chiark / gitweb /
5 years agopo: update French translation
Sylvain Plantefève [Thu, 19 Feb 2015 22:31:51 +0000 (23:31 +0100)]
po: update French translation

5 years agobuild-sys: bootctl depends on HAVE_BLKID
Marcel Holtmann [Sun, 22 Feb 2015 09:28:34 +0000 (10:28 +0100)]
build-sys: bootctl depends on HAVE_BLKID

5 years agologind: fix a typo in a polkit description
Piotr Drąg [Thu, 19 Feb 2015 15:12:38 +0000 (16:12 +0100)]
logind: fix a typo in a polkit description

5 years agoREADME: add pkg-config to required deps for autogen
Peter Waller [Mon, 22 Dec 2014 12:33:19 +0000 (12:33 +0000)]
README: add pkg-config to required deps for autogen

Otherwise, several macros are undefined.

5 years agoman: make bootup graph consistent
Chris Morin [Wed, 4 Feb 2015 19:54:34 +0000 (14:54 -0500)]
man: make bootup graph consistent

5 years agobuild-sys: update gitignore
Kay Sievers [Fri, 20 Feb 2015 14:42:32 +0000 (15:42 +0100)]
build-sys: update gitignore

5 years agoshared: handle unnamed sockets in socket_address_equal()
Michal Schmidt [Fri, 20 Feb 2015 01:25:16 +0000 (02:25 +0100)]
shared: handle unnamed sockets in socket_address_equal()

Make sure we don't inspect sun_path of unnamed sockets.
Since we cannot know if two unnamed sockets' adresses refer to the same
socket, just return false.

5 years agoshared: avoid semi-duplicating socket_address_equal()
Michal Schmidt [Fri, 20 Feb 2015 01:13:03 +0000 (02:13 +0100)]
shared: avoid semi-duplicating socket_address_equal()

Just call socket_address_equal() from socket_address_matches_fd()
instead of implementing similar comparing of addresses.

5 years agoshared: use SocketAddress in socket_address_matches_fd()
Michal Schmidt [Fri, 20 Feb 2015 01:04:05 +0000 (02:04 +0100)]
shared: use SocketAddress in socket_address_matches_fd()

Cleanup. No behavior change.

5 years agocore, shared: in deserializing, match same files reached via different paths
Michal Schmidt [Thu, 19 Feb 2015 22:12:38 +0000 (23:12 +0100)]
core, shared: in deserializing, match same files reached via different paths

When dbus.socket is updated like this:
... and daemon-reload is performed, bad things happen.
During deserialization systemd does not recognize that the two paths
refer to the same named socket and replaces the socket file with a new
one. As a result, applications hang when they try talking to dbus.

Fix this by finding a match not only when the path names are equal, but
also when they point to the same inode.
In socket_address_equal() it is necessary to move the address size
comparison into the abstract sockets branch. For path name sockets the
comparison must not be done and for other families it is redundant
(their sizes are constant and checked by socket_address_verify()).

FIFOs and special files can also have multiple pathnames, so compare the
inodes for them as well. Note that previously the pathname checks used
streq_ptr(), but the paths cannot be NULL.


5 years agobus: sync with kdbus.git (ABI break)
David Herrmann [Thu, 19 Feb 2015 19:02:40 +0000 (20:02 +0100)]
bus: sync with kdbus.git (ABI break)

This syncs up the new KDBUS_CMD_CONN_INFO behavior:
 - attach-flags are passed in .attach_flags, instead of directly merged
   with the command flags.

5 years agoman: add newlines to the pull-raw example in machinectl(1)
Benjamin Franzke [Thu, 19 Feb 2015 12:10:18 +0000 (13:10 +0100)]
man: add newlines to the pull-raw example in machinectl(1)

They were removed in commit 798d3a52 ("Reindent man pages to 2ch").

