summaryrefslogtreecommitdiff
path: root/util-linux/Config.in
diff options
context:
space:
mode:
Diffstat (limited to 'util-linux/Config.in')
-rw-r--r--util-linux/Config.in59
1 files changed, 24 insertions, 35 deletions
diff --git a/util-linux/Config.in b/util-linux/Config.in
index 7007915..7fde019 100644
--- a/util-linux/Config.in
+++ b/util-linux/Config.in
@@ -323,54 +323,43 @@ config CONFIG_UMOUNT
the tool to use. If you enabled the 'mount' utility, you almost certainly
also want to enable 'umount'.
-config CONFIG_FEATURE_MOUNT_FORCE
- bool " Support forced filesystem unmounting"
- default n
- depends on CONFIG_UMOUNT
- help
- This allows you to _force_ a filesystem to be umounted. This is generally
- only useful when you want to get rid of an unreachable NFS system.
-
comment "Common options for mount/umount"
depends on CONFIG_MOUNT || CONFIG_UMOUNT
config CONFIG_FEATURE_MOUNT_LOOP
- bool " Support for loop devices"
+ bool " Support loopback mounts"
default n
depends on CONFIG_MOUNT || CONFIG_UMOUNT
help
- Enabling this feature allows automatic loopback mounts, meaning you can mount
- filesystems contained in normal files as well as in block devices. The mount
- and umount commands will detect you are trying to mount a file instead of a
- block device, and transparently associate it with a loopback device (and free
- the loopback device on unmount) for you.
+ Enabling this feature allows automatic mounting of files (containing
+ filesystem images) via the linux kernel's loopback devices. The mount
+ command will detect you are trying to mount a file instead of a block
+ device, and transparently associate the file with a loopback device.
+ The umount command will also free that loopback device.
- You can still use the 'losetup' utility and mount the loopback device yourself
- if you need to do something advanced, such as specify an offset or cryptographic
- options to the loopback device.
-
-config CONFIG_FEATURE_MOUNT_LOOP_MAX
- int " max number of loop devices"
- default 7
- depends on CONFIG_FEATURE_MOUNT_LOOP
- help
- This option sets the highest numbered loop device to be used
- automatically by the '-o loop' feature of mount.
+ You can still use the 'losetup' utility (to manually associate files
+ with loop devices) if you need to do something advanced, such as
+ specify an offset or cryptographic options to the loopback device.
+ (If you don't want umount to free the loop device, use "umount -D".)
config CONFIG_FEATURE_MTAB_SUPPORT
- bool " Support for a /etc/mtab file (instead of symlink to /proc/mounts)"
+ bool " Support for the old /etc/mtab file"
default n
depends on CONFIG_MOUNT || CONFIG_UMOUNT
help
- If your root filesystem is writable and you wish to have the 'mount'
- utility create an mtab file listing the filesystems which have been
- mounted then you should enable this option. Most people that use
- BusyBox have a read-only root filesystem, so they will leave this
- option disabled and BusyBox will use the /proc/mounts file.
-
- Note that even non-embedded developers probably want to have /etc/mtab
- be a symlink to /proc/mounts, since otherwise mtab can get out of sync
- with the real kernel mount state in numerous ways.
+ Historically, Unix systems kept track of the currently mounted
+ partitions in the file "/etc/mtab". These days, the kernel exports
+ the list of currently mounted partitions in "/proc/mounts", rendering
+ the old mtab file obsolete. (In modern systems, /etc/mtab should be
+ a symlink to /proc/mounts.)
+
+ The only reason to have mount maintain an /etc/mtab file itself is if
+ your stripped-down embedded system does not have a /proc directory.
+ If you must use this, keep in mind it's inherently brittle (for
+ example a mount under chroot won't update it), can't handle modern
+ features like separate per-process filesystem namespaces, requires
+ that your /etc directory be writeable, tends to get easily confused
+ by --bind or --move mounts, and so on. (In brief: avoid.)
config CONFIG_READPROFILE
bool "readprofile"