]> err.no Git - util-linux/log
util-linux
17 years agoMerge commit 'origin/master'
LaMont Jones [Thu, 8 Nov 2007 03:07:28 +0000 (20:07 -0700)]
Merge commit 'origin/master'

Conflicts:

mount/Makefile.am
mount/my_dev_t.h

17 years agobuild-sys: non-linux support
Samuel Thibault [Sun, 4 Nov 2007 15:54:27 +0000 (15:54 +0000)]
build-sys: non-linux support

Here is a patch that disables linux-specific tools on non-linux targets.

Signed-Off-By: Samuel Thibault <samuel.thibault@ens-lyon.org>
Signed-Off-By: Karel Zak <kzak@redhat.com>
17 years agomore: non-linux support
Samuel Thibault [Sun, 4 Nov 2007 16:39:36 +0000 (16:39 +0000)]
more: non-linux support

Signed-Off-By: Samuel Thibault <samuel.thibault@ens-lyon.org>
17 years agomount: use blkdev_get_size()
Karel Zak [Wed, 7 Nov 2007 23:54:58 +0000 (00:54 +0100)]
mount: use blkdev_get_size()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agosfdisk: use get_linux_version()
Karel Zak [Wed, 7 Nov 2007 23:27:51 +0000 (00:27 +0100)]
sfdisk: use get_linux_version()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoelvtune: use get_linux_version()
Karel Zak [Wed, 7 Nov 2007 20:25:48 +0000 (21:25 +0100)]
elvtune: use get_linux_version()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: cleanup KERNEL_VERSION, remove my_dev_t.h
Karel Zak [Wed, 7 Nov 2007 20:13:59 +0000 (21:13 +0100)]
mount: cleanup KERNEL_VERSION, remove my_dev_t.h

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoblockdev: use lib/blkdev.c, fix --report
Karel Zak [Wed, 7 Nov 2007 19:59:28 +0000 (20:59 +0100)]
blockdev: use lib/blkdev.c, fix --report

This patch replaces duplicate code with blkdev_ functions from
lib/blkdev.c.

The patch also fix the output from "blockdev --report". The Size field
was in 512-byte sectors instead in real size in bytes.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agofdisk: cleanup BLK* ioctls usage
Karel Zak [Wed, 7 Nov 2007 18:05:31 +0000 (19:05 +0100)]
fdisk: cleanup BLK* ioctls usage

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomkswap: cleanup kB vs. KiB usage in error messages
Karel Zak [Wed, 7 Nov 2007 12:46:34 +0000 (13:46 +0100)]
mkswap: cleanup kB vs. KiB usage in error messages

mkswap(8) man page is talking about KiB (2^10), but the mkswap binary
uses kB (10^3) in error messages. That's strange. Let's use KiB
everywhere.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomkswap: BLKGETSIZE cleanup
Karel Zak [Wed, 7 Nov 2007 12:26:27 +0000 (13:26 +0100)]
mkswap: BLKGETSIZE cleanup

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agolib: add blkdev.{c,h}
Stefan Krah [Wed, 7 Nov 2007 11:37:30 +0000 (12:37 +0100)]
lib: add blkdev.{c,h}

Now we duplicate BLK* ioctls on many places... This patch also
fix BLKGETSIZE64 usage in dependence on kernel version.

Co-Author: Karel Zak <kzak@redhat.com>
Signed-off-by: Stefan Krah <stefan@bytereef.org>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomkswap: linux_version() code consolidation
Karel Zak [Tue, 6 Nov 2007 01:40:49 +0000 (02:40 +0100)]
mkswap: linux_version() code consolidation

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agolib: add linux_version.{c,h}
Stefan Krah [Tue, 6 Nov 2007 01:40:13 +0000 (02:40 +0100)]
lib: add linux_version.{c,h}

Signed-off-by: Stefan Krah <stefan@bytereef.org>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agopo: update fi.po (from translationproject.org)
Lauri Nurmi [Tue, 6 Nov 2007 00:35:42 +0000 (01:35 +0100)]
po: update fi.po (from translationproject.org)

17 years agodocs: fix ChangeLog URL
Pascal Terjan [Sat, 3 Nov 2007 10:55:52 +0000 (11:55 +0100)]
docs: fix ChangeLog URL

