Commit graph

27 commits

Author SHA1 Message Date
Dan Williams
cb7ca51e0a Spacing fix
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@2031 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-09-20 15:02:13 +00:00
Dan Williams
7d8f84edf6 2006-07-10 Dan Williams <dcbw@redhat.com>
Patch from Valentine Sinitsyn <e_val@inbox.ru>
	* src/nm-device.c
		- (real_act_stage3_ip_config_start): don't infinite loop when
			dhcdbd isn't running (Gnome #346845)


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1866 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-07-10 17:22:32 +00:00
Robert Love
33aec50e50 2006-05-25 Robert Love <rml@novell.com>
* src/nm-device.h: Introduce nm_ioctl_info(), which defines to
	  nm_info() if IOCTL_DEBUG is set and a no-op if not.  We can use this
	  instead of dumping ifdef's throughout the code.
	* src/NetworkManagerSystem.c, src/nm-device-802-11-wireless.c,
	  src/nm-device-802-3-ethernet.c, src/nm-device.c: Remove ifdef's and
	  use nm_ioctl_info() in lieu.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1768 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-05-25 22:37:41 +00:00
Jeremy Katz
927d2b3394 2006-04-27 Jeremy Katz <katzj@redhat.com>
* src/nm-device.c (discover_device_type): Actually use the hal
	device type instead of ioctl poking


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1710 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-04-27 15:49:40 +00:00
Robert Love
2a63ff03ec 2006-03-21 Robert Love <rml@novell.com>
* src/NetworkManagerSystem.c, src/NetworkManagerSystem.h: Add
	  nm_system_get_mtu(), which returns a user-provided or system-mandated
	  MTU value for a given device, if any, or zero if no such value
	  exists.  Add nm_system_set_mtu() to set the MTU for a given device
	  if we have a provided value.
	* src/nm-device.c: Set the MTU of devices.
	* src/backends/NetworkManagerSuSE.c: Read MTU, if any, from sysconfig.
	* src/backends/NetworkManagerDebian.c,
	  src/backends/NetworkManagerGentoo.c,
	  src/backends/NetworkManagerRedHat.c,
	  src/backends/NetworkManagerSlackware.c: Implement stub functions.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1626 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-03-21 17:57:01 +00:00
Peter Jones
4e97221b9b - fix a race the device objects refcount during creation of the worker thread.
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1603 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-03-14 20:51:58 +00:00
Peter Jones
efb46cb8c4 - kill an extra call to nm_device_stop(). The worker thread does this while
it's exiting, and this extra call causes it to bring the device down on
  the resume path.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1602 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-03-14 20:50:59 +00:00
Robert Love
32daa1d99a 2006-03-06 Robert Love <rml@novell.com>
* src/NetworkManagerSystem.h, src/nm-device.c, NetworkManagerDebian.c,
	  NetworkManagerRedHat.c, NetworkManagerGentoo.c,
	  NetworkManagerSlackware.c: Pass nm_system_device_get_system_config()
	  a second argument, NMData.
	* src/nm-ap-security.h, src/nm-ap-security.c: Export nm_ap_security_new.
	* src/backends/NetworkManagerSuSE.c: Add wireless networks from ifcfg-*
	  config files as trusted.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1564 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-03-08 18:33:44 +00:00
Dan Williams
a5919b696c 2006-03-05 Dan Williams <dcbw@redhat.com>
Process netlink messages in device subclasses rather than in
	NetworkManager.c.  Also add support for recognizing Wireless Events.

	* configure.in
		- Find GLIB_GENMARSHAL

	* src/Makefile.am
		- Since we're marshalling custom types for wireless event signals,
			we get to create our own marshallers using GLIB_GENMARSHAL

	* src/NetworkManager.c
		- (nm_monitor_wired_link_state): renamed to nm_monitor_setup
		- (nm_monitor_setup): renamed from nm_monitor_wired_link_state, and
			cut down somewhat.  We no longer process signals here.
		- (nm_data_new): create the netlink monitor here, and remove a
			useless call to nm_policy_schedule_device_change_check()
		- (nm_data_free): get rid of the netlink monitor here
		- (nm_device_link_activated, nm_device_link_deactivated): removed
		- (main): don't create the netlink monitor here, let nm_data_new
			do that.  Call nm_policy_schedule_device_change_check() right
			before we jump to the mainloop to figure out which device
			to use first

