platform/upstream/systemd.git
9 years agoman: various typos
Thomas Hindoe Paaboel Andersen [Mon, 2 Nov 2015 21:50:57 +0000 (22:50 +0100)]
man: various typos

9 years agoMerge pull request #1690 from evverx/run-runtime-directory
Lennart Poettering [Mon, 2 Nov 2015 19:16:18 +0000 (20:16 +0100)]
Merge pull request #1690 from evverx/run-runtime-directory

systemd-run can launch units with RuntimeDirectory

9 years agoMerge pull request #1736 from ronnychevalier/rc/test_execute_more_tests2
Lennart Poettering [Mon, 2 Nov 2015 19:06:48 +0000 (20:06 +0100)]
Merge pull request #1736 from ronnychevalier/rc/test_execute_more_tests2

test-execute: add more tests

9 years agoMerge pull request #1753 from fbuihuu/teach-delegate-for-runtime
Lennart Poettering [Mon, 2 Nov 2015 18:52:00 +0000 (19:52 +0100)]
Merge pull request #1753 from fbuihuu/teach-delegate-for-runtime

Teach bus_append_unit_property_assignment() about 'Delegate' property

9 years agoMerge pull request #1755 from ssahani/word-systemctl
Lennart Poettering [Mon, 2 Nov 2015 18:50:51 +0000 (19:50 +0100)]
Merge pull request #1755 from ssahani/word-systemctl

systemctl: port to extract_first_word

9 years agosystemctl: port to extract_first_word
Susant Sahani [Mon, 2 Nov 2015 16:17:40 +0000 (21:47 +0530)]
systemctl: port to extract_first_word

9 years agoMerge pull request #1751 from ssahani/word-machine
Daniel Mack [Mon, 2 Nov 2015 11:38:24 +0000 (12:38 +0100)]
Merge pull request #1751 from ssahani/word-machine

machine: fix #1739

9 years agoTeach bus_append_unit_property_assignment() about 'Delegate' property
Franck Bui [Mon, 2 Nov 2015 10:12:41 +0000 (11:12 +0100)]
Teach bus_append_unit_property_assignment() about 'Delegate' property

9 years agomachine: fix extract_first_word
Susant Sahani [Mon, 2 Nov 2015 11:25:59 +0000 (16:55 +0530)]
machine: fix extract_first_word

9 years agoMerge pull request #1722 from evverx/port-cap-bounding-set-to-extract-first-word
Daniel Mack [Mon, 2 Nov 2015 10:45:06 +0000 (11:45 +0100)]
Merge pull request #1722 from evverx/port-cap-bounding-set-to-extract-first-word

Port capabiliy bounding set parsing to extract_first_word

9 years agoMerge pull request #1750 from systemd/revert-1740-master
Daniel Mack [Mon, 2 Nov 2015 10:31:10 +0000 (11:31 +0100)]
Merge pull request #1750 from systemd/revert-1740-master

Revert "utf8.[ch]: use char32_t and char16_t instead of int, int32_t,…

9 years agoMerge pull request #1725 from ssahani/word-journal
Daniel Mack [Mon, 2 Nov 2015 10:23:32 +0000 (11:23 +0100)]
Merge pull request #1725 from ssahani/word-journal

sd-journal.c: port to extract_first_word

9 years agoRevert "utf8.[ch]: use char32_t and char16_t instead of int, int32_t, int16_t"
Lennart Poettering [Mon, 2 Nov 2015 10:21:25 +0000 (11:21 +0100)]
Revert "utf8.[ch]: use char32_t and char16_t instead of int, int32_t, int16_t"

9 years agoMerge pull request #1739 from ssahani/word-machine
Daniel Mack [Mon, 2 Nov 2015 10:21:18 +0000 (11:21 +0100)]
Merge pull request #1739 from ssahani/word-machine

machine: : port to extract_first_word

9 years agoMerge pull request #1740 from shawnl/master
Daniel Mack [Mon, 2 Nov 2015 09:57:38 +0000 (10:57 +0100)]
Merge pull request #1740 from shawnl/master

utf8.[ch]: use char32_t and char16_t instead of int, int32_t, int16_t

9 years agoMerge pull request #1738 from evverx/fix-paths-lookup-test
Daniel Mack [Mon, 2 Nov 2015 09:43:16 +0000 (10:43 +0100)]
Merge pull request #1738 from evverx/fix-paths-lookup-test