Signed-off-by: Pascal Terjan <pterjan@linuxfr.org>
17 years agosetarch: tweak the help text, and gettextize a forgotten message
Benno Schulenberg [Mon, 5 Nov 2007 23:26:59 +0000 (00:26 +0100)]
setarch: tweak the help text, and gettextize a forgotten message

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
17 years agomkfs.cramfs: remove unused header file
lizf [Thu, 1 Nov 2007 09:14:52 +0000 (17:14 +0800)]
mkfs.cramfs: remove unused header file

Remove including of assert.h

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
17 years agomount: remove MS_{REPLACE,AFTER,BEFORE,OVER}
Karel Zak [Tue, 6 Nov 2007 00:05:27 +0000 (01:05 +0100)]
mount: remove MS_{REPLACE,AFTER,BEFORE,OVER}

mount(8) and linux kernel don't support these mount flags. It's legacy
from an unimplemented stuff.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agosfdisk: cleanup 83 gcc warnings
Randy Dunlap [Tue, 16 Oct 2007 00:38:57 +0000 (17:38 -0700)]
sfdisk: cleanup 83 gcc warnings

Fix strict gcc warnings that come from using:
  ("-Wall -Wp,-D_FORTIFY_SOURCE=2")

83 warnings fixed:
sfdisk.c:249: warning: pointer targets in initialization differ in signedness
sfdisk.c:376: warning: pointer targets in passing argument 1 of 'chars_to_ulong' differ in signedness
sfdisk.c:702: warning: pointer targets in passing argument 1 of 'copy_to_int' differ in signedness
sfdisk.c:703: warning: pointer targets in passing argument 1 of 'copy_to_int' differ in signedness
sfdisk.c:1709: warning: pointer targets in passing argument 1 of 'fgets' differ in signedness
sfdisk.c:1709: warning: pointer targets in assignment differ in signedness
sfdisk.c:1714: warning: pointer targets in passing argument 1 of 'index' differ in signedness
sfdisk.c:1714: warning: pointer targets in assignment differ in signedness
sfdisk.c:1719: warning: pointer targets in passing argument 1 of 'index' differ in signedness
sfdisk.c:1719: warning: pointer targets in assignment differ in signedness
sfdisk.c:1723: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:1723: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1723: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:1723: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1723: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1723: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1729: warning: pointer targets in passing argument 1 of 'index' differ in signedness
sfdisk.c:1729: warning: pointer targets in assignment differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:1739: warning: pointer targets in passing argument 1 of 'strncmp' differ in signedness
sfdisk.c:2021: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2021: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2021: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2021: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2021: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2021: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2059: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2059: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2059: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2059: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2059: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2059: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2061: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2061: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2061: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2061: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2061: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2061: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2063: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2063: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2063: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2063: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2063: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2063: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2065: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2065: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2065: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2065: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2065: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2065: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2067: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2083: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2096: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2130: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2130: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2130: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2130: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2130: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2130: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2132: warning: pointer targets in passing argument 1 of '__builtin_strcmp' differ in signedness
sfdisk.c:2154: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2155: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2156: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2167: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2168: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness
sfdisk.c:2169: warning: pointer targets in passing argument 1 of 'get_ul' differ in signedness

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoscript: cleanup gcc warnings
Randy Dunlap [Tue, 16 Oct 2007 00:38:40 +0000 (17:38 -0700)]
script: cleanup gcc warnings

Fix strict gcc warnings that come from using:
  ("-Wall -Wp,-D_FORTIFY_SOURCE=2")

script.c:239: warning: ignoring return value of 'write', declared with attribute warn_unused_result
script.c:330: warning: ignoring return value of 'write', declared with attribute warn_unused_result
script.c:331: warning: ignoring return value of 'fwrite', declared with attribute warn_unused_result

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomore: cleanup gcc warnings
Randy Dunlap [Tue, 16 Oct 2007 00:38:22 +0000 (17:38 -0700)]
more: cleanup gcc warnings

