summaryrefslogtreecommitdiff
path: root/plugins
AgeCommit message (Collapse)AuthorFilesLines
2012-10-16wifi: Do not start autoscan if interface was not readyTomasz Bursztyka1-1/+4
2012-10-16gsupplicant: Add a new state according to 'interface_disabled'Tomasz Bursztyka1-0/+3
When soft rfkill is on, wpa_supplicant sets the interface disabled and sends a state named 'interface_disabled'. Taking this information into account fixes the following issue: - disable wifi (user setting) and hard rfkill it - then un-hard rfkill it, whereafter rfkill states (soft/hard) will go like this: * from 1/1 to 0/0 * from 0/0 to 1/0 when 0/0 occurs, connman will request to enable wifi The problem with this is that enabling wifi takes quite some time and in between ConnMan will soft block wifi to disable it (according to previous user setting). Thus it will request to disable wifi but since enabling is still going on, this request won't do anything. Meanwhile wpa_supplicant will also catch the soft rfkill event and wpa_supplicant will set the state to 'interface_disabled', but since it's not handled properly by ConnMan, the wifi_enable() callback will be called and the function will assume wifi got enabled.
2012-10-02session: Move the default config create part back to coreDaniel Wagner1-11/+1
2012-09-27session_policy: Use bearer free functionDaniel Wagner1-10/+1
2012-09-27session: Remove unused functionsDaniel Wagner1-30/+0
Basic types are not used for the configuration anymore.
2012-09-27session_policy: Implement create() and destroy()Daniel Wagner1-0/+63
2012-09-27session: Use session pointer instead of string idDaniel Wagner1-4/+6
Instead of passing in some string to identify we can use the connman_session pointer. This allows us to keep the way sessions are identfied away from the core, e.g. using the D-Bus owner id or something else.
2012-09-27session: Add plugin priorityDaniel Wagner1-0/+1
Support several session configuration plugins at runtime. Set the default priority to low for the current policy plugin.
2012-09-27session: Rename session_config to session_policyDaniel Wagner1-10/+10
2012-09-24wifi: Fix tethering with kernel 3.5Artem Bityutskiy1-3/+0
Wifi tethering works with kernel 3.4, but not with 3.5. Bisecting showed that the following kernel patch causes the breakage: "3edaf3e mac80211: manage AP netdev carrier state". Running connman with debugging enabled showed that in case of 3.4 we have the following sequence of RTM_NEWLINK events from the kernel: 1. IFF_UP not set, ifi_change=1 2. IFF_UP,IFF_LOWER_UP, ifi_change=1 which makes connman do the following: connmand[210]: plugins/wifi.c:wifi_newlink() index 4 flags 4098 change 1 connmand[210]: plugins/wifi.c:wifi_newlink() interface down connmand[210]: plugins/wifi.c:wifi_newlink() index 4 flags 69635 change 1 connmand[210]: plugins/wifi.c:wifi_newlink() interface up connmand[210]: plugins/wifi.c:wifi_newlink() carrier on connmand[210]: plugins/wifi.c:handle_tethering() index 4 bridge tether However, in 3.5 we have the following sequents of events from the kernel: 1. IFF_UP, ifi_change=1 2. IFF_UP,IFF_LOWER_UP, ifi_change=0 which makes connman do the following: connmand[493]: plugins/wifi.c:wifi_newlink() index 4 flags 4099 change 1 connmand[493]: plugins/wifi.c:wifi_newlink() interface up connmand[493]: plugins/wifi.c:wifi_newlink() index 4 flags 69635 change 0 The root-cause for it is that connman handles the "ifi_change" flag incorrectly. Connman interprets it as "if non-zero, there was some change", which is wrong. According to RFC 3549, it is "reserved for future use. Must be set to 0xFFFFFFFF". Thus, just remove that check, which makes tethering work.
2012-09-14session: Fix configuration plugin buildDaniel Wagner1-2/+2
Instead of allow the user configuring which plugin should build and used via issuing 'configure --with-configplugin=foo' add explicit enable flags 'configure --enable-session-policy'. Also with this patch, the name is changed from session-default to session-policy.
2012-09-04Use g_slist_prepend() where appropriatePatrik Flykt2-2/+2
Convert usage of g_slist_append() to g_slist_prepend() where appropriate. gdbus, dnsproxy, resolver, rtnl, session and session unit test have ordering requirements and thus not touched.
2012-08-30session: Move RoamingPolicy to configuration pluginDaniel Wagner1-2/+5
The RoamingPolicy setting which should be set by the platform integrator and not by the application developper is moved the configuration plugin.
2012-08-30session: Move EmergencyCall to configuration pluginDaniel Wagner1-0/+2
Instead fallback using the configuration plugin for selecting the emergency application. Now we don't have to update all other sessions which makes the whole code logic more straight forward. If a session is marked as an emergency call and this application calls Connect()/Disconnect() it will succedd and block all other sessions when they try to call Connect() or Disconnect().
2012-08-30session: Move Priority to configuration pluginDaniel Wagner1-2/+7
The Priority setting which should be set by the platform integrator and not by the application developer is moved the configuration plugin.
2012-08-30session: Add configuration pluginDaniel Wagner1-0/+76
2012-08-24pptp: Fix PPTP option typoPatrik Flykt1-1/+1
Fixes BMC#25660
2012-08-23pptp: Get VPN gateway resolved IP addressJukka Rissanen1-1/+1
We should not use the Host variable from provider as it can contain a non-resolved address. Fixes BMC#25658
2012-08-23l2tp: Get VPN gateway resolved IP addressJukka Rissanen1-1/+1
We should not use the Host variable from provider as it can contain a non-resolved address.
