chiark / gitweb /
elogind.git
10 years agocore,logind,networkd: don't pick up devices from udev before they finished udev initi...
Lennart Poettering [Wed, 18 Dec 2013 02:37:26 +0000 (03:37 +0100)]
core,logind,networkd: don't pick up devices from udev before they finished udev initialization

Managers shouldn't pick up the devices the manage before udev finished
initialization, hence check explicitly for that.

10 years agobusctl: output a single sorted list of names, including activatable and activated
Lennart Poettering [Wed, 18 Dec 2013 01:49:03 +0000 (02:49 +0100)]
busctl: output a single sorted list of names, including activatable and activated

10 years agobus: beef up driverd
Lennart Poettering [Wed, 18 Dec 2013 01:25:19 +0000 (02:25 +0100)]
bus: beef up driverd

Add validity checks for all parameters passed in.

Implement ReleaseName and StartServiceByName().

Fix access control in vtable.

10 years agobus: when releasing, taking, listing bus names check if they are valid
Lennart Poettering [Wed, 18 Dec 2013 01:23:07 +0000 (02:23 +0100)]
bus: when releasing, taking, listing bus names check if they are valid

Given that the kernel does not validate bus names we have to when using
them.

10 years agotests: fix build
Lennart Poettering [Wed, 18 Dec 2013 01:22:16 +0000 (02:22 +0100)]
tests: fix build

10 years agobus: sync with kdbus
Daniel Mack [Tue, 17 Dec 2013 23:20:55 +0000 (00:20 +0100)]
bus: sync with kdbus

Two new things here:

a) struct kdbus_notify_name_change now carries two struct
   kdbus_notify_id_change

b) a new KDBUS_CMD_BYEBYE ioctl that has no user yet in systemd

10 years agonetworkd: bridge - fix state machine
Tom Gundersen [Tue, 17 Dec 2013 18:56:09 +0000 (19:56 +0100)]
networkd: bridge - fix state machine

We were entering BRIDGE_STATE_CREATED rather than BRIDGE_STATE_READY.

10 years agonetworkd: link - remove useless states
Tom Gundersen [Sat, 14 Dec 2013 18:09:04 +0000 (23:39 +0530)]
networkd: link - remove useless states

Rework the state-machine a bit.

10 years agonetworkd: rename link_update_flags to link_update
Tom Gundersen [Tue, 17 Dec 2013 17:36:09 +0000 (18:36 +0100)]
networkd: rename link_update_flags to link_update

We are likely to track more than the flags in the future.

10 years agobus-util: return errors when encountered
Thomas Hindoe Paaboel Andersen [Fri, 13 Dec 2013 20:34:04 +0000 (21:34 +0100)]
bus-util: return errors when encountered

Introduced in:
9f6eb1cd58f2ddf2eb6ba0e4de056e13d938af75
313333b403439360c0396a50d77d0a1ee2bca4df

10 years ago_noreturn_ --> noreturn for C11 compat
Shawn Landden [Mon, 16 Dec 2013 16:53:53 +0000 (08:53 -0800)]
_noreturn_ --> noreturn for C11 compat

also define noreturn w/o <stdnoreturn.h>

10 years agoutil: no need for in_initrd() cache to be thread-local
Shawn Landden [Mon, 16 Dec 2013 00:56:21 +0000 (16:56 -0800)]
util: no need for in_initrd() cache to be thread-local

the process only has one working directory, and a race is
harmless

10 years ago__thread --> thread_local for C11 compat
Shawn Landden [Mon, 16 Dec 2013 00:24:14 +0000 (16:24 -0800)]
__thread --> thread_local for C11 compat

Also make thread_local available w/o including <threads.h>.
(as the latter hasn't been implemented, but this part is trivial)

10 years agoupdate TODO
Lennart Poettering [Tue, 17 Dec 2013 20:36:54 +0000 (21:36 +0100)]
update TODO

10 years agolibsystemd-bus: true/false instead of yes/no in msg dump
Lukasz Skalski [Tue, 17 Dec 2013 09:55:28 +0000 (10:55 +0100)]
libsystemd-bus: true/false instead of yes/no in msg dump

Due to this patch, message dump (for message which includes boolean
type) is more consistent with dbus-send (which display true/false
instead of yes/no for boolean). It's only simple 'cosmetics change'.