5 years agonspawn: chown basic device nodes to userns root
Lennart Poettering [Thu, 19 Feb 2015 11:03:39 +0000 (12:03 +0100)]
nspawn: chown basic device nodes to userns root

5 years agonspawn: fix build on non-selinux systems
Lennart Poettering [Thu, 19 Feb 2015 11:03:12 +0000 (12:03 +0100)]
nspawn: fix build on non-selinux systems

5 years agonspawn: add basic user namespacing support
Lennart Poettering [Thu, 19 Feb 2015 10:30:18 +0000 (11:30 +0100)]
nspawn: add basic user namespacing support

(This is incomplete, /proc and /sys are still owned by root from outside
the container, not inside)

5 years agosysv-generator: fix wrong "Overwriting existing symlink" warnings
Martin Pitt [Thu, 19 Feb 2015 10:06:24 +0000 (11:06 +0100)]
sysv-generator: fix wrong "Overwriting existing symlink" warnings

Fix result testing of is_symlink() to ignore negative results, which happen if
the file name does not exist at all. In this case we do not want a warning and
unlink the non-existing link.

5 years agosystemd-sysv-generator test: Adjust to dropped mapping
Martin Pitt [Thu, 19 Feb 2015 08:09:57 +0000 (09:09 +0100)]
systemd-sysv-generator test: Adjust to dropped mapping

Commit d5d8429a dropped the explicit mapping. Adjust the tests
accordingly to explicitly state the expected targets instead of runlevels.

5 years agohwdb: fix ThinkPad X* Tablet special keys
Lubomir Rintel [Wed, 18 Feb 2015 20:02:01 +0000 (21:02 +0100)]
hwdb: fix ThinkPad X* Tablet special keys

ThinkPad tablet firmware has DMI product name and version reversed:

Handle 0x0001, DMI type 1, 27 bytes
System Information
        Manufacturer: LENOVO
        Product Name: 7762AS1
        Version: ThinkPad X61 Tablet
        Serial Number: LKZCDH2
        UUID: 6ADBC681-4FC9-11CB-844F-B47CB9210BE2
        Wake-up Type: Power Switch
        SKU Number: Not Specified
        Family: ThinkPad X61 Tablet

5 years agopo: update Russian translation
Sergey Ptashnick [Thu, 19 Feb 2015 00:32:42 +0000 (03:32 +0300)]
po: update Russian translation

Add strings for fsckd.

5 years agoupdate TODO
Lennart Poettering [Wed, 18 Feb 2015 22:35:51 +0000 (23:35 +0100)]
update TODO

5 years agonspawn: when connected to pipes for stdin/stdout, pass them as-is to PID 1
Lennart Poettering [Wed, 18 Feb 2015 22:32:55 +0000 (23:32 +0100)]
nspawn: when connected to pipes for stdin/stdout, pass them as-is to PID 1

Previously we always invoked the container PID 1 on /dev/console of the
container. With this change we do so only if nspawn was invoked
interactively (i.e. its stdin/stdout was connected to a TTY). In all other
cases we directly pass through the fds unmodified.

This has the benefit that nspawn can be added into shell pipelines.

5 years agosystemd-boot: fix alignment of version in table
Thomas Hindoe Paaboel Andersen [Wed, 18 Feb 2015 21:18:12 +0000 (22:18 +0100)]
systemd-boot: fix alignment of version in table

5 years agoeverywhere: remove configurability of sysv runlevel to target mapping
Lennart Poettering [Wed, 18 Feb 2015 19:17:49 +0000 (20:17 +0100)]
everywhere: remove configurability of sysv runlevel to target mapping

With this change runlevel 2, 3, 4 are mapped to for
good, and 5 to This was already the previous mapping
but is now no longer reconfigurable, but hard-coded into the core.

This should generally simplify things, but also fix one bug: the
sysv-generator previously generated symlinks to runlevel[2-5].target
units, which possibly weren't picked up if these aliases were otherwise
only referenced by the real names "" and

We keep compat aliases "runlevel[2345].target" arround for cases where
this target name is explicitly requested.

