Documentation/TODO

   1
   2  Note that items with (!) have high priority.
   3
   4 cal
   5 ---
   6    - add option to print calendar in vertical way
   7      https://github.com/karelzak/util-linux/issues/604
   8    - support another --reforms, see for example freebsd version
   9      https://github.com/freebsd/freebsd/blob/master/usr.bin/ncal/ncal.c#L72
  10
  11 choom
  12 -----
  13    - add new tool to set/get process OOM setting
  14      https://github.com/karelzak/util-linux/issues/609
  15
  16 column
  17 ------
  18    - add option to NOT ignore empty lines
  19      https://github.com/karelzak/util-linux/issues/593
  20
  21 script
  22 ------
  23    - (!) add terminal type ($TERM), columns and lines to the header line, something like:
  24
  25                 Script started on 2018-03-05 13:02:08+0100 [term="xterm-256color", lines=53, columns=190]
  26
  27      see https://github.com/karelzak/util-linux/issues/583
  28
  29    - think about optional "event" records in timing file to save information
  30      about non-data changes like signals (SIGWINCH, SIGSTOP, SIGCONT, etc.)
  31      in format:
  32
  33              <time> 0 <name>[: var=data, ...]
  34
  35      for example:
  36
  37              0.001296 256
  38              0.001297   0 SIGWINCH: columns=50, lines=20
  39              0.000010 275
  40
  41 lsblk
  42 -----
  43    The problem is NVMe-multipath driver where in /sys/.../slaves/ are nodes
  44    that should be probably hidden for userspace. Note that kernel developers
  45    promised to fix this issue in kernel *** so hold off from that for now***
  46    [Mar 1, 2018].
  47
  48    Anyway, possible userspace solution is:
  49
  50    - verify that slave/holder nodes points to the real block devices
  51      (/sys/dev/block/<maj:min> for the node exists), otherwise silently ignore
  52      the slave/holder node -- see set_cxt().
  53
  54    - the same verification we necessary for nodes in /sys/block/<name>. It means
  55      verify result from sysfs_devname_to_devno() in set_cxt().
  56
  57
  58
  59 cleanup lib/path.c: [Karel Zak for v2.33]
  60 ------------------
  61    - always return rc from all functions
  62    - introduce control struct and use openat() from top-level path
  63      (e.g. top-level is /proc, etc)
  64    - use lib/path.c as low-level API for lib/sysfs.c
  65    - now we duplicate code in functions like {path,sysfs}_read_u32, etc.
  66    - it would be also nice to have a way how to redirect sysfs.c functions
  67      to /sys snapshots (e.g. lsblk regression tests)
  68
  69
  70 lscpu
  71 -----
  72   - read cpuid and uname information from file if --sysroot is specified, then
  73     we can prepare regression tests completely independent on hw and architecture.
  74
  75 lsblk
  76 -----
  77   - currently it does not show mountpoint for all devices in btrfs RAID. It's because
  78     /proc/#/mountinfo contains reference to the one device only. Maybe we can add some
  79     btrfs specific code to provide a better output for FS based stacks. Not sure.
  80     https://bugzilla.redhat.com/show_bug.cgi?id=1084453
  81
  82 nsenter(1)
  83 ----------
  84  - read the default UID and GID from the target process.
  85    http://thread.gmane.org/gmane.linux.utilities.util-linux-ng/9553/focus=9585
  86
  87 hwlock
  88 ------
  89  - use /var/lib/hwclock/drift to store hw-clock drift numbers.
  90  - use /etc/adjtime as read-only for UTC/LOCAL information only
  91  - the /var/lib/hwclock/drift should be implemented backwardly compatible,
  92    it means use the file only if exists, otherwise follow /etc/adjtime
  93
  94 bash completion
  95 ---------------
  96 - Optional argument handling requires user to press backspace to get
  97   argument completion.
  98 - Comma separated value, e.g., --output 'value1,value2', are not
  99   completed for users.
