment that sometimes occurs with NFS filesystems.

Submitted by: Ian <freebsd@damnhippie.dyndns.org>

Remove the private code for reading UFS superblocks, this does not belong
in df(1) when we have multiple filesystem types, and the complications of
handling UFS2 pushes this over the edge.

Use the .../mount/extern.h to get prototypes of the functions we
borrow from there. Constify things to match. (why aren't these
functions in a lib anyway ?)

Make everything static and set WARNS?=5.

The way the "df diskdevice" thing works for unmounted diskdevices
is not very general.

Sponsored by: DARPA & NAI Labs.

Use `The .Nm utility'

Make the columns in the output of df(1) line up, even for very large
filesystems. We now keep track of the maximum width required for
every variable-width field instead of just the first one.

PR: bin/15510
MFC after: 1 week

Warnings fixes inspired by lint, a commercial lint and WARNS=4.

o __P has been reoved
o Old-style K&R declarations have been converted to new C89 style
o register has been removed
o prototype for main() has been removed (gcc3 makes it an error)
o int main(int argc, char *argv[]) is the preferred main definition.
o Attempt to not break style(9) conformance for declarations more than
they already are.

Approved by: arch@, new style(9)

Make ``df -l'' work when no network filesystems are loaded.

PR: bin/32397

Do not dot terminate errx() strings

Default to WARNS=2. Binary builds that cannot handle this must explicitly
set WARNS=0.

Reviewed by: mike

Remove a misplaced space.

Fix style bugs introduced by rev 1.28. No functional changes.

mdoc(7) police: removed HISTORY info from the .Os call.

WARNS= -> WARNS?=

Submitted by: Mike Barcroft <mike@q9media.com>

Add parens to get the cast that was meant in previous commit.
While we're at it, this file seems to prefer `unsigned int'
over `u_int', so go with that.

Wrong. The size of size_t is *not* the same as the size of an integer.

MFC after: 2 weeks

Added the -l option to df, so to be compatable with other unicies.

PR: bin/27240
Reviewed by: GAWollman
MFC after: 2 weeks

Remove vestiges of MFS.

BDECFLAGS cleanup (modulo long long issues). Add WARNS. Tested on alpha.

Reviewed by: the great man himself (except alpha cleanups)

Fix operation of df on unmounted filesystems, and add the ability to run df
on unmounted non-UFS filesystem using '-t'

Submitted by: bde

GC some dead code relating to running df on unmounted block devices,
and remove the setgid operator bit from the installed binary: if you want
to view free disk space on an unmounted device, you should have read
permissions to access it.

Reviewed by: phk

mdoc(7) police: split punctuation characters + misc fixes.

mdoc(7) police: use the new features of the Nm macro.

Add -g for gigabyte sizes.

Approved by: jkh

Fix miscellaneous mdoc macro argument limit infringements.

PR: 18465
Reported by: Kazu TAKAMUNE <takamune@avrl.mei.co.jp>

Fixed LDADD. Using ${LIBM} instead of -lm gave the wrong libm in most
cases and broke the world in some cases.

Fixed some style bugs (the usual ones for DPADD and LDADD, misplacement
of DPADD and LDADD, and misplacement of $FreeBSD$).

add human readable output (-h and -H)

Obtained from: parts of human readable code from OpenBSD
Reviewed by: obrien

add POSIX, byte and megabyte block size ouput flags

PR: 13579 (POSIX flag)
Submitted by: Mike Meyer <mwm@phone.net>

Most modern OSs have the ability to flag certain mounts as ones to
be ignored by default by the df(1) program. This is used mostly to
avoid stat()-ing entries that do not represent "real" disk mount
points (such as those made by an automounter such as amd.) It is
also useful not to have to stat() these entries because it takes
longer to report them that for other file systems, being that these
mount points are served by a user-level file server and resulting in
several context switches. Worse, if the automounter is down
unexpectedly, a causal df(1) will hang in an interruptible way.

PR: kern/9764
Submitted by: Erez Zadok <ezk@cs.columbia.edu>

$Id$ -> $FreeBSD$

Don't suggest sysctl(8) as a means for discovering what filesystem types
are available; suggest lsvfs(1) instead.