Fix strict gcc warnings that come from using:
  ("-Wall -Wp,-D_FORTIFY_SOURCE=2")

more.c:185: warning: passing argument 1 of 'setupterm' discards qualifiers from pointer target type
more.c:205: warning: passing argument 1 of 'tparm' discards qualifiers from pointer target type
more.c:812: warning: pointer targets in passing argument 2 of 'mbrtowc' differ in signedness
more.c:931: warning: pointer targets in passing argument 2 of 'mbrtowc' differ in signedness
more.c:1285: warning: pointer targets in passing argument 1 of 'ttyin' differ in signedness
more.c:1486: warning: pointer targets in passing argument 1 of 'ttyin' differ in signedness
more.c:1879: warning: pointer targets in passing argument 2 of 'mbrtowc' differ in signedness

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
17 years agobuild: don't deliver col* and ul as part of bsdutils for now. Closes: #446939
LaMont Jones [Mon, 29 Oct 2007 20:07:55 +0000 (14:07 -0600)]
build: don't deliver col* and ul as part of bsdutils for now.  Closes: #446939

Bsdmainutils delivers these files, so we'll not conflict with them for now.

17 years agoMerge branch 'stable/v2.13'
LaMont Jones [Tue, 30 Oct 2007 14:27:59 +0000 (08:27 -0600)]
Merge branch 'stable/v2.13'

17 years agochangelog: release
LaMont Jones [Mon, 29 Oct 2007 20:30:37 +0000 (14:30 -0600)]
changelog: release

17 years agobuild: don't deliver col* and ul as part of bsdutils for now. Closes: #446939
LaMont Jones [Mon, 29 Oct 2007 20:07:55 +0000 (14:07 -0600)]
build: don't deliver col* and ul as part of bsdutils for now.  Closes: #446939

Bsdmainutils delivers these files, so we'll not conflict with them for now.

17 years agoMerge commit 'origin/stable/v2.13' into stable/v2.13
LaMont Jones [Fri, 26 Oct 2007 12:48:29 +0000 (06:48 -0600)]
Merge commit 'origin/stable/v2.13' into stable/v2.13

17 years agoMerge commit 'origin/master'
LaMont Jones [Fri, 26 Oct 2007 12:48:18 +0000 (06:48 -0600)]
Merge commit 'origin/master'

Conflicts:

mount/Makefile.am
mount/mount.c

17 years agodocs: add info about .bugfix releases and branches
Karel Zak [Mon, 22 Oct 2007 08:47:46 +0000 (10:47 +0200)]
docs: add info about .bugfix releases and branches

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomkswap: possible to crash with SELinux relabeling support
KaiGai Kohei [Mon, 22 Oct 2007 08:30:19 +0000 (10:30 +0200)]
mkswap: possible to crash with SELinux relabeling support

When fgetfilecon() is failed with -ENODATA, this process does not
exit.  However, "oldcontext" is not initialized in this case, so
context_new() will be called with uninitialized "oldcontext" at the
next.

Finally, it makes a segmentation fault, because context_new() have to
refer an incorrect memory region.

The attached patch fixes this matter using matchpathcon().  If we
cannot obtain actual file context due to -ENODATA, a context which is
returned by matchpathcon() is applied as oldcontext.  Then, the type
of the context is relabeled to "swapfile_t" explicitly.

Signed-off-by: KaiGai Kohei <kaigai@kaigai.gr.jp>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agotests: use losetup -s
Karel Zak [Thu, 25 Oct 2007 22:55:36 +0000 (00:55 +0200)]
tests: use losetup -s

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agotailf: opened file leaving unclosed
lizf [Mon, 22 Oct 2007 03:27:15 +0000 (11:27 +0800)]
tailf: opened file leaving unclosed

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
17 years agoblockdev: fix: opened file leaving unclosed
lizf [Mon, 22 Oct 2007 03:26:27 +0000 (11:26 +0800)]
blockdev: fix: opened file leaving unclosed

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
17 years agosfdisk: opened files leaving unclosed
Karel Zak [Mon, 22 Oct 2007 03:25:39 +0000 (11:25 +0800)]
sfdisk: opened files leaving unclosed

Co-Author: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agosetterm: opened file leaving unclosed
Karel Zak [Thu, 25 Oct 2007 20:10:07 +0000 (22:10 +0200)]
setterm: opened file leaving unclosed

Co-Author: lizf <lizf@cn.fujitsu.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoagetty: ungettextize several debugging messages.
Benno Schulenberg [Sat, 29 Sep 2007 13:18:28 +0000 (15:18 +0200)]
agetty: ungettextize several debugging messages.

