summaryrefslogtreecommitdiffstats
path: root/shlibs
Commit message (Collapse)AuthorAgeFilesLines
* build-sys: use top-level directory for libmount rather than shlibs/mountKarel Zak2011-06-0935-14266/+0Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* build-sys: use top-level directory for libuuid rather than shlibs/uuidKarel Zak2011-06-0932-2662/+0Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: fix uninitialized variable in sampleKarel Zak2011-06-082-2/+3
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: use mnt_table_get_fs_root() in utab codeKarel Zak2011-06-024-115/+57Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: add mnt_table_is_mounted()Karel Zak2011-05-314-1/+215
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: use sysfs_init() more carefullyKarel Zak2011-05-303-17/+16Star
| | | | | Reported-by: Francesco Cosoleto <cosoleto@gmail.com> Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: add some debug messagesKarel Zak2011-05-301-2/+13
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: improve blkid__scan_dirKarel Zak2011-05-241-2/+6
|
* libblkid: add ID_PART_ENTRY_{OFFSET,SIZE,DISK}Karel Zak2011-05-232-1/+10
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: use partno for partitions mapped by DMKarel Zak2011-05-231-4/+44
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: found whole-disk for partitions mapped by kpartxKarel Zak2011-05-231-97/+119
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* lib: [sysfs.c] make sysfs_read_* function more robustKarel Zak2011-05-183-7/+17
| | | | | | | The functions does not modify result if the requested sysfs attribute does not exist. Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: use stuff from sysfs.h and at.hKarel Zak2011-05-166-167/+67Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: use cached buffers for nested PT probingKarel Zak2011-05-162-23/+41
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: make whole disk probing more robustKarel Zak2011-05-163-63/+103
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: cleanup prober initializationKarel Zak2011-05-162-2/+12
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: cleanup flagsKarel Zak2011-05-164-20/+21
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* blkid: don't read past end of FAT32 cluster chainJohn Lindgren2011-05-121-1/+3
| | | | | | | | | | | | | | | | | | | | | | | When looking for a volume label on a FAT32 filesystem, blkid does not correctly handle special cluster values in the range 0x0FFFFFF8 to 0x0FFFFFFF, which mark the last cluster in a chain. As a result, it begins to read [what it treats as] FAT entries from past the end of the FAT. Depending on the data read, it may then try to parse random data from the filesystem (including user files, free space, and other directories) as though it were part of the root directory. Because parsing stops early when a volume label is found, the problem only occurs on filesystems without a volume label. When it occurs, it may result in a long sequence of lseek and read calls; on an older IDE drive with a 2 GB FAT32 partition, this amounted to around 3 seconds of disk I/O. After patching, blkid stop parsing as soon as it reaches a cluster value greater than or equal to the number of entries in the FAT. [kzak@redhat.com: - add le32_to_cpu()] Signed-off-by: John Lindgren <john.lindgren@tds.net>
* libblkid: ignore hfsplus superblocks with blocksize < 512Karel Zak2011-05-061-2/+4
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: validate hfs blocksize != 0Timo Warns2011-05-061-0/+3
| | | | | | | Validate that blocksize != 0 as a blocksize == 0 can cause a division-by-zero error. Signed-off-by: Timo Warns <warns@pre-sense.de>
* libmount: minor fix to mnt_tabdiff_*Karel Zak2011-04-274-7/+11
| | | | | | | - tabdiff has to support NULL fs->source - disable some debug messages Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: add mnt_tabdiff_* functionsKarel Zak2011-04-265-2/+398
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: add mnt_reset_table()Karel Zak2011-04-263-5/+26
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: support NULL source path for mnt_table_find_* functionsKarel Zak2011-04-263-16/+33
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: add support for x-* mount commentsKarel Zak2011-04-212-3/+14
| | | | | | | | | | | | | | | | | | | | | | | The current way how add any comment or extra information to /etc/fstab is to use comment="" mount option. This is not too elegant and readable if you want to add more than one comment. It seems better to add any generic prefix that will be used for all 3rd-party userspace mount options. Expected syntax is: x-<application>-<options>[=<data>] For example: x-systemd-bar=managed All x-* options (as well as comment=) are ignored by libmount and not stored to the mtab file. Requested-by: Lennart Poettering <lennart@poettering.net> Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: use libmnt_lock for utab flockKarel Zak2011-04-133-102/+32Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: merge mtab and utab locking codeKarel Zak2011-04-132-94/+178