	* src/NetworkManagerSystem.[ch]
		- (nm_system_get_rtnl_index_from_iface, nm_system_get_iface_from_rtnl_index):
			convert back and forth from interface names to interface
			indexes

	* src/nm-device-802-11-wireless.c
		- (real_init): connect to wireless-event signals from the netlink
			monitor object
		- (nm_device_802_11_wireless_event): new function, schedule handler
			for wireless event signals from the netlink monitor object.  We
			want the handler to run in the device's context
		- (wireless_event_helper): handle wireless-event signals from netlink
		- (nm_device_802_11_wireless_dispose): disconnect wireless-event
			signal handler

	* src/nm-device-802-11-wireless.h
		- remove unused prototype for nm_device_802_11_wireless_new

	* src/nm-device-802-3-ethernet.c
		- (real_init): new function; set up signal handlers for link events
		- (nm_device_802_3_ethernet_link_activated): new function, schedule
			handler for netlink link activated events on device's main loop
		- (link_activated_helper): when we get a link activated event, set
			the device's link to be active
		- (nm_device_802_3_ethernet_link_deactivated): new function; schedule
			handler for netlink link deactivated events on device's main loop
		- (link_deactivated_helper): when we get a link deactivated event, set
			the device's link to be inactive
		- (nm_device_802_3_ethernet_dispose): disconnect signal handler on
			dispose

	* src/nm-device-802-3-ethernet.h
		- remove unused prototype for nm_device_802_3_ethernet_new

	* src/nm-device.[ch]
		- (nm_get_device_by_iface_locked): variant of nm_get_device_by_iface
			but locks the device list
		- (nm_device_set_active_link): a little bit of cleanup and de-indenting

	* src/nm-netlink-monitor.[ch]
		- (nm_netlink_monitor_class_install_signals): New signal
			"wireless-event"
		- (nm_netlink_monitor_new): keep reference to NMData so we can get
			at the device list
		- (nm_netlink_monitor_event_handler): expand for wireless events too

	* src/nm-marshal-main.c
		- Include generated nm-marshal.c and nm-marshal.h

	* src/nm-marshal.list
		- List of custom marshal functions


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1555 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-03-06 01:10:58 +00:00
Dan Williams
ef9674ce75 2006-03-04 Dan Williams <dcbw@redhat.com>
Clean up activation cancellation.  Should be a lot faster now.  Observed
	an issue with wireless devices between stage 2 and 3 of activation, where
	activation would be cancelled, but the device thread wouldn't notice until
	the supplicant association timed out.  Reorganize activation such that
	a cancellation handler gets immediately scheduled in the device's thread,
	and devices have a chance to perform any custom cleanup too.

	* src/nm-device.[ch]
		- (activation_cancel_handler): new device-type-specific function
			for cleaning up device-type-specific stuff on cancellation
		- (cancel_activation): removed
		- (nm_device_activation_cancel): subsume functionality of
			real_cancel_activation, but instead of doing anything, punt
			operation to a handler that's run in device-thread context
		- (nm_device_schedule_activation_handle_cancel): fix spelling of
			a warning message
		- (activation_handle_cancel_helper): cancellation handler run in
			device-thread context, calls device-type-specific cancelation,
			then tears down the activation request
		- (real_activation_cancel_handler): generic cancellation handler,
			deals with cancelling any in-process DHCP request
		- (nm_device_activate_stage1_device_prepare,
		   nm_device_activate_stage2_device_config,
		   nm_device_activate_stage3_ip_config_start,
		   nm_device_activate_stage4_ip_config_get,
		   nm_device_activate_stage4_ip_config_timeout,
		   nm_device_activate_stage5_ip_commit): don't call
			nm_device_schedule_activation_handle_cancel() any more, since
			cancellation will have been already scheduled for us by
			nm_device_activation_cancel().  Just exit the function and
			assume that the cancel handler will be called next.