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
17 years agodocs: tweak a few messages for clarity
Benno Schulenberg [Sat, 29 Sep 2007 13:14:42 +0000 (15:14 +0200)]
docs: tweak a few messages for clarity

Add a missing period, a missing space, a comma and a word for clarity,
plus a period and an uppercase letter to match surrounding messages.
Further add a missing call to gettext, and undo an unneeded linewrap.

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
17 years agocfdisk: slightly increase the size of menu buttons
Benno Schulenberg [Sat, 29 Sep 2007 13:04:50 +0000 (15:04 +0200)]
cfdisk: slightly increase the size of menu buttons

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
17 years agocfdisk: translate partition-type names when they are printed.
Benno Schulenberg [Sat, 29 Sep 2007 13:02:28 +0000 (15:02 +0200)]
cfdisk: translate partition-type names when they are printed.

Signed-off-by: Benno Schulenberg <bensberg@justemail.net>
17 years agohwclock: do not create a zero adjfile
Alain Guibert [Mon, 24 Sep 2007 10:57:22 +0000 (12:57 +0200)]
hwclock: do not create a zero adjfile

When hwclock --hctosys is started very early during the system startup,
with / still mounted read-only, and there was no /etc/adjtime file,
hwclock fails creating a default adjfile full of zeroes, and prints an
error message. I believe that such zero adjfile is not necessary,
because it means exactly the same as no adjfile at all.

The attached patch prevents creation of a zero adjfile, of course unless
something gets changed (this never happens during a --hctosys).

Signed-off-by: Alain Guibert <alguibert+ulng@free.fr>
17 years agomount: prevent loop mounting the same file twice
Karel Zak [Tue, 11 Sep 2007 12:35:34 +0000 (14:35 +0200)]
mount: prevent loop mounting the same file twice

The mount syscall prevents mounting the same device twice
to the same mountpoint. When loop mounting a file, for each
file a new loop device gets allocated, which prevents the detection
of loop mounting the same file to the same mountpoint twice.
The patch adds a check to prevent double mounts, if the same loopfile
is going to be mounted with the same offset to the same mountpoint.

Co-Author: Matthias Koenig <mkoenig@suse.de>
Signed-off-by: Matthias Koenig <mkoenig@suse.de>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agolosetup: canonicalize loopfile name
Karel Zak [Thu, 25 Oct 2007 10:29:51 +0000 (12:29 +0200)]
losetup: canonicalize loopfile name

When setting up a loop device, canonicalize the loop file
name. This simplifies a later identification of loop file names
when querying the loop devices.

Co-Author: Matthias Koenig <mkoenig@suse.de>
Signed-off-by: Matthias Koenig <mkoenig@suse.de>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agolosetup: fix errno usage
Karel Zak [Thu, 25 Oct 2007 10:10:31 +0000 (12:10 +0200)]
losetup: fix errno usage

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agorev: use warn() in errs.h
Li Zefan [Mon, 10 Sep 2007 06:45:37 +0000 (14:45 +0800)]
rev: use warn() in errs.h

The function warn() in rev.c is actually duplicate code,
so here we remove it.

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoswapon: cleanup usage()
Karel Zak [Thu, 25 Oct 2007 08:49:53 +0000 (10:49 +0200)]
swapon: cleanup usage()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: clean up global variables
Karel Zak [Thu, 25 Oct 2007 08:39:23 +0000 (10:39 +0200)]
mount: clean up global variables

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agolosetup: remove duplicate xstrdup() and error()
Karel Zak [Thu, 25 Oct 2007 08:19:12 +0000 (10:19 +0200)]
losetup: remove duplicate xstrdup() and error()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: cleanup error() and die()
Karel Zak [Thu, 25 Oct 2007 08:12:51 +0000 (10:12 +0200)]
mount: cleanup error() and die()

 * moves error() and die() to sundries.h
 * removes at_die
 * adds __attribute__ ((__format__ (__printf__ )))

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoumount: use atexit() rather than (*at_die)()
Karel Zak [Sat, 24 Nov 2007 16:55:57 +0000 (17:55 +0100)]
umount: use atexit() rather than (*at_die)()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: use atexit() rather than (*at_die)()
Karel Zak [Tue, 23 Oct 2007 19:14:59 +0000 (21:14 +0200)]
mount: use atexit() rather than (*at_die)()

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: remove redundant fflush
Karel Zak [Tue, 23 Oct 2007 18:45:22 +0000 (20:45 +0200)]
mount: remove redundant fflush

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: cleanup canonicalize() usage
Karel Zak [Mon, 22 Oct 2007 13:49:26 +0000 (15:49 +0200)]
mount: cleanup canonicalize() usage

This patch renames canonicalize() to canonicalize_mountpoint() and
moves this function to realpath.c where is all cannonicalize code.
The canonicalize_mountpoint() function checks for special "none",
"proc", "swap" pseudo mointpoint.

