We generally document the suggested paths, not the paths possible in
weird, non-standard setups. We do this in order to not confuse
administrators/users unnecessarily and to push people to install things
into the same directories on all distributions.

We are PID 1 after all, the really basic building block of the OS.
Unlike for an app there's very little benefit in being entirely
relocatable.

cgroup: make cg_pid_get_path() work properly for co-mounted controllers and normalized named hierarchies

Also, always accept both our simple hexdump syntax and UUID syntax.

- Consistent use of $VAR vs ${VAR}
- Consistent use of && vs 'if'
- Add error checking to some places
- Consistent error messages ("Can't" vs "Cannot", etc.)
- Function declarations at the top
- Miscellaneous adjustments

Related to https://bugzilla.redhat.com/show_bug.cgi?id=957135.

This was missed in commit 7027ff61 and
means that the --machine option would work but not its shorthand, -M.

"critical-chain" prints a tree of the critical chain of units

$ ./systemd-analyze critical-chain
graphical.target @1.226s
└─multi-user.target @1.226s
  └─nfs-lock.service @961ms +265ms
    └─rpcbind.service @958ms +1ms
      └─network.target @957ms
        └─NetworkManager.service @434ms +522ms
          └─basic.target @428ms
            └─sockets.target @428ms
              └─dbus.socket @428ms
                └─sysinit.target @427ms
                  └─systemd-tmpfiles-setup.service @411ms +15ms
                    └─local-fs.target @410ms
                      └─boot-efi.automount @410ms
                        └─boot.mount @397ms +9ms
                          └─local-fs-pre.target @192ms
                            └─systemd-udevd.service @187ms +5ms
                              └─systemd-udevd-control.socket @140ms
                                └─-.mount

With the "--fuzz=<ms>" parameter one can display more units around
the critical units.

$ ./systemd-analyze --fuzz=10ms critical-chain
└─multi-user.target @1.226s
  └─nfs-lock.service @961ms +265ms
    ├─rpcbind.service @958ms +1ms
    │ └─network.target @957ms
    │   └─NetworkManager.service @434ms +522ms
    │     ├─basic.target @428ms
    │     │ ├─sockets.target @428ms
    │     │ │ ├─dbus.socket @428ms
    │     │ │ │ └─sysinit.target @427ms
    │     │ │ │   ├─systemd-tmpfiles-setup.service @411ms +15ms
    │     │ │ │   │ └─local-fs.target @410ms
    │     │ │ │   │   └─boot-efi.automount @410ms
    │     │ │ │   │     └─boot.mount @397ms +9ms
    │     │ │ │   │       └─local-fs-pre.target @192ms
    │     │ │ │   │         └─systemd-udevd.service @187ms +5ms
    │     │ │ │   │           ├─systemd-udevd-control.socket @140ms
    │     │ │ │   │           │ └─-.mount
    │     │ │ │   │           └─systemd-udevd-kernel.socket @140ms
    │     │ │ │   └─swap.target @421ms
    │     │ │ │     └─dev-disk-by\x2duuid-....swap @414ms +6ms
    │     │ │ │       └─systemd-journald.socket
    │     │ │ ├─rpcbind.socket @428ms
    │     │ │ └─cups.socket @428ms
    │     │ ├─paths.target @428ms
    │     │ │ └─cups.path @428ms
    │     │ ├─timers.target @427ms
    │     │ │ └─systemd-tmpfiles-clean.timer @427ms
    │     │ └─sysinit.target @427ms
    │     │   └─...
    │     └─dbus.socket @428ms
    │       └─...
    └─network.target @957ms
      └─...

I typically run VMs with 1024MiB allocated; systemd is unable to write
coredumps in this scenario at all because the default kernel
configuration will only overcommit 50% of available RAM.

Avoid this failure by using a realloc() loop.

See: http://lists.freedesktop.org/archives/systemd-devel/2013-April/010709.html

Distributions may have selinux but not sushell or might
need to set a custom debug shell.

Defaults to /sbin/sushell if selinux is enabled, /bin/sh if not.

[zj: Renamed --with-debugshelltty to --with-debug-tty, and
     added a line in output showing DEBUGSHELL and DEBUGTTY.
     I figure that debug shell is pretty useful, and I hope
     the extra line in configure status will draw attention
     to it.]

This was needed with log_struct_unit() but log_notice_unit() adds it
anyway.

clang emits warnings about unused attribute _saved_errno_, which drown
out other—potentially useful—warnings. gcc documentation is not exactly
verbose about the effects of __attribute__((unused)) on variables, but
let's assume that it works if the unit test passes.

(struct stat).st is off_t, which usually is a long, or a long long.
There's no good format string modifier for it, so use a cast.

gcc (and other compilers) sometimes generate spurious warnings, and
thus users of public headers must be able to disable warnings.

Printf format attributes can be disabled by setting
    #define _sd_printf_attr_
before including the header file.

Also, add similar logic for sentinel attribute:
    #define _sd_sentinel_attr_
before including the header file disables the attribute.

It is imperative that open source code be well attributed.
Sprinkle attribute((alloc_size)) here and there, telling gcc
how much memory we are actually allocating.

According to gcc documentation, returned pointer "cannot alias any
other pointer valid when the function returns" and "the memory has
undefined content". This second part is (hopefully) untrue for all
those functions.

Arbitrary fields can be attached at the level of the handler,
and they'll be sent with all messages from this handler.

This facility is used to attach SYSLOG_IDENTIFIER to all messages,
since otherwise journald attaches SYSLOG_IDENTIFIER=python or
something similar, which is completely useless.

When a trigger unit wants to know if a stop is queued for it, we should
just check precisely that and do not check whether it is actually
stopped already. This is because we use these checks usually from state
change calls where the state variables are not updated yet.

This change splits unit_pending_inactive() into two calls
unit_inactive_or_pending() and unit_stop_pending(). The former checks
state and pending jobs, the latter only pending jobs.

Partially revert 2b3c81b0, which
tried to avoid inconsistent rules about when and how to create the
/dev/rtc symlink.

Instead of conditionally or not creating the /dev/rtc link at all,
now always create it with additional and more reliable udev rules.

First try to find the "system rtc" with the hctosys flag, if this
is not found, fall back to create the link for /dev/rtc0.

Our code now never actively searches for the "system rtc" it can
always use /dev/rtc.

The time for systemd initialization and selinux policy loading
is accounted to the initrd or the kernel, which is wrong.

Instead of:

Startup finished in 5.559s (firmware) + 36ms (loader) + 665ms (kernel) +
975ms (initrd) + 1.410s (userspace) = 8.647s

the more correct output is:

Startup finished in 5.559s (firmware) + 36ms (loader) + 665ms (kernel) +
475ms (initrd) + 1.910s (userspace) = 8.647s

This adds some syntactic sugar with a macro RUN_WITH_LOCALE() that reset
the thread-specific locale temporarily.

Disallow recursive .include, and make it unavailable in anything but
unit files.

A small patch to remove a build warnining when SELinux is disabled.

Let's better be safe than sorry.

This makes sure nss-myhostname not only resolves the local host name to
127.0.0.2/::1 but also the host name 'localhost: to 127.0.0.1/::1. This
makes installation of /etc/passwd optional, as it usually only includes
a mapping for 'localhost'.

This change also resolves ::1 to the local hostname (as before), but
also lists 'localhost' as an alias. This means look-ups are now fully
reversible, even though they are 1:n mappings.

Finally, the module will no longer erroneously claim that local IP
addresses which aren't on the loopback device were.