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: block signals when update utabKarel Zak2011-04-131-48/+81
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: more robust mtab and utab update (CVE-2011-1676, CVE-2011-1677)Karel Zak2011-04-131-39/+73
| | | | | | | | | | | | | | http://thread.gmane.org/gmane.comp.security.oss.general/4374 Changes: - always use temporary file - use fflush() for the temporary file - check fprintf() return value Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: remove test for non-zero head count in FAT superblock probe.Nick Holloway2011-04-121-2/+0Star
| | | | | | | | | | | | | | This change enables the auto-detection of the internal storage on a VTech Kidizoom camera, which has a zero head count value in the superblock. Microsoft documentation, "FAT: General Overview of On-Disk Format", states that the sectors and heads fields are only relevant for media that have a geometry. The Linux kernel also does not require a non-zero head count, and can manually mount the filesystem. Signed-off-by: Nick Holloway <Nick.Holloway@pyrites.org.uk>
* libmount: avoid redundant declaration of mnt_context_do_umountJan Engelhardt2011-04-121-2/+0Star
| | | | Signed-off-by: Jan Engelhardt <jengelh@medozas.de>
* libmount: fix parsing of mountinfo from 2.6.39Karel Zak2011-04-051-8/+21
| | | | | | | | | | The /proc/self/mountinfo file uses " - " field as a separator between optional fields and next fields in the file. The '-' char could be used in the fields (for example in UUIDs), so it's necessary to check for whole " - " string rather than for '-' char only. Reported-by: "Aneesh Kumar K. V" <aneesh.kumar@linux.vnet.ibm.com> Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: add mountpoint(1) implementation to samples/Karel Zak2011-04-013-1/+176
|
* umount: use helper= for all UIDsKarel Zak2011-04-013-5/+5
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: support /run/mount rather than /dev/.mountKarel Zak2011-03-317-16/+28
| | | | | | | | | The /run directory should be preferred on distributions where this directory exists. http://thread.gmane.org/gmane.linux.redhat.fedora.devel/146976 Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: plug memory leak in sample programPetr Uzel2011-03-311-0/+1
| | | | | Signed-off-by: Petr Uzel <petr.uzel@suse.cz> Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: fix some typos and copy&paste mistakes in commentsPetr Uzel2011-03-313-20/+19Star
| | | | Signed-off-by: Petr Uzel <petr.uzel@suse.cz>
* libmount: fix memory leak in cacheKarel Zak2011-03-301-2/+2
| | | | | Reported-by: Petr Uzel <petr.uzel@suse.cz> Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: add phelper= supportKarel Zak2011-03-303-17/+45
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: small lock code cleanupKarel Zak2011-03-303-25/+29
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: use chdir() and NOFOLLOW umount flag for umount operationKarel Zak2011-03-304-3/+114
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: fix testKarel Zak2011-03-302-2/+1Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: cleanup umount codeKarel Zak2011-03-301-38/+29Star
| | | | | | | This patch removes obsolete stuff. The minimal requirement for libmount umount code is Linux >= 2.4.11. Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: block signals when writing to mtabKarel Zak2011-03-305-26/+73
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: fix leak in sample programKarel Zak2011-03-281-1/+1
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: create a default lock for mtab updateKarel Zak2011-03-281-1/+10
| | | | | | | | | Now it's necessary to explicitly create a lock for mnt_update_table(). It seems better to create the lock in mnt_update_table() by default to skip this extra step. Reported-by: Jan Engelhardt <jengelh@medozas.de> Signed-off-by: Karel Zak <kzak@redhat.com>
* libmount: rename mount option "quiet" to "silent"Karel Zak2011-03-182-3/+3
| | | | | | See commit 2ac1890aa0d8ce404528587134fcabe0e2834963. Signed-off-by: Karel Zak <kzak@redhat.com>
* libblkid: add blkid_evaluate_spec()Karel Zak2011-03-173-8/+50
| | | | | | | | | | | | The function blkid_evaluate_tag() is useful for tags only (e.g. LABEL=foo). But we also need to address devices by tags OR paths in many utils. The function blkid_evaluate_spec() support this functionality without extra care about the way how device is addressed. The tags as well as paths are converted to the standardized device path. Signed-off-by: Karel Zak <kzak@redhat.com>
* build-sys: include <uuid.h> rather than <uuid/uuid.h>Karel Zak2011-03-161-5/+1Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>
* build-sys: don't support external (e2fsprogs) libuuidKarel Zak2011-03-161-6/+1Star
| | | | Signed-off-by: Karel Zak <kzak@redhat.com>