login: remove obsolete and confusing comment

Reported-by: Jan Pazdziora <jpazdziora@redhat.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

mkswap: fix holes detection (infinite loop and/or stack-buffer-underflow)

Reported-by: Brian Lane <bcl@redhat.com>
Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=1971877
Fix: https://github.com/karelzak/util-linux/issues/1348
Signed-off-by: Karel Zak <kzak@redhat.com>

lib/loopdev: perform retry on EAGAIN

Suggested-by: Ingo Franzki <ifranzki@linux.ibm.com>
Signed-off-by: Karel Zak <kzak@redhat.com>

mount.8.adoc: document SELinux use of nosuid mount flag

Using mount flag `nosuid` also affects SELinux domain transitions but
this has not been documented well.

Signed-off-by: Topi Miettinen <toiwoton@gmail.com>

dmesg: fix indentation in man page

Double-semicolon at the end of the option line results in description
being indented extra tab to the right. Replacing them with double-colons
allows all options to be displayed at the same indent level.

lib/path: fix possible leak when use ul_path_read_string() [coverity scan]

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

hwclock: close adjtime on write error [coverity scan]

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

newgrp: fix memory leak [coverity scan]

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

dmesg: fix possible memory leak [coverity scan]

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

findmnt: (verify) fix memory leak [asan]

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

findmnt: (verify) fix cache related memory leaks on --nocanonicalize [coverity scan]

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

libblkid: Add hyphens to UUID string representation in Stratis superblock parsing

[kzak@redhat.com: - small code cleanup]

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

hexdump: correctly display signed single byte integers

When using the format string '/1 "%d"', the byte did not display as a
signed integer as expected, it was interpreted as unsigned.

lsblk: fix formatting in -e option

uclampset: Fix left over optind++

The code was changed to use ':' in getopt_long() but these were left
over by mistake causing weird random errors when using these options
depending on the order they were fed.

Signed-off-by: Qais Yousef <qais.yousef@arm.com>

rfkill: Set scols table name to make the json output valid

[kzak@redhat.com: - s/rfkill/rfkilldevices/]

Fixes: https://github.com/karelzak/util-linux/issues/1339
Signed-off-by: Nicolai Dagestad <nicolai@dagestad.fr>
Signed-off-by: Karel Zak <kzak@redhat.com>

libsmartcols: fix bare array on JSON output

The bare array ({[...]}) for top level list of entries is invalid. It
seems better to print empty string than nothing. This is workaround
for broken utils, better is to define a table name.

Addresses: https://github.com/karelzak/util-linux/issues/1339
Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: make re-use of generated man-pages more robust

We need to ignore man-pages at all if not available (not generated
yet), but asciidoctor is not installed.

Fix: https://github.com/karelzak/util-linux/issues/1334
Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: add generated man-pages to distribution tarball

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

lscpu: fix build on powerpc

fails with error: label at end of compound statement
and fix typo as bonus.

Signed-off-by: Georgy Yakovlev <gyakovlev@gentoo.org>

lsblk: use ID_MODEL_ENC is possible

Fix: https://github.com/karelzak/util-linux/issues/1098
Signed-off-by: Karel Zak <kzak@redhat.com>

include/strutils: fix heap-buffer-overflow in normalize_whitespace()

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

build-sys: fix {release-version} man pages

* define {release-version} as $VERSION

* do not use {docdir} in adoc; it's built-in ascidoctor variable
  https://docs.asciidoctor.org/asciidoc/latest/attributes/document-attributes-reference/#intrinsic-attributes

Addresses: https://github.com/karelzak/util-linux/issues/1327
Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: install hardlink bash-completion

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

docs: fix typo in v2.37-ReleaseNotes

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

docs: add uclampset to AUTHORS file

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

build-sys: release++ (v2.37)

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

docs: update v2.37-ReleaseNotes

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

docs: update AUTHORS file

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

po: merge changes

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

po: add ko.po (from translationproject.org)

po: update uk.po (from translationproject.org)

po: update pt.po (from translationproject.org)

po: update pl.po (from translationproject.org)

po: update hr.po (from translationproject.org)

po: update fr.po (from translationproject.org)

po: update de.po (from translationproject.org)

po: update cs.po (from translationproject.org)

libblkid: fix docs

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

docs: update copyright years

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

build-sys: remove with-cryptsetup from tools/config-gen.d/all.conf

Unfortunately libtools is not smart enough to link libblkid
dynamically if we link in-tree static libmount.a for libmount tests.
In this case libtools always uses also libblkid.a, but it's problem
for libcryptsetup which requires versioned symbols from libblkid.so

This is no problem for normal binaries, but for libmount tests only
(where we need static library to by-pass public library API).