** For dbus-send **

dbus-send --system --dest=org.freedesktop.DBus --type=method_call --print-reply / org.freedesktop.DBus.NameHasOwner string:org.freedesktop.login1

method return sender=org.freedesktop.DBus -> dest=:1.97 reply_serial=2
   boolean true

** For libsystemd-bus (without this patch) **

‣ Type=method_call  Endian=l  Flags=0  Version=2 Serial=8
  Destination=org.freedesktop.DBus  Path=/org/freedesktop/DBus  Interface=org.freedesktop.DBus  Member=NameHasOwner
  MESSAGE "s" {
   STRING "org.freedesktop.login1";
  };

‣ Type=method_return  Endian=l  Flags=1  Version=2 Serial=51  ReplySerial=8
  Sender=:1.59  Destination=:1.67
  UniqueName=:1.59  WellKnownNames={org.freedesktop.DBus}
  MESSAGE "b" {
   BOOLEAN yes;
  };

For me true/false seems to be better readable than yes/no for BOOLEAN.

10 years agojournal-file.c: remove redundant assignment of variable
Thomas Hindoe Paaboel Andersen [Tue, 17 Dec 2013 19:15:45 +0000 (20:15 +0100)]
journal-file.c: remove redundant assignment of variable

we also do 'last_index = (uint64_t) -1;' at the end of the while
loop so there is no reason to also do it here.

10 years agobus: properly return an error code when release_name fails abnormally
Lennart Poettering [Tue, 17 Dec 2013 19:39:15 +0000 (20:39 +0100)]
bus: properly return an error code when release_name fails abnormally

10 years agobus-driverd: Fix return code in driver_request_name
Lukasz Skalski [Tue, 17 Dec 2013 16:31:46 +0000 (17:31 +0100)]
bus-driverd: Fix return code in driver_request_name

RequestName return codes should be consistent with Dbus Specification.

VALUE - DESCRIPTION
1-The caller is now the primary owner of the name, replacing any previous owner,
2-The name already had an owner (QUEUE flag was not specified),
3-The name already has an owner (QUEUE flag was specified),
4-Application trying to request ownership of a name is already the owner of it.

10 years agojournald: before closing /dev/kmsg let's unregister the event source
Lennart Poettering [Tue, 17 Dec 2013 19:02:21 +0000 (20:02 +0100)]
journald: before closing /dev/kmsg let's unregister the event source

10 years agojournald: close /dev/kmsg again if we cannot make use of it
Lennart Poettering [Tue, 17 Dec 2013 18:57:13 +0000 (19:57 +0100)]
journald: close /dev/kmsg again if we cannot make use of it

10 years agojournald: downgrade warning message when /dev/kmsg does not exist
Lennart Poettering [Tue, 17 Dec 2013 18:56:06 +0000 (19:56 +0100)]
journald: downgrade warning message when /dev/kmsg does not exist

10 years agodhcp: Properly handle unsuccessful cases for DHCP Ack
Patrik Flykt [Tue, 17 Dec 2013 15:24:16 +0000 (16:24 +0100)]
dhcp: Properly handle unsuccessful cases for DHCP Ack

Ignore DHCP Ack packets if they are found to be erroneous or have
an unexpected xid by setting the result to zero instead of propagating
an error. Also remember to report a DHCP Nak and stop the DHCP client.

10 years agoudev link-config: add asserts to silence scan-build
Thomas Hindoe Paaboel Andersen [Tue, 17 Dec 2013 15:09:52 +0000 (16:09 +0100)]
udev link-config: add asserts to silence scan-build

10 years agobus: removed unused variable in driverd
Thomas Hindoe Paaboel Andersen [Tue, 17 Dec 2013 15:07:00 +0000 (16:07 +0100)]
bus: removed unused variable in driverd

acked by Daniel Mack

10 years agobus: make driverd code more similar to other daemons, and make it exit on idle among...
Lennart Poettering [Tue, 17 Dec 2013 14:44:05 +0000 (15:44 +0100)]
bus: make driverd code more similar to other daemons, and make it exit on idle among other things

