Rafał Miłecki [Fri, 13 Apr 2018 11:26:14 +0000 (13:26 +0200)]
libfstools: fix foreachdir() to pass dir with a trailing slash
Commit
cc63723d886fd ("overlay: use lstat rather than stat and make sure
there are no trailing spaces") changed behavior of foreachdir() breaking
some callbacks. Before that modification all callbacks were getting
directory with a trailing slash. Above commit started removing them.
This broke handle_whiteout() which doesn't work at all since then. It
constructs file paths incorrectly: slash is missing between directory
and a file name. It seems noone noticed it for years because this issue
got hidden by switch2jffs() which also handles whiteouts with its system
command "cp -a" call.
Fix that regression by setting trailing slash back - right after calling
lstat(). Also to keep code simple just skip all entries that aren't
directories. This keeps conditions for removing/setting trailing slash
trivial. A side effect is not calling callbacks for files which is a
free bonus optimization.
Fixes: cc63723d886fd ("overlay: use lstat rather than stat and make sure there are no trailing spaces")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Acked-by: John Crispin <john@phrozen.org>
(cherry picked from commit
79721f0376974859f39699a2237dbfa691233523)
Rafał Miłecki [Thu, 4 Jan 2018 13:58:18 +0000 (14:58 +0100)]
libfstools: support file paths longer than 255 chars
Alloc globdir buffer dynamically and simply use realloc when needed.
This fixes e.g. segmentation fault in jffs2reset due to an infinite
recurrency when dealing with longs paths.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Daniel Golle [Fri, 30 Jun 2017 15:20:22 +0000 (17:20 +0200)]
libfstools: fix matching device name
compare strlen()+1 characters to make sure we match the trailing \0 as
well. Otherwise things get fishy when using lvm2, see this example:
/dev/mapper/data: UUID="xxxxxx-xxxx-xxxx-xxxx-xxxx-xxxx-xxxxxx" MOUNT="/mnt" TYPE="LVM2_member"
/dev/mapper/data-fs: UUID="xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx" LABEL="xxxxxxxx" VERSION="1.0" MOUNT="/mnt" TYPE="ext4"
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Felix Fietkau [Tue, 9 May 2017 10:37:42 +0000 (12:37 +0200)]
fstools: use -Wno-format-truncation instead of -Wno-error=format-truncation
Fixes build error with older gcc
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Pieter Smith [Wed, 29 Mar 2017 16:21:56 +0000 (18:21 +0200)]
libfstools: fix multiple volume_identify usages with the same volume
This fixes e.g. factory-flashed startup issue with jffs2 on ubi overlay
Commit
ba019965 ("libfstools: accept volume as argument in most calls")
broke startup for factory-flashed jffs2 on ubi systems, causing substantial
slowdown in factory environments.
When starting up with a factory-flashed jffs2 on ubi system, the "rootfs_data"
volume contains a deadcode marker. In the start phase, mount_root then mounts a
tmpfs overlay, and postpones remounting of the jffs2 overlay until the done
phase of the startup.
The refactoring in
ba019965 eliminated an unneeded call to volume_find() when
done() called jffs2_switch(). Unfortunately the refactoring did not take into
account that volume_identify() does not function correctly when called twice in
a row on the same struct volume when using an mtd driver.
mtd_volume_identify() uses mtd_volume_load() to open an fd to the mtd device
and reads a potential deadcode marker from the fd. The first time this works,
and FS_DEADCODE is returned.
When volume_identify() is called a second time however, mtd_volume_load()
notices that we already have an open fd, does nothing further and returns 0
without resetting the file offset to 0. mtd_volume_identify() now reads past
the deadcode marker and now returns FS_JFFS2 if the mtd device is a UBIVOLUME.
jffs2_switch() then handles the wrong case, either pulling the root out from
under user-space in Chaos Calmer, or indefinitely sticking to a tmpfs overlay
in later OpenWRT builds.