5 years agoupdate TODO
Lennart Poettering [Wed, 18 Feb 2015 18:39:54 +0000 (19:39 +0100)]
update TODO

5 years agorun: if we fail to set a property assignment then really fail
Lennart Poettering [Wed, 18 Feb 2015 18:39:40 +0000 (19:39 +0100)]
run: if we fail to set a property assignment then really fail

5 years agonspawn: add support for --property= to set scope properties
Lennart Poettering [Wed, 18 Feb 2015 18:38:55 +0000 (19:38 +0100)]
nspawn: add support for --property= to set scope properties

This is similar to systemd-run's --property= setting.

5 years agoshared: introduce cmsg_close_all() call
Lennart Poettering [Wed, 18 Feb 2015 18:20:47 +0000 (19:20 +0100)]
shared: introduce cmsg_close_all() call

The call iterates through cmsg list and closes all fds passed via

This patch also ensures the call is used wherever appropriate, where we
might get spurious fds sent and we should better close them, then leave
them lying around.

5 years agosystemctl: allow interactive authorization for all bus calls
Lennart Poettering [Wed, 18 Feb 2015 17:54:38 +0000 (18:54 +0100)]
systemctl: allow interactive authorization for all bus calls

Make use of the new sd_bus_set_allow_interactive_authorization() call to
globally enable interactive authorization. Also, turn on PK agent for
more calls.

This allows us to make use of the sd_bus_call_method() convencience
helper at more places.

5 years agosystemctl: let's make use of FOREACH_STRING() where we can
Lennart Poettering [Wed, 18 Feb 2015 17:27:32 +0000 (18:27 +0100)]
systemctl: let's make use of FOREACH_STRING() where we can

5 years agocore: rework policykit hookup
Lennart Poettering [Wed, 18 Feb 2015 16:40:57 +0000 (17:40 +0100)]
core: rework policykit hookup

- Always issue selinux access check as early as possible, and PK check
  as late as possible.

- Introduce a new policykit action for altering environment

- Open most remaining bus calls to unprivileged clients via PK

5 years agopo: Update German translation
Martin Pitt [Wed, 18 Feb 2015 16:09:15 +0000 (17:09 +0100)]
po: Update German translation

5 years agoAdd mock fsck process
Didier Roche [Mon, 26 Jan 2015 16:46:36 +0000 (17:46 +0100)]
Add mock fsck process

5 years agoAdd man page and references to it.
Didier Roche [Mon, 26 Jan 2015 16:34:59 +0000 (17:34 +0100)]
Add man page and references to it.

Add man page explaining the plymouth theme protocol, usage of the daemon
as well as the socket activation part.
Adapt existing fsck man page.

5 years agoAdd fsckd service and socket, retarget systemd-fsck
Didier Roche [Mon, 26 Jan 2015 16:30:00 +0000 (17:30 +0100)]
Add fsckd service and socket, retarget systemd-fsck

systemd-fsckd can be socket-activated by systemd-fsck process. Reflect that
in the different unit files.

5 years agoRefresh po files
Didier Roche [Mon, 26 Jan 2015 16:23:08 +0000 (17:23 +0100)]
Refresh po files

Also, add new plymouth fsckd translated strings in french.
Refreshed with "make update-po".

5 years agoTranslate fsckd messages for plymouth
Didier Roche [Mon, 26 Jan 2015 16:12:54 +0000 (17:12 +0100)]
Translate fsckd messages for plymouth

For plymouth themes not supporting i18n (like .script), send translated
messages to display to user, which is equivalent to the sent machine
readable data.

5 years agoAdd gettext support
Didier Roche [Thu, 29 Jan 2015 15:12:58 +0000 (16:12 +0100)]
Add gettext support

5 years agoConnect to plymouth and support cancellation of in progress fsck
Didier Roche [Thu, 5 Feb 2015 16:08:18 +0000 (17:08 +0100)]
Connect to plymouth and support cancellation of in progress fsck

Try to connect and send to plymouth (if running) some checked report progress,
using direct plymouth protocole.

