Thomas Hindoe Paaboel Andersen [Fri, 21 Feb 2014 19:54:30 +0000 (20:54 +0100)]
man: networkd typo fixes
Tom Gundersen [Fri, 21 Feb 2014 15:54:00 +0000 (16:54 +0100)]
udev - link-setup - expose ID_NET_DRIVER
This is the same as shown by 'ethtool -i <ifname>', and is sometimes
set even though DRIVER is not.
Jasper St. Pierre [Fri, 21 Feb 2014 16:22:42 +0000 (11:22 -0500)]
logind-session: Fix invalid free in the error case
utf8 needs to be initialized to NULL for the free for the early return,
otherwise we try to free invalid data.
Jason A. Donenfeld [Fri, 21 Feb 2014 15:36:59 +0000 (16:36 +0100)]
install: networkd.service depend on !--disable
It doesn't make any sense to symlink this unit file into /etc when the
unit file itself isn't even installed, with --disable-networkd. This
moves the GENERAL_ALIASES logic into the right "if" block.
Lennart Poettering [Fri, 21 Feb 2014 18:22:00 +0000 (19:22 +0100)]
update TODO
Lennart Poettering [Fri, 21 Feb 2014 18:19:31 +0000 (19:19 +0100)]
man: suffix networkd config file options with "="
That's what we do for all options in the other man pages. It helps
clarifying that these are options that values need to be assigned to.
Lennart Poettering [Fri, 21 Feb 2014 17:57:47 +0000 (18:57 +0100)]
man: split settings documentaiton in systemd.unit(5) into two sections
Lennart Poettering [Fri, 21 Feb 2014 17:57:15 +0000 (18:57 +0100)]
man: don't document ".include" in configuration files anymore as first step to deprecate them one day
Lennart Poettering [Fri, 21 Feb 2014 17:54:59 +0000 (18:54 +0100)]
update TODO
Dave Reisner [Fri, 21 Feb 2014 15:31:26 +0000 (10:31 -0500)]
journald: ignore failure to watch hostname_fd on older kernels
Prior to 3.2, /proc/sys/kernel/hostname isn't a pollable file and
sd_event_add_io will return EPERM. Ignore this failure, since it isn't
critical to journald operation.
Reported and tested by user sraue on IRC.
Lennart Poettering [Fri, 21 Feb 2014 17:04:29 +0000 (18:04 +0100)]
ptyfwd: reset nonblocking mode
Apparently bash doesn't turn off non-blocking mode on stdin/stdout when
reading from it, so be nice to bash. Ideally bash would do this on its
own for robustness reasons, though.
https://bugs.freedesktop.org/show_bug.cgi?id=70622
Kay Sievers [Fri, 21 Feb 2014 17:11:47 +0000 (18:11 +0100)]
NEWS: add name_to_handle_at() requirement
Samuli Suominen [Fri, 21 Feb 2014 14:14:51 +0000 (16:14 +0200)]
build-sys: make network files respect --rootprefix
With --rootprefix= systemd-udevd gets installed to /lib/systemd, and since
the network configuration is also required during early boot, it should be
available there with it. Using --prefix= is not an option since it would
put everything, including pkg-config files, man pages, documentation, to /
which is not wanted. This commit puts 99-default.link to
/lib/systemd/network/ when required.
Tom Gundersen [Fri, 21 Feb 2014 13:51:19 +0000 (14:51 +0100)]
.network/.netdev/.link: allow to match on architecture
Jason A. Donenfeld [Fri, 21 Feb 2014 14:21:21 +0000 (15:21 +0100)]
configure: show if networkd is enabled in status
Lennart Poettering [Fri, 21 Feb 2014 14:06:44 +0000 (15:06 +0100)]
man: there is no ExecStopPre= for service units
https://bugs.freedesktop.org/show_bug.cgi?id=73177
Tom Gundersen [Fri, 21 Feb 2014 13:20:14 +0000 (14:20 +0100)]
install: enable networkd by default
This treats it similarly to getty@tty1.service and remote-fs.target, which can be
enabled/disabled, but defaults to enabled.
Jason A. Donenfeld [Fri, 21 Feb 2014 03:38:17 +0000 (04:38 +0100)]
install: do not statically enable systemd-networkd
[tomegun: pruned the commit message as not to contradict the follow-up commit]
Thomas Bächler [Fri, 21 Feb 2014 10:20:24 +0000 (11:20 +0100)]
sd-event: Fix systemd crash when using timer units.