tests: lookup_paths_init depends on SYSTEMD_UNIT_PATH. Test it

9 years agoMerge pull request #1749 from zonque/cmsg
Lennart Poettering [Mon, 2 Nov 2015 09:31:24 +0000 (10:31 +0100)]
Merge pull request #1749 from zonque/cmsg

core: two trivial debug/error message improvements

9 years agoMerge pull request #1711 from mustrumr/network-use-after-free
Daniel Mack [Mon, 2 Nov 2015 09:17:24 +0000 (10:17 +0100)]
Merge pull request #1711 from mustrumr/network-use-after-free

network: fix use-after-free in link_free

9 years agoMerge pull request #1748 from karelzak/todo
Daniel Mack [Mon, 2 Nov 2015 09:16:04 +0000 (10:16 +0100)]
Merge pull request #1748 from karelzak/todo

update TODO

9 years agocore: be more verbose when NameOwnerChanged subscriptions fail
Daniel Mack [Thu, 15 Oct 2015 15:39:14 +0000 (17:39 +0200)]
core: be more verbose when NameOwnerChanged subscriptions fail

9 years agocore: service: fix error message
Daniel Mack [Tue, 13 Oct 2015 16:04:13 +0000 (18:04 +0200)]
core: service: fix error message

The directive is called FileDescriptorStoreMax=, not FDStoreMax=.

9 years agoMerge pull request #1745 from poettering/journal-deadlock
Daniel Mack [Mon, 2 Nov 2015 09:10:44 +0000 (10:10 +0100)]
Merge pull request #1745 from poettering/journal-deadlock

Make sure journald never blocks on sd_notify() to PID 1

9 years agoupdate TODO
Karel Zak [Mon, 2 Nov 2015 08:59:27 +0000 (09:59 +0100)]
update TODO

- inotify utab has been removed and we use libmount now
- x-systemd.requires= fstab option has been implemented to specify
  dependencies in fstab

9 years agojournal-remote: remove unused variable warning when building without GNUTLS.
Lennart Poettering [Sun, 1 Nov 2015 20:59:17 +0000 (21:59 +0100)]
journal-remote: remove unused variable warning when building without GNUTLS.

9 years agojournald: never block when sending messages on NOTIFY_SOCKET socket
Lennart Poettering [Sun, 1 Nov 2015 20:50:24 +0000 (21:50 +0100)]
journald: never block when sending messages on NOTIFY_SOCKET socket

Otherwise we might run into deadlocks, when journald blocks on the
notify socket on PID 1, and PID 1 blocks on IPC to dbus-daemon and
dbus-daemon blocks on logging to journald. Break this cycle by making
sure that journald never ever blocks on PID 1.

Note that this change disables support for event loop watchdog support,
as these messages are sent in blocking style by sd-event. That should
not be a big loss though, as people reported frequent problems with the
watchdog hitting journald on excessively slow IO.

Fixes: #1505.

9 years agosd-daemon: verify NOTIFY_SOCKET path length
Lennart Poettering [Sun, 1 Nov 2015 20:49:19 +0000 (21:49 +0100)]
sd-daemon: verify NOTIFY_SOCKET path length

Better generate a real error then simply connect to the wrong socket.

9 years agoutf8.[ch]: use char32_t and char16_t instead of int, int32_t, int16_t
Shawn Landden [Sun, 1 Nov 2015 03:39:15 +0000 (20:39 -0700)]
utf8.[ch]: use char32_t and char16_t instead of int, int32_t, int16_t

rework C11 utf8.[ch] to use char32_t instead of uint32_t when referring
to unicode chars, to make things more expressive.

9 years agomachine: : port to extract_first_word
Susant Sahani [Sat, 31 Oct 2015 22:06:07 +0000 (03:36 +0530)]
machine: : port to extract_first_word

9 years agosd-journal.c: port to extract_first_word
Susant Sahani [Sat, 31 Oct 2015 22:02:45 +0000 (03:32 +0530)]
sd-journal.c: port to extract_first_word

9 years agoman: MANAGER_SYSTEM understands SYSTEMD_UNIT_PATH too
Evgeny Vereshchagin [Sat, 31 Oct 2015 20:07:47 +0000 (23:07 +0300)]
man: MANAGER_SYSTEM understands SYSTEMD_UNIT_PATH too

9 years agotests: lookup_paths_init depends on SYSTEMD_UNIT_PATH. Test it
Evgeny Vereshchagin [Sat, 31 Oct 2015 19:00:32 +0000 (22:00 +0300)]
tests: lookup_paths_init depends on SYSTEMD_UNIT_PATH. Test it

9 years agosd-daemon: increase sd_notify() socket buffer size
Lennart Poettering [Fri, 30 Oct 2015 10:27:29 +0000 (11:27 +0100)]
sd-daemon: increase sd_notify() socket buffer size

Let's make sure we don't start blocking on sd_notify() earlier than
necessary, let's bump the socket buffer sizes to 8M.