Update message is the following:
* num_devices corresponds to the current number of devices being checked (int)
* progress corresponds to the current minimum percentage of all devices being
  checked (float, from 0 to 100)
* string is a translated message ready to be displayed by the plymouth theme
  displaying the information above. It can be overriden by plymouth themes
  supporting i18n.

Grab in fsckd plymouth watch key Control+C, and propagate this cancel request
to systemd-fsck which will terminate fsck.

Send a message to signal to user what key we are grabbing for fsck cancel.

Message is: fsckd-cancel-msg:<string>
Where string is a translated string ready to be displayed by the plymouth theme
indicating that Control+C can be used to cancel current checks. It can be
overriden (matching only fsckd-cancel-msg prefix) for themes supporting i18n.

5 years agosystemd-fsck: always connect to systemd-fsckd
Didier Roche [Mon, 26 Jan 2015 15:01:11 +0000 (16:01 +0100)]
systemd-fsck: always connect to systemd-fsckd

Remove the plymouth running or show-status checks from systemd-fsck. Instead,
always connect to systemd-fsckd socket, and let this one decide if we display
progress or not.

5 years agofsckd daemon for inter-fsckd communication
Didier Roche [Wed, 4 Feb 2015 15:42:47 +0000 (16:42 +0100)]
fsckd daemon for inter-fsckd communication

Add systemd-fsckd multiplexer which accepts multiple systemd-fsck
instances to connect to it and sends progress report. systemd-fsckd then
computes and writes to /dev/console the number of devices currently being
checked and the minimum fsck progress. This will be used for interactive
progress report and cancelling in plymouth.

systemd-fsckd stops on idle when no systemd-fsck is connected.

Make the necessary changes to systemd-fsck to connect to the systemd-fsckd

5 years agotest: utf8 - fix utf16 tests on BE machines
Tom Gundersen [Wed, 18 Feb 2015 13:33:50 +0000 (14:33 +0100)]
test: utf8 - fix utf16 tests on BE machines

5 years agogummiboot/sd-boot/systemd-boot: rename galore
Tom Gundersen [Tue, 17 Feb 2015 19:12:50 +0000 (20:12 +0100)]
gummiboot/sd-boot/systemd-boot: rename galore

What used to be gummiboot, was renamed sd-boot when it was merged into
systemd. Let's try to be a bit more consistent with the rest of systemd
and rename it again as follows:

The EFI bootloader is now called 'systemd-bootx64.efi', and its sources are in
'src/boot/efi/'. The drop-in directory where bootctl will find EFI loaders
is now /usr/lib/systemd/boot/efi/.

5 years compile with gcc v5
Harald Hoyer [Wed, 18 Feb 2015 13:45:12 +0000 (14:45 +0100)] compile with gcc v5

Set -std=gnu90 to let efibind.h define the standard types.

[tomegun: retyped the patch as the originl would not apply]

5 years agoupdate TODO
Lennart Poettering [Wed, 18 Feb 2015 12:08:23 +0000 (13:08 +0100)]
update TODO

5 years agologind: open up most bus calls for unpriviliged processes, using PolicyKit
Lennart Poettering [Wed, 18 Feb 2015 11:55:25 +0000 (12:55 +0100)]
logind: open up most bus calls for unpriviliged processes, using PolicyKit

Also, allow clients to alter their own objects without any further
priviliges. i.e. this allows clients to kill and lock their own sessions
without involving PK.

