Samuel Ortiz [Fri, 10 Dec 2010 23:44:00 +0000 (00:44 +0100)]
ofono: Set regulatory domain from network MCC
The regulatory domains can be set based on the GSM Mobile Country Code.
Samuel Ortiz [Fri, 10 Dec 2010 20:13:04 +0000 (21:13 +0100)]
technology: Export set_regdom to plugins
The oFono plugin will be the first user.
Samuel Ortiz [Thu, 9 Dec 2010 00:37:02 +0000 (01:37 +0100)]
wifi: Regulatory domain setting implementation
Samuel Ortiz [Thu, 9 Dec 2010 00:33:40 +0000 (01:33 +0100)]
gsupplicant: Country setting implementation
This implements the Country wpa_supplicant global property setting, in
order to be able to set a regulatory domain.
Samuel Ortiz [Thu, 9 Dec 2010 00:31:54 +0000 (01:31 +0100)]
gsupplicant: Support empty property set responses
Henri Bragge [Wed, 15 Dec 2010 12:23:41 +0000 (14:23 +0200)]
TODO: Claim ownership of the available EAP tasks
Samuel Ortiz [Wed, 15 Dec 2010 14:16:36 +0000 (15:16 +0100)]
doc: Fix the backtrace document
Jean-Grégoire Foulon [Wed, 15 Dec 2010 09:15:53 +0000 (10:15 +0100)]
doc: Typo fixes
Leena Gunda [Wed, 15 Dec 2010 04:00:19 +0000 (09:30 +0530)]
iwmx: Fix build failure
connman_network_set_method() has been replaced by an IPv4 or
an IPv6 version.
Fixes BMC#11315
Samuel Ortiz [Wed, 15 Dec 2010 09:17:25 +0000 (10:17 +0100)]
network: Don't set an IPv6 address if method is OFF or UNKNOWN
Samuel Ortiz [Tue, 14 Dec 2010 18:22:36 +0000 (19:22 +0100)]
ipconfig: Remove ipconfig_driver
Samuel Ortiz [Tue, 14 Dec 2010 18:15:25 +0000 (19:15 +0100)]
ipconfig: Remove connman_ipconfig_get_ipv6config
Samuel Ortiz [Tue, 14 Dec 2010 18:06:40 +0000 (19:06 +0100)]
AUTHORS: Mention Jukka's contributions
Jukka Rissanen [Thu, 9 Dec 2010 11:11:54 +0000 (13:11 +0200)]
ipconfig: Check for a valid IPv6 config before appending D-Bus data
Jukka Rissanen [Thu, 9 Dec 2010 11:11:53 +0000 (13:11 +0200)]
ipconfig: Check for a valid IPv4 config before appending D-Bus data
Jukka Rissanen [Thu, 9 Dec 2010 11:11:52 +0000 (13:11 +0200)]
ipconfig: Set the IPv6 gateway address only from an IPv6 config
Jukka Rissanen [Thu, 9 Dec 2010 11:11:51 +0000 (13:11 +0200)]
ipconfig: Return a valid IPv6 configuration
Jukka Rissanen [Thu, 9 Dec 2010 11:11:50 +0000 (13:11 +0200)]
ipconfig: Define 2 separate pointers for IPv4 and IPv6
This patch does not yet enable IPv6 only network support but it is
required for later work.
Jukka Rissanen [Thu, 9 Dec 2010 11:11:49 +0000 (13:11 +0200)]
service: __connman_service_notify() prototype change
In preparation for the IPv6 only support, this makes the code more
readable.
Jukka Rissanen [Thu, 9 Dec 2010 11:11:48 +0000 (13:11 +0200)]
ipconfig: Allow more than one IP address per device
Jukka Rissanen [Thu, 9 Dec 2010 11:11:47 +0000 (13:11 +0200)]
ipconfig: Do not add duplicate IP address to the address list
Jukka Rissanen [Thu, 9 Dec 2010 11:11:46 +0000 (13:11 +0200)]
network: Remove ipconfig from the network structure
This is not used at all.