Signed-off-by: Pieter Smith <pieter.smith@philips.com>
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Felix Fietkau [Thu, 4 May 2017 14:13:34 +0000 (16:13 +0200)]
build: disable the format-truncation warning error to fix gcc 7 build errors
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Daniel Golle [Thu, 20 Apr 2017 17:50:41 +0000 (19:50 +0200)]
libfstools: silence mkfs.{ext4,f2fs}
Reduce noise during firstboot when creating overlay fs on block
rootdisk devices.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Rafał Miłecki [Mon, 27 Mar 2017 11:17:40 +0000 (13:17 +0200)]
libfstools: add basic documentation of mount functions
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Acked-by: John Crispin <john@phrozen.org>
Felix Fietkau [Sat, 11 Feb 2017 15:01:50 +0000 (16:01 +0100)]
add missing includes
On glibc 2.25, sys/sysmacros.h needs to be included for makedev, major
and minor.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Florian Fainelli [Sun, 4 Dec 2016 23:13:49 +0000 (15:13 -0800)]
libfstools: Check return values for fread and system
libfstools/rootdisk.c: In function 'rootdisk_volume_identify':
libfstools/rootdisk.c:172:7: error: ignoring return value of 'fread', declared with attribute warn_unused_result [-Werror=unused-result]
fread(&magic, sizeof(magic), 1, f);
^
libfstools/rootdisk.c:179:7: error: ignoring return value of 'fread', declared with attribute warn_unused_result [-Werror=unused-result]
fread(&magic, sizeof(magic), 1, f);
^
libfstools/rootdisk.c: In function 'rootdisk_volume_init':
libfstools/rootdisk.c:268:9: error: ignoring return value of 'system', declared with attribute warn_unused_result [-Werror=unused-result]
system(str);
^
cc1: all warnings being treated as errors
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Alberto Bursi [Thu, 27 Oct 2016 20:59:27 +0000 (20:59 +0000)]
fstools: added f2fs to block
added code to block so it can recognize and operate the filesystem checker of f2fs
added f2fs to the filesystem whitelist of block so it can mount it on
/overlay at boot.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Jo-Philipp Wich [Fri, 21 Oct 2016 14:53:57 +0000 (16:53 +0200)]
block: fall back to external mount helper
If the mount(2) syscall fails with ENOENT, attempt to mount the filesystem
using an external "/sbin/mount.$fstype" command.
This allows filesystems which do not have direct kernel support, like fuse-
or network filesystems.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Fri, 21 Oct 2016 13:46:25 +0000 (15:46 +0200)]
block: fix error reporting
The current block code wrongly reported the return value of the mount() and
umount2() syscalls, which is always -1 in case the call failed.
Use errno and strerror(errno) instead to propagate the correct error code to
the user.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Mon, 17 Oct 2016 11:12:31 +0000 (13:12 +0200)]
libblkid-tiny: remove unused name member
The "name" member of struct blkid_struct_probe is not used anywhere anymore
so remove it from the definition to save some stack and heap space.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Mon, 17 Oct 2016 11:10:07 +0000 (13:10 +0200)]
probe: stop handling name property
In the full liblkid there is no tag called "NAME" and the name value reported
by libblkid-tiny was just the kind of uuid passed to blkid_probe_set_uuid_as().
Strip any handling of this property to allow removing it from libblkid-tiny.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Mon, 17 Oct 2016 11:06:57 +0000 (13:06 +0200)]
block: remove handling of name property
There is no concept of a "NAME" property within libblkid, the value previously
used was the name parameter of blkid_probe_set_uuid_as() which does not denote
the file system name, but the kind of UUID to store.
Since the value never makes sense, stop using it when reporting block
information.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Mon, 17 Oct 2016 09:40:14 +0000 (11:40 +0200)]
libblkid-tiny: fix SquashFS version detection on different endian systems
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Mon, 17 Oct 2016 09:28:31 +0000 (11:28 +0200)]
libblkid-tiny: avoid setting phantom UUIDs
When blkid_probe_set_uuid_as() is invoked with a non-NULL name parameter
then the name parameter denotes the kind of UUID (like "EXT_JOURNAL") not
the name of the file system.
Only copy the UUID value to the probe uuid member if the given name is
either NULL or "UUID".
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Sun, 16 Oct 2016 22:05:15 +0000 (00:05 +0200)]
probe: add full libblkid support
Attempt to dlopen() libblkid.so at runtime and use it for proping
filesystems if available.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Sun, 16 Oct 2016 20:45:59 +0000 (22:45 +0200)]
block: add probe abstraction layer
Add an abstraction layer which separates block.c from libblkid-tiny implementation
details in order to prepare support for optionally using the full libblkid.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Daniel Golle [Fri, 23 Sep 2016 03:10:42 +0000 (05:10 +0200)]
libfstools: properly label ext4 overlay
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Fri, 23 Sep 2016 03:10:05 +0000 (05:10 +0200)]
block: also probe loop devices
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Sun, 24 Jul 2016 18:44:38 +0000 (20:44 +0200)]
block: include mountpoint in info output
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Sun, 11 Sep 2016 14:52:52 +0000 (16:52 +0200)]
libfstools: gather mountpoints from /proc/self/mountinfo
This allows identifying /dev/root by its major:minor number which
are part of /proc/self/mountinfo but aren't contained in /proc/mounts.