10 years agobusctl: on kdbus the bus driver is a service like any other, hence don't skip it
Lennart Poettering [Tue, 17 Dec 2013 02:55:59 +0000 (03:55 +0100)]
busctl: on kdbus the bus driver is a service like any other, hence don't skip it

10 years agobus: remarshal messages when necessary before sending
Lennart Poettering [Tue, 17 Dec 2013 01:43:45 +0000 (02:43 +0100)]
bus: remarshal messages when necessary before sending

10 years agoactivate: fix error checking on epoll_ctl()
Lennart Poettering [Tue, 17 Dec 2013 02:14:20 +0000 (03:14 +0100)]
activate: fix error checking on epoll_ctl()

10 years agobus: install unit files for new driver service
Lennart Poettering [Tue, 17 Dec 2013 01:43:38 +0000 (02:43 +0100)]
bus: install unit files for new driver service

10 years agobus: sync with kdbus
Daniel Mack [Tue, 17 Dec 2013 12:45:02 +0000 (13:45 +0100)]
bus: sync with kdbus

kdbus_cmd_{ep,ns,bus}_make are now consolidated, and bloom_size has
become an item of the dynamically sized list.

10 years agobus: sync up with kdbus
Daniel Mack [Tue, 17 Dec 2013 11:25:20 +0000 (12:25 +0100)]
bus: sync up with kdbus

10 years agobus-driverd: Fix unique name return in driver_get_name_owner
Lukasz Skalski [Tue, 17 Dec 2013 08:50:16 +0000 (09:50 +0100)]
bus-driverd: Fix unique name return in driver_get_name_owner

10 years agoman: networkd - clarify Address/Gateway keys in [Network] section
Tom Gundersen [Sat, 7 Dec 2013 22:23:18 +0000 (23:23 +0100)]
man: networkd - clarify Address/Gateway keys in [Network] section

10 years agonetworkd: add support for Route sections
Tom Gundersen [Sat, 7 Dec 2013 22:03:19 +0000 (23:03 +0100)]
networkd: add support for Route sections

10 years agonetworkd: correct logging message
Tom Gundersen [Wed, 4 Dec 2013 15:21:36 +0000 (16:21 +0100)]
networkd: correct logging message

10 years agortnl: fix for 32bits
Marc-Antoine Perennou [Tue, 17 Dec 2013 05:13:57 +0000 (14:13 +0900)]
rtnl: fix for 32bits

Commit 0a0dc69b655cfb10cab39133f5d521e7b35ce3d5 broke tests for 32 bits

10 years agobuild-sys: fix distcheck
Zbigniew Jędrzejewski-Szmek [Tue, 17 Dec 2013 04:58:51 +0000 (23:58 -0500)]
build-sys: fix distcheck

10 years agojournal: fix against (theoretical) undefined behavior
Shawn Landden [Mon, 16 Dec 2013 23:41:00 +0000 (15:41 -0800)]
journal: fix against (theoretical) undefined behavior

While all the libc implementations I know return NULL when memchr's size
parameter is 0, without accessing any memory, passing NULL to memchr is
still invalid:

C11 7.24.1p2: Where an argument declared as "size_t n" specifies the length
of the array for a function, n can have the value zero on a call to that
function. Unless explicitly stated otherwise in the description of a
particular function in this subclause, pointer arguments on such a call
shall still have valid values, as described in 7.1.4. On such a call, a
function that locates a character finds no occurrence, a function that
compares two character sequences returns zero, and a function that copies
characters copies zero characters.

see http://llvm.org/bugs/show_bug.cgi?id=18247

10 years agoman: describe journalctl --show-cursor
Zbigniew Jędrzejewski-Szmek [Tue, 17 Dec 2013 04:40:00 +0000 (23:40 -0500)]
man: describe journalctl --show-cursor

10 years agobus-driverd: actually add --version
Zbigniew Jędrzejewski-Szmek [Tue, 17 Dec 2013 03:40:09 +0000 (22:40 -0500)]
bus-driverd: actually add --version