	* src/nm-device-802-3-ethernet.c
		- (real_act_stage2_config): remove; didn't do anything anyway

	* src/nm-device-802-11-wireless.c
		- (supplicant_status_cb): ensure we don't do anything if the activation
			got cancelled
		- (real_activation_cancel_handler): implement; cancel user key request
			on activation cancellation


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1549 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-03-04 06:44:05 +00:00
Robert Love
8cd691824c s/Stage x (foo) of y/Stage x of y (foo)/g
git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1475 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-02-15 21:19:09 +00:00
Robert Love
ff9ec8e962 2006-02-15 Robert Love <rml@novell.com>
* src/nm-device-802-11-wireless.c, src/nm-device.c: When printing debug
	  information about what connection stage we are at, provide the total
	  number of stages, e.g. "Stage 2 of 5", so users know how far along
	  they are if they experience problems.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1474 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-02-15 18:33:28 +00:00
Robert Love
b60c770dab 2006-01-26 Robert Love <rml@novell.com>
* configure.in: Change '-Wno-unused' to '-Wno-unused-parameter'
	* gnome/applet/applet-compat.c, gnome/applet/applet-dbus-devices.c,
	  gnome/applet/applet-dbus-info.c, gnome/applet/applet-dbus.c,
	  gnome/applet/applet.c, gnome/applet/applet.h, src/nm-dbus-device.c,
	  gnome/applet/nm-gconf-wso-wep.c, gnome/applet/nm-gconf-wso-wpa-psk.c,
	  gnome/applet/nm-gconf-wso.c, gnome/applet/nm-gconf-wso.h,
	  gnome/applet/other-network-dialog.c, src/nm-device.c, test/nm-tool.c,
	  gnome/applet/passphrase-dialog.c, src/nm-device-802-11-wireless.c,
	  gnome/applet/wireless-security-manager.c, src/nm-ip4-config.c,
	  gnome/applet/wireless-security-option.c, src/nm-ap-security.c,
	  gnome/applet/wso-wep-ascii.c, gnome/applet/wso-wep-hex.c,
	  gnome/applet/wso-wep-passphrase.c, gnome/applet/wso-wpa-psk.c,
	  libnm-util/dbus-helpers.c, src/NetworkManagerAP.c, src/nm-dbus-nmi.c,
	  src/NetworkManagerSystem.c, src/nm-ap-security-wep.c,
	  src/nm-device-802-11-wireless.h, test/libnm-util/test-ciphers.c,
	  src/named-manager/nm-named-manager.c, test/test-common/test-common.c:
	  Kill unused variables, labels, and static functions.  Don't pass
	  string literals as the format string for printf-like functions.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1391 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-26 21:08:44 +00:00
Robert Love
d1f6f5dde4 2006-01-23 Robert Love <rml@novell.com>
* dhcp-manager/nm-dhcp-manager.c, nm-device.c, nm-ip4-config.c,
	  nm-ip4-config.h, NetworkManagerSystem.h: Save the hostname reported
	  by DHCP and pass it to the backends, allowing distribution-specific
	  behavior with respect to the DHCP-supplied hostname (if nothing else,
	  some distributions might not want to set the hostname).
	* backends/NetworkManagerSuSE.c: Set the hostname if the variable
	  DHCLIENT_SET_HOSTNAME is set to "yes" in /etc/sysconfig/network/dhcp.
	  Also update our NIS behavior.
	* backends/NetworkManagerDebian.c, backends/NetworkManagerGentoo.c,
	  backends/NetworkManagerRedHat.c, backends/NetworkManagerSlackware.c:
	  Add stub functions.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1382 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-23 21:02:39 +00:00
Dan Williams
2864ede525 2006-01-22 Dan Williams <dcbw@redhat.com>
* src/NetworkManagerAP.[ch]
		- (nm_ap_get_matched, nm_ap_set_matched): remove

	* src/NetworkManagerAPList.[ch]
		- (nm_ap_list_diff): removed
		- (nm_ap_list_merge_scanned_ap): move AP dbus signal logic here,
			deal with access points changing essids on us

	* src/nm-device-802-11-wireless.c
		- (add_new_ap_to_device_list): move AP dbus signal logic to
			src/NetworkManagerAPList.c
		- (real_can_interrupt_activation): new function; allow interruption
			of device activation if we are waiting for a network key

	* src/NetworkManagerPolicy.c
		- (nm_policy_device_change_check): allow interruption of currently
			activating devices if the device allows it.  Previous behavior
			would refuse to activate a just-plugged wired device if a
			wireless device was waiting for a key.

