* '2012wk23' of git://github.com/kerolasa/lelux-utiliteetit:
lsblk: use blkdev_scsi_type_to_name()
blkdev: add blkdev_scsi_type_to_name()
wipefs: use symbolic value for markup mode
eject: inform if CD-ROM drive is not ready
docs: clean up partx.8 manual
include: fix void pointer arithmetics warnings
sysfs: fix printf format warnings
build: fix unused parameter warnings
build: fix redundant redeclaration warnings
include: fix spurious list.h warnings
uuidd: use output redirection which works [checkbashisms]
blkid: fix realloc memory leak [cppcheck]
setarch: do not use -1 as array index [cppcheck]
valgrind --leak-check=full ./sys-utils/mount -t cifs //127.0.0.1/users /mnt/smb -o user=root,password=linux
....
==21359== Invalid read of size 1
==21359== at 0x415AC6: mnt_optstr_remove_option_at (optstr.c:310)
==21359== by 0x416358: mnt_optstr_apply_flags (optstr.c:716)
==21359== by 0x40DFBF: mnt_context_prepare_mount (context_mount.c:86)
==21359== by 0x40EB5A: mnt_context_mount (context_mount.c:782)
==21359== by 0x4058B0: main (mount.c:918)
==21359== Address 0x51cd5bf is 1 bytes before a block of size 10 alloc'd
==21359== at 0x4C297CD: malloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==21359== by 0x4C29957: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==21359== by 0x415780: __mnt_optstr_append_option (optstr.c:188)
==21359== by 0x412822: mnt_fs_append_options (fs.c:764)
==21359== by 0x409288: mnt_context_append_options (context.c:733)
==21359== by 0x4053F0: main (mount.c:776)
Signed-off-by: Petr Uzel <petr.uzel@suse.cz>
valgrind --leak-check=full ./sys-utils/mount -t cifs //127.0.0.1/users /mnt/smb -o user=root,password=linux
....
==21359== 28 bytes in 1 blocks are definitely lost in loss record 1 of 1
==21359== at 0x4C298B2: realloc (in /usr/lib64/valgrind/vgpreload_memcheck-amd64-linux.so)
==21359== by 0x415780: __mnt_optstr_append_option (optstr.c:188)
==21359== by 0x415CB5: mnt_optstr_set_option (optstr.c:387)
==21359== by 0x40D778: do_mount (context_mount.c:192)
==21359== by 0x40E6A9: mnt_context_do_mount (context_mount.c:685)
==21359== by 0x40EB7B: mnt_context_mount (context_mount.c:786)
==21359== by 0x4058B0: main (mount.c:918)
Signed-off-by: Petr Uzel <petr.uzel@suse.cz>
mount -t cifs //127.0.0.1/users /mnt/smb -o user=root,password=linux
is incorrectly translated to
mount.cifs -o noexec,nosuid,nodev,user=root,password=linux ...
The command mount(8) should be sensitive to "user" (without "=<name>")
only. The correct cifs command line is:
mount.cifs -o user=root,password=linux
Addresses: https://bugzilla.novell.com/show_bug.cgi?id=766157
Signed-off-by: Karel Zak <kzak@redhat.com>
mount -t foo something /mnt/test -o user=root,password=linux
generates
"rw,noexec,nosuid,nodev,password=linux,,user=root"
options string for /sbin/mount.foo, the ",," is incorrect.
Signed-off-by: Karel Zak <kzak@redhat.com>
Fixes a bug in option string parsing wherein a line such as:
ro,relatime,,nosuid,nodev
Will be seen as only the tokens "ro" and "relatime" after the parser
encounters a zero length (and erroneously declared NULL) option.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
It's the responsibility of anyone adding to this list in the future to
ensure that the list remains sorted.
While we're at it, expand this list of known pseudofs types.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Note that mountpoint (target_ paths in /proc/mounts and /proc/self/mountinfo
are always canonicalized by kernel.
* for umount we don't have to canonicalize target
by default if the mountpoint is found in /proc/self/mountinfo
* in mnt_table_find_target() is unnecessary to canonicalize target paths
if the table of the filesystems is read from /proc/self/mountinfo
Addresses: https://bugzilla.redhat.com/show_bug.cgi?id=820707
Signed-off-by: Karel Zak <kzak@redhat.com>
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>
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.
Signed-off-by: Karel Zak <kzak@redhat.com>
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>
Provide the recursive flag of propagation mounting. Recursive flag could
be used together with propagation flag.
[kzak@redhat.com: - add MS_SILENT to acceptable propagation flags,
- fix new code in libmount too]
Signed-off-by: Karel Zak <kzak@redhat.com>
Signed-off-by: Dong Hao <haodong@linux.vnet.ibm.com>
For systems with regular mtab the fs->root should be ignored in
"already mounted" heuristic.
Reported-by: Matt Burgess <matthew@linuxfromscratch.org>
Signed-off-by: Karel Zak <kzak@redhat.com>
In early userspace, the rootfs is mounted with itself as its parent.
Example /proc/self/mountinfo:
1 1 0:1 / / rw - rootfs rootfs rw
14 1 0:3 / /proc rw,nosuid,nodev,noexec,relatime - proc proc rw
15 1 0:13 / /sys rw,nosuid,nodev,noexec,relatime - sysfs sys rw
16 1 0:5 / /dev rw,nosuid,relatime - devtmpfs dev rw,size=506956k
17 1 0:14 / /run rw,nosuid,nodev,relatime - tmpfs run rw,mode=755
This causes an infinite loop in mnt_table_next_child_fs, and is
evidenced by a crash via infinite recursion in findmnt. Simply catch the
condition where the ID of the parent is the ID of the current fs we're
examining and skip over it to the next mount in the table.
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
We have to be careful with "none" or another dummy sources for pseudo
filesystems. These strings should be canonicalized or compared as a
paths.
The function is not exported by library API.
Signed-off-by: Karel Zak <kzak@redhat.com>
This causes more problems than it solves. In the latest edition:
# mount -t proc none foo
mount: foo: mount failed: Invalid argument
A check for source and target fails in mnt_context_apply_fstab()
because, even though they were indeed specified on the cmdline,
__mnt_fs_set_source_ptr() altered this and NULL'd out the source.
If you're able to mount this device via other means, other tools start
reporting oddities, such as mount's output:
(null) on /foo type proc (rw,relatime)
or findmnt:
TARGET SOURCE FSTYPE OPTIONS
/foo proc rw,relatime
Simply treat "none" like any other source when passed in.
[kzak@redhat.com: - don't translate NULL to "none" in mnt_fs_set_source()]
Signed-off-by: Dave Reisner <dreisner@archlinux.org>
Signed-off-by: Karel Zak <kzak@redhat.com>
Several horizontal lists are turned to vertical, and sorted to
alphabetical order. Additionally spaces are converted to tabs where
ever possible.
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
Unfortunately, it seems that for example mount.cifs don't care about
the API, so we need exception like the original mount(8).
Signed-off-by: Karel Zak <kzak@redhat.com>
mnt_table_find_target() canonicalizes paths by libmnt cache, but it's
overkill if the difference between paths is tailing slash only.
Signed-off-by: Karel Zak <kzak@redhat.com>
Karel Zak
Petr Uzel
Sami Kerola
Dave Reisner
Bernhard Voelker
Dong Hao
Dave Reisner