Neil Williams [Sun, 10 Feb 2008 20:51:41 +0000 (13:51 -0700)]
debian/rules: allow cross-building
In line with the other cross-building support bugs:
http://lists.debian.org/debian-devel/2007/11/msg00116.html
This patch is necessary to allow util-linux to cross-build in Debian,
following recommendations in autotools-dev. The cache file will be
handled separately.
Addresses-Debian-Bug: 465123 Signed-off-by: Neil Williams <codehelp@debian.org> Signed-off-by: LaMont Jones <lamont@debian.org>
LaMont Jones [Sun, 13 Jan 2008 05:35:50 +0000 (22:35 -0700)]
renice: detect errors in arguments, add -v, -h and long options
* renice was using atoi(), which does no error detection, meaning
that: "renice +20 blah" was accepted as valid.
* add -h | --help
* add -v | --version
* add long options for -p, -u and -g
* cleanup coding style
Addresses-Debian-Bug: #385245
Co-Author: Karel Zak <kzak@redhat.com> Signed-off-by: LaMont Jones <lamont@debian.org> Signed-off-by: Karel Zak <kzak@redhat.com>
H. Peter Anvin [Wed, 16 Jan 2008 18:53:56 +0000 (13:53 -0500)]
fdisk: better fallback for get_random_id()
When /dev/urandom is not available, we have to use some kind of a hack
to generate a random MBR identifier. Use a better fallback that
incorporates the clock down to microsecond granularity.
Karel Zak [Mon, 28 Jan 2008 13:02:35 +0000 (14:02 +0100)]
login: fix a small memory leak and remove unnecessary zeroing
This patch fix a small memory leak (rh#251539) and also remove
unnecessary pam_set_item(). We needn't to zeroing PAM_USER when the
value is already NULL, it doesn't make sense.
Addresses-Red-Hat-Bugzilla: #251539 Signed-off-by: Karel Zak <kzak@redhat.com>
Karel Zak [Mon, 28 Jan 2008 12:13:25 +0000 (13:13 +0100)]
umount: don't print duplicate error messages
umount(8) is trying to umount a device two times to prevent some
obscure scenarios. It's maybe a nice feature, but it also produces
duplicate error messages.
# umount /home
umount: /home: device is busy
umount: /home: device is busy
Pádraig Brady [Fri, 4 Jan 2008 10:44:53 +0000 (10:44 +0000)]
cal: fix weekday alignment for certain locales
For example this had too much padding: LANG=zh_CN.utf8 cal -j
while this had too little padding: LANG=hu_HU.utf8 cal
This had invalid chars: LANG=li_BE.utf8 cal
This had too few chars: LANG=si_LK.utf8 cal
Note some locales may display with slightly worse alignment
(fa_IR.utf8 for example), but that is only because the terminal
is not merging the combining characters. This happens on
gnome-terminal-2.18.3-1.fc8 at least.
Signed-off-by: Pádraig Brady <P@draigBrady.com> Signed-off-by: Karel Zak <kzak@redhat.com>
This patch also fixes the "Last cylinder..." hint message. The "+number"
without any suffix is not a size at all. It's number of cylinders/sectors.
Note, the 10^N suffixes are not proposed to end-uses in the hint message.
Frédéric Bothamy [Sun, 13 Jan 2008 05:35:51 +0000 (22:35 -0700)]
sys-utils: correct setarch.8 manpage link creation
At least on Debian, .so commands are relative to the man directory
(e.g., /usr/share/man), not to the subdirectory:
% man i386
man: can't open /usr/share/man/setarch.8: No such file or directory
No manual entry for i386
See also http://bugs.debian.org/453245
Addresses-Debian-Bug: #453245 Signed-off-by: Frédéric Bothamy <frederic.bothamy@free.fr> Signed-off-by: LaMont Jones <lamont@debian.org> Signed-off-by: Karel Zak <kzak@redhat.com>
Frédéric Bothamy [Sun, 13 Jan 2008 05:35:51 +0000 (22:35 -0700)]
sys-utils: correct setarch.8 manpage link creation
At least on Debian, .so commands are relative to the man directory
(e.g., /usr/share/man), not to the subdirectory:
% man i386
man: can't open /usr/share/man/setarch.8: No such file or directory
No manual entry for i386
See also http://bugs.debian.org/453245
Addresses-Debian-Bug: #453245 Signed-off-by: Frédéric Bothamy <frederic.bothamy@free.fr> Signed-off-by: LaMont Jones <lamont@debian.org> Signed-off-by: Karel Zak <kzak@redhat.com>
Matthias Koenig [Mon, 17 Dec 2007 10:33:20 +0000 (11:33 +0100)]
mount: "nofail" mount option
It would be nice to have a mount option "nofail" indicating that mount
should not return an error if the device does not exit. This is useful
for hotpluggable devices which are configured in fstab and __might__
not exist at boot time.
Co-Author: Karel Zak <kzak@redhat.com> Signed-off-by: Matthias Koenig <mkoenig@suse.de> Signed-off-by: Karel Zak <kzak@redhat.com>
projects / util-linux / log