We already do something similar for our logging socket buffers, hence
apply a similar bump here.

9 years agosysctl.d: bump number of queueable AF_UNIX/SOCK_DGRAM datagrams
Lennart Poettering [Fri, 30 Oct 2015 10:25:00 +0000 (11:25 +0100)]
sysctl.d: bump number of queueable AF_UNIX/SOCK_DGRAM datagrams

The default of 16 is pretty low, let's bump this to accomodate for more
queued datagrams. This is useful for AF_UNIX/SOCK_DGRAM logging and
sd_notify() sockets as this allows queuing more datagrams before things
start to block, thus improving parallelization and logging performance.

9 years agosysv-generator: modernize
Lennart Poettering [Wed, 28 Oct 2015 23:19:31 +0000 (00:19 +0100)]
sysv-generator: modernize

- Make sure we log each error at least once, and at most once

- Replace FOREACH_WORD loops by extract_first_word() loops

- Use FOREACH_DIRENT() for directory loops

- Use free_and_strdup() where appropriate

- Do not operate on half-loaded SysV files

- Always properly free all memory

9 years agocore: bail our earlier when doing audit
Lennart Poettering [Wed, 28 Oct 2015 17:23:26 +0000 (18:23 +0100)]
core: bail our earlier when doing audit

Let's make sure we don't even try to create the audit socket

9 years agocore,nspawn: minor coding style fixes
Lennart Poettering [Wed, 28 Oct 2015 17:23:07 +0000 (18:23 +0100)]
core,nspawn: minor coding style fixes

9 years agonspawn: fix minor memory leak
Lennart Poettering [Wed, 28 Oct 2015 17:22:23 +0000 (18:22 +0100)]
nspawn: fix minor memory leak

When rebooting nspawn containers about 400 times we'd otherwise hit the
fd limit and refuse further reboots.

9 years agocore: constify a few things
Lennart Poettering [Wed, 28 Oct 2015 11:43:52 +0000 (12:43 +0100)]
core: constify a few things

9 years agosystemctl: add a way to explicitly request client-side unit installing
Lennart Poettering [Thu, 8 Oct 2015 14:15:38 +0000 (16:15 +0200)]
systemctl: add a way to explicitly request client-side unit installing

This adds support for a new environment variable
SYSTEMCTL_INSTALL_CLIENT_SIDE, that ensures that systemctl executes
install operations client-side instead of passing them to PID1. This is
useful in debugging situations, but even beyond that. However, we don't
want to make it official API, hence let's just make it an undocumented
environment variable.

Similar, add a second variable, SYSTEMCTL_SKIP_SYSV which allows
skipping the SysV chkconfig fall-back if set. This is useful for similar
reasons, and exposed as undocumented as environment variable for similar
reasons, too.

9 years agoMerge pull request #1734 from evverx/overwrite-sysemd-unit-path
Ronny Chevalier [Sat, 31 Oct 2015 16:22:20 +0000 (17:22 +0100)]
Merge pull request #1734 from evverx/overwrite-sysemd-unit-path

core: set_unit_path overwrites SYSTEMD_UNIT_PATH

9 years agotest-path: move all related test files to a specific directory
Ronny Chevalier [Fri, 30 Oct 2015 17:12:31 +0000 (18:12 +0100)]
test-path: move all related test files to a specific directory

To avoid polluting test/

9 years agotest-execute: move all files related to a specific directory
Ronny Chevalier [Fri, 30 Oct 2015 17:03:18 +0000 (18:03 +0100)]
test-execute: move all files related to a specific directory

To avoid polluting test/

9 years agobuild-sys: add missing test files to EXTRA_DIST
Ronny Chevalier [Sat, 31 Oct 2015 13:46:45 +0000 (14:46 +0100)]
build-sys: add missing test files to EXTRA_DIST