Also fix jffs2reset when using an ext4 overlay by adding it to the
list of filesystems allowed if the root_only parameter of
find_mount_point is set.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Felix Fietkau [Tue, 6 Sep 2016 14:37:13 +0000 (16:37 +0200)]
libfstools: add rootdisk overlay volume support
Uses either ext4 or f2fs, depending on the overlay volume size
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 11:03:26 +0000 (13:03 +0200)]
libfstools: replace hardcoded mentions of jffs2 in a few places
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 10:09:34 +0000 (12:09 +0200)]
libfstools: add ext4 filesystem type
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 10:02:08 +0000 (12:02 +0200)]
libfstools: add f2fs filesystem type and simplify fs type code
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Tue, 6 Sep 2016 14:14:35 +0000 (16:14 +0200)]
libfstools: use container_of for volume private data
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Wed, 7 Sep 2016 15:30:09 +0000 (17:30 +0200)]
libfstools: call volume_init() before accessing v->blk
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Tue, 6 Sep 2016 11:52:33 +0000 (13:52 +0200)]
libfstools: remove obsolete enum
Signed-off-by: Felix Fietkau <nbd@nbd.name>
John Crispin [Wed, 31 Aug 2016 16:16:31 +0000 (18:16 +0200)]
fix logic bug inside extroot uuid verification code
Signed-off-by: John Crispin <john@phrozen.org>
Преподобный Гомер [Wed, 17 Aug 2016 11:51:31 +0000 (14:51 +0300)]
libfstools: "lowerdir" variable name in fopivot
Hi all!
In fopivot in libfstools variable name "lowerdir" is confusing and
doesn't make any sense.
There is a patch renaming that var to "mount_options".
John Crispin [Tue, 16 Aug 2016 10:06:33 +0000 (12:06 +0200)]
fix return code of overlay_mount_fs
Signed-off-by: John Crispin <john@phrozen.org>
Daniel Golle [Sun, 24 Jul 2016 15:54:31 +0000 (17:54 +0200)]
block: best-effort in find_mount_point
Don't immediatly fail if block device cannot be stat'ed.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Sun, 24 Jul 2016 13:16:16 +0000 (15:16 +0200)]
block: get mountpoints from /proc/self/mountinfo
Instead of using two different implementations reading /proc/mounts
and /proc/self/mountinfo, gather all information on mountpoints from
/proc/self/mountinfo.
While at it, tokenize mountinfo inline instead of using strtok_r
because it's hard to tell whether strtok_r allocated new memory or
merely returns a pointer to the (last) token. This might later on be
relevant once we want to free that memory...
Tokenizing mountinfo using strchr inline works without allocating any
new memory by scanning and re-writing the string returned by gets().
As a result, the returned string needs to be duplicated to be safe for
use even after the fclose().
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Thu, 21 Jul 2016 02:04:34 +0000 (04:04 +0200)]
block: print mountpoint if already mounted
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Thu, 21 Jul 2016 01:55:48 +0000 (03:55 +0200)]
block: also check /proc/self/mountinfo to find mountpoint
Matching only the device name doesn't always work, e.g. in case of
/dev/root. Thus also check the device(minor,major) which can be scraped
from /proc/self/mountinfo.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Thu, 21 Jul 2016 01:19:22 +0000 (03:19 +0200)]
block: don't truncate mountpoint path
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Wed, 20 Jul 2016 23:57:50 +0000 (01:57 +0200)]
block: allow block info /dev/ubi?_?
Allow querying /dev/ubi?_? devices though they are not block devices.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Wed, 20 Jul 2016 22:58:08 +0000 (00:58 +0200)]
block: avoid ubi{,block} duplicates
Skip ubi?_? device if ubiblock?_? is present.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Wed, 20 Jul 2016 23:20:10 +0000 (01:20 +0200)]
block: fix ubi?_? glob pattern
The glob pattern intended for ubi?_? devices was matching
ubiblock?_? devices as well. Change the glob to actually return only
ubi devices.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Florian Fainelli [Fri, 1 Jul 2016 23:06:42 +0000 (16:06 -0700)]
cmake: Link against libjson-c
block uses libblob_msgjson which requires us to link against libjson-c.