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

lscpu: fix NVIDIA ARM hw implementer spelling case

Ref respective vendor sites.

Signed-off-by: Ville Skyttä <ville.skytta@iki.fi>
Signed-off-by: Karel Zak <kzak@redhat.com>

lscpu: recognize more ARM implementers

Add just the implementer string, no parts yet.

Refs https://developer.arm.com/documentation/ddi0595/2021-03/AArch64-Registers/MIDR-EL1--Main-ID-Register
Refs https://github.com/torvalds/linux/blob/b90e90f40b4ff23c753126008bf4713a42353af6/arch/arm64/include/asm/cputype.h#L54

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

po: add xgettext hint for non-c-format string

Reported-by: Petr Pisar <petr.pisar@atlas.cz>
Signed-off-by: Karel Zak <kzak@redhat.com>

lscpu: cleanup tab vs. space

Do not mix tab and space in structs initializations.

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

lscpu: Replace space with tabs

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

lscpu: add MHZ to the -e output

Addresses: https://github.com/karelzak/util-linux/issues/1314
Signed-off-by: Karel Zak <kzak@redhat.com>

fstrim: do not start the timer in initrd

I'm working on building initramfs images directly from normal
packages, which means that the pristine system rpms should behave
correctly as much as possible also in the initrd. There usually isn't
enough time for the timer to actually fire, but starting it gives a
line on the console and generally looks confusing and sloppy. So let's
skip the timer if it ever ends up being enabled in the initrd.

Checking for /etc/initrd-release is the standard condition that
systemd's initrd units use.

Merge branch 'more-floating-point' of https://github.com/kerolasa/util-linux

* 'more-floating-point' of https://github.com/kerolasa/util-linux:
  more: fix floating point exception core dump

fstrim: fix paths comparison

Fix: https://github.com/karelzak/util-linux/issues/1312
Signed-off-by: Karel Zak <kzak@redhat.com>

more: fix floating point exception core dump

Make the code avoid divided by zero.  This can happen when file has content
but is zero in size.  Such files can be found from procfs, possibly some
other pseudo-filesystems.  To reproduce the issue run the following.

    $ more /proc/crypto
    ...
    Floating point exception (core dumped)

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

build-sys: fix po-man/ make check

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

blkdiscard: do not probe for signatures on --force

The command-line option --force is defined as "disable all checks",
but the current code does not follow this idea. We need a way how to
disable read from the device (for example for dm-integrity devices).

Fixes: https://github.com/karelzak/util-linux/issues/1308
Signed-off-by: Karel Zak <kzak@redhat.com>

Merge branch 'whitespace' of https://github.com/eworm-de/util-linux

build-sys: update util-linux-man.pot on 'make dist'

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

fix whitespace issue in ducumentation

This removes trailing whitespaces and fixes minor issues.

build-sys: release++ (v2.37-rc2)

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

docs: update v2.37-ReleaseNotes

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

docs: update AUTHORS file

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

po: merge changes

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

po: update sv.po (from translationproject.org)

tests: update build tests

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

build-sys: fix libblkid dependence

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

column: add placeholder '0' to specify all columns

Fixes: https://github.com/karelzak/util-linux/issues/1306
Signed-off-by: Karel Zak <kzak@redhat.com>

blkdiscard: fix compilation without libblkid

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

mkswap: remove unused variable when compile without libblkid

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

docs: add #1266 to TODO file

References: https://github.com/karelzak/util-linux/issues/1266
Signed-off-by: Karel Zak <kzak@redhat.com>

unshare: Fix error message when setting proc mount propagation

The mount() command associated with this error message is not about
unmounting the proc fs, but rather about changing the propagation
of mount events for the proc fs. Rewrite the error message to
reflect this.

rfkill: make RFKILL_EVENT_SIZE_V1 use more portable

The old linux/rfkill.h uses "int", new versions use sizeof() for the
macro ...