10 years agosystemctl: highlight filenames in cat
Zbigniew Jędrzejewski-Szmek [Tue, 17 Dec 2013 03:34:07 +0000 (22:34 -0500)]
systemctl: highlight filenames in cat

10 years agosystemctl: fix return value from cat
Zbigniew Jędrzejewski-Szmek [Tue, 17 Dec 2013 03:33:45 +0000 (22:33 -0500)]
systemctl: fix return value from cat

10 years agoman: reword awkward phrase
Zbigniew Jędrzejewski-Szmek [Tue, 17 Dec 2013 03:10:21 +0000 (22:10 -0500)]
man: reword awkward phrase

10 years agobus: use gperf to map error string to errno
Zbigniew Jędrzejewski-Szmek [Mon, 16 Dec 2013 01:35:22 +0000 (20:35 -0500)]
bus: use gperf to map error string to errno

10 years agobus: update kdbus.h
Kay Sievers [Tue, 17 Dec 2013 01:50:19 +0000 (02:50 +0100)]
bus: update kdbus.h

10 years agobus: when connecting to a container's system bus, double fork after joining the conta...
Lennart Poettering [Tue, 17 Dec 2013 00:57:27 +0000 (01:57 +0100)]
bus: when connecting to a container's system bus, double fork after joining the container's namespace

setns(CLONE_NEWPID) only applies to child processes, hence let's fork
once after joining the container's namespace.

10 years agobuild-sys: build bus-driverd if kdbus support is enabled
Lennart Poettering [Tue, 17 Dec 2013 00:36:59 +0000 (01:36 +0100)]
build-sys: build bus-driverd if kdbus support is enabled

10 years agobus: when entering an existing namespace to connect to a container's system bus also...
Lennart Poettering [Tue, 17 Dec 2013 00:03:09 +0000 (01:03 +0100)]
bus: when entering an existing namespace to connect to a container's system bus also switch over PID namespace

This is necessary to ensure that kdbus can collect creds of the
destination namespace when connecting.

10 years agocore: always create /dev/kdbus/ns (and make it private 0700) after setting up the...
Lennart Poettering [Tue, 17 Dec 2013 00:02:13 +0000 (01:02 +0100)]
core: always create /dev/kdbus/ns (and make it private 0700) after setting up the kdbus system bus

10 years agojournal-file: protect against alloca(0)
Thomas Hindoe Paaboel Andersen [Mon, 16 Dec 2013 22:35:30 +0000 (23:35 +0100)]
journal-file: protect against alloca(0)

10 years agobus: fix typo in systemd-bus-proxyd
Kay Sievers [Mon, 16 Dec 2013 22:25:32 +0000 (23:25 +0100)]
bus: fix typo in systemd-bus-proxyd

10 years agoAdd bus-driverd
Daniel Mack [Fri, 29 Nov 2013 23:45:53 +0000 (00:45 +0100)]
Add bus-driverd

systemd-bus-driverd is a small daemon that connects to kdbus and
implements the org.freedesktop.DBus interface. IOW, it provides the bus
functions  traditionally taken care for by dbus-daemon.

Calls are proxied to kdbus, either via libsystemd-bus (were applicable)
or with the open-coded use of ioctl().

Note that the implementation is not yet finished as the functions to
add and remove matches and to start services by name are still missing.

10 years agobus: when a busname unit refuses to activate a service it should flush the queue
Lennart Poettering [Mon, 16 Dec 2013 20:26:21 +0000 (21:26 +0100)]
bus: when a busname unit refuses to activate a service it should flush the queue

10 years agobus: also mask dbus.service in generator if kdbus is found
Lennart Poettering [Mon, 16 Dec 2013 20:06:34 +0000 (21:06 +0100)]
bus: also mask dbus.service in generator if kdbus is found

10 years agounits: properly make bus proxy socket of type Accept=yes
Lennart Poettering [Mon, 16 Dec 2013 20:02:55 +0000 (21:02 +0100)]
units: properly make bus proxy socket of type Accept=yes

10 years agosystemctl: properly initialize and free sd_bus_error in "systemctl cat"
Lennart Poettering [Mon, 16 Dec 2013 19:53:55 +0000 (20:53 +0100)]
systemctl: properly initialize and free sd_bus_error in "systemctl cat"

We need to properly initialize all error structs before use and free
them after use.

