BluetoothEventLoop has an instance of BluetoothDeviceService which you
could call disable() on.

On Mon, Mar 2, 2009 at 11:11 PM, pavan savoy <[email protected]> wrote:
> Hey Nick,
>
> I am kind a stuck in here, find attached the patch...
> I got the AdapterRemoved Signal, from dbus.. I sent it out to
> EventLoop over "method_onBluetoothUnexpectedOff" and then what ?
>
> I don't have an instance of DeviceService to do an 'disable' ? so I am
> currently trying to do stop hcid and see what happens.
>
> I hope at least that would be updated to UI.
>
> regards,
> Pavan
>
> On 3/3/09, pavan savoy <[email protected]> wrote:
>> Oh, I wish I could do all that ...
>>
>> anyway from the initial look of things ....In
>> frameworks/base/core/jni/android_server_BluetoothEventLoop.cpp
>>
>> in function event_filter, I add a line say,
>> dbus_message_is_signal(msg,"org.bluez.Manager","AdapterRemoved"); ?
>>
>> and then what else ?
>> I don't how I can tell the server_bluetoothDevice that adapter has
>> been removed ?
>>
>> regards,
>> Pavan
>>
>>
>> On 3/3/09, Nick Pelly <[email protected]> wrote:
>>>
>>> Hi Pavan,
>>>
>>> It would be nice if the UI could be updated when hcid hits this
>>> particular failure case. Feel free to send a patch.
>>>
>>> Nick
>>>
>>> On Mon, Mar 2, 2009 at 5:37 AM, pavan savoy <[email protected]>
>>> wrote:
>>>> Just to put in a bit more info.
>>>>
>>>> I//system/bin/hciattach: Found a --vendor-- chip!
>>>> I//system/bin/hciattach: Firmware file : /lib/-----------
>>>> I//system/bin/hciattach( 1341): Loaded script version 1
>>>> I/bluedroid( 1007): Starting hcid deamon
>>>> I//system/bin/hcid( 1345): hcid[1346]: Bluetooth HCI daemon
>>>> I//system/bin/hcid( 1345): hcid[1346]: Enabling debug information
>>>> I//system/bin/hcid( 1186): hcid[1187]: HCI dev 0 registered
>>>> I//system/bin/hcid( 1186): hcid[1187]: HCI dev 0 already up
>>>> I//system/bin/hcid( 1186): hcid[1187]: Device hci0 has been added
>>>> I//system/bin/hcid( 1186): hcid[1187]: Starting security manager 0
>>>> I//system/bin/hcid( 1186): hcid[1187]: Can't read simple pairing mode on
>>>> hci0: I/O error (5)
>>>> I//system/bin/hcid( 1186): hcid[1187]: child 1189 exited
>>>> I//system/bin/hcid( 1186): hcid[1187]: child 1190 exited
>>>> ----------------
>>>> I//system/bin/hciattach( 1181): Initialization timed out.
>>>>
>>>> After hci0 has been added my hciattach timed out !!!! Although hcid came
>>>> back telling...
>>>>
>>>> I//system/bin/hcid( 1186): hcid[1187]: HCI device open failed: hci0
>>>> I//system/bin/hcid( 1186): hcid[1187]: HCI dev 0 down
>>>> I//system/bin/hcid( 1186): hcid[1187]: Stopping security manager 0
>>>> I//system/bin/hcid( 1186): hcid[1187]: Device hci0 has been disabled
>>>> I/logwrapper( 1181): /system/bin/hciattach terminated by exit(1)
>>>> I//system/bin/hcid( 1186): hcid[1187]: HCI dev 0 unregistered
>>>> I//system/bin/hcid( 1186): hcid[1187]: Unregister path: /org/bluez/hci0
>>>> I//system/bin/hcid( 1186): hcid[1187]: Device hci0 has been removed
>>>>
>>>> even if the hci0 is un-registered, just because it was registered
>>>> once....
>>>> The UI doesn't seem to get it .... or is not programmed to get it ?
>>>>
>>>>
>>>>
>>>>
>>>> On Mon, Mar 2, 2009 at 5:42 PM, pavan savoy <[email protected]>
>>>> wrote:
>>>>>
>>>>> Nick,
>>>>>
>>>>> Some strange scenarios,
>>>>>
>>>>> Suppose hciattach has 2 steps
>>>>>    1. send some part of chip vendor script through serial port &
>>>>>    2. send some part of vendor script through socket [hci0] interface.
>>>>>
>>>>> However inside bluetooth.c, once bt_enable is able to create socket, it
>>>>> tries to start hcid.
>>>>>
>>>>> If suppose hciattach at that point of time, fails for some "strange"
>>>>> reason, [or say I do a hciconfig hci0 down]...
>>>>>
>>>>> How is the UI / bluetooth.c supposed to know that bluetooth is not on ?
>>>>>
>>>>> regards,
>>>>> Pavan
>>>>
>>>>
>>>> >
>>>>
>>>
>>> >>
>>>
>>
>
> >
>

--~--~---------~--~----~------------~-------~--~----~
unsubscribe: [email protected]
website: http://groups.google.com/group/android-porting
-~----------~----~----~----~------~----~------~--~---

Reply via email to