9 years agotest-execute: add tests for IOSchedulingClass
Ronny Chevalier [Fri, 30 Oct 2015 16:46:32 +0000 (17:46 +0100)]
test-execute: add tests for IOSchedulingClass

9 years agotest-execute: add tests for OOMScoreAdjust
Ronny Chevalier [Fri, 30 Oct 2015 16:09:22 +0000 (17:09 +0100)]
test-execute: add tests for OOMScoreAdjust

9 years agotest-execute: add test for EnvironmentFile
Ronny Chevalier [Fri, 30 Oct 2015 15:55:44 +0000 (16:55 +0100)]
test-execute: add test for EnvironmentFile

9 years agotest-execute: add test for PrivateNetwork
Ronny Chevalier [Fri, 30 Oct 2015 12:51:51 +0000 (13:51 +0100)]
test-execute: add test for PrivateNetwork

9 years agocore: set_unit_path overwrites SYSTEMD_UNIT_PATH
Evgeny Vereshchagin [Sat, 31 Oct 2015 11:14:37 +0000 (14:14 +0300)]
core: set_unit_path overwrites SYSTEMD_UNIT_PATH

9 years agocore: port config_parse_bounding_set to extract_first_word
Evgeny Vereshchagin [Fri, 30 Oct 2015 06:25:12 +0000 (09:25 +0300)]
core: port config_parse_bounding_set to extract_first_word

9 years agoMerge pull request #1729 from endocode/dongsu/jr-mhd-memlimit-v3
Lennart Poettering [Fri, 30 Oct 2015 14:48:49 +0000 (15:48 +0100)]
Merge pull request #1729 from endocode/dongsu/jr-mhd-memlimit-v3

journal-remote: increase memory limit per connection to avoid errors …

9 years agojournal-remote: increase memory limit per connection to avoid errors with HTTPS
Dongsu Park [Wed, 21 Oct 2015 12:51:10 +0000 (14:51 +0200)]
journal-remote: increase memory limit per connection to avoid errors with HTTPS

Explicitly set MHD_OPTION_CONNECTION_MEMORY_LIMIT to a larger value,
when setting up microhttpd, to give more memory per HTTP(S) connection.

This way systemd-journal-remote can now prevent microhttpd from failing
in creating response headers with messages like "Not enough memory for
write", especially when lots of HTTPS requests arrive. That's precisely
because MHD_OPTION_CONNECTION_MEMORY_LIMIT in libmicrohttpd defaults to
32768, which is in practice insufficient in this case.
See also https://gnunet.org/bugs/view.php?id=4007 for more details.

Fixes: https://github.com/coreos/bugs/issues/927

9 years agoMerge pull request #1720 from evverx/shell-completion-global-mode
Daniel Mack [Fri, 30 Oct 2015 09:56:38 +0000 (10:56 +0100)]
Merge pull request #1720 from evverx/shell-completion-global-mode

shell-completion: systemctl: show completions for `user` in `global` …

9 years agotests: add test for capability bounding set parsing
Evgeny Vereshchagin [Fri, 30 Oct 2015 06:22:53 +0000 (09:22 +0300)]
tests: add test for capability bounding set parsing

9 years agohwdb: Update database of Bluetooth company identifiers
Marcel Holtmann [Fri, 30 Oct 2015 03:22:54 +0000 (04:22 +0100)]
hwdb: Update database of Bluetooth company identifiers

9 years agoshell-completion: systemctl: show completions for `user` in `global` mode
Evgeny Vereshchagin [Thu, 29 Oct 2015 07:29:29 +0000 (10:29 +0300)]
shell-completion: systemctl: show completions for `user` in `global` mode

9 years agoMerge pull request #1709 from evverx/set-location-completion
Daniel Mack [Thu, 29 Oct 2015 18:47:02 +0000 (19:47 +0100)]
Merge pull request #1709 from evverx/set-location-completion

shell-completion: hostnamectl: add the command set-location

9 years agoMerge pull request #1717 from ronnychevalier/rc/test_capabilityboundingset2
Daniel Mack [Thu, 29 Oct 2015 18:46:01 +0000 (19:46 +0100)]
Merge pull request #1717 from ronnychevalier/rc/test_capabilityboundingset2

test: add tests to ensure that the capabilities are properly set

9 years agotest: add tests to ensure that the capabilities are properly set
Ronny Chevalier [Thu, 29 Oct 2015 13:22:23 +0000 (14:22 +0100)]
test: add tests to ensure that the capabilities are properly set