	* src/nm-device.[ch]
		- (nm_device_can_interrupt_activation): new function; ask devices
			whether their activation can be interrupted


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1375 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-22 22:40:14 +00:00
Robert Love
cbd27080ed 2006-01-18 Robert Love <rml@novell.com>
* src/NetworkManager.c, src/NetworkManagerSystem.h, src/nm-device.c:
	  Allow backends to flag a device (in whatever distro-dependent way
	  they so desire) as disabled.  NM will ignore any such device.
	* src/backends/NetworkManagerDebian.c,
	  src/backends/NetworkManagerRedHat.c,
	  src/backends/NetworkManagerSlackware.c: Add stub function
	  nm_system_device_get_disabled() that always returns FALSE (enabled).
	* src/backends/NetworkManagerSuSE.c: Add system_disabled field to the
	  SUSE-specific configuration structure.  Fill it in from the
	  NM_CONTROLLED variable in the system networking scripts.  If this var
	  exists and is "no", we ignore the device.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1361 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-19 18:00:48 +00:00
Dan Williams
135da08fc5 2006-01-17 Dan Williams <dcbw@redhat.com>
* src/nm-device.c
		- (nm_device_class_init): connect a default act_stage4_ip_config_timeout
			function.  Fixes crash when wired DHCP fails.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1355 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-18 02:27:38 +00:00
Robert Love
8eb99d62fc 2006-01-16 Robert Love <rml@novell.com>
* gnome/applet/eggtrayicon.c, src/nm-device.c, src/nm-ap-security.c,
	  gnome/applet/nm-gconf-wso-wep.c, gnome/applet/nm-gconf-wso-wpa-psk.c,
	  gnome/applet/nm-gconf-wso.c, src/nm-device-802-3-ethernet.c,
	  gnome/vpn-properties/nm-vpn-properties.c, src/nm-ap-security-wep.c,
	  src/nm-ap-security-wpa-psk.c, src/nm-device-802-11-wireless.c,
	  src/nm-netlink-monitor.c: Don't miss any initializers on structure
	  declarations, ever.
	* gnome/applet/applet.c: Remove useless check (NMState is unsigned and
	  NM_STATE_DISCONNECTED is zero).


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1351 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-17 18:22:56 +00:00
Dan Williams
8b1b8ee7a5 2006-01-13 Dan Williams <dcbw@redhat.com>
* src/nm-device.c
	  src/nm-device.h
		- Allow subclasses to implement deactivate_quickly()
		- (nm_device_deactivate_quickly): call subclass
			deactivate_quickly() methods
		- (nm_device_set_active_link): small cleanups, and don't
			deactivate the device right away because we risk a deadlock
			when called from device thread, waiting for the device
			thread to cancel activation

	* src/nm-device-802-11-wireless.c
		- (real_deactivate_quickly): new function
		- (nm_device_802_11_wireless_class_init): hook in real_deactivate_quickly
		- (real_deactivate): move supplicant cleanup to real_deactivate_quickly
			so that we kill the supplicant when we sleep too
		- (supplicant_interface_init): work around naive naming attempts of
			wpa_ctrl when naming sockets


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1326 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-13 16:53:55 +00:00
Robert Love
5b8589dfb0 2006-01-10 Robert Love <rml@novell.com>
* src/nm-device-802-11-wireless.c, src/nm-device-802-11-wireless.h,
	  src/nm-device-802-3-ethernet.c, src/nm-device-802-3-ethernet.h,
	  src/nm-device.c: Fix a FIXME!  Reimplement the function
	  nm_device_update_hw_address() in device subclass variants,
	  nm_device_802_3_ethernet_set_address() and
	  nm_device_802_11_wireless_set_address(), hook them up, and use them.
	  This fixes the existing bug where MAC addresses are all zeros.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1309 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-10 21:15:54 +00:00
Dan Williams
30cc1984f7 2006-01-07 Dan Williams <dcbw@redhat.com>
* src/nm-device-private.h
	  src/nm-device.c
		- (nm_device_activate_schedule_stage3_ip_config_start): make this
			function available to subclasses