There was a copy-paste error introduced in commit
c2ba3ad6604ef2e189d7e0a36d6911116e84d3ab
which causes the following error when using timer units:
Assertion '(x->type == SOURCE_MONOTONIC && y->type == SOURCE_MONOTONIC) || (x->type == SOURCE_REALTIME && y->type == SOURCE_REALTIME)'
failed at src/libsystemd/sd-event/sd-event.c:264, function latest_time_prioq_compare(). Aborting.
Daniel Mack [Fri, 21 Feb 2014 11:05:38 +0000 (12:05 +0100)]
systemctl: fix compiler warning in list_timers()
get_next_elapse() will always fill 'next' with values when it
returns >= 0. Hence, the compiler is wrong about this warning.
Initialize 'next' nevertheless.
src/systemctl/systemctl.c: In function ‘list_timers’:
src/systemctl/systemctl.c:953:43: warning: ‘next.monotonic’ may be used
uninitialized in this function [-Wmaybe-uninitialized]
converted = nw.realtime - (nw.monotonic - next.monotonic);
^
In file included from ./src/shared/log.h:30:0,
from src/systemctl/systemctl.c:46:
./src/shared/macro.h:137:38: warning: ‘next.realtime’ may be used
uninitialized in this function [-Wmaybe-uninitialized]
_a < _b ? _a : _b; \
^
src/systemctl/systemctl.c:933:32: note: ‘next.realtime’ was declared here
dual_timestamp next;
^
Daniel Mack [Fri, 21 Feb 2014 10:39:38 +0000 (11:39 +0100)]
shared/architecture.h: fix preprocessor directives
Use 'if defined()', not 'ifdef defined()'. Fixes the following warning.
CC src/shared/architecture.lo
In file included from src/shared/architecture.c:24:0:
src/shared/architecture.h:89:17: warning: extra tokens at end of #ifdef
directive [enabled by default]
# ifdef defined(WORDS_BIGENDIAN)
^
Zbigniew Jędrzejewski-Szmek [Fri, 21 Feb 2014 03:39:26 +0000 (22:39 -0500)]
man: xinclude pkg-config note
Zbigniew Jędrzejewski-Szmek [Fri, 21 Feb 2014 00:47:03 +0000 (19:47 -0500)]
man: xinclude --host/--machine
As usual, those common options are pushed to the end.
Zbigniew Jędrzejewski-Szmek [Thu, 13 Feb 2014 19:40:40 +0000 (14:40 -0500)]
man: xinclude --user/--system
Zbigniew Jędrzejewski-Szmek [Thu, 13 Feb 2014 19:33:51 +0000 (14:33 -0500)]
man: xinclude --help/--version/--no-pager
Zbigniew Jędrzejewski-Szmek [Wed, 19 Feb 2014 21:04:30 +0000 (16:04 -0500)]
build-sys: add conditionals and regenerate manpage list
The list of man pages is auto generated, based on conditonal='...'
attributes in the man page itself.
Zbigniew Jędrzejewski-Szmek [Wed, 19 Feb 2014 21:06:10 +0000 (16:06 -0500)]
man: ignore non-manpage files when generating makefile
Zbigniew Jędrzejewski-Szmek [Thu, 13 Feb 2014 22:10:12 +0000 (17:10 -0500)]
manager: be more precise in message about restart
Lennart Poettering [Fri, 21 Feb 2014 03:04:47 +0000 (04:04 +0100)]
update TODO
Lennart Poettering [Fri, 21 Feb 2014 03:02:12 +0000 (04:02 +0100)]
nspawn: when adding a veth interface to a bridge, use the "vb-" rather than "ve-" interface name prefix
This way we can recognize the interfaces later on to apply different
host-side configuration to them.
Jason St. John [Thu, 20 Feb 2014 23:09:27 +0000 (18:09 -0500)]
replace tabs with spaces in some files
Files:
* hwdb/60-keyboard.hwdb
* shell-completion/zsh/_systemd-coredumpctl
* src/test/test-helper.h
Michael Scherer [Thu, 20 Feb 2014 15:19:44 +0000 (16:19 +0100)]
core: Add AppArmor profile switching
This permit to switch to a specific apparmor profile when starting a daemon. This
will result in a non operation if apparmor is disabled.
It also add a new build requirement on libapparmor for using this feature.
Colin Walters [Fri, 21 Feb 2014 02:29:00 +0000 (03:29 +0100)]
selinux: Don't attempt to load policy in initramfs if it doesn't exist
Currently on at least Fedora, SELinux policy does not come in the
initramfs. systemd will attempt to load *both* in the initramfs and
in the real root.