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

tests: update libfdisk JSON outputs

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

libfdisk: (script) print bootable flag only when set

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

rfkill: fix compiler warning [-Wsign-compare]

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

lib/jsonwrt: don't use ctype.h for ASCII chars

tolower() does not work "as expected" for tr_TR.UTF-8 (Turkish).
Fortunately, we need to convert only objects and variables names in
JSON output, and this is always old good ASCII.

Anyway, for more details:

$ cat a.c
#include <ctype.h>
#include <stdio.h>
#include <locale.h>

int main(void)
{
int in, out;

setlocale(LC_ALL, "");

in ='I';
out = tolower(in);

printf("%1$c [%1$d] --> %2$c [%2$d]\n", in, out);
return 0;
}

$ make a
cc     a.c   -o a

$ LANG=en_US.utf8 ./a
I [73] --> i [105]

$ LANG=tr_TR.UTF-8 ./a
I [73] --> I [73]

Fixes: https://github.com/karelzak/util-linux/issues/1302
Signed-off-by: Karel Zak <kzak@redhat.com>

lib/jsonwrt: remove fputs_quoted_json_* functions from include/carefulputc.h

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

lib/jsonwrt: remove 'islast' from API

The code should be able to keep track about previous content and print
JSON objects separator automatically.

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

libfdisk: use lib/jsonwrt.s for JSON formatting

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

rfkill: fix compiler warning [-Wformat=]

* `RFKILL_EVENT_SIZE_V1` is defined as sizeof(struct rfkill_event), so let's use %zu

* cast `len` to size_t to make it usable with %zu

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

lib/fileutils: close fd if fdopen is failed

Signed-off-by: Masatake YAMATO <yamato@redhat.com>

hwclock: follow timespec and use long int for nsec

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

script: fix time_t=long assumptions

References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

rtcwake: fix time_t=long assumptions

References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

lib/strutils: assume 64-bit time_t

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

hardlink: fix time_t=long assumptions

References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

test_uuid_parser: fix time_t=long assumptions

References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

fsck: fix time_t=long assumptions

References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

scriptplay: fix time_t=long assumptions

Fixes: https://github.com/karelzak/util-linux/issues/1069
References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

lib/pty-session: fix time_t=long assumptions

Fixes: https://github.com/karelzak/util-linux/issues/1069
References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

flock: fix time_t=long assumptions

This also fixes flock for archs (e.g. sparc64) where suseconds_t is not long.

Fixes: https://github.com/karelzak/util-linux/issues/1069
References: http://github.com/karelzak/util-linux/commit/ce3355cc54d97711bc240783324f7ab51fd6e371
Signed-off-by: Karel Zak <kzak@redhat.com>

build-sys: keep adoc files in dist_noinst_DATA

* rename MANPAGES_EXTRA= to ADOCFILES_COMMON=

* keep track about individual adoc files by dist_noinst_DATA=
  This variable is not effected by automake conditions, so the files
  are always distributed.

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

tests: update sparc lscpu tests

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

lscpu: assume L1d, L1i, L2, L3 for sparc

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

lscpu: read Sparc caches files

Fixes: https://github.com/karelzak/util-linux/issues/1296
Signed-off-by: Karel Zak <kzak@redhat.com>

meson: fix systemd dependence

Fixes: https://github.com/karelzak/util-linux/issues/1301
Signed-off-by: Karel Zak <kzak@redhat.com>

Merge branch 'master' of https://github.com/arbego/util-linux

* 'master' of https://github.com/arbego/util-linux:
  Fixed wrongful time_t=long assumptions in hwclock.c and timeutils.c
  Changed int64_t casts to long long int casts
  Fixed format strings and type casts in hwclock to work with 64-bit time_t on 32-bit linux

Fixed wrongful time_t=long assumptions in hwclock.c and timeutils.c

build-sys: add configure options to disable individual utils

lscpu: fix "caches" header

It's possible that lscpu does not print any caches if full caches size
is zero. In this case we do not need the header.

Addresses: https://github.com/karelzak/util-linux/issues/1207
Signed-off-by: Karel Zak <kzak@redhat.com>

docs/TODO: Minor update and fix typo

Changed int64_t casts to long long int casts

Fixed format strings and type casts in hwclock to work with 64-bit time_t on 32-bit linux