Jukka Rissanen [Thu, 9 Dec 2010 11:11:45 +0000 (13:11 +0200)]
ipconfig: Make sure ipconfig is not null before de referencing it
Julien Massot [Fri, 10 Dec 2010 11:00:35 +0000 (11:00 +0000)]
dhcp: Add function to clear DHCP configuration
Julien Massot [Fri, 10 Dec 2010 11:00:33 +0000 (11:00 +0000)]
gdhcp: Add IPv4 Link-Local support
IPv4 Link-Local negociation starts on gdhcp client failure.
Mohamed Abbas [Thu, 9 Dec 2010 00:37:16 +0000 (16:37 -0800)]
device: Reset device->connections and device->scanning on device enable
Make sure we reset these value since the device just enabled.
Mohamed Abbas [Mon, 13 Dec 2010 15:32:35 +0000 (16:32 +0100)]
gsupplicant: Dont allow scanning while associating
In gsupplicant defer any scanning if association is in progress.
Mohamed Abbas [Thu, 9 Dec 2010 00:37:12 +0000 (16:37 -0800)]
network: Keep device->connections counter in sync
Calling __connman_device_decrease_connections from set_connnected
can decrease the connections counter twice on disconnect.
Samuel Ortiz [Fri, 10 Dec 2010 10:51:33 +0000 (11:51 +0100)]
Mention Julien's contributions
Samuel Ortiz [Fri, 10 Dec 2010 10:50:19 +0000 (11:50 +0100)]
gdhcp: Fix server listener watch events list
Add G_IO_NVAL, G_IO_ERR and G_IO_HUP to the events list.
Julien Massot [Fri, 10 Dec 2010 10:47:01 +0000 (11:47 +0100)]
gdhcp: Fix client listener watch events list
Add G_IO_NVAL, G_IO_ERR and G_IO_HUP to the events list.
Julien Massot [Fri, 10 Dec 2010 10:04:05 +0000 (10:04 +0000)]
gdhcp: Fix spelling mistake
Daniel Wagner [Fri, 10 Dec 2010 08:53:58 +0000 (09:53 +0100)]
tools: Rename stats-ringbuffer-dump to stats-tool
Due to new functionality added to the stats-ringbuffer-dump tool
it is now called stats-tool.
Besides to dump statistic ring buffer files it also allows to create
them for testing.
Furthermore, it create history files now too.
Daniel Wagner [Thu, 9 Dec 2010 12:49:33 +0000 (13:49 +0100)]
stats: Sum stats ring buffer up info history file
When the ring buffer (SERVICE_NAME.data file) is full then it
is summarized into a history (SERVICE_NAME.history) file.
The history file has the same format as the data file.
Daniel Wagner [Mon, 6 Dec 2010 17:12:21 +0000 (18:12 +0100)]
configure: Decrease default size of stats files
Johan Hedberg [Thu, 9 Dec 2010 19:52:54 +0000 (21:52 +0200)]
gdbus: fix accessing freed callback data
cb->disc_func or cb->conn_func could remove the callback so this needs
to be checked for before continuing processing.
David Woodhouse [Tue, 7 Dec 2010 23:29:49 +0000 (23:29 +0000)]
dhcp: Handle proxy PAC option correctly
We're requesting the unofficial proxy option (252) but then throwing
away the response because we don't recognise the type for it.
This fixes proxy detection via DHCP.
Martin Xu [Thu, 9 Dec 2010 15:10:54 +0000 (16:10 +0100)]
rtnl: Call __connman_technology_add_interface() after rtnl->newlink()
The device is created and the underlying technology attached from
the rtnl newlink hook, so __connman_technology_add_interface() needs
to be called after it.
Samuel Ortiz [Thu, 9 Dec 2010 00:28:15 +0000 (01:28 +0100)]
technology: Add support for regulatory domain notification
Samuel Ortiz [Wed, 8 Dec 2010 18:39:58 +0000 (19:39 +0100)]
technology: Set technology driver to NULL when unregistering
Samuel Ortiz [Wed, 8 Dec 2010 18:35:44 +0000 (19:35 +0100)]
Revert "main: Stop rfkill after the plugins"
This reverts commit
b2a4246439738b5e9a262da77d56d99d348da36a.
