chiark / gitweb /
Lennart Poettering [Wed, 26 Nov 2014 14:43:20 +0000 (15:43 +0100)]
core: reindent mount/kmod tables
Lennart Poettering [Wed, 26 Nov 2014 14:42:54 +0000 (15:42 +0100)]
update TODO
Lennart Poettering [Wed, 26 Nov 2014 14:01:50 +0000 (15:01 +0100)]
update TODO
Lennart Poettering [Wed, 26 Nov 2014 13:59:12 +0000 (14:59 +0100)]
sd-bus: given that the kernel now passes the auxgroups list as 32bit array to us, no need to convert to uid_t manually
This way, we can save one allocation and avoid copying the array
unnecesarily.
Lennart Poettering [Wed, 26 Nov 2014 13:14:17 +0000 (14:14 +0100)]
update TODO
Lennart Poettering [Wed, 26 Nov 2014 11:11:39 +0000 (12:11 +0100)]
update TODO
David Herrmann [Wed, 26 Nov 2014 10:11:54 +0000 (11:11 +0100)]
bus: update kdbus.h (ABI break)
We changed creds to u32, so fix systemd sd-bus to acknowledge that.
Lennart Poettering [Wed, 26 Nov 2014 02:18:17 +0000 (03:18 +0100)]
reorder TODO a bit
Zbigniew Jędrzejewski-Szmek [Wed, 26 Nov 2014 01:32:03 +0000 (20:32 -0500)]
udevadm trigger: allow matching by device name
This makes udevadm trigger mirror udevadm info, except that multiple
device names can be specified. Instructions in 60-keyboard.hwdb should
now actually work.
udevadm(8) is updated, but it could use a bit more polishing.
https://bugs.freedesktop.org/show_bug.cgi?id=82311
Zbigniew Jędrzejewski-Szmek [Wed, 26 Nov 2014 00:39:19 +0000 (19:39 -0500)]
udevadm: split out find_device helper
The idea is to unify the way that devices can be specified.
Lennart Poettering [Wed, 26 Nov 2014 01:55:42 +0000 (02:55 +0100)]
update TODO
Lennart Poettering [Wed, 26 Nov 2014 01:54:41 +0000 (02:54 +0100)]
bus-kernel: when installing an activator, ask for any kind of metadata to be attached to incoming messages
We don't know what the activated service might want in the end, hence
enable everything current and future, just to be sure.
Lennart Poettering [Wed, 26 Nov 2014 01:54:11 +0000 (02:54 +0100)]
sd-bus: don't clobber return values on failure in bus_kernel_open_bus_fd()
Lennart Poettering [Wed, 26 Nov 2014 01:50:49 +0000 (02:50 +0100)]
sd-bus: set per-bus attach flag requirement mask to ANY
On the system and user busses we create it's the receiver that chooses
which metadata is attched, not the sender, hence set the requirement
mask to ANY, to allow any current of future credential bit to be
attached.
Lennart Poettering [Wed, 26 Nov 2014 01:21:15 +0000 (02:21 +0100)]
update TODO
Lennart Poettering [Wed, 26 Nov 2014 01:20:28 +0000 (02:20 +0100)]
sd-bus: add suppot for renegotiating message credential attach flags
Lennart Poettering [Wed, 26 Nov 2014 00:45:14 +0000 (01:45 +0100)]
sd-bus: use free_and_strdup() where appropriate
This simplifies things a bit and makes sure we free any previously set
creds component before writing in a new one.
Tom Gundersen [Tue, 25 Nov 2014 21:30:29 +0000 (22:30 +0100)]
resolved: fix typo in sd_notify() call
Lennart Poettering [Tue, 25 Nov 2014 19:50:55 +0000 (20:50 +0100)]
update TODO
Lennart Poettering [Tue, 25 Nov 2014 18:32:48 +0000 (19:32 +0100)]
core: show log message about process triggering kdbus service activation
Lennart Poettering [Tue, 25 Nov 2014 18:54:18 +0000 (19:54 +0100)]
sd-bus: react properly to EOVERFLOW by generating a log message about dropped broadcast messages and proceeding
Lennart Poettering [Tue, 25 Nov 2014 18:25:33 +0000 (19:25 +0100)]
kdbus: update header file to current upstream version
Lennart Poettering [Tue, 25 Nov 2014 18:25:19 +0000 (19:25 +0100)]
kdbus: minor simplification
Lennart Poettering [Tue, 25 Nov 2014 18:25:03 +0000 (19:25 +0100)]
update TODO
Lennart Poettering [Tue, 25 Nov 2014 18:24:45 +0000 (19:24 +0100)]
util: mark page_size() as pure
Lennart Poettering [Tue, 25 Nov 2014 12:24:08 +0000 (13:24 +0100)]
update TODO
Lennart Poettering [Tue, 25 Nov 2014 12:23:44 +0000 (13:23 +0100)]
sd-bus: fix error handling when receiving invalid service name
Also, properly keep track of incoming additional service names.
