docs: update TODO

Signed-off-by: Karel Zak <kzak@redhat.com>

blkid: small fix for safe_print()

This small fix makes it possible to print the first unprintable character
with code 128 from ASCII table.

Signed-off-by: Sergey Gusarov <laborer2008@gmail.com>

umount: use UMOUNT_NOFOLLOW for non-root users

The UMOUNT_NOFOLLOW umount2() flag is supported since kernel 2.6.34.

Signed-off-by: Karel Zak <kzak@redhat.com>

blkid: add -d option to print non-printable chars

Reported-by: laborer2008 laborer <laborer2008@gmail.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

partx: add fallback for openat() to be usable on old systems

Reported-by: "Gabor Z. Papp" <gzp@papp.hu>
Signed-off-by: Karel Zak <kzak@redhat.com>

include: [c.h] add definition wrappers for old libc versions

Signed-off-by: Davidlohr Bueso <dave@gnu.org>
Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: fix leak in sample program

Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: create a default lock for mtab update

Now it's necessary to explicitly create a lock for mnt_update_table().
It seems better to create the lock in mnt_update_table() by default to
skip this extra step.

Reported-by: Jan Engelhardt <jengelh@medozas.de>
Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: do not ignore datarootdir

Fix to the ./configure warning which has been printed since
autotools version 2.59c (released at April 2006).

WARNING:  'po/Makefile.in.in' seems to ignore the --datarootdir setting

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>

docs: update TODO

Signed-off-by: Karel Zak <kzak@redhat.com>

lib: [fsprobe.c] fix blkid_evaluate_spec() call

lib/fsprobe.c:56:2: warning: passing argument 2 of ‘blkid_evaluate_spec’ from incompatible pointer type
shlibs/blkid/src/blkid.h:189:14: note: expected ‘struct blkid_struct_cache **’ but argument is of type ‘blkid_cache’

Signed-off-by: Karel Zak <kzak@redhat.com>

ddate: use ARRAY_SIZE

Signed-off-by: Karel Zak <kzak@redhat.com>

ddate: remove non-ascii chars from ddate

Signed-off-by: Karel Zak <kzak@redhat.com>

tailf: support --lines 0

Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: rename mount option "quiet" to "silent"

 See commit 2ac1890aa0d8ce404528587134fcabe0e2834963.

Signed-off-by: Karel Zak <kzak@redhat.com>

fdisk: check index before access to array [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: add blkid_evaluate_spec()

The function blkid_evaluate_tag() is useful for tags only (e.g.
LABEL=foo). But we also need to address devices by tags OR paths in
many utils.

The function blkid_evaluate_spec() support this functionality without
extra care about the way how device is addressed. The tags as well as
paths are converted to the standardized device path.

Signed-off-by: Karel Zak <kzak@redhat.com>

lib: [fsprobe.c] remove obsolete <blkid/blkid.h>

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: don't use HAVE_LIBBLKID_INTERNAL macro

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: clean up partx Makefile

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: don't support external (e2fsprogs) libblkid

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: include <uuid.h> rather than <uuid/uuid.h>

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: don't support external (e2fsprogs) libuuid

Signed-off-by: Karel Zak <kzak@redhat.com>

lib: [fsprobe.c] use internal libblkid only

fsck, mount and swapon will be compilable with libblkid from
util-linux package. The old libblkid from e2fsprogs will not be
supported.

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: use blkid_probe_get_buffer() more carefully

Signed-off-by: Karel Zak <kzak@redhat.com>

Revert "umount: call /sbin/umount.<uhelper> for root user too"

This reverts commit c56226697342ddd663492f77768e7a7cb8e579a1.

umount: call /sbin/umount.<uhelper> for root user too

Originally the "uhelper=" was ignored for root, now the umount command
calls /sbin/umount.<uhelper> always.

This change is necessary for things like pam_mount where a device has
to be uninitialized by uhelper rather than directly umounted by
umount(2).

Note that all uhelpers has to be linked with libmount or execute

umount -fake -i /mountpoint

to remove mountpoint entry from /etc/mtab or /dev/.mount/utab.

Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: cleanup *_do_[u]mount() return codes and docs

Signed-off-by: Karel Zak <kzak@redhat.com>

mount: rename "quiet" to "silent" (MS_SILENT)

The quiet mount option was in collision with vfat and hfs mount
options. The option was also undocumented -- so it's probably safe to
rename the option than add some ugly hacks or exceptions for vfat/hfs
to the mount(8) code.

Reported-by: https://bugzilla.redhat.com/show_bug.cgi?id=622089
Signed-off-by: Karel Zak <kzak@redhat.com>

tests: update GPT test

Signed-off-by: Karel Zak <kzak@redhat.com>

blkid: fix double free [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

lib: [tt.c] use mbs_truncate() from mbsalign.c

Signed-off-by: Karel Zak <kzak@redhat.com>

lib: [tt.c] dereferencing data before a null check [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

tailf: fix open() return value check [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

fsck.minix: use lib/ismounted.c

Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: keep code more readable for analyzers [coverity scan]

This is not a bug ('ent' is always non-NULL is 'm' is non-NULL), but
let's keep static analyzes and humans who read the code happy.