Lucas De Marchi [Sat, 27 Nov 2010 19:39:01 +0000 (17:39 -0200)]
gdbus: explicitly compare pointers to NULL
This patch was generated by the following semantic patch
(http://coccinelle.lip6.fr/)
// <smpl>
@fix disable is_null,isnt_null1@
expression *E;
@@
- !E
+ E == NULL
// </smpl>
Samuel Ortiz [Wed, 8 Dec 2010 15:52:19 +0000 (16:52 +0100)]
technology: Fix technology refcount
The blocked getter was bumping the technology refcount by calling
technology_get() instead of technology_find().
Samuel Ortiz [Wed, 8 Dec 2010 15:04:22 +0000 (16:04 +0100)]
main: Stop rfkill after the plugins
In order to catch rfkill device deletion and to propagate those events
to the technology layer properly.
Samuel Ortiz [Wed, 8 Dec 2010 14:57:12 +0000 (15:57 +0100)]
dnsproxy: Stop polling a nameserver when connect fails
The server data structure also needs to be cleaned up.
Martin Xu [Wed, 8 Dec 2010 11:27:13 +0000 (12:27 +0100)]
ethernet: Add USB CDC tethering support
Martin Xu [Wed, 8 Dec 2010 08:28:40 +0000 (16:28 +0800)]
Add CONNMAN_DEVICE_TYPE_GADGET and CONNMAN_SERVICE_TYPE_GADGET
Martin Xu [Wed, 8 Dec 2010 08:28:39 +0000 (16:28 +0800)]
inet: Add connman_inet_[add|remove]_to_bridge() routines
Ian Lee [Wed, 8 Dec 2010 00:13:12 +0000 (00:13 +0000)]
Fix SYSTEMD_UNITDIR typo
Martin Xu [Fri, 3 Dec 2010 11:06:47 +0000 (19:06 +0800)]
tethering: Set tethering_status earlier
With synchronous technologies (e.g. ethernet),
connman_technology_tethering_notify() is called immediatly after
enabling or disabling tethering. If tethering_status is not properly
set, then that will fail.
Samuel Ortiz [Tue, 7 Dec 2010 15:52:48 +0000 (16:52 +0100)]
dnsproxy: Support multiple search domain per service
The server data structure now takes a domain list instead of a single
domain string.
Henri Bragge [Tue, 7 Dec 2010 08:49:38 +0000 (10:49 +0200)]
resolver: Add support for multiple search domains
Domains are appended to entry list just like nameservers, and finally exported
to /etc/resolv.conf. Domains are written to resolv.conf in reverse order so
that the most recently appended domain will have the highest priority.
Search domains should only be useful for resolvfile_resolver and should be
ignored by dnsproxy_resolver.
Henri Bragge [Fri, 3 Dec 2010 12:42:30 +0000 (14:42 +0200)]
dnsproxy: Ignore adding and removing of NULL servers
Samuel Ortiz [Tue, 7 Dec 2010 11:16:57 +0000 (12:16 +0100)]
gresolv: Destroy query after removing it from the resolv queue
Fixes BMC#10958
Samuel Ortiz [Tue, 7 Dec 2010 11:15:43 +0000 (12:15 +0100)]
gresolv: Remove ipv6 query from the resolv queue
Samuel Ortiz [Mon, 6 Dec 2010 23:14:23 +0000 (00:14 +0100)]
wifi: Fix technology driver name and type
Samuel Ortiz [Fri, 3 Dec 2010 23:25:38 +0000 (00:25 +0100)]
ofono: Fix signal strength report
David Woodhouse [Fri, 3 Dec 2010 01:20:31 +0000 (01:20 +0000)]
rtnl: Receive notification of RDNSS from IPv6 router advertisements
David Woodhouse [Fri, 3 Dec 2010 01:20:13 +0000 (01:20 +0000)]
resolver: Handle temporary nameservers with automatic expiry
We'll need this for servers discovered by RDNSS in IPv6 RA
Samuel Ortiz [Thu, 2 Dec 2010 16:56:16 +0000 (17:56 +0100)]
Remove selftest file and option
David Woodhouse [Thu, 2 Dec 2010 14:16:51 +0000 (14:16 +0000)]
dnsproxy: Fix handling of IPv6 nameservers
Samuel Ortiz [Thu, 2 Dec 2010 16:00:53 +0000 (17:00 +0100)]
service: Initiate scan when not retrying a failed connection
Patrik Flykt [Thu, 2 Dec 2010 14:09:16 +0000 (16:09 +0200)]
test: simple-agent support for ReportError method call
simple-agent receives the ReportError method call and prompts whether to
retry the service.