Now, the selinux_init_load_policy() API has a regular error return
value, as well as an "enforcing" boolean. To determine enforcing
state, it looks for /etc/selinux/config as well as the presence of
"enforcing=" on the kernel command line.
Ordinarily, neither of those exist in the initramfs, so it will return
"unknown" for enforcing, and systemd will simply ignore the failure to
load policy.
Lennart Poettering [Fri, 21 Feb 2014 02:13:57 +0000 (03:13 +0100)]
build-sys: don't use -Wcast-align anymore
We do a lot of these casts, and they are all OK, so let's get rid of the
warning, to make things quieter.
http://lists.freedesktop.org/archives/systemd-devel/2014-February/017174.html
Michael Scherer [Thu, 20 Feb 2014 15:07:08 +0000 (16:07 +0100)]
build-sys: Fix compilation of nspawn when seccomp is not enabled
Lennart Poettering [Fri, 21 Feb 2014 02:07:42 +0000 (03:07 +0100)]
missing: simplifications
Holger Schurig [Thu, 20 Feb 2014 13:39:13 +0000 (14:39 +0100)]
build-sys: Add setns() functions if not in the C library.
Debian Stable is still using glibc 2.13, which doesn't provide the setns().
So we detect this and provide a tiny wrapper that issues the setns syscall
towards the kernel.
Lennart Poettering [Fri, 21 Feb 2014 01:28:54 +0000 (02:28 +0100)]
hostnamectl: should the sanitized arch, not the native uname() one
Lennart Poettering [Fri, 21 Feb 2014 01:06:04 +0000 (02:06 +0100)]
core: add new ConditionArchitecture() that checks the architecture returned by uname()'s machine field.
Tom Gundersen [Thu, 20 Feb 2014 22:26:07 +0000 (23:26 +0100)]
man: networkd/udev - add to [Match] documentation
Tom Gundersen [Thu, 20 Feb 2014 21:55:42 +0000 (22:55 +0100)]
net-util: match - allow globbing
For now support globbing for interface name and path.
Thomas Hindoe Paaboel Andersen [Thu, 20 Feb 2014 22:12:02 +0000 (23:12 +0100)]
bootctl: add bash completion
Umut Tezduyar Lindskog [Thu, 20 Feb 2014 20:04:03 +0000 (21:04 +0100)]
sd-dhcp-client: prevent timer related memory leaks
Tom Gundersen [Thu, 20 Feb 2014 18:56:50 +0000 (19:56 +0100)]
networkd: netdev - allow filtering on kernel cmdline, host and virt
Tom Gundersen [Thu, 20 Feb 2014 18:39:49 +0000 (19:39 +0100)]
network/link: Match - filter on kernel cmdline, host and virt
Tom Gundersen [Thu, 20 Feb 2014 18:26:54 +0000 (19:26 +0100)]
condition: split out most of condition handling into libsystemd-shard
The parts that require linknig to libcap, libselinux and friends stays in libsystemd-core.
Tom Gundersen [Thu, 20 Feb 2014 18:09:49 +0000 (19:09 +0100)]
condition: refactor a bit
No functional change, just move code around a bit to make it simpler to split out
some functions.
Lennart Poettering [Thu, 20 Feb 2014 18:14:52 +0000 (19:14 +0100)]
python: reindent _reader.c
All files should follow our coding style, and that means 8ch indenting.
Let's correct that.
Lennart Poettering [Thu, 20 Feb 2014 17:44:51 +0000 (18:44 +0100)]
macro: add nice macro for disabling -Wnonnull temporarily
Lennart Poettering [Thu, 20 Feb 2014 17:35:03 +0000 (18:35 +0100)]
macro: introduce nice macro for disabling -Wmissing-prototypes warnigs
Lennart Poettering [Thu, 20 Feb 2014 17:18:32 +0000 (18:18 +0100)]
macro: introduce a nice macro for disabling -Wformat-nonliteral temporarily
Lennart Poettering [Thu, 20 Feb 2014 17:06:35 +0000 (18:06 +0100)]
update TODO
Lennart Poettering [Thu, 20 Feb 2014 17:05:56 +0000 (18:05 +0100)]
util: get rid of warnings around assert_cc() macro
Suggested by Holger Schurig.
Daniel Mack [Thu, 20 Feb 2014 16:24:36 +0000 (17:24 +0100)]
journal: fix compiler warning in journal_file_append_data()
gcc (4.8.2, arm) does not understand that journal_file_append_field()
will always set 'fo' when it returns 0, so this warning is bogus.
Anyway, fix it by initialiting fo = NULL.