Some external toolchains would be failing to find that library unless
specified explicitly.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Florian Fainelli [Fri, 1 Jul 2016 23:06:41 +0000 (16:06 -0700)]
cmake: Find libubox/ulog.h
Add a CMake FIND_PATH and INCLUDE_DIRECTORIES searching for
libubox/ulog.h. Some external toolchains which do not include standard
locations would fail to find the header otherwise.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Daniel Dickinson [Sun, 22 May 2016 09:30:38 +0000 (05:30 -0400)]
block.c: Add ability to mount with ACL and XATTR support
Some users will want to use OpenWrt/LEDE devices as NAS
devices and have full POSIX ACL and user_xattr support
(along with other possible use cases), therefore add
support to mount with POSIX ACLs and/or user XATTR
support.
Signed-off-by: Daniel Dickinson <lede@daniel.thecshore.com>
Daniel Dickinson [Sun, 22 May 2016 09:30:37 +0000 (05:30 -0400)]
block.c: Use <linux/fs.h> instead of defining mount flags ourselves
Let's use the cannonical source of mount flags instead of
defining the flags ourselves.
Signed-off-by: Daniel Dickinson <lede@daniel.thecshore.com>
Daniel Dickinson [Sun, 22 May 2016 09:22:48 +0000 (05:22 -0400)]
block.c: Add support for checking vfat filesystems
vfat is a common filesystem which users may want to mount on an
OpenWrt/LEDE device, so support peforming filesystem checks
before mount for vfat.
Signed-off-by: Daniel Dickinson <lede@daniel.thecshore.com>
Daniel Dickinson [Sun, 22 May 2016 09:22:47 +0000 (05:22 -0400)]
block.c: Make static string a const char * instead char *
There is no reason for e2fsck string to be altered, and the
only places where is used take const char * as parameters
so make e2fsck a const char *.
Signed-off-by: Daniel Dickinson <lede@daniel.thecshore.com>
John Crispin [Sun, 15 May 2016 14:20:20 +0000 (16:20 +0200)]
mount_root: check for preinit sentinel file
Signed-off-by: John Crispin <john@phrozen.org>
John Crispin [Sun, 15 May 2016 14:13:30 +0000 (16:13 +0200)]
Revert "mount_root: code failed in failsafe"
This reverts commit
09f8e0097fe1370700658542e7bda831c31de4bf.
John Crispin [Wed, 11 May 2016 16:07:36 +0000 (18:07 +0200)]
Revert "fstools: support for ext4fs overlay"
This reverts commit
bf4f08af9b8fe991216023a906182744eb1e4645.
this patch was breaking devices that have a normal ext4 rootfs
Signed-off-by: John Crispin <john@phrozen.org>
John Crispin [Tue, 10 May 2016 13:39:37 +0000 (15:39 +0200)]
mount_root: code failed in failsafe
the code checks if we are in PREINIT before mounting root.
change this to check if root is actually mounted.
this is a regression caused by ssh login during failsafe.
Signed-off-by: John Crispin <john@phrozen.org>
John Crispin [Wed, 27 Apr 2016 05:34:04 +0000 (07:34 +0200)]
snapshot: md5sum return codes are not properly handled
Signed-off-by: John Crispin <john@phrozen.org>
John Crispin [Wed, 27 Apr 2016 01:30:40 +0000 (03:30 +0200)]
block: add remount paramter
Signed-off-by: John Crispin <john@phrozen.org>
Ram Chandra Jangir [Fri, 11 Mar 2016 16:31:42 +0000 (22:01 +0530)]
fstools: support for ext4fs overlay
This change will enables eMMC (ext4 fs) boot support, when we try to boot
from eMMC card then it will read partition names from
/sys/block/mmcblkX/mmcblkXY/uevent
file and will mount the rootfs_data partition as ext4fs overlay.
Signed-off-by: Ram Chandra Jangir <rjangi@codeaurora.org>
Jo-Philipp Wich [Mon, 25 Apr 2016 15:49:38 +0000 (17:49 +0200)]
block: respect mount flags for /overlay
Instead of hardcoding no-flags in the mount syscall, use the user configured
values, this is useful when mounting external overlays with "sync" or "ro"
flags.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Felix Fietkau [Sun, 10 Jan 2016 17:35:36 +0000 (18:35 +0100)]
fstools: check existing filesystem state before writing the new one
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Roman Yeryomin [Sat, 19 Dec 2015 21:38:02 +0000 (23:38 +0200)]
fstools: allow to stack another overlay on top of existing one
`mount_root ram' will pivot existing root to ram even if it was
overlayfs already. Useful when playing with new configurations
as it allows to preserve existing/stable configuration.