Lennart Poettering [Tue, 25 Nov 2014 12:22:55 +0000 (13:22 +0100)]
sd-bus: properly handle non-initialized audit records attached to incoming kernel messages
Lennart Poettering [Tue, 25 Nov 2014 12:22:19 +0000 (13:22 +0100)]
sd-bus: properly copy selinux label and description field when duplicating creds object
Lennart Poettering [Tue, 25 Nov 2014 12:21:23 +0000 (13:21 +0100)]
sd-bus: add supplementary groups list to creds object
Lennart Poettering [Tue, 25 Nov 2014 01:11:57 +0000 (02:11 +0100)]
busctl: if no parameter is specified for "busctl status" show credentials of bus owner
Lennart Poettering [Tue, 25 Nov 2014 00:59:48 +0000 (01:59 +0100)]
sd-bus: properly handle uninitialized audit creds from kdbus
Lennart Poettering [Mon, 24 Nov 2014 21:11:53 +0000 (22:11 +0100)]
sd-bus: don't fail when querying creds and dbus1 refuses to tell us the selinux context
Lennart Poettering [Mon, 24 Nov 2014 20:53:29 +0000 (21:53 +0100)]
busctl: add new --augment-creds= switch for controlling whether shown credential data shall be augment with data from /proc
Lennart Poettering [Mon, 24 Nov 2014 20:41:40 +0000 (21:41 +0100)]
sd-bus: update to current kernel version, by splitting off the extended KDBUS_ITEM_PIDS structure from KDBUS_ITEM_CREDS
Also:
- adds support for euid, suid, fsuid, egid, sgid, fsgid fields.
- makes augmentation of creds with data from /proc explicitly
controllable to give apps better control over this, given that this is
racy.
- enables augmentation for kdbus connections (previously we only did it
for dbus1). This is useful since with recent kdbus versions it is
possible for clients to control the metadata they want to send.
- changes sd_bus_query_sender_privilege() to take the euid of the client
into consideration, if known
- when we don't have permissions to read augmentation data from /proc,
don't fail, just don't add the data in
Lennart Poettering [Mon, 24 Nov 2014 20:39:18 +0000 (21:39 +0100)]
busctl: improve readability a bit
Lennart Poettering [Mon, 24 Nov 2014 20:37:26 +0000 (21:37 +0100)]
bus: change creds dumping order to be more close to internal storage order
Lennart Poettering [Mon, 24 Nov 2014 20:21:24 +0000 (21:21 +0100)]
util: skip incomplete ucred information in getpeersec()
Lennart Poettering [Mon, 24 Nov 2014 20:20:53 +0000 (21:20 +0100)]
sd-bus: don't blindly take incomplete ucred bits from AF_UNIX when constructing message
Lennart Poettering [Mon, 24 Nov 2014 16:55:44 +0000 (17:55 +0100)]
update TODO
Peter Hutterer [Tue, 25 Nov 2014 10:45:50 +0000 (20:45 +1000)]
hwdb: fix a typo
Martin Pitt [Tue, 25 Nov 2014 10:39:30 +0000 (11:39 +0100)]
tmpfiles.d: Fix directory name
The .service uses "/var/lib/container", not "containers".
David Herrmann [Tue, 25 Nov 2014 09:26:08 +0000 (10:26 +0100)]
localed: verify xkb keymaps after pk-queries
Make sure not to run xkb-keymap validation twice if pk-queries are used.
Move it below pk-checks.
David Herrmann [Tue, 25 Nov 2014 09:24:39 +0000 (10:24 +0100)]
terminal/idev: forward xkb-messages
Properly forward all XKB messages. You can use XKB_LOG_VERBOSITY= to
control the amount of messages sent by XKB. We explicitly set
XKB_LOG_LEVEL to 7 you can use SYSTEMD_LOG_LEVEL to control the log-level
generically.
David Herrmann [Tue, 25 Nov 2014 09:24:08 +0000 (10:24 +0100)]
terminal/idev: avoid magic numbers
Use XKB_CONTEXT_NO_FLAGS instead of magic 0.