Patrik Flykt [Thu, 2 Dec 2010 14:09:15 +0000 (16:09 +0200)]
service: Call __connman_agent_report_error when a service fails to connect
When a service fails, call __connman_agent_report_error to send an
ReportError method call to the agent. Sending of ReportError is requested
only if the service was connected due to a D-Bus request. Thus automatic
connection creation will be done silently without invoking an agent.
Patrik Flykt [Thu, 2 Dec 2010 14:09:14 +0000 (16:09 +0200)]
agent: Agent API ReportError method call
The __connman_agent_report_error function creates and sends a ReportError
method call to the registered agent. The agent can reply with D-Bus error
org.moblin.connman.Agent.Error.Retry if the same service is to be retried.
Patrik Flykt [Thu, 2 Dec 2010 14:09:13 +0000 (16:09 +0200)]
service: Create service_complete helper function
The service_complete helper function factors out reply sending to clients,
running autoconnect and updating service modification time.
Patrik Flykt [Thu, 2 Dec 2010 14:09:12 +0000 (16:09 +0200)]
doc: Fix Agent API error definitions
D-Bus errors renamed to org.moblin.connman.Agent.* since object paths are
not allowed to be used as error name components.
Samuel Ortiz [Tue, 30 Nov 2010 19:12:34 +0000 (20:12 +0100)]
wifi: Register technology driver
Samuel Ortiz [Tue, 30 Nov 2010 19:11:20 +0000 (20:11 +0100)]
technology: Add regulatory domain framework
David Woodhouse [Thu, 2 Dec 2010 11:00:03 +0000 (11:00 +0000)]
gresolv: Implement RFC3484 rule 9 (prefer longest matching prefix)
David Woodhouse [Thu, 2 Dec 2010 10:25:58 +0000 (10:25 +0000)]
gresolv: Add scope handling to RFC3484 sort
David Woodhouse [Thu, 2 Dec 2010 09:14:51 +0000 (09:14 +0000)]
gresolv: First partial implementation for RFC3484 sorting
David Woodhouse [Thu, 2 Dec 2010 02:47:53 +0000 (02:47 +0000)]
gresolv: Calculate precedence/label/etc required for RFC3484 sorting
David Woodhouse [Thu, 2 Dec 2010 00:25:30 +0000 (00:25 +0000)]
gresolv: Don't convert results to strings so early
We're going to want results in sockaddr form for sorting, so let's put
them directly into the array we'll want for the sort process, even though
we haven't actually implemented sorting yet.
We can ditch the support for arbitrary callbacks from the individual
queries, too; there's no need for that. The callback to user code comes
from the *lookup*, having combined the A and AAAA query results.
David Woodhouse [Wed, 1 Dec 2010 21:49:16 +0000 (21:49 +0000)]
gresolv: Send A and AAAA queries, merge results for lookup callback
David Woodhouse [Wed, 1 Dec 2010 16:49:05 +0000 (16:49 +0000)]
gresolv: Start handling AAAA results in query
David Woodhouse [Wed, 1 Dec 2010 16:40:52 +0000 (16:40 +0000)]
gresolv: Fix handling of IPv6 nameservers
If IPv6 nameservers were specified in /etc/resolv.conf, we would end up
sending NS queries to 0.0.0.0 because we weren't parsing the __res_state
structure correctly. We were assuming that all listed nameservers would
be Legacy IP.
Also fix connect_udp_channel() to generate the address correctly instead
of just asssuming Legacy IP.
David Woodhouse [Wed, 1 Dec 2010 16:35:58 +0000 (16:35 +0000)]
gresolv: Handle POLLERR on DNS UDP socket
If we get an error on the UDP socket, we'll currently go into an endless
loop eating CPU with poll() returning POLLERR and us ignoring it.
This at least hooks things up so that our callback gets called, and we
stop looping. But the callback's handling of POLLERR isn't correct. It
should close the socket and open a new one. As it is, it's just going to
try to use the same broken socket again for the next request. But at least
with this patch it'll stop eating CPU.