Also, there's no point in flushing stdout if we output a \n anyway...

10 years agobus: fix symlink to bus proxy socket
Lennart Poettering [Mon, 16 Dec 2013 19:39:44 +0000 (20:39 +0100)]
bus: fix symlink to bus proxy socket

10 years agobus: use GREEDY_REALLOC() when allocating message queues
Lennart Poettering [Mon, 16 Dec 2013 19:32:37 +0000 (20:32 +0100)]
bus: use GREEDY_REALLOC() when allocating message queues

10 years agobus: let's use GREEDY_REALLOC() when allocating space for containers
Lennart Poettering [Mon, 16 Dec 2013 19:00:25 +0000 (20:00 +0100)]
bus: let's use GREEDY_REALLOC() when allocating space for containers

10 years agoexecute: also set SO_SNDBUF when spawning a service with stdout/stderr connected...
Lennart Poettering [Mon, 16 Dec 2013 19:00:09 +0000 (20:00 +0100)]
execute: also set SO_SNDBUF when spawning a service with stdout/stderr connected to journald

10 years agoloopback-setup: remove stray hunk
Tom Gundersen [Mon, 16 Dec 2013 18:31:50 +0000 (19:31 +0100)]
loopback-setup: remove stray hunk

10 years agonetwork: more asserts to shut up scan-build
Thomas Hindoe Paaboel Andersen [Mon, 16 Dec 2013 17:55:59 +0000 (18:55 +0100)]
network: more asserts to shut up scan-build

10 years agokdbus.h: add alignment requirements
Kay Sievers [Mon, 16 Dec 2013 16:47:28 +0000 (17:47 +0100)]
kdbus.h: add alignment requirements

10 years agonetwork: use SETLINK to bring up interfaces
Tom Gundersen [Mon, 16 Dec 2013 13:37:51 +0000 (14:37 +0100)]
network: use SETLINK to bring up interfaces

10 years agortnl: replace message_append by typesafe versions
Tom Gundersen [Sun, 15 Dec 2013 13:00:20 +0000 (14:00 +0100)]
rtnl: replace message_append by typesafe versions

10 years agortnl: support interleaved reading and writing, and rewind
Tom Gundersen [Fri, 6 Dec 2013 16:19:55 +0000 (17:19 +0100)]
rtnl: support interleaved reading and writing, and rewind

10 years agortnl: simplify route_new()
Tom Gundersen [Sat, 7 Dec 2013 20:18:44 +0000 (21:18 +0100)]
rtnl: simplify route_new()

Drop most of the arguments and instead introduce set_dst_prefixlen().

10 years agortnl: simplify link_new()
Tom Gundersen [Fri, 6 Dec 2013 17:16:16 +0000 (18:16 +0100)]
rtnl: simplify link_new()

Drop most of the arguments and instead introduce link_set_{flags,type}.

10 years agortnl: match - only match on one type at a time
Tom Gundersen [Fri, 6 Dec 2013 14:20:36 +0000 (15:20 +0100)]
rtnl: match - only match on one type at a time

10 years agortnl: clean up/add asserts
Tom Gundersen [Fri, 6 Dec 2013 14:13:34 +0000 (15:13 +0100)]
rtnl: clean up/add asserts

10 years agortnl: add support for receiving route messages
Tom Gundersen [Fri, 6 Dec 2013 15:26:17 +0000 (16:26 +0100)]
rtnl: add support for receiving route messages

10 years agobus: increase the bus socket buffer to 8 MB similar, to the log socket buffers
Lennart Poettering [Mon, 16 Dec 2013 16:05:51 +0000 (17:05 +0100)]
bus: increase the bus socket buffer to 8 MB similar, to the log socket buffers

10 years agoupdate TODO
Lennart Poettering [Mon, 16 Dec 2013 16:05:42 +0000 (17:05 +0100)]
update TODO

10 years agoutil: try harder to increase the send/recv buffers of sockets
Lennart Poettering [Mon, 16 Dec 2013 16:04:36 +0000 (17:04 +0100)]
util: try harder to increase the send/recv buffers of sockets

If we have the priviliges we will try SO_SNDBUFFORCE/SO_RCVBUFFORCE and
only fall back to SO_SNDBUF/SO_RCVBUF if that fails.