5 years agomachinectl: issue all bus commands while allowing interactive auth
Lennart Poettering [Wed, 18 Feb 2015 10:42:03 +0000 (11:42 +0100)]
machinectl: issue all bus commands while allowing interactive auth

5 years agomachined: open up most of machined's commands to unprivileged clients via PolicyKit
Lennart Poettering [Wed, 18 Feb 2015 10:41:28 +0000 (11:41 +0100)]
machined: open up most of machined's commands to unprivileged clients via PolicyKit

5 years agosd-bus: allow setting a per-connection default value for the "allow-interactive-authe...
Lennart Poettering [Tue, 17 Feb 2015 19:30:33 +0000 (20:30 +0100)]
sd-bus: allow setting a per-connection default value for the "allow-interactive-authentication" message flag

Most of our client tools want to set this bit for all their method
calls, even though it defaults to off in sd-bus, and rightfully so.
Hence, to simplify thing, introduce a per sd_bus-object flag that sets
the default value for all messages created on the connection.

5 years agoimport: remove unused variable
Thomas Hindoe Paaboel Andersen [Tue, 17 Feb 2015 19:06:13 +0000 (20:06 +0100)]
import: remove unused variable

5 years agomachined: make "machinectl copy-to" and "machinectl copy-from" server side operations
Lennart Poettering [Tue, 17 Feb 2015 18:22:03 +0000 (19:22 +0100)]
machined: make "machinectl copy-to" and "machinectl copy-from" server side operations

This way, any bus client can make use of these calls.

5 years agomachined: various simplifications
Lennart Poettering [Tue, 17 Feb 2015 17:54:33 +0000 (18:54 +0100)]
machined: various simplifications

5 years agobus: sync with kdbus
David Herrmann [Tue, 17 Feb 2015 17:05:32 +0000 (18:05 +0100)]
bus: sync with kdbus

Pull in new kdbus changes, namely:
 - ENOMSG is merged with EAGAIN for consistency

5 years agobus: don't rely on static IDs in tests
David Herrmann [Tue, 17 Feb 2015 17:03:50 +0000 (18:03 +0100)]
bus: don't rely on static IDs in tests

We should never put any constraints on the IDs returned by the kernel. Use
sd_bus_get_unique_name() to retrieve the ID of a peer.

5 years agobuild-sys: add missing HAVE_GNUEFI condition
Kay Sievers [Tue, 17 Feb 2015 17:01:10 +0000 (18:01 +0100)]
build-sys: add missing HAVE_GNUEFI condition

5 years agomachined: move logic for bind mounting into containers from machinectl to machined
Lennart Poettering [Tue, 17 Feb 2015 16:19:57 +0000 (17:19 +0100)]
machined: move logic for bind mounting into containers from machinectl to machined

This extends the bus interface, adding BindMountMachine() for bind
mounting directories from the host into the container.

5 years agobootctl: add sd-boot support
Kay Sievers [Sun, 8 Feb 2015 16:18:30 +0000 (17:18 +0100)]
bootctl: add sd-boot support

5 years agosd-boot: add EFI boot manager and stub loader
Kay Sievers [Sun, 8 Feb 2015 11:25:35 +0000 (12:25 +0100)]
sd-boot: add EFI boot manager and stub loader

5 years agotmpfiles: Fix parse_acl error message
Martin Pitt [Tue, 17 Feb 2015 11:47:51 +0000 (12:47 +0100)]
tmpfiles: Fix parse_acl error message

parse_acl() returns the error instead of setting errno.

5 years agojournal-remote: fix client_cert memory leak
Michal Schmidt [Tue, 17 Feb 2015 09:36:57 +0000 (10:36 +0100)]
journal-remote: fix client_cert memory leak

Found by Valgrind while testing the previous memory leak fix.

5 years agojournal-remote: fix certificate status memory leak
Michal Schmidt [Tue, 17 Feb 2015 09:33:01 +0000 (10:33 +0100)]
journal-remote: fix certificate status memory leak

The output of gnutls_certificate_verification_status_print() needs to be

Noticed this while staring at verify_cert_authorized() to see what could
possibly confuse gcc5 on armv7hl to segfault during compilation.