Samuel Ortiz [Mon, 29 Nov 2010 16:44:52 +0000 (17:44 +0100)]
gdhcp: Initialise socket structure before binding
Samuel Ortiz [Mon, 29 Nov 2010 16:43:28 +0000 (17:43 +0100)]
device: Support already powered up devices
Leena Gunda [Thu, 25 Nov 2010 15:11:38 +0000 (20:41 +0530)]
device: Support already powered devices
When a device is already powered up, the enable hook will return -EALREADY.
In that case, ConnMan should just set it to powered and enable its
underlying technology.
Fixes BMC#10397
Gustavo F. Padovan [Thu, 25 Nov 2010 19:09:10 +0000 (17:09 -0200)]
ofono: Parse the new Lockdown property
ConnMan should not try power up the modem if some is holding the modem
lock. If it does an error will be returned in that case so this is just an
optimization.
Connman will re-power the modem if the lock is released.
Marcel Holtmann [Fri, 26 Nov 2010 10:16:40 +0000 (11:16 +0100)]
Release 0.64
Marcel Holtmann [Fri, 26 Nov 2010 10:10:09 +0000 (11:10 +0100)]
Remove support for dhclient plugin
Marcel Holtmann [Fri, 26 Nov 2010 10:01:30 +0000 (11:01 +0100)]
Remove moblin.org domain from resolver selftest
Marcel Holtmann [Fri, 26 Nov 2010 09:57:42 +0000 (10:57 +0100)]
Change service name from org.moblin.connman to net.connman
Marcel Holtmann [Fri, 26 Nov 2010 09:49:09 +0000 (10:49 +0100)]
Revert "Return to mainloop before starting DHCP procedure"
This reverts commit
c678775fd7e36250be2c036e4746d8af0164fea9.
Tomasz Bursztyka [Thu, 25 Nov 2010 12:17:21 +0000 (14:17 +0200)]
Update TODO list regarding WPS support
Jukka Rissanen [Thu, 25 Nov 2010 12:09:19 +0000 (13:09 +0100)]
Update TODO list about IPv6 enhancements
Patrik Flykt [Thu, 25 Nov 2010 12:08:35 +0000 (13:08 +0100)]
Update TODO list regarding Agent API
Samuel Ortiz [Thu, 25 Nov 2010 12:02:26 +0000 (13:02 +0100)]
TODO update
Samuel Ortiz [Thu, 25 Nov 2010 11:40:49 +0000 (12:40 +0100)]
Mention Gustavo Padovan's contributions
Gustavo F. Padovan [Thu, 25 Nov 2010 01:01:27 +0000 (23:01 -0200)]
ofono: Fix Modem properties parsing
Rename properties to prop to keep code under 80 columns.
Gustavo F. Padovan [Wed, 24 Nov 2010 18:21:39 +0000 (16:21 -0200)]
test: Providers is no longer a manager property
Martin Xu [Thu, 25 Nov 2010 06:03:14 +0000 (14:03 +0800)]
gdhcp: Support hostname with space characters
Samuel Ortiz [Wed, 24 Nov 2010 14:45:47 +0000 (15:45 +0100)]
tethering: Remove bridge asynchronously
The tethering bridge should be removed once the technology signals that
tethering is actually disabled.
Samuel Ortiz [Wed, 24 Nov 2010 14:36:56 +0000 (15:36 +0100)]
tethering: Properly check for tethering refcount hitting 0
Samuel Ortiz [Wed, 24 Nov 2010 14:24:05 +0000 (15:24 +0100)]
tethering: Disable forward delay on the tethering bridge
Marcel Holtmann [Wed, 24 Nov 2010 13:53:35 +0000 (14:53 +0100)]
Return to mainloop before starting DHCP procedure
Just before starting the DHCP procedure return to mainloop first. This
fixes some cases where the interface is not yet up. This happens mainly
when using Bluetooth PAN connections where the network interface will be
newly created every single time.
Marcel Holtmann [Wed, 24 Nov 2010 13:48:02 +0000 (14:48 +0100)]
Increase number of DHCP discover attempts for interoperability
Currently the DHCP discover is retried five times with a timeout of
three seconds. This is not always enough if a bridge forward delay
is longer than 15 seconds is configured. So increase DHCP discover
attempts to ten times.