The patch also adds a new generic canonicalize() function.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agodocs: add info about .bugfix releases and branches
Karel Zak [Mon, 22 Oct 2007 08:47:46 +0000 (10:47 +0200)]
docs: add info about .bugfix releases and branches

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomkswap: possible to crash with SELinux relabeling support
KaiGai Kohei [Mon, 22 Oct 2007 08:30:19 +0000 (10:30 +0200)]
mkswap: possible to crash with SELinux relabeling support

When fgetfilecon() is failed with -ENODATA, this process does not
exit.  However, "oldcontext" is not initialized in this case, so
context_new() will be called with uninitialized "oldcontext" at the
next.

Finally, it makes a segmentation fault, because context_new() have to
refer an incorrect memory region.

The attached patch fixes this matter using matchpathcon().  If we
cannot obtain actual file context due to -ENODATA, a context which is
returned by matchpathcon() is applied as oldcontext.  Then, the type
of the context is relabeled to "swapfile_t" explicitly.

Signed-off-by: KaiGai Kohei <kaigai@kaigai.gr.jp>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoMerge branch 'stable/v2.13'
LaMont Jones [Sun, 21 Oct 2007 04:19:34 +0000 (22:19 -0600)]
Merge branch 'stable/v2.13'

Conflicts:

debian/changelog

17 years agochangelog: release
LaMont Jones [Sun, 21 Oct 2007 02:59:59 +0000 (20:59 -0600)]
changelog: release

17 years agoMerge branch 'stable/v2.13'
LaMont Jones [Sun, 21 Oct 2007 02:24:41 +0000 (20:24 -0600)]
Merge branch 'stable/v2.13'

17 years agoMerge commit 'origin/master'
LaMont Jones [Sun, 21 Oct 2007 02:23:09 +0000 (20:23 -0600)]
Merge commit 'origin/master'

17 years agomount.8: Make package references be the actual binary package name in the distro...
LaMont Jones [Fri, 19 Oct 2007 16:48:14 +0000 (10:48 -0600)]
mount.8:  Make package references be the actual binary package name in the distro.  LP: #154399

Signed-off-by: LaMont Jones <lamont@debian.org>
17 years agoMerge commit 'origin/stable/v2.13' into stable/v2.13
LaMont Jones [Fri, 19 Oct 2007 12:46:56 +0000 (06:46 -0600)]
Merge commit 'origin/stable/v2.13' into stable/v2.13

17 years agosetarch: generate groff links in a better way
Karel Zak [Mon, 15 Oct 2007 11:41:18 +0000 (13:41 +0200)]
setarch: generate groff links in a better way

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agobuild-sys: add v2.14 to NEWS
Karel Zak [Thu, 18 Oct 2007 13:07:07 +0000 (15:07 +0200)]
build-sys: add v2.14 to NEWS

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agobuild-sys: remove errs.h
Karel Zak [Mon, 15 Oct 2007 13:08:31 +0000 (15:08 +0200)]
build-sys: remove errs.h

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agotailf: replace errs.h with libc err.h
Karel Zak [Thu, 18 Oct 2007 12:52:46 +0000 (14:52 +0200)]
tailf: replace errs.h with libc err.h

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agosetarch: generate groff links in a better way
Karel Zak [Mon, 15 Oct 2007 11:41:18 +0000 (13:41 +0200)]
setarch: generate groff links in a better way

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agocal: replace errs.h with libc err.h
Karel Zak [Mon, 15 Oct 2007 13:07:05 +0000 (15:07 +0200)]
cal: replace errs.h with libc err.h

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agocolumn: replace errs.h with libc err.h
Karel Zak [Mon, 15 Oct 2007 13:04:41 +0000 (15:04 +0200)]
column: replace errs.h with libc err.h