Daniel Mack [Thu, 20 Feb 2014 16:21:13 +0000 (17:21 +0100)]
journal: fix compiler warning in real_journal_next()
gcc (4.8.2, arm) does not understand that next_beyond_location() will
always set 'p' when it returns > 0. Initialize p in order to fix this.
Daniel Mack [Thu, 20 Feb 2014 16:14:50 +0000 (17:14 +0100)]
libsystemd: fix compiler warning in property_get_set_callbacks_run()
gcc (4.8.2, arm) doesn't understand that vtable_property_get_userdata()
will always set 'u' when it returns > 0. Hence, the warning is bogus,
but anyway.
src/libsystemd/sd-bus/bus-objects.c:510:19: warning: 'u' may be used uninitialized in this function [-Wmaybe-uninitialized]
(and yes, indeed, even the reported line numbers are bogus in this case)
Dave Reisner [Thu, 20 Feb 2014 14:02:39 +0000 (09:02 -0500)]
readd journald.so install hooks under libsystemd.so
Inadvertently removed in
add00dfd1f05.
Daniel Mack [Thu, 20 Feb 2014 12:29:54 +0000 (13:29 +0100)]
core/unit: fix unit_add_target_dependencies() for units with no dependencies
For units without any dependencies, r needs to be initialized to 0.
Otherwise, the return value of unit_add_target_dependencies() is
unspecified.
Lennart Poettering [Thu, 20 Feb 2014 02:55:22 +0000 (03:55 +0100)]
NEWS: fixes that probably should have been made before the 209 release
Lennart Poettering [Thu, 20 Feb 2014 00:22:12 +0000 (01:22 +0100)]
build-sys: missing EXTRA_DIST
Lennart Poettering [Wed, 19 Feb 2014 23:54:42 +0000 (00:54 +0100)]
build-sys: install compat .pc file
Lennart Poettering [Wed, 19 Feb 2014 23:25:37 +0000 (00:25 +0100)]
build-sys: add linkwarning.h to tarball
Lennart Poettering [Wed, 19 Feb 2014 23:03:00 +0000 (00:03 +0100)]
update TODO
Lennart Poettering [Wed, 19 Feb 2014 22:54:58 +0000 (23:54 +0100)]
api: in constructor function calls, always put the returned object pointer first (or second)
Previously the returned object of constructor functions where sometimes
returned as last, sometimes as first and sometimes as second parameter.
Let's clean this up a bit. Here are the new rules:
1. The object the new object is derived from is put first, if there is any
2. The object we are creating will be returned in the next arguments
3. This is followed by any additional arguments
Rationale:
For functions that operate on an object we always put that object first.
Constructors should probably not be too different in this regard. Also,
if the additional parameters might want to use varargs which suggests to
put them last.
Note that this new scheme only applies to constructor functions, not to
all other functions. We do give a lot of freedom for those.
Note that this commit only changes the order of the new functions we
added, for old ones we accept the wrong order and leave it like that.
Jason A. Donenfeld [Tue, 18 Feb 2014 18:11:08 +0000 (19:11 +0100)]
sd-bus: the bus returned should be the first arg
This matches the API of previous headers, such as sd-journal.h.
Tom Gundersen [Wed, 19 Feb 2014 21:37:13 +0000 (22:37 +0100)]
udev: net-config - allow interface names to be set from the hwdb
Kay Sievers [Wed, 19 Feb 2014 20:03:59 +0000 (21:03 +0100)]
build-sys: install experimental man pages only with --enable-kdbus
Lennart Poettering [Wed, 19 Feb 2014 20:01:12 +0000 (21:01 +0100)]
build-sys: bump release in preparation for release
Kay Sievers [Wed, 19 Feb 2014 19:52:48 +0000 (20:52 +0100)]
build-sys: enable compat-libs for distcheck
Kay Sievers [Wed, 19 Feb 2014 19:16:31 +0000 (20:16 +0100)]
build-sys: export experimental symbols only with --enable-kdbus
Lennart Poettering [Wed, 19 Feb 2014 19:35:33 +0000 (20:35 +0100)]
bus: fix unreffing logic
Tom Gundersen [Wed, 19 Feb 2014 18:44:06 +0000 (19:44 +0100)]
sd-dhcp-client: fix mixup between raw and udp sockets
Kay Sievers [Wed, 19 Feb 2014 17:30:14 +0000 (18:30 +0100)]
build-sys: add gcrypt CFLAGS
Lennart Poettering [Wed, 19 Feb 2014 17:30:41 +0000 (18:30 +0100)]
man: fix references to .pc files which aren't separate anymore
Kay Sievers [Wed, 19 Feb 2014 17:21:19 +0000 (18:21 +0100)]
sd-damon is LGPL now
Lennart Poettering [Wed, 19 Feb 2014 17:20:04 +0000 (18:20 +0100)]
update TODO
Lennart Poettering [Wed, 19 Feb 2014 17:19:06 +0000 (18:19 +0100)]
man: don't advertise sd-daemon as embeddable anymore
It's now part of libsystemd, and should be used like any other API.