David Herrmann [Mon, 24 Nov 2014 17:25:30 +0000 (18:25 +0100)]
terminal/idev: use compose tables
Before forwarding keyboard events, feed them into possible compose tables.
This enables Compose-key and Dead-key features.
Few notes:
* REPEAT events are never fed into compose tables. It just doesn't make
sense and is usually not wanted. Compose-sequences are usually hard to
remember and take time to type. Thus, the REPEAT event of the
Compose-key itself would often cancel the compose sequence already.
* Stop resolving symbols for UP events. Anything but keycodes is never
associated to a physical key, but is a one-time action. There is
nothing like UP events for key-symbols!
* Cancel compose-sequences on Multi-Key UP. See the inline comment. We
should make this configurable!
David Herrmann [Mon, 24 Nov 2014 15:32:54 +0000 (16:32 +0100)]
terminal/idev: add compose-file support
Add support for compose files to idev-keyboard. This requires
libxkbcommon-0.5.0, which is pretty new, but should be fine.
We don't use the compose-files, yet. Further commits will put life into
them.
WaLyong Cho [Mon, 24 Nov 2014 11:46:20 +0000 (20:46 +0900)]
smack: introduce new SmackProcessLabel option
In service file, if the file has some of special SMACK label in
ExecStart= and systemd has no permission for the special SMACK label
then permission error will occurred. To resolve this, systemd should
be able to set its SMACK label to something accessible of ExecStart=.
So introduce new SmackProcessLabel. If label is specified with
SmackProcessLabel= then the child systemd will set its label to
that. To successfully execute the ExecStart=, accessible label should
be specified with SmackProcessLabel=.
Additionally, by SMACK policy, if the file in ExecStart= has no
SMACK64EXEC then the executed process will have given label by
SmackProcessLabel=. But if the file has SMACK64EXEC then the
SMACK64EXEC label will be overridden.
[zj: reword man page]
Zbigniew Jędrzejewski-Szmek [Mon, 24 Nov 2014 01:01:40 +0000 (20:01 -0500)]
gpt-auto-generator: properly ignore value
A negative return code was treated as a true value.
David Herrmann [Mon, 24 Nov 2014 14:39:00 +0000 (15:39 +0100)]
build-sys: support local ./configure arguments
I often want to use the awesome "./autogen.sh [cmd]" arguments, but have
to append some custom ./configure options. For now, I always had to edit
autogen.sh manually, or copy the full commands out of it and run it
myself.
As I think this is super annoying, this commit adds support for
".config.args" files in $topdir. If it exists, any content is just
appended to $args, thus to any ./configure invokation of autogen.sh.
Maybe autotools provide something similar out-of-the-box. In that case,
feel free to revert this and lemme know!
David Herrmann [Mon, 24 Nov 2014 14:12:42 +0000 (15:12 +0100)]
localed: validate xkb keymaps
Introduce a new optional dependency on libxkbcommon for systemd-localed.
Whenever the x11 keymap settings are changed, use libxkbcommon to compile
the keymap. If the compilation fails, print a warning so users will get
notified.
On compilation failure, we still update the keymap settings for now. This
patch just introduces the xkbcommon infrastructure to have keymap
validation in place. We can later decide if/how we want to enforce this.
Zbigniew Jędrzejewski-Szmek [Mon, 24 Nov 2014 14:11:12 +0000 (09:11 -0500)]
cryptsetup: default to no hash when keyfile is specified
For plain dm-crypt devices, the behavior of cryptsetup package is to
ignore the hash algorithm when a key file is provided. It seems wrong
to ignore a hash when it is explicitly specified, but we should default
to no hash if the keyfile is specified.
https://bugs.freedesktop.org/show_bug.cgi?id=52630
Zbigniew Jędrzejewski-Szmek [Thu, 6 Nov 2014 03:39:13 +0000 (22:39 -0500)]
manager: only issue overmount warning when the check succeeded
If for any reason the check failed (selinux?), we would still issue
the warning. Check the return status.
Zbigniew Jędrzejewski-Szmek [Sat, 8 Nov 2014 15:06:12 +0000 (10:06 -0500)]
manager: let manager_free() handle NULLs
This makes the calling code a bit simpler.
Lukas Nykryn [Thu, 20 Nov 2014 15:06:18 +0000 (16:06 +0100)]
dbus: return non-zero return value in the case that prefix won't match
strv_extend returns 0 in the case of success which means that
else if (bus_track_deserialize_item(&m->deserialized_subscribed, l) == 0)
log_warning("Unknown serialization item '%s'", l);
will be printed when value is added correctly.