Reported by: Neil Blakey-Milner <nbm@rucus.ru.ac.za>

Removed occurrences of consecutive repeated words (such as "the the").

Cleaning out old stuff from one of my source trees:
use mkdtemp() rather than mktemp() and fix a trivial memory leak.

.Sh ENVIRONMENT VARIABLES -> .Sh ENVIRONMENT

PR: 6599
Reviewed by: phk
Submitted by: Josh Gilliam <josh@quick.net>

Correct use of .Nm.

Removed definition of _NEW_VFSCONF. The new vfsconf interface is now
the default.

Style police: keep a variable list in alphabetical order, and add
pointless paren's to return statements.

Nitpicked by: bde :)

Make df(1) exit with non-success status if something went wrong.

PR: bin/4415

environmental -> environment.

Add Bruce's _NEW_VFSCONF flag to make this compile cleanly and
sccsid -> rcsid.

compare return value from getopt against -1 rather than EOF, per the final
posix standard on the topic.

Merged from Lite2. The -t option is now handled better. The only visible
change should be that the usage message now actually matches the man page.

Merged from Lite2. The man page no longer attempts to list all the
vfs types.

Fixed ordering of FreeBSD `k' option in synopsis.

Merge from Lite2. Straightforward changes.

This is a funny one: df for a device that was not mounted used to fail
for root only.

Revert $FreeBSD$ to $Id$

Include <ufs/ufs/ufsmount.h>.

Make the long-awaited change from $Id$ to $FreeBSD$

This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.

Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.

Sort cross references.

Merge Lite2 mods, cleanup man page, and -Wall cleaning.

The manual page of df called cd9660 file systems by the name
isofs while the df command itself used the name iso9660fs or
cdfs. Both of these were inconsistent with the name cd9660 which
is used by the mount command. I modified df to recognize all of
the names cd9660, cdfs, isofs, and iso9660fs, and take them all
to refer to the same thing. Naturally I added a note of this
behaviour in the manual page too.

Submitted-By: Jukka Ukkonen <jau@jau.csc.fi>

[HISTORY] command appeared in Version 1 AT&T UNIX
Obtained from: A Quarter Century of UNIX, Peter H. Salus, page 41

Partially fixed negative and truncated "Avail" counts in df output.
This fixes PR943.

ffs/ffs_vfsops.c:
ffs_statfs() multiplied by (100 - minfree) as part of calculating the
minfree percentage (complemented in 100%), so with the standard minfree
of 8, it was broken for file systems of size >= 1TB/92 = 11GB. Use the
standard freespace() macro instead. This also fixes a rounding bug (the
"Avail" count was sometimes 1 too small).

ffs/* (not fixed):
The freespace() macro multiplies by minfree, so with the standard
minfree of 8, it is broken for file systems of size >= 1TB/8 = 128GB.
This bug is more serious since it affects block allocation.

ffs/ffs_alloc.c (not fixed):
Ordinary users are sometimes allowed to allocate 1 (partial) block
too many so that the "Avail" count goes negative. E.g., if there is
1 fragment available and the file is fairly large, one more full
block is allocated.

df/df.c:
ufs_df() used/uses essentially the same code as ffs_statfs(), so it
had/has the same bugs.

ufs_df() gratuitously replaced "Avail" counts of < 0 by 0, so it
gave different results for non-mounted file systems in this case.

Remove trailing whitespace.

Reviewed by: phk

You will find enclosed some changes to make gcc -Wall more happy in
/usr/src/bin. Note that some patches are still needed in that directory.

I (Joerg) finished most of Philippe's cleanup. /bin/sh will still
need *allot* of work, however.

Submitted by: charnier@lirmm.fr (Philippe Charnier)

Added $Id$

The last commit was bogus...the putenv doesn't affect the parent process,
so the BLOCKSIZE doesn't need to be preserved.
Also initialized the flags variables, and used 1k instead of 1024 for
BLOCKSIZE.

Implemented the -k flag more properly...don't destroy the user's setting
of the BLOCKSIZE environment variable.

Gimme back my `-k' option!
Also document `-t' option in the usage message.

This commit was generated by cvs2svn to compensate for changes in r1556,
which included commits to RCS files with non-trunk default branches.