Kay Sievers [Wed, 19 Feb 2014 15:33:09 +0000 (16:33 +0100)]
build-sys: merge libsystemd-journal into libsystemd
Thomas Hindoe Paaboel Andersen [Wed, 19 Feb 2014 16:45:13 +0000 (17:45 +0100)]
busctl: install bash completion
Lennart Poettering [Wed, 19 Feb 2014 16:52:41 +0000 (17:52 +0100)]
tmpfiles: simplification
Lennart Poettering [Wed, 19 Feb 2014 16:52:28 +0000 (17:52 +0100)]
util: simplify get_ctty()
Lennart Poettering [Wed, 19 Feb 2014 16:51:54 +0000 (17:51 +0100)]
timedatectl: fix minor memory leak
Lennart Poettering [Wed, 19 Feb 2014 16:51:08 +0000 (17:51 +0100)]
udev: make sure we always return a valid error code in trie_store()
Lennart Poettering [Wed, 19 Feb 2014 16:50:15 +0000 (17:50 +0100)]
sd-id128: use new dev_urandom() call
Lennart Poettering [Wed, 19 Feb 2014 16:49:50 +0000 (17:49 +0100)]
journal: don't clobber return parameters of sd_journal_get_cutoff_realtime_usec() on failure
Lennart Poettering [Wed, 19 Feb 2014 16:49:00 +0000 (17:49 +0100)]
execute: modernizations
Lennart Poettering [Wed, 19 Feb 2014 16:47:11 +0000 (17:47 +0100)]
make gcc shut up
If -flto is used then gcc will generate a lot more warnings than before,
among them a number of use-without-initialization warnings. Most of them
without are false positives, but let's make them go away, because it
doesn't really matter.
Lennart Poettering [Wed, 19 Feb 2014 16:43:40 +0000 (17:43 +0100)]
NEWS: Mention the new Personality= switch in unit files
Thomas Hindoe Paaboel Andersen [Wed, 19 Feb 2014 16:35:35 +0000 (17:35 +0100)]
busctl: add --no-legend and use in bash completion
Thomas Hindoe Paaboel Andersen [Wed, 19 Feb 2014 16:12:17 +0000 (17:12 +0100)]
busctl: update bash completion
Thomas Hindoe Paaboel Andersen [Wed, 19 Feb 2014 16:02:45 +0000 (17:02 +0100)]
man: busctl typo fix
Tom Gundersen [Wed, 19 Feb 2014 15:40:52 +0000 (16:40 +0100)]
sd-rtnl: test - include missing.h to build on old kernels
Tom Gundersen [Wed, 19 Feb 2014 15:34:34 +0000 (16:34 +0100)]
networkd: dhcp - log when lease is lost
Tom Gundersen [Wed, 19 Feb 2014 14:59:31 +0000 (15:59 +0100)]
networkd: link - only reset transient hostname if it was set by us
Maciej Wereski [Wed, 19 Feb 2014 10:39:01 +0000 (11:39 +0100)]
systemctl: remove erroneous return in runlevel_parse_argv()
Maciej Wereski [Wed, 19 Feb 2014 10:39:00 +0000 (11:39 +0100)]
logind: remove redundant check in manager_new()
Lennart Poettering [Wed, 19 Feb 2014 15:23:30 +0000 (16:23 +0100)]
update TODO
Lennart Poettering [Wed, 19 Feb 2014 15:13:42 +0000 (16:13 +0100)]
label: minor optimizations
Tom Gundersen [Wed, 19 Feb 2014 14:45:54 +0000 (15:45 +0100)]
missing.h: add some IFLA_* constants
These were added to the kernel between 3.5 and 3.9, let's not require such
recent kernels (yet).
David Herrmann [Wed, 19 Feb 2014 12:24:38 +0000 (13:24 +0100)]
event: fix crash on child-source state modifications
Setting a child-source state to anything but SD_EVENT_OFF currently does
nothing. The condition logic is flawed. Move the state update *below* the
test for the previous state.
Fixes a crash if you call:
sd_event_source_set_enabled(source, SD_EVENT_ON);
sd_event_source_unref(source);
on a child-source in its own callback.