Karel Zak [Fri, 27 Apr 2012 08:30:08 +0000 (10:30 +0200)]
libmount: close device fd (to avoid mount(2) EBUSY)
libmount uses libblkid to detect filesystem type. Unfortunately, the
blkid probe struct is not freed before mount(2), it means that the
device is still open and mount(2) may return EBUSY.
We don't need persistent blkid stuff in libmount, so let's close
all immediately after device probing.
Reported-by: David Zeuthen <zeuthen@gmail.com> Signed-off-by: Karel Zak <kzak@redhat.com>
Dave Reisner [Thu, 26 Apr 2012 00:30:52 +0000 (20:30 -0400)]
findmnt: add match_by_file to do within-device matching
Use the newly exported mnt_get_mountpoint to determine the device that a
given file resides on, in case the supplied source or target is not
explicitly a mount point.
Karel Zak [Tue, 24 Apr 2012 09:52:04 +0000 (11:52 +0200)]
libmount: fix mount by pattern
mount /foo /bar
without entry in /etc/fstab the mount command tries all filesystems
from /{etc,proc}/filesystems. We should NOT call mount(2) more then
once if the syscall returns for example ENOENT, acceptable is only
EINVAL.
Petr Uzel [Mon, 23 Apr 2012 08:43:38 +0000 (10:43 +0200)]
lscpu: fix possibly undefined operation
With -Wall -Werror, compilation of lscpu.c fails with:
Making all in sys-utils
make[2]: Entering directory `/home/petr/upstream/util-linux/sys-utils'
gcc -std=gnu99 -DHAVE_CONFIG_H -I. -I.. -include ../config.h -I../include -DLOCALEDIR=\"/usr/share/locale\" -fsigned-char -Wall -Werror -MT lscpu.o -MD -MP -MF .deps/lscpu.Tpo -c -o lscpu.o lscpu.c
lscpu.c: In function ‘print_parsable’:
lscpu.c:971:7: error: operation on ‘p’ may be undefined [-Werror=sequence-point]
cc1: all warnings being treated as errors
Fix by splitting the pointer increment to separate statement.
Petr Uzel [Sun, 22 Apr 2012 14:16:39 +0000 (16:16 +0200)]
libblkid: befs: declare functions static
Since commit 0b0c231f64d02e682
(swapon: move generic code to swapon-common.c), if util-linux is
configured with --enable-static --disable-shared --enable-libmount-mount
options, the build fails with the following error:
...
libtool: link: gcc -std=gnu99 -fsigned-char -ggdb -O1 -Wall -D_FORTIFY_SOURCE=2 -o switch_root switch_root.o
/home/petr/upstream/util-linux/libblkid/src/.libs/libblkid.a(befs.o): In function `get_uuid':
/home/petr/upstream/util-linux/libblkid/src/superblocks/befs.c:333: multiple definition of `get_uuid'
swapon-swapon-common.o:/home/petr/upstream/util-linux/sys-utils/swapon-common.c:99: first defined here
collect2: ld returned 1 exit status
/bin/sh ../libtool --tag=CC --mode=link gcc -std=gnu99 -fsigned-char -ggdb -O1 -Wall -D_FORTIFY_SOURCE=2 -o flock flock.o strutils.o
libtool: link: gcc -std=gnu99 -fsigned-char -ggdb -O1 -Wall -D_FORTIFY_SOURCE=2 -o pivot_root pivot_root.o
make[2]: *** [swapon] Error 1
make[2]: *** Waiting for unfinished jobs....
/home/petr/upstream/util-linux/libblkid/src/.libs/libblkid.a(befs.o): In function `get_uuid':
/home/petr/upstream/util-linux/libblkid/src/superblocks/befs.c:333: multiple definition of `get_uuid'
swapoff-swapon-common.o:/home/petr/upstream/util-linux/sys-utils/swapon-common.c:99: first defined here
getdtablesize() is the number of descriptors and since decriptors start
at 0, its value isnt the largest descriptors possible. One should use
getdtablesize() - 1 instead.
Signed-off-by: Chas Williams - CONTRACTOR <chas@cmf.nrl.navy.mil>
Bernhard Voelker [Thu, 19 Apr 2012 22:10:44 +0000 (00:10 +0200)]
Fix typos found by misspellings
The tool misspellings (https://github.com/lyda/misspell-check)
detected several typos. Command used:
$ git ls-files | grep -v ^po/ | misspellings -f -
* isosize: Fix typo in usage string.
* configure.ac: Fix typo in help string of --enable-most-builds option.
* fdisk: Fix typo in man page.
* libblkid, blkid, mount: Likewise.
* Fix various typos in docs and in source code comments.
Signed-off-by: Bernhard Voelker <mail@bernhard-voelker.de>
fdisk: do not call sgi and sun code when creating a new dos label
When creating a new DOS label, SGI and SUN pt logic have no business being
there. Remove sun/sgi_nolabel(), since the default amount of partitions are
previously set to 4 and the sun/sgi magic nums are not used in this context.
When some program' output exceeds the terminal's dimensions, it is a nice
feature to call a pager that acts as calling 'less' to allow better user
navigation. This patch adds this functionality, based on what perf and git
have (ie: git log).
Karel Zak [Wed, 18 Apr 2012 09:43:42 +0000 (11:43 +0200)]
swapoff: move code from swapon.c to swapoff.c
* multipurpose binaries (symlinks) suck when used with autotools
* swapoff shares small subset of the code with swapon
* usage() and 'struct option' is unnecessary complex when shared
* shorter and simple code is easer to maintain...