9 years agoMerge pull request #1713 from evverx/test-cap-bounding-set-merging
Ronny Chevalier [Thu, 29 Oct 2015 12:44:23 +0000 (13:44 +0100)]
Merge pull request #1713 from evverx/test-cap-bounding-set-merging

Another fix of the CapabilityBoundingSet parsing

9 years agocore: fix capability bounding set parsing
Evgeny Vereshchagin [Thu, 29 Oct 2015 11:13:04 +0000 (14:13 +0300)]
core: fix capability bounding set parsing

bug: CapabilityBoundingSet= doesn't reset all caps

9 years agotest: add test for capability bounding set parsing
Evgeny Vereshchagin [Thu, 29 Oct 2015 11:12:22 +0000 (14:12 +0300)]
test: add test for capability bounding set parsing

9 years agonetwork: fix use-after-free in link_free
Hristo Venev [Thu, 29 Oct 2015 10:14:44 +0000 (12:14 +0200)]
network: fix use-after-free in link_free

Freeing a link removes it both from addresses and addresses_foreign,
causing SIGSEGV if one of the sets is freed.

9 years agoshell-completion: hostnamectl: add the command set-location
Evgeny Vereshchagin [Thu, 29 Oct 2015 04:47:16 +0000 (07:47 +0300)]
shell-completion: hostnamectl: add the command set-location

9 years agoMerge pull request #1700 from ssahani/word
Lennart Poettering [Wed, 28 Oct 2015 18:32:35 +0000 (19:32 +0100)]
Merge pull request #1700 from ssahani/word

Port to extract_first_word

9 years agoMerge pull request #1707 from dvdhrm/notify-prio
Lennart Poettering [Wed, 28 Oct 2015 18:31:37 +0000 (19:31 +0100)]
Merge pull request #1707 from dvdhrm/notify-prio

core: fix priority ordering in notify-handling

9 years agocore: fix priority ordering in notify-handling
David Herrmann [Wed, 28 Oct 2015 18:11:36 +0000 (19:11 +0100)]
core: fix priority ordering in notify-handling

Currently, we dispatch NOTIFY messages in a tight loop. Regardless how
much data is incoming, we always dispatch everything that is queued.
This, however, completely breaks priority event-handling of sd-event.
When dispatching one NOTIFY event, another completely different event
might fire, or might be queued by the NOTIFY handling. However, this
event will not get dispatched until all other further NOTIFY messages are
handled. Those might even arrive _after_ the other event fired, and as
such completely break priority ordering of sd-event (which several code
paths rely on).

Break this by never dispatching multiple messages. Just return after each
message that was read and let sd-event handle everything else.

(The patch looks scarier that it is. It basically just drops the for(;;)
 loop and re-indents the loop-content.)

9 years agonspwan: port to extract_first_word
Susant Sahani [Wed, 28 Oct 2015 17:29:01 +0000 (22:59 +0530)]
nspwan: port to extract_first_word

9 years agoinstall: port to extract_first_word
Susant Sahani [Wed, 28 Oct 2015 17:28:24 +0000 (22:58 +0530)]
install: port to extract_first_word

9 years agojournal-remote: port to extract_first_word
Susant Sahani [Wed, 28 Oct 2015 17:27:59 +0000 (22:57 +0530)]
journal-remote: port to extract_first_word

9 years agoMerge pull request #1705 from filbranden/testutil1
Lennart Poettering [Wed, 28 Oct 2015 17:19:03 +0000 (18:19 +0100)]
Merge pull request #1705 from filbranden/testutil1

Move tests for extract-word and parse-util into separate test cases

9 years agotest-parse-util: Move parse-util tests into their own test case
Filipe Brandenburger [Wed, 28 Oct 2015 15:14:50 +0000 (08:14 -0700)]
test-parse-util: Move parse-util tests into their own test case

Tests for the functions defined in src/basic/parse-util.c. Reorder them
to match the order in which the functions are defined in the source
file. Adjusted the list of include files to remove the ones no longer
needed in test-util.c.

Tested that `make check` still passes as expected. Also checked the
number of lines removed from test-util.c matches the expected, as an
additional verification that no tests were dropped or duplicated in the
move.