5 years agoupdate TODO v219
Lennart Poettering [Mon, 16 Feb 2015 20:54:57 +0000 (21:54 +0100)]
update TODO

5 years agoRevert "build-sys: re-add old symbols for abi compat"
Michal Schmidt [Mon, 16 Feb 2015 20:17:48 +0000 (21:17 +0100)]
Revert "build-sys: re-add old symbols for abi compat"

This reverts commit 93a1e66efd4b0f4cda29c467d20d0f7510c0b3a8.

It broke the builds for Walters.
Kay hates it and "just knows" there are no actual users.

5 years agoNEWS: typo fixes
Thomas Hindoe Paaboel Andersen [Mon, 16 Feb 2015 20:21:16 +0000 (21:21 +0100)]
NEWS: typo fixes

5 years agoTranspose args in strv_fnmatch() to be more oo
Zbigniew Jędrzejewski-Szmek [Mon, 16 Feb 2015 19:04:36 +0000 (14:04 -0500)]
Transpose args in strv_fnmatch() to be more oo

5 years agotest-network: fix segfault with NULL dev_path/driver/type/name
Zbigniew Jędrzejewski-Szmek [Mon, 16 Feb 2015 18:56:31 +0000 (13:56 -0500)]
test-network: fix segfault with NULL dev_path/driver/type/name

5 years agoupdate TODO
Lennart Poettering [Mon, 16 Feb 2015 18:36:56 +0000 (19:36 +0100)]
update TODO

5 years agoNEWS: fix typo
Lennart Poettering [Mon, 16 Feb 2015 18:36:51 +0000 (19:36 +0100)]
NEWS: fix typo

5 years agoanalyze: change behaviour of combined --to/from--pattern
Zbigniew Jędrzejewski-Szmek [Fri, 13 Feb 2015 23:38:33 +0000 (18:38 -0500)]
analyze: change behaviour of combined --to/from--pattern

We would require a match against all three: patterns specified
with --to, with --from, and as positional arguments to show an
edge. This does not seem useful. Let instead the positional args
behave like they were specified in both --to and --from, which is
fairly intuitive and should be more useful.

5 years agonetwork-internal: chain matches with AND in net_match_config()
Zbigniew Jędrzejewski-Szmek [Fri, 13 Feb 2015 23:38:22 +0000 (18:38 -0500)]
network-internal: chain matches with AND in net_match_config()

The test would treat the first non-empty set of matches in
match_paths, match_drivers, match_types, match_names as definitive
(essentially chaining them with OR). Make those tests instead match
like other tests and require all to pass if the set of patterns is

5 years agoAdd helper for fnmatch over strv
Zbigniew Jędrzejewski-Szmek [Fri, 13 Feb 2015 23:37:43 +0000 (18:37 -0500)]
Add helper for fnmatch over strv

5 years agohwdb: Update database of Bluetooth company identifiers
Marcel Holtmann [Mon, 16 Feb 2015 18:10:39 +0000 (19:10 +0100)]
hwdb: Update database of Bluetooth company identifiers

5 years agotimesync: Use UINT64_C for OFFSET_1900_1970
Cristian Rodríguez [Mon, 16 Feb 2015 16:24:29 +0000 (13:24 -0300)]
timesync: Use UINT64_C for OFFSET_1900_1970

So it matches what the comment says in both 32 and 64 bit systems.

5 years agoupdate TODO
Lennart Poettering [Mon, 16 Feb 2015 17:41:49 +0000 (18:41 +0100)]
update TODO

5 years agobuild-sys: re-add old symbols for abi compat
Zbigniew Jędrzejewski-Szmek [Mon, 16 Feb 2015 17:13:20 +0000 (12:13 -0500)]
build-sys: re-add old symbols for abi compat

For two releases those exported with version 183 by mistake, and then
they were fixed to have version 215 (015419c0df libudev: fix symbol
version for udev_queue_flush() and udev_queue_get_fd()). But that
breaks ABI compatibility for binaries compiled with udev from before
that commit. There most likely very few such binaries, if any, but as
a matter of principle we should export the old symbols too, in order
to keep full compatibility.