 100
 101 libmount (mount/umount)
 102 -----------------------
 103
 104  - remount all filesystems given their type, for example:
 105         mount -a -t tmpfs -o remount
 106    It should be probably enough to add remount_all() to mount(8) in way we
 107    already have mount_all().  https://github.com/karelzak/util-linux/issues/589
 108
 109  - add --onlyonce to force mount(8) to check if mountpoint is already used. Now
 110    "already mounted" detection is used for --all only. The problem is if you
 111    call "mount <mountpoint>" more than once for in fstab defined tmpfs (or network
 112    filesystem etc.). In this case kernel does not return EBUSY, but a new instance
 113    of the FS is created. https://github.com/karelzak/util-linux/issues/448
 114    (... just idea, maybe wrong idea)
 115
 116  - support CAP_SYS_ADMIN; for mount(2) syscall the CAP_SYS_ADMIN is good
 117    enough. Unfortunately, mount(8) does more things like check for filesystem
 118    type (but it's usually done by udev, so root perms are unnecessary), create
 119    loop devices, write to /run/mount/utab or /etc/mtab, etc.
 120
 121    It would be nice to improve libmount to check for CAP_SYS_ADMIN if suid not
 122    set and allow to use it for simple tasks where no another operation is
 123    necessary.
 124
 125  - allow to execute mount(2) in another namespace, something like:
 126         mount --namespace=/proc/$n/ns/mnt  /dev/sda2 /bar
 127    see https://bugzilla.redhat.com/show_bug.cgi?id=1199554 for more details.
 128
 129  - add options to control fstab/mtab mount options usage, something like:
 130
 131    --options-mode={ignore,append,prepend,replace}      MNT_OMODE_{IGNORE, ...}
 132    --options-source={fstab,mtab,disable}               MNT_OMODE_{FSTAB,MTAB,NOTAB}
 133    --options-source-force                              MNT_OMODE_FORCE
 134
 135    (all this already supported by libmount)
 136
 137
 138 partx
 139 -----
 140
 141  - support mapping by device-mapper if argv[0] is "kpartx" or --dm option is used.
 142
 143
 144 docs
 145 ----
 146
 147  - (!) use something better than gtk-doc for libmount and libblkid (doxyden?)
 148
 149  - (!) add API documentation to libuuid
 150
 151 exit codes
 152 ----------
 153
 154   - we need coreutils-like exit codes for failed exec(). See:
 155     https://www.spinics.net/lists/util-linux-ng/msg12776.html
 156     and original attempt:
 157     https://github.com/karelzak/util-linux/pull/311
 158
 159 login-utils:
 160 -----------
 161
 162   - consolidate newgrp(1)
 163     * we have "su --group/--supp-group" to switch between groups, newgrp(1) in
 164       util-linux and shadow-utils (and sg(1) alias in shadow-utils)
 165     * the unique functionality provided by newgpr(1) is support for group
 166       passwords [/etc/gshadow] -- do we really need this functionality?
 167     * maybe we can mark group-passwords as deprecated, and replace sg(1) and
 168       newgpr(1) with su(1) code. The another way is to ask for group password in
 169       su --group too.
 170     * note that shadow-utils newgpr(1) provides support for syslog and audit log.
 171
 172
 173 libblkid
 174 --------
 175
 176  - (!) don't use internally blkid_loff_t, rather use off_t, size_t, ssize_t,
 177        stdint.h types and so on...
 178
 179  - add FSSIZE value  -- filesystem size (klibc requirement)
 180
 181  - (!) add support for dasd PT (used for example on s390)
 182
 183 libfdisk
 184 --------
 185
 186  - add support for Apple Partition Map (see libblkid/src/partitions/mac.c)
 187    http://en.wikipedia.org/wiki/Apple_Partition_Map
 188
 189 misc
 190 ----
 191
 192  - add mllockall() and SCHED_FIFO to hwclock,
 193    see http://lkml.org/lkml/2008/10/12/132
 194
 195
 196 ---------------
 197 exotic requests
 198 ---------------
 199
 200  - add SELinux security contexts support to the 'ipcs' utility
 201    http://bugzilla.redhat.com/show_bug.cgi?id=225342
 202
 203    Would be great to list the current system IPC Objects with their respective
 204    security labels (where allowed) with something like 'ipcs -Z' - following the
 205    way other tools reports those.
 206
 207