9 years agotest-extract-word: Move extract-word tests into their own test case
Filipe Brandenburger [Wed, 28 Oct 2015 14:57:11 +0000 (07:57 -0700)]
test-extract-word: Move extract-word tests into their own test case

Tests for the functions defined in src/basic/extract-word.c.

Tested that `make check` still passes as expected.

9 years agobuild-sys: Keep .gitignore sorted
Filipe Brandenburger [Wed, 28 Oct 2015 14:42:53 +0000 (07:42 -0700)]
build-sys: Keep .gitignore sorted

Let's try to keep it that way! :-)

9 years agoMerge pull request #1703 from ronnychevalier/rc/test_execute_check_nobody2
Lennart Poettering [Wed, 28 Oct 2015 12:07:48 +0000 (13:07 +0100)]
Merge pull request #1703 from ronnychevalier/rc/test_execute_check_nobody2

test-execute: check if nobody exists before running some tests

9 years agoMerge pull request #1699 from filbranden/cpuaffinity9
Lennart Poettering [Wed, 28 Oct 2015 12:06:56 +0000 (13:06 +0100)]
Merge pull request #1699 from filbranden/cpuaffinity9

cpu-set-util: Support ranges in parse_cpu_set_and_warn (v4)

9 years agotest-execute: check if nobody exists before running some tests
Ronny Chevalier [Wed, 28 Oct 2015 10:30:40 +0000 (11:30 +0100)]
test-execute: check if nobody exists before running some tests

9 years agoMerge pull request #1701 from hosiet/l10n-cn
Daniel Mack [Wed, 28 Oct 2015 09:23:36 +0000 (10:23 +0100)]
Merge pull request #1701 from hosiet/l10n-cn

l10n: add Simplified Chinese translation.

9 years agol10n: add zh_CN translation.
Boyuan Yang [Wed, 28 Oct 2015 07:23:58 +0000 (15:23 +0800)]
l10n: add zh_CN translation.

9 years agoman: Update man page documentation for CPUAffinity
Filipe Brandenburger [Tue, 13 Oct 2015 07:12:39 +0000 (00:12 -0700)]
man: Update man page documentation for CPUAffinity

Document support for commas as a separator and possibility of specifying
ranges of CPU indices.

Tested by regenerating the manpages locally and reading them on man.

9 years agocpu-set-util: Support ranges in parse_cpu_set_and_warn
Filipe Brandenburger [Fri, 25 Sep 2015 12:23:23 +0000 (05:23 -0700)]
cpu-set-util: Support ranges in parse_cpu_set_and_warn

Tested CPUAffinity ranges on both a service unit and in system.conf and
confirmed they work as expected (by inspecting /proc/PID/status, for the
main pid of the service and for pid 1).  Also mixed ranges with both
spaces, commas, trailing commas and spaces.

Added new tests to increase coverage of ranges and prevent regressions.

9 years agoparse-util: Introduce new parse_range function
Filipe Brandenburger [Tue, 13 Oct 2015 06:55:31 +0000 (23:55 -0700)]
parse-util: Introduce new parse_range function

This function will be useful for CPUAffinity settings that involve
ranges of CPUs.

Make it generic and include test coverage to prevent regressions.

9 years agocpu-set-util: Accept commas as separators in parse_cpu_set_and_warn
Filipe Brandenburger [Fri, 25 Sep 2015 11:45:22 +0000 (04:45 -0700)]
cpu-set-util: Accept commas as separators in parse_cpu_set_and_warn

Tested CPUAffinity settings on both a service unit and in system.conf
and confirmed they work as expected.

Added a new test to confirm that trailing commas and spaces work and to
prevent any regressions in that area.

9 years agoMerge pull request #1698 from poettering/pid-is-unwaited-fix
Tom Gundersen [Tue, 27 Oct 2015 17:09:32 +0000 (18:09 +0100)]
Merge pull request #1698 from poettering/pid-is-unwaited-fix

fixup for previous pr

9 years agoservice: exiting early is nicer than deeply-indented code blocks
Lennart Poettering [Tue, 27 Oct 2015 17:02:24 +0000 (18:02 +0100)]
service: exiting early is nicer than deeply-indented code blocks

9 years agocore: all unit types that watch control PIDs should use the same logic
Lennart Poettering [Tue, 27 Oct 2015 16:59:45 +0000 (17:59 +0100)]
core: all unit types that watch control PIDs should use the same logic

When coldplugging the unit state, make sure to follow the same basic
logic for all unit types: always verify whether the control PID is still
a waitable process before proceeding.