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomesg: replace errs.h with libc err.h
Karel Zak [Mon, 15 Oct 2007 12:51:28 +0000 (14:51 +0200)]
mesg: replace errs.h with libc err.h

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agobuild-sys: add err.h check
Karel Zak [Mon, 15 Oct 2007 12:17:14 +0000 (14:17 +0200)]
build-sys: add err.h check

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoMerge commit 'origin/stable/v2.13' into stable/v2.13
LaMont Jones [Mon, 15 Oct 2007 10:56:42 +0000 (04:56 -0600)]
Merge commit 'origin/stable/v2.13' into stable/v2.13

17 years agopo: update po files
Karel Zak [Mon, 15 Oct 2007 10:05:59 +0000 (12:05 +0200)]
po: update po files

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agopo: fix typo in de.po
Karel Zak [Thu, 11 Oct 2007 12:31:41 +0000 (14:31 +0200)]
po: fix typo in de.po

de.po:9010: internationalized messages should not contain the `\r' escape sequence

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agopo: update nl.po (from translationproject.org)
Benno Schulenberg [Thu, 11 Oct 2007 09:59:48 +0000 (11:59 +0200)]
po: update nl.po (from translationproject.org)

17 years agopo: update de.po (from translationproject.org)
Michael Piefel [Fri, 5 Oct 2007 10:43:51 +0000 (12:43 +0200)]
po: update de.po (from translationproject.org)

17 years agoscript: dies on SIGWINCH
Karel Zak [Fri, 5 Oct 2007 10:22:13 +0000 (12:22 +0200)]
script: dies on SIGWINCH

The "doinput" process doesn't make a difference between SIGWINCH and
SIGCHILD.  This process also sends unnecessary SIGWINCH to child (the
signal is ignored by child). Fixed.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoionice: add a note about permissions to ionice.1
Karel Zak [Thu, 20 Sep 2007 12:33:31 +0000 (14:33 +0200)]
ionice: add a note about permissions to ionice.1

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agochsh: should use pam_end function to terminate the PAM transaction
Yu Zhiguo [Thu, 6 Sep 2007 06:28:49 +0000 (14:28 +0800)]
chsh: should use pam_end function to terminate the PAM transaction

chsh will use PAM transaction if macros 'REQUIRE_PASSWORD' and
'HAVE_SECURITY_PAM_MISC_H' are defined, but there is no pam_end function be
used when the PAM transaction be terminated.

This patch also cleanup PAM code in chsh(1).

Signed-off-by: Yu Zhiguo <yuzg@cn.fujitsu.com>
Co-Author: Karel Zak <kzak@redhat.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agochfn: add pam_end() call and cleanup PAM code
Karel Zak [Wed, 3 Oct 2007 22:22:36 +0000 (00:22 +0200)]
chfn: add pam_end() call and cleanup PAM code

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agoMerge branch 'stable/v2.13'
LaMont Jones [Fri, 12 Oct 2007 03:02:43 +0000 (21:02 -0600)]
Merge branch 'stable/v2.13'

Conflicts:

debian/changelog
disk-utils/mkfs.cramfs.c

17 years agoMerge commit 'origin/master'
LaMont Jones [Fri, 12 Oct 2007 02:51:36 +0000 (20:51 -0600)]
Merge commit 'origin/master'

17 years agobuild: don't deliver (emtpy) /usr/share/util-linux. Closes: #445908
LaMont Jones [Fri, 12 Oct 2007 02:51:01 +0000 (20:51 -0600)]
build: don't deliver (emtpy) /usr/share/util-linux.  Closes: #445908

Signed-off-by: LaMont Jones <lamont@debian.org>
17 years agoscript: dies on SIGWINCH. Closes: #445956
Karel Zak [Fri, 5 Oct 2007 10:22:13 +0000 (12:22 +0200)]
script: dies on SIGWINCH.  Closes: #445956

The "doinput" process doesn't make a difference between SIGWINCH and
SIGCHILD.  This process also sends unnecessary SIGWINCH to child (the
signal is ignored by child). Fixed.