10 years agocatalog: fix language detection
Zbigniew Jędrzejewski-Szmek [Mon, 16 Dec 2013 02:07:47 +0000 (21:07 -0500)]
catalog: fix language detection

Detection would fail if language was not specified in the filename
but a dot appeared somewhere higher in the path.

10 years agocore: refuse doing %h, %s, %U specifier resolving in PID 1
Lennart Poettering [Mon, 16 Dec 2013 03:59:31 +0000 (04:59 +0100)]
core: refuse doing %h, %s, %U specifier resolving in PID 1

These specifiers require NSS lookups to work, and we really shouldn't do
them from PID 1 hence. With this change they are now only supported for
user systemd instance, or when the configured user for a unit is root.

10 years agobuild-sys: warn if builds are not byte-by-byte reproducible due to usage of __DATE__...
Lennart Poettering [Mon, 16 Dec 2013 03:43:42 +0000 (04:43 +0100)]
build-sys: warn if builds are not byte-by-byte reproducible due to usage of __DATE__ and suchlike

10 years agoFix segv introduced by 2fd069b18e525860514a70d3ea08410ca122d3e2
Colin Walters [Mon, 16 Dec 2013 01:23:23 +0000 (20:23 -0500)]
Fix segv introduced by 2fd069b18e525860514a70d3ea08410ca122d3e2

n->path is pointing to the value now, we set s = NULL above.

10 years agobuild-sys: add cppcheck target
Lennart Poettering [Sun, 15 Dec 2013 23:04:40 +0000 (00:04 +0100)]
build-sys: add cppcheck target

This uses --enable=all mode. Should be taken with a grain of salt
though. While many recommendations make sense we should probably keep
"int r" always on function scope, and many of the portability warnings
really don't matter to us because we only care for Linux/glibc.

10 years agoFix a few signed/unsigned format string issues
Zbigniew Jędrzejewski-Szmek [Sun, 15 Dec 2013 21:26:27 +0000 (16:26 -0500)]
Fix a few signed/unsigned format string issues

Since numbers involved are all small, behaviour was correct already.

https://bugzilla.redhat.com/show_bug.cgi?id=1043304

10 years agoFix a few resource leaks in error paths
Zbigniew Jędrzejewski-Szmek [Sun, 15 Dec 2013 21:25:04 +0000 (16:25 -0500)]
Fix a few resource leaks in error paths

https://bugzilla.redhat.com/show_bug.cgi?id=1043304

10 years agobus: _printf_ attributes should be on prototypes not function definitions for non...
Lennart Poettering [Sun, 15 Dec 2013 22:43:47 +0000 (23:43 +0100)]
bus: _printf_ attributes should be on prototypes not function definitions for non-static functions

10 years agobus: update kdbus monitoring interface
Kay Sievers [Sun, 15 Dec 2013 15:56:28 +0000 (16:56 +0100)]
bus: update kdbus monitoring interface

10 years agoudev-builtin-path: fix printf specifiers
Zbigniew Jędrzejewski-Szmek [Sun, 15 Dec 2013 04:09:14 +0000 (23:09 -0500)]
udev-builtin-path: fix printf specifiers

10 years agobus: return negative errno on error
Zbigniew Jędrzejewski-Szmek [Sun, 15 Dec 2013 03:18:49 +0000 (22:18 -0500)]
bus: return negative errno on error

https://bugs.freedesktop.org/show_bug.cgi?id=72688

10 years agotest-journal-syslog: use streq_ptr since we pass in NULL
Thomas Hindoe Paaboel Andersen [Sun, 15 Dec 2013 01:29:38 +0000 (02:29 +0100)]
test-journal-syslog: use streq_ptr since we pass in NULL

10 years agosystemctl: add the --plain option to the help message
Djalal Harouni [Sat, 14 Dec 2013 23:05:38 +0000 (00:05 +0100)]
systemctl: add the --plain option to the help message

10 years agoman: remove advice to avoid setting the same var more than once
Zbigniew Jędrzejewski-Szmek [Sat, 14 Dec 2013 22:30:25 +0000 (17:30 -0500)]
man: remove advice to avoid setting the same var more than once

So far the compatibility with .desktop settings hasn't been imporant
at all, and we do not want people to write convoluted unit
files.