9 years agocore: don't pass uninitialzed PIDs to pid_is_unwaited()
Lennart Poettering [Tue, 27 Oct 2015 16:56:12 +0000 (17:56 +0100)]
core: don't pass uninitialzed PIDs to pid_is_unwaited()

Since 5fd9b2c5467b0a42ccdabc7eb8e516d512609a8e passing a pid of 0 to
pid_is_unwaited() and pid_is_live() is considered as a request on the
current process, similar how the other calls in process-util.c handle a
PID of 0. This broke service.c, which passes a 0 PID and expects it to
be considered an unwaited process.

This fix make sure we can boot again.

9 years agoMerge pull request #1695 from evverx/fix-cap-bounding-merging
Lennart Poettering [Tue, 27 Oct 2015 16:09:05 +0000 (17:09 +0100)]
Merge pull request #1695 from evverx/fix-cap-bounding-merging

core: fix CapabilityBoundingSet merging

9 years agoMerge pull request #1691 from poettering/util-lib-3
Tom Gundersen [Tue, 27 Oct 2015 15:55:16 +0000 (16:55 +0100)]
Merge pull request #1691 from poettering/util-lib-3

split up util.h and macro.h into even more bits

9 years agoutil-lib: move inotify-related definitions to fs-util.[ch]
Lennart Poettering [Tue, 27 Oct 2015 13:58:05 +0000 (14:58 +0100)]
util-lib: move inotify-related definitions to fs-util.[ch]

9 years agoutil-lib: move fdname_is_valid() to fd-util.[ch]
Lennart Poettering [Tue, 27 Oct 2015 13:57:44 +0000 (14:57 +0100)]
util-lib: move fdname_is_valid() to fd-util.[ch]

9 years agoautomount: move generically userful call repeat_mount() into mount-util.[ch]
Lennart Poettering [Tue, 27 Oct 2015 13:25:58 +0000 (14:25 +0100)]
automount: move generically userful call repeat_mount() into mount-util.[ch]

9 years agoprocess-util: move a couple of process-related calls over
Lennart Poettering [Tue, 27 Oct 2015 13:24:58 +0000 (14:24 +0100)]
process-util: move a couple of process-related calls over

9 years agoprocess-util: make some minor corrections to PID live detection
Lennart Poettering [Tue, 27 Oct 2015 13:02:45 +0000 (14:02 +0100)]
process-util: make some minor corrections to PID live detection

9 years agoprocess-util: rename get_parent_of_pid() → get_process_ppid()
Lennart Poettering [Tue, 27 Oct 2015 13:01:48 +0000 (14:01 +0100)]
process-util: rename get_parent_of_pid() → get_process_ppid()

In order to match the other get_process_xyz() calls.

9 years agoprocess-util: actually move rename_process() over
Lennart Poettering [Tue, 27 Oct 2015 12:56:40 +0000 (13:56 +0100)]
process-util: actually move rename_process() over

The prototype was moved long ago, actually move the definition over now,
too.

9 years agomount-util: move fstype_is_network() and name_to_handle_at() definitions over
Lennart Poettering [Tue, 27 Oct 2015 12:45:00 +0000 (13:45 +0100)]
mount-util: move fstype_is_network() and name_to_handle_at() definitions over

9 years agoupdate TODO
Lennart Poettering [Mon, 26 Oct 2015 21:02:13 +0000 (22:02 +0100)]
update TODO

9 years agoutil-lib: split out allocation calls into alloc-util.[ch]
Lennart Poettering [Tue, 27 Oct 2015 02:01:06 +0000 (03:01 +0100)]
util-lib: split out allocation calls into alloc-util.[ch]

9 years agoutil-lib: split out globbing related calls into glob-util.[ch]
Lennart Poettering [Tue, 27 Oct 2015 00:48:17 +0000 (01:48 +0100)]
util-lib: split out globbing related calls into glob-util.[ch]

9 years agostat-util.h: move F_TYPE_EQUAL() macro definition to stat-util.h
Lennart Poettering [Tue, 27 Oct 2015 00:26:52 +0000 (01:26 +0100)]
stat-util.h: move F_TYPE_EQUAL() macro definition to stat-util.h