Signed-off-by: Karel Zak <kzak@redhat.com>
Signed-off-by: LaMont Jones <lamont@debian.org>
17 years agoMerge commit 'origin/stable/v2.13' into stable/v2.13
LaMont Jones [Fri, 12 Oct 2007 02:45:52 +0000 (20:45 -0600)]
Merge commit 'origin/stable/v2.13' into stable/v2.13

17 years agotests: fix blkid cache usage
Karel Zak [Thu, 11 Oct 2007 22:33:01 +0000 (00:33 +0200)]
tests: fix blkid cache usage

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: -L|-U segfault when label or uuid doesn't exist
Karel Zak [Thu, 11 Oct 2007 14:18:29 +0000 (16:18 +0200)]
mount: -L|-U segfault when label or uuid doesn't exist

 # mount -L foo
 Segmentation fault

mount(8) calls strcmp() with NULL argument.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agotests: fix blkid cache usage
Karel Zak [Thu, 11 Oct 2007 22:33:01 +0000 (00:33 +0200)]
tests: fix blkid cache usage

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomount: -L|-U segfault when label or uuid doesn't exist
Karel Zak [Thu, 11 Oct 2007 14:18:29 +0000 (16:18 +0200)]
mount: -L|-U segfault when label or uuid doesn't exist

 # mount -L foo
 Segmentation fault

mount(8) calls strcmp() with NULL argument.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agotests: exactly define a time format in ls -l output
Karel Zak [Thu, 11 Oct 2007 12:52:40 +0000 (14:52 +0200)]
tests: exactly define a time format in ls -l output

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agopo: fix typo in de.po
Karel Zak [Thu, 11 Oct 2007 12:31:41 +0000 (14:31 +0200)]
po: fix typo in de.po

de.po:9010: internationalized messages should not contain the `\r' escape sequence

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agopo: update po files
Karel Zak [Thu, 11 Oct 2007 12:30:01 +0000 (14:30 +0200)]
po: update po files

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agortcwake: fix typo
Karel Zak [Thu, 11 Oct 2007 12:27:08 +0000 (14:27 +0200)]
rtcwake: fix typo

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agocal: add support for highlighting an arbitrary date
Pádraig Brady [Mon, 1 Oct 2007 23:22:45 +0000 (00:22 +0100)]
cal: add support for highlighting an arbitrary date

This is done by calling cal with the extra day parameter like:

cal 14 9 1752

Note the tests were updated to use the new syntax.

Note also that this patch changes the -y option
to always print a full year, even if a month or
the -[13] options are specified.
This matches the cal operation from bsdmainutils on debian
and also allows one to print a full year while
highlighting a particular date.

Signed-off-by: Pádraig Brady <P@draigBrady.com>
17 years agotailf: add option -n to specifying output lines
Li Zefan [Thu, 11 Oct 2007 11:57:35 +0000 (13:57 +0200)]
tailf: add option -n to specifying output lines

It will be useful if we can print out the last n lines instead of the last
10, just like tail.

There are examples:

tailf -n 5 file1
tailf --lines 10 file2
tailf -20 file3

Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agotailf: clean up gcc warnings & fix use of errno
Karel Zak [Wed, 3 Oct 2007 21:15:03 +0000 (14:15 -0700)]
tailf: clean up gcc warnings & fix use of errno

Fix strict gcc warnings in tailf that come from using:
  ("-Wall -Wp,-D_FORTIFY_SOURCE=2")

tailf.c:111: warning: ignoring return value of 'fwrite', declared with attribute warn_unused_result

Also, tailf uses perror() for error reporting, but it inserts
an fprintf call first, so perror() is actually reporting the
result of the fprintf() call, not the failing call; change
the code to print the message by using strerror() instead.

Builds cleanly on x86_32 and x86_64.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agolosetup: clean up gcc warnings
Randy Dunlap [Mon, 8 Oct 2007 16:59:18 +0000 (09:59 -0700)]
losetup: clean up gcc warnings

Fix strict gcc warnings that come from using:
  ("-Wall -Wp,-D_FORTIFY_SOURCE=2")

lomount.c:98: warning: pointer targets in initialization differ in signedness
lomount.c:111: warning: format '%04x' expects type 'unsigned int', but argument 4 has type '__kernel_old_dev_t'
lomount.c:300: warning: pointer targets in passing argument 1 of 'xstrncpy' differ in signedness
lomount.c:307: warning: pointer targets in passing argument 1 of '__builtin___snprintf_chk' differ in signedness
lomount.c:339: warning: pointer targets in passing argument 1 of 'xstrncpy' differ in signedness

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
17 years agosetterm: dump by TIOCLINUX is deprecated since linux 1.1.92.
Karel Zak [Thu, 11 Oct 2007 10:45:53 +0000 (12:45 +0200)]
setterm: dump by TIOCLINUX is deprecated since linux 1.1.92.

Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agopo: update nl.po (from translationproject.org)
Benno Schulenberg [Thu, 11 Oct 2007 09:59:48 +0000 (11:59 +0200)]
po: update nl.po (from translationproject.org)

17 years agomount: improve chmod & chown usage and clean up gcc warnings (fstab.c)
Karel Zak [Wed, 3 Oct 2007 21:15:18 +0000 (14:15 -0700)]
mount: improve chmod & chown usage and clean up gcc warnings (fstab.c)

Fix strict gcc warnings in tailf that come from using:
  ("-Wall -Wp,-D_FORTIFY_SOURCE=2")

fstab.c:770: warning: ignoring return value of 'chown', declared with attribute warn_unused_result

The patch makes chmod() and chown() mandatory. We cannot rename()
temporary mtab to the final mtab when owner is not the same user as
owner of the original mtab. It's security risk.

Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agomkfs.cramfs: clean up gcc warnings
Randy Dunlap [Sat, 29 Sep 2007 08:12:06 +0000 (01:12 -0700)]
mkfs.cramfs: clean up gcc warnings

* clean up code, gcc warnings
  (try compilation with "-Wall -Wp,-D_FORTIFY_SOURCE=2")

Builds cleanly on x86_32 and x86_64.

mkfs.cramfs.c:211: warning: pointer targets in passing argument 2 of 'MD5Update' differ in signedness
mkfs.cramfs.c:336: warning: format '%u' expects type 'unsigned int', but argument 4 has type 'size_t'
mkfs.cramfs.c:355: warning: pointer targets in assignment differ in signedness
mkfs.cramfs.c:448: warning: pointer targets in passing argument 1 of '__builtin___strncpy_chk' differ in signedness
mkfs.cramfs.c:448: warning: pointer targets in passing argument 1 of '__strncpy_ichk' differ in signedness
mkfs.cramfs.c:450: warning: pointer targets in passing argument 1 of '__builtin___strncpy_chk' differ in signedness
mkfs.cramfs.c:450: warning: pointer targets in passing argument 1 of '__strncpy_ichk' differ in signedness
mkfs.cramfs.c:490: warning: pointer targets in passing argument 1 of 'strlen' differ in signedness
mkfs.cramfs.c:620: warning: pointer targets in passing argument 1 of 'compress' differ in signedness
mkfs.cramfs.c:620: warning: pointer targets in passing argument 3 of 'compress' differ in signedness
mkfs.cramfs.c:671: warning: pointer targets in passing argument 3 of 'do_compress' differ in signedness
mkfs.cramfs.c:824: warning: format '%Ld' expects type 'long long int', but argument 4 has type 'loff_t'
mkfs.cramfs.c:872: warning: format '%d' expects type 'int', but argument 3 has type 'ssize_t'
mkfs.cramfs.c:880: warning: format '%d' expects type 'int', but argument 3 has type 'ssize_t'
mkfs.cramfs.c:885: warning: format '%d' expects type 'int', but argument 3 has type 'long unsigned int'
mkfs.cramfs.c:889: warning: pointer targets in passing argument 2 of 'crc32' differ in signedness
mkfs.cramfs.c:896: warning: format '%Ld' expects type 'long long int', but argument 4 has type 'loff_t'
mkfs.cramfs.c:896: warning: format '%d' expects type 'int', but argument 5 has type 'ssize_t'
mkfs.cramfs.c:909: warning: format '%d' expects type 'int', but argument 4 has type 'ssize_t'
mkfs.cramfs.c:909: warning: format '%d' expects type 'int', but argument 5 has type 'ssize_t'

Co-Author: Karel Zak <kzak@redhat.com>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: Karel Zak <kzak@redhat.com>
17 years agodocs: add a note about minix v3 to TODO file
Karel Zak [Wed, 10 Oct 2007 22:22:38 +0000 (00:22 +0200)]
docs: add a note about minix v3 to TODO file

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