5 years agohwdb: update
Lennart Poettering [Mon, 16 Feb 2015 16:28:35 +0000 (17:28 +0100)]
hwdb: update

5 years agosd-bus: add missing format attribute
Cristian Rodríguez [Mon, 16 Feb 2015 04:09:55 +0000 (01:09 -0300)]
sd-bus: add missing format attribute

5 years agoupdate NEWS
Lennart Poettering [Mon, 16 Feb 2015 16:17:07 +0000 (17:17 +0100)]
update NEWS

5 years agohwdb: fix syntax
Lennart Poettering [Mon, 16 Feb 2015 16:14:20 +0000 (17:14 +0100)]
hwdb: fix syntax

5 years agosd-bus: make sure we always serialize the reply cookie as 64bit on dbus2, and 32bit...
Lennart Poettering [Mon, 16 Feb 2015 14:59:45 +0000 (15:59 +0100)]
sd-bus: make sure we always serialize the reply cookie as 64bit on dbus2, and 32bit on dbus1

5 years agoRevert "Revert "sd-bus: change serialization of kdbus messages to qualify in their...
Lennart Poettering [Fri, 13 Feb 2015 16:32:38 +0000 (17:32 +0100)]
Revert "Revert "sd-bus: change serialization of kdbus messages to qualify in their entirety as gvariant objects""

This reverts commit 954871d8ba15911d014f76ed2c7a9492953cf39d.

5 years agopo: Update German translation
Christian Kirbach [Sun, 15 Feb 2015 13:30:05 +0000 (14:30 +0100)]
po: Update German translation

5 years agohwdb: Microsoft Microsoft 3-Button Mouse
Lukas Märdian [Sat, 14 Feb 2015 19:37:40 +0000 (14:37 -0500)]
hwdb: Microsoft Microsoft 3-Button Mouse

5 years agocore: make RuntimeDirectory honor SELinux labels
Zbigniew Jędrzejewski-Szmek [Sat, 14 Feb 2015 19:11:24 +0000 (14:11 -0500)]
core: make RuntimeDirectory honor SELinux labels

5 years agobus-proxy: be more verbose if messages cannot be forwarded
David Herrmann [Sat, 14 Feb 2015 18:01:21 +0000 (19:01 +0100)]
bus-proxy: be more verbose if messages cannot be forwarded

If we cannot forward messages, include information on the peer and message
just like the xml-policy does. This helps debugging such situations and
figuring out what exactly is going wrong.

5 years agobootchart: display each CPU utilization/wait
WaLyong Cho [Fri, 13 Feb 2015 05:59:39 +0000 (14:59 +0900)]
bootchart: display each CPU utilization/wait

5 years agonetwork: fix 'LinkLocal' -> 'LinkLocalAddressing' in network files
David Herrmann [Sat, 14 Feb 2015 14:31:26 +0000 (15:31 +0100)]
network: fix 'LinkLocal' -> 'LinkLocalAddressing' in network files

'LinkLocal' was renamed to 'LinkLocalAddressing' in:

    commit 56fd6bf795926409b087bce406ea851ad89f9fe8
    Author: Tom Gundersen <>
    Date:   Sat Feb 14 00:32:26 2015 +0100

        networkd: .network - rename LinkLocal to LinkLocalAddressing

..but apparently the network files were not updated. Fix this.

5 years agobus-proxy: don't fake data we don't have
David Herrmann [Sat, 14 Feb 2015 14:13:38 +0000 (15:13 +0100)]
bus-proxy: don't fake data we don't have

UDS sockets transmit EUID+EGID only. Don't try to fake data we don't know!
Otherwise, this might be used to override user-limits by non-root setuid
programs (by faking UID==EUID).

Now that sd-bus is fixed to always use EUID even on UDS, we can safely set
all other UID/GID fields to INVALID.