Iago López Galeiras [Thu, 20 Nov 2014 20:18:23 +0000 (21:18 +0100)]
test: support empty environment variables in unit files
Also update TODO, empty environment variables in Environment= and
EnvironmentFile= options work.
Zbigniew Jędrzejewski-Szmek [Sun, 23 Nov 2014 16:07:15 +0000 (11:07 -0500)]
man: reword list-dependencies description
The previous version was a bit too vague. It is better
to simply list all dependency types that are followed.
Previous version also made an emphasis on dependencies introduced by
configuration. But this command (or systemd) don't care about this
distinction between configured and automatically added dependencies at
all. This distinctionis removed from the main description, and an
explanatory paragraph is added to remind the user that all
dependencies are shown, no matter where they come from.
Zbigniew Jędrzejewski-Szmek [Sun, 23 Nov 2014 01:21:56 +0000 (20:21 -0500)]
systemctl: show BindsTo/BoundBy in list-dependencies
Suggested-by: Peter Mattern <matternp@arcor.de>
http://lists.freedesktop.org/archives/systemd-devel/2014-November/025437.html
Richard Schütz [Fri, 21 Nov 2014 20:44:48 +0000 (21:44 +0100)]
nspawn: ignore EEXIST when mounting tmpfs
commit
79d80fc1466512d0ca211f4bfcd9de5f2f816a5a introduced a regression that
prevents mounting a tmpfs if the mount point already exits in the container's
root file system. This commit fixes the problem by ignoring EEXIST.
Chris Atkinson [Sat, 22 Nov 2014 20:58:47 +0000 (15:58 -0500)]
man/logind.conf.xml: make consistent with default logind.conf
https://bugs.freedesktop.org/show_bug.cgi?id=86263
Zbigniew Jędrzejewski-Szmek [Sat, 22 Nov 2014 23:00:07 +0000 (18:00 -0500)]
Do not check for existence of remote binaries
systemd-run would fail when run with -M or -H and an absolute path,
if this path did not exists locally. Allow it to continue, since we
don't have a nice way of checking if the binary exists remotely.
The case where -M or -H is used and a local path is unchanged, and we
still iterate over $PATH to find the binary. We need to convert to an
absolute path, and we don't have a nice mechanism to check remotely,
so we assume that the binary will be located in the same place locally
and remotely.
http://lists.freedesktop.org/archives/systemd-devel/2014-November/025418.html
Zbigniew Jędrzejewski-Szmek [Sat, 22 Nov 2014 22:12:44 +0000 (17:12 -0500)]
missing: add IFLA_BRPORT_LEARNING and friends
IFLA_BRPORT_LEARNING was added in
v3.10-rc4-583-g9ba18891f7,
and IFLA_BRPORT_UNICAST_FLOOD in
v3.10-rc4-584-g867a59436f.
Zbigniew Jędrzejewski-Szmek [Sat, 22 Nov 2014 17:04:50 +0000 (12:04 -0500)]
missing: add __NR_getrandom for a few more archs
Mips has getrandom() too, but there's just too many variants
for me too care. Either someone who cares does it, or they get
compile-time warnings with old kernel headers.
Lukas Nykryn [Fri, 21 Nov 2014 16:02:15 +0000 (17:02 +0100)]
selinux-access: fix broken ternary operator
it seems to be a typo introduced by
ebcf1f97de4f6b1580ae55eb56b1a3939fe6b602
- _r = selinux_access_check(_b, _m, _u->source_path ?:_u->fragment_path, (permission), &_error); \
+ ({ Unit *_unit = (unit); selinux_generic_access_check(bus,message, _unit->fragment_path ?: _unit->fragment_path, permission,error); })
Lennart Poettering [Fri, 21 Nov 2014 19:46:01 +0000 (20:46 +0100)]
man: don't mix 2ch and 8ch indentation in busctl.xml
Lennart Poettering [Fri, 21 Nov 2014 19:44:48 +0000 (20:44 +0100)]
man: add emacs header to get correct indention in nxml-mode for the manpage XML files that use 2ch indenting
In the long run we really should figure out if we want to stick with 8ch
or 2ch indenting, and not continue with half-and-half. For now, just
make emacs aware of the files that use 2ch indenting.