Signed-off-by: Roman Yeryomin <roman@advem.lv>
Hauke Mehrtens [Sun, 22 Nov 2015 14:38:01 +0000 (15:38 +0100)]
fix gcc format security error.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Martin Blumenstingl [Sat, 31 Oct 2015 15:38:42 +0000 (16:38 +0100)]
libblkid-tiny: Add F2FS support
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Martin Blumenstingl [Sat, 31 Oct 2015 15:38:41 +0000 (16:38 +0100)]
libblkid-tiny: Update the code from util-linux's libblkid
This updates the relevant source files to util-linux changeset
4419ffb9eff5801fdbd385a4a6199b3877f802ad.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
John Crispin [Wed, 26 Aug 2015 01:54:46 +0000 (03:54 +0200)]
overlay: use lstat rather than stat and make sure there are no trailing spaces
with this patch even uclibc wont deleted symlinked folders
Signed-off-by: John Crispin <blogic@openwrt.org>
John Crispin [Tue, 25 Aug 2015 20:43:18 +0000 (22:43 +0200)]
make jffs2reset not iterate over symlinked folders
Signed-off-by: John Crispin <blogic@openwrt.org>
John Crispin [Mon, 24 Aug 2015 08:11:47 +0000 (10:11 +0200)]
jffs2reset: add a -r (reboot) option to jffs2reset
Signed-off-by: John Crispin <blogic@openwrt.org>
Felix Fietkau [Mon, 25 May 2015 21:55:07 +0000 (23:55 +0200)]
mount_root: use symlink instead of xattr to store initialization state
xattr seems to be unreliable with jffs2
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Rafał Miłecki [Sat, 11 Apr 2015 07:56:44 +0000 (09:56 +0200)]
jffs2reset: use jffs2_mark if rootfs_data isn't mounted
Erasing all rootfs_data blocks may cause some problems with partition
identification. It won't contain MAGIC, but will be successfully mounted
with delayed blocks marking. This may be really confusing when user
reboots before JFFS2 finishes its blocks management. During the next
boot rootfs_data will be a valid partition (possibly with data) but
libblkid won't detect it.
Also adjust message in jffs2_mark to make more sense when used together
with jffs2_reset.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Rafał Miłecki [Sat, 11 Apr 2015 07:56:43 +0000 (09:56 +0200)]
jffs2reset: avoid code duplication in jffs2_reset and jffs2_mark
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Jo-Philipp Wich [Sun, 12 Apr 2015 20:18:15 +0000 (22:18 +0200)]
block: reject mount sections with relative targets
Specifying relative target paths leads to undefined results and confuses
users. Explicitely reject relative paths with a warning instead of doing
random stuff.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
John Crispin [Thu, 9 Apr 2015 20:53:59 +0000 (22:53 +0200)]
make sure to also scan for leading and ending " when looking for mtd devices
Signed-off-by: John Crispin <blogic@openwrt.org>
John Crispin [Sat, 28 Mar 2015 17:32:22 +0000 (18:32 +0100)]
properly handle return codes
Signed-off-by: John Crispin <blogic@openwrt.org>
Felix Fietkau [Sun, 22 Mar 2015 14:59:33 +0000 (15:59 +0100)]
mount_root: keep track of overlay initialization state (via xattr)
Remove all files (except sysupgrade.tgz) if /etc/init.d/done was not called
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Felix Fietkau [Sun, 22 Mar 2015 13:11:39 +0000 (14:11 +0100)]
overlay: call volume_init() before mount
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Felix Fietkau [Sun, 22 Mar 2015 13:08:27 +0000 (14:08 +0100)]
overlay: create a common function for deleting all overlay data (optionally excluding sysupgrade.tgz)
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
John Crispin [Thu, 12 Mar 2015 12:57:41 +0000 (13:57 +0100)]
switch to _DEFAULT_SOURCE for modern glibc compat
Signed-off-by: Jeff Waugh <jdub@bethesignal.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 21:57:57 +0000 (22:57 +0100)]
mount_root: better log messages for ram overlay cases
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 21:57:34 +0000 (22:57 +0100)]
libfstools: clarify message when switching to internal overlay
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 21:56:51 +0000 (22:56 +0100)]
libfstools: make mtd_volume_identify() less chatty
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 21:42:17 +0000 (22:42 +0100)]
libfstools: remove dead source file
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 20:43:54 +0000 (21:43 +0100)]
libfstools: convert to ulog() api
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 20:32:45 +0000 (21:32 +0100)]
ubi: convert to ulog() api
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 20:28:12 +0000 (21:28 +0100)]
snapshot: convert to ulog() api
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 20:25:35 +0000 (21:25 +0100)]
jffs2reset: convert to ulog() api