Karel Zak [Wed, 11 Apr 2012 10:37:59 +0000 (12:37 +0200)]
Merge branch 'close_stream' of git://github.com/kerolasa/lelux-utiliteetit
* 'close_stream' of git://github.com/kerolasa/lelux-utiliteetit:
disk-utils: verify writing to streams was successful
fdisk: verify writing to streams was successful
getopt: verify writing to streams was successful
hwclock: verify writing to streams was successful
login-utils: verify writing to streams was successful
misc-utils: verify writing to streams was successful
mount: verify writing to streams was successful
partx: verify writing to streams was successful
schedutils: verify writing to streams was successful
sys-utils: verify writing to streams was successful
term-utils: verify writing to streams was successful
text-utils: verify writing to streams was successful
include: add stream error checking facility
Add a random number(s) generator specific file. The intial functions are based
on what libuuid provide. I did some modifications like avoid WIN32 checks - this
is util-LINUX.
[kzak@redhat.com: - move jrand_seed to lib/randutils.c
- use TLS for jrand_seed (like original code from libuuid)
- use size_t for buffer sizes
- add close() to random_get_bytes]
Signed-off-by: Davidlohr Bueso <dave@gnu.org> Signed-off-by: Karel Zak <kzak@redhat.com>
Petr Uzel [Fri, 6 Apr 2012 14:53:13 +0000 (16:53 +0200)]
sfdisk: fix calculation due to type mismatch (ix86)
Instructing sfdisk to create one partition spanning
across entire disk (",,") on 32bit system, if the disk
is sufficiently large (~2TB) leads to wrong calculation
in compute_start_sect() due to type mismatch.
Can be reproduced as:
--------------------------
linux-3ln5:~ # modprobe scsi_debug virtual_gb=2000
linux-3ln5:~ # sfdisk /dev/sda <<< ',,'
Checking that no-one is using this disk right now ...
OK
Disk /dev/sda: 261083 cylinders, 255 heads, 63 sectors/track
sfdisk: ERROR: sector 0 does not have an msdos signature
sfdisk: /dev/sda: unrecognized partition table type
Old situation:
sfdisk: No partitions found
no room for partition descriptor
sfdisk: bad input
--------------------------
Addresses: https://bugzilla.novell.com/show_bug.cgi?id=754789 Reported-by: Dan Mares <dan.mares@norcrossgroup.com> Signed-off-by: Petr Uzel <petr.uzel@suse.cz>
Sami Kerola [Thu, 5 Apr 2012 20:40:49 +0000 (22:40 +0200)]
docs: eject.1: align with Documentation/howto-man-page.txt
* Add spaces after dots (English spacing).
* Add line break macros.
* Use URL and email macros where appropriate.
* Fix typo at the end of the manual (s/lsblk/eject/).
Mike Frysinger [Thu, 5 Apr 2012 15:50:46 +0000 (11:50 -0400)]
eject: use CDROM_DRIVE_STATUS if available for tray toggling
One some platforms, the -T option can be unreliable (see reference bug
report for some examples). Instead, if the kernel supports the cdrom
status ioctl, use that to ask explicitly for the current tray status
and then open/close accordingly.
The eject_cdrom() func was reworked slightly, but none of the existing
callers care about the explicit normalization to [0,1] values, so have
it return the raw value so we can convert toggle_tray() over to using
that.
Finally, now that toggle_tray() uses a lot of helper functions, drop
the check on CDROMCLOSETRAY. The sub-functions take care of that.
Reference: https://bugs.gentoo.org/261880 Signed-off-by: Mike Frysinger <vapier@gentoo.org>
Dave Reisner [Tue, 27 Mar 2012 12:53:58 +0000 (08:53 -0400)]
fsck: don't free fstab table on parse failure
Even if we fail parsing, fstab gets referenced later in the code (and
will subsequently crash via heap corruption). Take the easy way out and
simply avoid deallocating this table, as it will be freed for us on
program exit regardless.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Karel Zak [Thu, 5 Apr 2012 13:58:02 +0000 (15:58 +0200)]
Merge branch 'eject'
* eject: (47 commits)
tests: use eject --force
eject: use BUILD_EJECT, move to sys-utils
eject: clean up usage()
eject: clean up man page
eject: add --force option
eject: check for hotplug/removable attribute
eject: remove obsolete code, use EXIT_* macros
tests: add umount-by-eject tests
eject: improve work with partitioned devices
lib/sysfs: improve sysfs_is_partition_dirent()
eject: call umount <mountpoint> rather than <device>
eject: use libmount to detect if cdrom is mounted
eject: make the code robust
eject: use SG_IO ioctl for scsi
eject: support CDIOCEJECT ioctl
eject: new close_tray code from Fedora
eject: new auto_eject code from Fedora
eject: add -X from Fedora
eject: add --manualeject from fedora
eject: clean up devname usage
...
Karel Zak [Thu, 5 Apr 2012 10:06:23 +0000 (12:06 +0200)]
eject: check for hotplug/removable attribute
* based on code from Fedora
* supported are only:
- /sys/dev/block/<maj:min>/removable devices
- devices connected by usb, ieee1394, pcmcia, mmc or ccw subsystem.
Frankly, this heuristic sucks -- all is hot-pluggable in this days,
but it's probably good idea to restrict the default behavior to avoid
user's mistakes.
projects / thirdparty / util-linux.git / log