5 years agoutil: use a shared lookup function for string tables
Bruno Bottazzini [Fri, 13 Feb 2015 20:40:50 +0000 (18:40 -0200)]
util: use a shared lookup function for string tables

Macro DEFINE_STRING_TABLE_LOOKUP expands to a new function for each
of the almost 120 tables throghout the code.
Move the its implementation to a function (guaranteed to never be inlined),
and make the macro expand to an inlined function that calls this function.
This saves a few kilobytes from the systemd binary

(David: - fix coding-style
        - use 'ssize_t' to fix 32bit to 64bit propagation
        - use streq_ptr())

5 years agonetworkd: network - rename ipv6token parser
Tom Gundersen [Fri, 13 Feb 2015 23:32:56 +0000 (00:32 +0100)]
networkd: network - rename ipv6token parser

Too generic name.

5 years agonetworkd: .network - rename LinkLocal to LinkLocalAddressing
Tom Gundersen [Fri, 13 Feb 2015 23:32:26 +0000 (00:32 +0100)]
networkd: .network - rename LinkLocal to LinkLocalAddressing

Makes it a bit less ambiguous.

5 years agoman: - document IPv6Token
Tom Gundersen [Fri, 13 Feb 2015 23:30:51 +0000 (00:30 +0100)]
man: - document IPv6Token

5 years agosd-rtnl: add missing IFLA_INET6_ADDR_GEN_MODE definition
Djalal Harouni [Fri, 13 Feb 2015 22:39:13 +0000 (23:39 +0100)]
sd-rtnl: add missing IFLA_INET6_ADDR_GEN_MODE definition

Add missing IFLA_INET6_ADDR_GEN_MODE definition so we can build with
kernel headers < 3.17

5 years agoTODO
Tom Gundersen [Fri, 13 Feb 2015 23:14:52 +0000 (00:14 +0100)]

5 years agobus-proxy: ECONNRESET/ENOTCONN can hit us on every step, hence handle it on every...
Lennart Poettering [Fri, 13 Feb 2015 16:17:45 +0000 (17:17 +0100)]
bus-proxy: ECONNRESET/ENOTCONN can hit us on every step, hence handle it on every step

5 years agobus-proxy: a few simplifications
Lennart Poettering [Fri, 13 Feb 2015 16:17:28 +0000 (17:17 +0100)]
bus-proxy: a few simplifications

5 years agosd-bus: initialize a few structs at time or declaration
Lennart Poettering [Fri, 13 Feb 2015 16:16:51 +0000 (17:16 +0100)]
sd-bus: initialize a few structs at time or declaration

5 years agobus-proxy: close each connection fd only once
Lennart Poettering [Fri, 13 Feb 2015 16:15:41 +0000 (17:15 +0100)]
bus-proxy: close each connection fd only once

After passing the fds over to the sd_bus object, we should forget them,
so that we don't close them a second time when the object goes away.

5 years agobus-proxy: also consider ENOTCONN a clean termination condition
Lennart Poettering [Fri, 13 Feb 2015 15:11:45 +0000 (16:11 +0100)]
bus-proxy: also consider ENOTCONN a clean termination condition

Sometimes, when we try to reply to messages we don't check return
values. This means we might miss a ECONNRESET, and will get a ENOTCONN
on next command. Treat both the same hence.

5 years agoexit-on-idle: only exit if actually idle
Tom Gundersen [Fri, 13 Feb 2015 15:20:45 +0000 (16:20 +0100)]
exit-on-idle: only exit if actually idle

sd_event_wait() returning 0 usually means that it timed out, which means it must
have been idle. However, sd_event_wait() may return 0 in case an event was triggered
but it turned out there was nothing to do. Make the check for idle explicit to avoid
this edge-case.

5 years agobus-proxy: whenever we cannot forward a message, report this back to caller, but...
Lennart Poettering [Fri, 13 Feb 2015 14:38:38 +0000 (15:38 +0100)]
bus-proxy: whenever we cannot forward a message, report this back to caller, but don't exit

Errors like EPERM from the kernel should certainly not be reason to
exit. Let's try to be defensive here, and try to continue on most send
errors, but possibly tell the sender about it.