10 years agoman: beef up ExecStart description
Zbigniew Jędrzejewski-Szmek [Sat, 14 Dec 2013 22:21:55 +0000 (17:21 -0500)]
man: beef up ExecStart description

We have lots of questions from people who assume that shell syntax works
here, so let's be very explicit what is allowed and what is not. A few
examples should also help.

http://bugs.debian.org/732156

10 years agobuild: use -ftrapv for development
Shawn Landden [Sat, 14 Dec 2013 17:27:44 +0000 (09:27 -0800)]
build: use -ftrapv for development

We want to find these bugs if they exist.

10 years agoAdd more _printf_'s for format-nonliterals
Thomas Hindoe Paaboel Andersen [Sat, 14 Dec 2013 12:09:07 +0000 (13:09 +0100)]
Add more _printf_'s for format-nonliterals

Clang is a bit more strict wrt format-nonliterals:
http://clang.llvm.org/docs/LanguageExtensions.html#format-string-checking

Adding these extra printf attributes also makes gcc able to find more
problems. E.g. this patch uncovers a format issue in udev-builtin-path_id.c

Some parts looked intetional about breaking the format-nonliteral check.
I added some supression for warnings there.

10 years agoevent: some snake-oil to speed up impossible error checks
Lennart Poettering [Sat, 14 Dec 2013 04:08:47 +0000 (05:08 +0100)]
event: some snake-oil to speed up impossible error checks

10 years agoevent: instead of reset the revents field when we get new revents data from epoll...
Lennart Poettering [Sat, 14 Dec 2013 04:08:15 +0000 (05:08 +0100)]
event: instead of reset the revents field when we get new revents data from epoll, OR it in

10 years agoevent: dynamically adjust size of events array instead of pre-allocating it possibly...
Lennart Poettering [Sat, 14 Dec 2013 04:07:13 +0000 (05:07 +0100)]
event: dynamically adjust size of events array instead of pre-allocating it possibly too large or too small

10 years agoutil: port last code over to new namespace utility calls
Lennart Poettering [Sat, 14 Dec 2013 04:06:40 +0000 (05:06 +0100)]
util: port last code over to new namespace utility calls

10 years agoutil: when joining a namespace make sure to reset all uids to 0 after
Lennart Poettering [Sat, 14 Dec 2013 04:04:49 +0000 (05:04 +0100)]
util: when joining a namespace make sure to reset all uids to 0 after
the transition

10 years agobus: when connecting to a container's kdbus instance, enter namespace first
Lennart Poettering [Fri, 13 Dec 2013 21:02:47 +0000 (22:02 +0100)]
bus: when connecting to a container's kdbus instance, enter namespace first

Previously we'd open the connection in the originating namespace, which
meant most peers of the bus would not be able to make sense of the
PID/UID/... identity of us since we didn't exist in the namespace they
run in. However they require this identity for privilege decisions,
hence disallowing access to anything from the host.

Instead, when connecting to a container, create a temporary subprocess,
make it join the container's namespace and then connect from there to
the kdbus instance. This is similar to how we do it for socket
conections already.

THis also unifies the namespacing code used by machinectl and the bus
APIs.

10 years agogetty-generator: fix stripping /dev/
Thomas Hindoe Paaboel Andersen [Fri, 13 Dec 2013 22:21:35 +0000 (23:21 +0100)]
getty-generator: fix stripping /dev/

10 years agobus: install systemd-bus-proxyd unit files for compatibility with dbus1
Lennart Poettering [Fri, 13 Dec 2013 19:29:35 +0000 (20:29 +0100)]
bus: install systemd-bus-proxyd unit files for compatibility with dbus1

10 years agogetty-generator: fix segfault when $container_ttys is not set
Lennart Poettering [Fri, 13 Dec 2013 19:36:02 +0000 (20:36 +0100)]
getty-generator: fix segfault when $container_ttys is not set