Lennart Poettering [Fri, 21 Nov 2014 19:31:50 +0000 (20:31 +0100)]
busctl: add --timeout= option to specify method call timeout
Lennart Poettering [Fri, 21 Nov 2014 19:13:26 +0000 (20:13 +0100)]
busctl: add options to control message header flags when invoking methods
Martin Pitt [Thu, 20 Nov 2014 13:30:52 +0000 (14:30 +0100)]
nspawn: Add try-{host,guest} journal link modes
--link-journal={host,guest} fail if the host does not have persistent
journalling enabled and /var/log/journal/ does not exist. Even worse, as there
is no stdout/err any more, there is no error message to point that out.
Introduce two new modes "try-host" and "try-guest" which don't fail in this
case, and instead just silently skip the guest journal setup.
Change -j to mean "try-guest" instead of "guest", and fix the wrong --help
output for it (it said "host" before).
Change systemd-nspawn@.service.in to use "try-guest" so that this unit works
with both persistent and non-persistent journals on the host without failing.
https://bugs.debian.org/770275
Lennart Poettering [Fri, 21 Nov 2014 00:20:57 +0000 (01:20 +0100)]
units: make sure rfkill service is bount to the actual hardware
Lennart Poettering [Fri, 21 Nov 2014 00:17:52 +0000 (01:17 +0100)]
Revert "systemd-logind.service: set Type=notify"
This reverts commit
a4962513c555fe3ac4b5bebf97a71701361a45b0.
logind.service is a D-Bus service, hence we should use the dbus name as
indication that we are up. Type=dbus is implied if BusName= is
specified, as it is in this case.
This removes a warning that is printed because a BusName= is specified
for a Type=notify unit.
Martin Pitt [Thu, 20 Nov 2014 13:37:08 +0000 (14:37 +0100)]
tmpfiles.d: Create /var/lib/containers
Create /var/lib/containers so that it exists with an appropriate mode. We want
0700 by default so that users on the host aren't able to call suid root
binaries in the container. This becomes a security issue if a user can enter a
container as root, create a suid root binary, and call that from the host.
(This assumes that containers are caged by mandatory access control or are
started as user).
Lennart Poettering [Thu, 20 Nov 2014 22:12:29 +0000 (23:12 +0100)]
busctl: show property values in "introspect" output, add "set-property" command, and support both a terse and a verbose output format
Lennart Poettering [Thu, 20 Nov 2014 20:01:06 +0000 (21:01 +0100)]
busctl: when --address is specified do not assume we connect to a full bus
Lennart Poettering [Thu, 20 Nov 2014 19:59:57 +0000 (20:59 +0100)]
busctl: use canned error message on parse failure
Lennart Poettering [Thu, 20 Nov 2014 19:58:39 +0000 (20:58 +0100)]
sd-bus: create clean error when a property Set() call with incorrect signature is passed in
Lennart Poettering [Thu, 20 Nov 2014 19:58:02 +0000 (20:58 +0100)]
build: avoid trailing whitespace in feature string
Jakub Filak [Wed, 19 Nov 2014 10:01:19 +0000 (11:01 +0100)]
util: add functions getting proc cwd and root
/proc/[pid]/cwd and /proc/[pid]/root are symliks to corresponding
directories
The added functions returns values of that symlinks.
Thomas Haller [Wed, 19 Nov 2014 12:57:08 +0000 (13:57 +0100)]
sd-dhcp-lease: fix copy-paste error asserting wrong function argument
Lennart Poettering [Wed, 19 Nov 2014 23:00:22 +0000 (00:00 +0100)]
busctl: pass error output to stdout (rather than stderr) when generate tree for all objects
This is a ton more useful when some services fail, since we continue
crawling then and output everything to a pager.
Lennart Poettering [Wed, 19 Nov 2014 22:59:26 +0000 (23:59 +0100)]
busctl: improve output of service creds
Lennart Poettering [Wed, 19 Nov 2014 20:12:54 +0000 (21:12 +0100)]
busctl: add new "introspect" verb for introspecting objects
Lennart Poettering [Wed, 19 Nov 2014 19:52:47 +0000 (20:52 +0100)]
sd-bus: refuse properties that claim to be both writable and constant at the same time
Lennart Poettering [Wed, 19 Nov 2014 19:52:23 +0000 (20:52 +0100)]
core: watchdog bus properties cannot be both writable and constant
Lennart Poettering [Wed, 19 Nov 2014 15:43:41 +0000 (16:43 +0100)]
tests: fix minor memory leak
Lennart Poettering [Wed, 19 Nov 2014 15:43:03 +0000 (16:43 +0100)]
set: make set_consume() actually free the allocated string if the string already is in the set
Lennart Poettering [Wed, 19 Nov 2014 15:42:21 +0000 (16:42 +0100)]
busctl: split out introspection parser from tree logic so that we can reuse it for a future "busctl introspect" command
Dave Reisner [Wed, 19 Nov 2014 13:13:34 +0000 (08:13 -0500)]
systemd-logind.service: set Type=notify
The code already calls sd_notify("READY=1"), so we may as well take
advantage of the startup behavior in the unit. The same was done for
the journal in
a87a38c20.