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1283 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-07 16:22:17 +00:00
Dan Williams
7f07df2466 2006-01-03 Dan Williams <dcbw@redhat.com>
* src/NetworkManagerPolicy.c
	  src/nm-device.[ch]
	  src/nm-device-802-11-wireless.c
		- Move wireless-specific activation failure and success code
			into wireless device class


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1259 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-03 22:11:35 +00:00
Dan Williams
1c5f300910 2006-01-03 Dan Williams <dcbw@redhat.com>
* src/dhcp-manager/nm-dhcp-manager.c
		- Recognize activation cancellation when waiting for DHCP
			configuration from dhcdbd
		- Ignore non-dhcdbd messages

	* src/nm-device.c
		- (real_act_stage3_ip_config_start): return to correct behavior
			of letting the dhcp-manager notify us of failure or
			success rather than incorrectly doing that ourselves
		- (nm_device_activate_stage4_ip_config_get): deal with
			activation cancellation a bit earlier


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1253 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-03 18:18:05 +00:00
Dan Williams
d242edf282 2006-1-3 Dan Williams <dcbw@redhat.com>
* src/nm-device-802-11-wireless.c
	  src/nm-device.[ch]
		- Add hooks to subclasses for stage3_ip_config_start and
			stage4_ip_config_timeout

	* src/nm-device-802-3-ethernet.c
		- (real_get_generic_capabilities): make devices NM-supported
			by default


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1251 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-03 17:47:38 +00:00
Dan Williams
d639dbeac0 2006-1-3 Dan Williams <dcbw@redhat.com>
* src/NetworkManager.c
		- Move link-checking/probing into the device subclasses
			themselves

	* src/nm-device.[ch]
	  src/nm-device-802-11-wireless.c
	  src/nm-device-802-3-ethernet.c
		- Do periodic link checking in device subclasses rather
			than being triggered from NetworkManager.c
		- discover_wireless_capabilities -> get_wireless_capabilities
		- discover_generic_capabilities -> get_generic_capabilities
		- Device subclass activation routines now return a value of type
			NMActStageReturn to indicate what step to perform next
		- Devices now override stage4_get_ip4_config if they choose


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1249 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-03 17:07:07 +00:00
Dan Williams
6a35ad489c 2006-1-1 Dan Williams <dcbw@redhat.com>
* src/nm-device-802-11-wireless.c
		- (real_init): don't chain up to parent init because we don't
			need to do that anymore

	* src/nm-device.c
		- (discover_device_type): fix arguments to ioctl() to correctly
			pass interface name
		- (nm_device_new): consolidate generic device initialization into
			nm_device_new()
		- (real_init): remove, consolidated to nm_device_new()
		- (nm_device_deactivate, real_deactivate): consolidate


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1247 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2006-01-02 02:50:47 +00:00
Dan Williams
4ed4b491fa 2005-12-31 Dan Williams <dcbw@redhat.com>
* refactor NMDevice into a GObject-based framework with separate
		objects for wired and wireless.  The following files are no
		longer used but should stick around for a bit so we don't
		loose code through the cracks:
			NetworkManagerDevice.c
			NetworkManagerDevice.h
			NetworkManagerWireless.c
			NetworkManagerWireless.h

	The intent here is to allow each device type to manage its own
	connection & activation life-cycle, ie to allow wireless devices
	to interface with wpa_supplicant, etc.  There's a fair bit of
	encapsulation breakage right now that should gradually get pulled
	back into each device, along with things like periodic property
	updates and link probing.


git-svn-id: http://svn-archive.gnome.org/svn/NetworkManager/trunk@1244 4912f4e0-d625-0410-9fb7-b9a5a253dbdc
2005-12-31 08:21:24 +00:00