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 20:16:56 +0000 (21:16 +0100)]
mount_root: convert to ulog() api
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 20:09:25 +0000 (21:09 +0100)]
block: convert to ulog() api
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 18:50:40 +0000 (19:50 +0100)]
block: don't declare find_root_dev with UBIFS_EXTROOT
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 25 Feb 2015 12:44:40 +0000 (13:44 +0100)]
libfstools: extroot: simplify kmodloader invocation
Now that kmodloader supports LD_LIBRARY_PATH and ignores argv[2] we can
remove the extra argument and save one intermediate pointer.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 18 Feb 2015 18:50:37 +0000 (19:50 +0100)]
libfstools: extroot: rework libdir and block tool discovery
Account for new upper/ directory component in recent overlayfs versions.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 18 Feb 2015 18:28:59 +0000 (19:28 +0100)]
block: improve extroot error reporting
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 18 Feb 2015 18:28:35 +0000 (19:28 +0100)]
block: rework config loading
Account for the new upper/ directory component in recent overlayfs version.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Wed, 18 Feb 2015 18:25:36 +0000 (19:25 +0100)]
block: add support for logging to dmesg
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Tue, 17 Feb 2015 19:27:59 +0000 (20:27 +0100)]
block: support finding root device by mount point
When checking extroot, fall back to using stat(/) to determine the root
device node if no rootfs MTD partition can be found, e.g. on systems using
an ext4 root filesystem.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Tue, 17 Feb 2015 19:19:59 +0000 (20:19 +0100)]
block: extroot: remove unneeded rootfs tests
The code testing for a "rootfs" MTD partition serves no actual purpose and
the test results are not used either. On top of that it prevents working
extroot on non-MTD systems, so drop the code.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Tue, 17 Feb 2015 19:13:43 +0000 (20:13 +0100)]
block: support builtin fstab config
Fall back to /etc/config/fstab if the file on the overlay cannot be loaded.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Tue, 17 Feb 2015 18:50:00 +0000 (19:50 +0100)]
libfstools: extroot: support builtin block
Fix regression introduced with migration to fs-tools.
The old ubox based exroot did a fall back to /sbin/block if $cfg/sbin/block
did not exist.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Jo-Philipp Wich [Tue, 17 Feb 2015 18:18:27 +0000 (19:18 +0100)]
mount_root: try extroot even without mtd volume
This is a precondition for supporting external overlays on ext4 root fs.
Also rename volume pointer variables for better clarity.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Felix Fietkau [Sun, 11 Jan 2015 16:52:46 +0000 (17:52 +0100)]
jffs2reset: fix recursive file/directory removal
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Daniel Golle [Mon, 22 Dec 2014 11:34:42 +0000 (12:34 +0100)]
libblkid-tiny: add btrfs probe info
This only adds the magic for btrfs to show btrfs volumes in
block detect.
In order to properly support btrfs with more than one device and avoid
trying to mount the same device group once for each instance we have to
make sure that only one device per group will be mounted (btrfs then
automatically detects all other devices of the group).
For unmounting, the opposite should be applied: umount if called for
any of the devices in a mounted device group, unmount the filesystem.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Rafał Miłecki [Tue, 23 Dec 2014 21:20:13 +0000 (22:20 +0100)]
libfstools: ubi: rework reading volumes
UBI volume IDs don't have to be contiguous, this may happen if there
were few and the not last one was removed. In such case we may have
e.g. ubi0_0, ubi0_2.
It means we can't simply read value from "volumes_count" and iterate
from 0 to the count - 1.
This patch rewrites ubi_part_match to read whole directory and look
for ubi%d_%d entries in it.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Rafał Miłecki [Tue, 23 Dec 2014 20:33:49 +0000 (21:33 +0100)]
libfstools: ubi: avoid crash in volume finding function
In case /sys/devices/virtual/ubi/ubi?/ubi?_?/name doesn't exist volname
will be NULL and calling strlen for NULL isn't a good idea.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Rafał Miłecki [Tue, 23 Dec 2014 11:56:52 +0000 (12:56 +0100)]
libfstools: fix find_mount to return device path
So far we were returning mount point which doesn't make sense for a
function that already takes mount point as an argument.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>