Dan Williams [Tue, 18 Nov 2014 23:13:12 +0000 (17:13 -0600)]
sd-dhcp-lease: load/save client ID
The lease is usually tied to the client ID, so users of the
lease may want to know what client ID it was acquired with.
Dan Williams [Tue, 18 Nov 2014 23:01:20 +0000 (17:01 -0600)]
sd-dhcp-client: allow getting/setting the client ID
The client identifier can be in many different formats, not just
the one that systemd creates from the Ethernet MAC address. Non-
ethernet interfaces may have different client IDs formats. Users
may also have custom client IDs that the wish to use to preserve
lease options delivered by servers configured with the existing
client ID.
Lennart Poettering [Tue, 18 Nov 2014 22:35:46 +0000 (23:35 +0100)]
update TODO
Tom Gundersen [Tue, 18 Nov 2014 18:10:37 +0000 (19:10 +0100)]
sd-dhcp-lease: expose load/save functions
Based on patch by Dan Williams.
Dan Williams [Tue, 4 Nov 2014 17:20:43 +0000 (11:20 -0600)]
sd-dhcp-client: fix REBOOT state handling
client->secs wasn't getting set in the REBOOT state, causing
an assertion. REBOOT should work the same way as INIT, per
RFC 2131:
secs 2 Filled in by client, seconds elapsed since client
began address acquisition or renewal process.
REBOOT is necessary because some DHCP servers (eg on
home routers) do not hand back the same IP address unless the
'ciaddr' field is filled with that address, which DISCOVER
cannot do per the RFCs. This leads to multiple leases
on machine reboot or DHCP client restart.
Susant Sahani [Mon, 17 Nov 2014 10:58:57 +0000 (16:28 +0530)]
namespace: unchecked return value from library
fix:
CID
1237553 (#1 of 6): Unchecked return value from library
(CHECKED_RETURN
CID
1237553 (#3 of 6): Unchecked return value from library
(CHECKED_RETURN)
CID
1237553 (#4 of 6): Unchecked return value from library
(CHECKED_RETURN)
CID
1237553 (#5 of 6): Unchecked return value from library
(CHECKED_RETURN
CID
1237553 (#6 of 6): Unchecked return value from library
(CHECKED_RETURN)
Susant Sahani [Tue, 11 Nov 2014 08:06:33 +0000 (13:36 +0530)]
bus-socket: fix CID#996290 Unchecked return value
Susant Sahani [Tue, 11 Nov 2014 08:06:32 +0000 (13:36 +0530)]
resolve: fix CID#
1237549 Unchecked return value
Susant Sahani [Tue, 11 Nov 2014 08:06:31 +0000 (13:36 +0530)]
log: unchecked return value from library
fix
1237557 Unchecked return value from library
Ronny Chevalier [Sun, 9 Nov 2014 14:42:23 +0000 (15:42 +0100)]
udev: silence TOCTOU warning when creating a directory
CID#979416. There is no real race here to fix, but lets make coverity
happy and rework the code.
Note that we still fail if the directory is removed _after_ we ran
mkdir(), so the same race is still there. Coverity is complaining, though.
Rewrite the code to make it happy.
(David: rewrote the commit-message to note that this is not a race. If I'm
wrong, blame me, not Ronny!)
Thomas Hindoe Paaboel Andersen [Sat, 15 Nov 2014 22:43:09 +0000 (23:43 +0100)]
pam_systemd: remove unused null check
username was already checked with isempty() and cannot be null at
this point.
CID#
1237766
Thomas Hindoe Paaboel Andersen [Sat, 15 Nov 2014 20:32:59 +0000 (21:32 +0100)]
autogen: drop unnecessary -Wno-gnu for the clang shortcut
It is no longer necessary after
65e3a2cf7c3b399853dd309f702ca5078b7d16ea
where -Wno-gnu-variable-sized-type-not-at-end is added if we detect
that the compiler is clang.