2012-08-23vpn: Accept also "true" or "1" as a boolean valueJukka Rissanen3-3/+9
Currently user is able to set the boolean VPN config value only by setting the variable to "yes". This is not very intuitive so accept also the value of "true" or "1".
2012-08-23device: Refactor device scan functionJukka Rissanen2-84/+65
In wifi plugin this means that we combine three wifi scan functions (normal, fast, hidden) into one scan function.
2012-08-22wifi: Ignore duplicate hidden APsJukka Rissanen1-23/+46
We must not add duplicate entries to list of hidden AP. There can be more that one service file for each AP if we have multiple wifi cards in the system. Fixes BMC#24779
2012-08-22wifi: Always Scan hidden APsJukka Rissanen1-7/+0
We should scan hidden AP even if the AP is not marked as autoconnectable so that the user is able to connect to it manually.
2012-08-17dundee: Fix error code handling in add_device()Daniel Wagner1-4/+2
If no error happens then we should just return and not run into the cleanup code. And when we cleanup we should not double free pointers.
2012-08-16bluetooth: Add the device into hash table when relevantTomasz Bursztyka1-2/+2
Fixes BMC#25215
2012-08-15wifi: Try only twice when connecting to a non-favorite serviceJulien Massot1-2/+13
2012-08-14dundee Create ipconfig before setting indexDaniel Wagner1-0/+8
Since connman_network_set_index() does not create magically the ipconfig for us anymore, we need to create it inside set_connected().
2012-08-14bluetooth: Defer add_networks() if adapter is not powered yetDaniel Wagner1-3/+63
In the case we add a new adapter and the adapter is powered off we would ignore the reported devices. Later when the adapter was powered on it would just not add those networks. Instead silently ignoring the devices, store the D-Bus paths in a hash table and process them as soon the adapter is powered. Fixes BMC#25322
2012-08-14bluetooth: Remove unused argument in check_networks()Daniel Wagner1-4/+3
2012-08-14bluetooth: Remove unused argument in add_network()Daniel Wagner1-3/+3
2012-08-14ofono: Set IPv6 nameserversDaniel Wagner1-2/+17
2012-08-14ofono: Create ipconfig before setting indexDaniel Wagner1-41/+26
Since connman_network_set_index() does not create magically the ipconfig for us anymore, we need to create it inside set_connected().
2012-08-10dundee: Clean up device in case of setup failureGustavo Padovan1-9/+33
If the creation and registration of Bluetooth DUN device fails we need to clean up before return, otherwise we leak memory.
2012-08-10bluetooth: Add device to hash before registrationGustavo Padovan1-2/+3
During the connman_device_register() procedure a lookup to the bluetooth_devices hash table happens, however the device is not on the hash at this point and the look out fails. If the registration fails, technology_disable() returns the Failed message on D-Bus with the error status zero. That happens because we don't have any device registered. This patch moves the insertion of the device to before the device registration.
2012-07-24wifi: Return -EBUSY when tetheringTomasz Bursztyka1-1/+1
Fixes BMC#25561
2012-07-15plugins: Avoid shadowing connection variableMarcel Holtmann3-38/+38
2012-07-13wifi: Debug print fixedJukka Rissanen1-1/+1
Convert a warn message to debug one because it is printed too often to be usefull as a warning message.
2012-07-13wifi: Scan all hidden and provisioned APJukka Rissanen1-13/+59
2012-07-05wifi: Add support for autoscan requestTomasz Bursztyka1-1/+17
2012-07-05wifi: Add capability to set regulatory domain through device's interfaceTomasz Bursztyka1-0/+31
Fixes BMC#25244
2012-07-05gsupplicant: Adapt set country callback to return a resultTomasz Bursztyka1-4/+5
This is necessary for the next coming patches, to fix country setting by interface
2012-06-30ofono: Watch for signals only on OFONO_SERVICELucas De Marchi1-15/+18
2012-06-30dundee: Watch for signals only on DUNDEE_SERVICELucas De Marchi1-5/+5
2012-06-30bluetooth: Watch for signals only on BLUEZ_SERVICELucas De Marchi1-8/+9
2012-06-30ofono: Fix potential crashGuillaume Zajac1-0/+6
If a dummy context is created, it might happen that its index is equal to -1. In this particular case the IP address has not been copied to context->address during ipv4 extraction. If we force from oFono a context activation, it will make ConnMan crashing on set_connected().
2012-06-28wifi: Add wifi pointer NULL checksPatrik Flykt1-3/+24
Always check what connman_device_get_data() returns and act accordingly.
2012-06-25wifi: Check we have valid wifi pointer in autoscanJukka Rissanen1-0/+3
The wifi pointer in device might be NULL if wpa_supplicant dies between autoscans.
2012-06-18wifi: Use common scan callback handlingPatrik Flykt1-2/+1
2012-06-18wifi: Do a hidden WiFi network scan also for user scansPatrik Flykt1-3/+3
Rename the callback function since it's not used only for autoscan.
2012-06-18wifi: Use scan_callback() instead of hidden_scan_callback()Patrik Flykt1-36/+26
The functionality of the callbacks are the same, thus start using scan_callback() only. As a result move code around.