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: fix typo in *_to_cpu() usage in raid detection [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

lsblk: fix strtoul() usage [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: add PART_ENTRY_* to docs

Signed-off-by: Karel Zak <kzak@redhat.com>

findmnt: check mnt_tab_next_fs() return code [coverity scan]

Signed-off-by: Karel Zak <kzak@redhat.com>

hwclock: build on non-Linux

Thanks to the direct ISA method and by disabling the RTC get/set epoch
functionality, hwclock can work fine on non-Linux systems which provide
ioperm or iopl.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>

cfdisk: add home and end keys movements in partition list

Signed-off-by: Francesco Cosoleto <cosoleto@gmail.com>

cfdisk: move keys related to the cursor of the partition list outside menuselect()

Signed-off-by: Francesco Cosoleto <cosoleto@gmail.com>

cfdisk: move "No more partition" warning to draw_cursor()

Signed-off-by: Francesco Cosoleto <cosoleto@gmail.com>

cfdisk: use keypad() in menu selection function

This makes the source code looks cleaner, and works a bit better than
the previous way with hardcoded escape sequences.

Signed-off-by: Francesco Cosoleto <cosoleto@gmail.com>

lsblk: use ssize_t for readlink() return code

Signed-off-by: Karel Zak <kzak@redhat.com>

lsblk: add TYPE column to lsblk output

[kzak@redhat.com: - fix leaks
                  - use lowercase for TYPEs]

Signed-off-by: Milan Broz <mbroz@redhat.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

tailf: fix memmove() usage

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: print helpful error when pkg.m4 is missing

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

build-sys: fix unportable Makefile.am assignment

Makefile.am:36: warning: `:='-style assignments are not portable

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

build-sys: Enable automake's -Wno-portability option.

The util-linux depends on GNU make.

shlibs/blkid/src/Makefile.am:58: warning: `%'-style pattern rules are a GNU make extension
shlibs/mount/src/Makefile.am:62: warning: `%'-style pattern rules are a GNU make extension

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

build-sys: introduce AM_PROG_CC_C_O macro

Fix to warning which will appear when using fairly new automake
(automake commit id fd22dca).

disk-utils/Makefile.am:31: warning: compiling `mkswap.c' with per-target flags requires `AM_PROG_CC_C_O' in `configure.ac'

Reference documentation
http://www.gnu.org/software/hello/manual/automake/Public-Macros.html#index-AM_005fPROG_005fCC_005fC_005fO-296

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

build-sys: use AC_LANG_SOURCE to suppress warnings

Messages bellow be supressed by using AC_LANG_SOURCE macro around
code test blocks.

m4/tls.m4:13: AX_CHECK_TLS is expanded from...
configure.ac:335: the top level
configure.ac:335: warning: AC_LANG_CONFTEST: no AC_LANG_SOURCE call detected in body
../../lib/autoconf/lang.m4:193: AC_LANG_CONFTEST is expanded from...
../../lib/autoconf/general.m4:2660: _AC_LINK_IFELSE is expanded from...
../../lib/autoconf/general.m4:2677: AC_LINK_IFELSE is expanded from...
../../lib/m4sugar/m4sh.m4:605: AS_IF is expanded from...
../../lib/autoconf/general.m4:2728: _AC_RUN_IFELSE is expanded from...
../../lib/m4sugar/m4sh.m4:605: AS_IF is expanded from...
../../lib/autoconf/general.m4:2747: AC_RUN_IFELSE is expanded from...
../../lib/m4sugar/m4sh.m4:605: AS_IF is expanded from...
../../lib/autoconf/general.m4:2030: AC_CACHE_VAL is expanded from...
../../lib/autoconf/general.m4:2051: AC_CACHE_CHECK is expanded from...
m4/tls.m4:13: AX_CHECK_TLS is expanded from...
configure.ac:335: the top level

Reference to documentation;
http://www.flameeyes.eu/autotools-mythbuster/forwardporting/autoconf.html#id1168723648533

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

rtcwake: do not duplicate argument strings

This is not necessary and prevents memory leaks.

Reported-by: Marek Otahal <markotahal@gmail.com>
Signed-off-by: Davidlohr Bueso <dave@gnu.org>

blkid: don't ignore -i

Signed-off-by: Karel Zak <kzak@redhat.com>

blkid: uniformize the synopsis, remove "the the"

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>

textual: tweak several manpages, mainly the blkid one

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>

raw: remove constants from message strings; undo some abbrevs

Having a constant embedded within a message string prevents gettext from
picking up the rest of the string.  Also, translators may wish to change
word order; using a constant as a prefix or postfix makes this impossible.

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>

uuidd: add "-q" to synopsis, order options alphabetically

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>

include: [c.h]: remove strings.h include dublicate

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

include: [c.h]: typeof to __typeof__ fix

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

docs: update TODO

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

tailf: tailf mention new help & version options in man page

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

tailf: use long options

This was TODO item from commit 947a7c9c. The patch also
introduces version and help switches.

[kzak@redhat.com: rewrite old_style_option()]

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>

logger: mention long option in logger man page

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

logger: use libc error printing facilities

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>

logger: support long options

Use getopt_long and usage output changed to match long options.
This patch will also scrutiny argument of formerly undocumented
-P option.

[kzak@redhat.com: - include c.h]

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>

logger: fix variable type compiler warning

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

fsck: fix clang compiler warning

fsck.c:874:22: warning: format string is not a string literal
      (potentially insecure) [-Wformat-security]
errx(EXIT_USAGE, _(fs_type_syntax_error));
 ^~~~~~~~~~~~~~~~~~~~~~~

The issue was introduced in my commit 0a09eb4e, sorry.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>

man page syntax fixes

Signed-off-by: Ville Skyttä <ville.skytta@iki.fi>

mount: update mtab on "-f" and compiled with libmount

Reported-by: silvan minghetti <bu1137@gmail.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

umount: segfaults with inconsistent entry in /etc/fstab

Addresses: https://bugs.launchpad.net/ubuntu/+source/util-linux/+bug/726283
Reported-by: Greg Brockman <gdb@gregbrockman.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

docs: update TODO

Signed-off-by: Karel Zak <kzak@redhat.com>

move struct option to .rodata

It does not make sense to have writable large arrays of "struct
option" on the stack.

Signed-off-by: Karel Zak <kzak@redhat.com>

docs: update TODO file

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: add term-utils/.gitignore

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: move wall to term-utils/

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: move mesg to term-utils/

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: move reset to term-utils/

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: move script and scriptreplay to term-utils/

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: move setterm to term-utils/

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: move agetty to term-utils

Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: add term-utils/

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: fix gcc warnings

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: fix EFI GPT uuid byte order

Intel uses little-endians for UUID, the rest of the sane world uses 16
byte big-endian array...

Reported-by: Andrew Lutomirski <luto@mit.edu>
Signed-off-by: Karel Zak <kzak@redhat.com>

kill: add -q sigval to use sigqueue(2)

Signed-off-by: Karel Zak <kzak@redhat.com>

kill: translate "-l <num>" to RT<n>

Signed-off-by: Karel Zak <kzak@redhat.com>

kill: add support for real-time signals

Newly supported syntax for RT signals:

 RT<n>   = SIGRTMIN + n
 RTMIN+<n> = SIGRTMIN + n
 RTMAX-<n> = SIGRTMAX - n

the final signal number has to be in interval [SIGRTMIN,SIGRTMAX].

Note that the "SIG" prefix is also supported for RT signals of course.

Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=676297
Signed-off-by: Karel Zak <kzak@redhat.com>

rtcwake: test for available suspend modes

Signed-off-by: Lawrence Rust <lawrence@softsystem.co.uk>

dmesg.c: cleanups -- use err(), xalloc()

Signed-off-by: Marek Polacek <mpolacek@redhat.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

docs: update TODO

Signed-off-by: Karel Zak <kzak@redhat.com>

libblkid: try to detect if PT is newer than LVM

LVM (pvcreate) wipes the begin of the device. If there is a PT in the
wiped area then LVM signature is obsolete.

 # pvcreate /dev/sdb
 # fdisk /dev/sdb

old version:
 # blkid -p -o udev /dev/sdb
 ID_FS_TYPE=LVM2_member

new version:
 # blkid -p -o udev /dev/sdb
 ID_PART_TABLE_TYPE=dos

Reported-by: Matej Cepl <mcepl@redhat.com>
Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=679799
Signed-off-by: Karel Zak <kzak@redhat.com>

docs: update TODO

Signed-off-by: Karel Zak <kzak@redhat.com>

tests: update libmount tests (add optstr output)

Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: better "user" evaluation

Signed-off-by: Karel Zak <kzak@redhat.com>

libmount: don't export functions for vfs/fs/userspace mount options

Signed-off-by: Karel Zak <kzak@redhat.com>

ipcs: fix typo

Reported-by: Francesco Cosoleto <cosoleto@gmail.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

docs: update TODO file

Signed-off-by: Karel Zak <kzak@redhat.com>

lsblk: bitwise or TT_FL_ASCII with tt_flags instead of setting

Signed-off-by: Dave Reisner <d@falconindy.com>

hwclock: make RTC default to UTC time

If /etc/adjtime doesn't specify UTC or LOCAL, rtcwake defaults to UTC
and hwclock defaults to LOCAL.

Switch hwclock to meet the behaviour of rtcwake (default=UTC), also
matching the kernel's CONFIG_RTC_HCTOSYS behaviour.

The user impact of this change should be minimal, as anyone who has run
"hwclock --systohc" before will have their UTC/LOCAL choice already
recorded in /etc/adjtime.

Signed-off-by: Daniel Drake <dsd@laptop.org>