**** BEGIN LOGGING AT Mon Dec 27 02:59:57 2010 **** BEGIN LOGGING AT Mon Dec 27 18:49:46 2010 Dec 27 18:59:09 denkenz: hi, ping Dec 27 19:56:04 denkenz: holtmann : you suggested to split set_gps() into two methods: enable and disable (for the public gps atom header). And, also, enable() should provide to the core the NMEA/whatever device node to use and the type of the device. Should this information provided be used to report the readonly properties Device and Type via dbus right?. Another question I have is what exactly is reporting the readonly Device property (or examples). For mbm /dev/ttyACM2 wou Dec 27 19:56:33 Why can't the device node be a property? Dec 27 19:56:51 Would be a lot more consistent with all the other interfaces. Dec 27 20:03:42 holtmann: Ah, I see now. I am okey with the device node, but I was thinking that "device node" readonly property was just a general advice and that I should check the property values. Now it is clear Dec 27 23:43:17 have anyone tested ofono 0.36 on the freerunner/calypso? Dec 27 23:44:16 i cannot list that modem, the modem.conf looks good Dec 27 23:44:39 fgau: modem.conf is gone. Run 0.37 and use a udev rule. Dec 27 23:45:39 holtmann: then no udev rule exists for the calypso Dec 27 23:46:39 You have to create one for yourself. it has been discussed on the mailing list why we not include one. Dec 27 23:46:54 Same reason we don't include one for IFX modems. Dec 27 23:47:43 holtmann: ok, thanks! we will do Dec 28 00:13:46 holtmann: would be a cool idea to include a sample one, comented to help Dec 28 00:14:53 emdete: Yes, has been discussed as well. Waiting for the patch for it ;) Dec 28 00:15:17 holtmann: :D Dec 28 00:15:37 my change doesn't work, so it would save my time alot ;) Dec 28 00:15:42 I'm still waiting for someone to confirm the udev way works Dec 28 00:16:15 denkenz: do you have a sample of the 'udev way'? Dec 28 00:17:28 There's a thread about calypso on the ML Dec 28 00:17:34 I posted the udev rule there Dec 28 00:18:11 denkenz: For IFX it just works fine. Dec 28 00:18:26 I know, I don't see any reason for it not to work Dec 28 00:18:48 But I want it confirmed before I put in the docs Dec 28 00:21:51 denkenz: can you paste it here or a link to that mail? Dec 28 00:24:09 KERNEL=="ttySAC0", ENV{OFONO_DRIVER}="calypso" Dec 28 00:24:28 is it really KERNEL? not NAME? Dec 28 00:25:51 no idea Dec 28 00:26:51 I have not put this into the docs for a reason ;) Dec 28 00:27:29 In case of IFX it is kernel. Dec 28 00:27:30 KERNEL=="ttyIFX[0-9]*", ENV{OFONO_DRIVER}="ifx" Dec 28 00:28:38 however, at least the udev manpage hints that KERNEL is indeed the proper rule Dec 28 00:29:19 denkenz: the freerunner has a rule like KERNEL=="s3c2410_serial0", NAME=="ttySAC0", ENV{OFONO_DRIVER}="calypso" already Dec 28 00:29:34 sry, KERNEL=="s3c2410_serial[0-9]", NAME="ttySAC%n" Dec 28 00:29:56 the first one was the one i tried Dec 28 00:30:08 Then the calypso might need to use an alternative udev rule Dec 28 00:30:48 KERNEL=="s3c2410_serial0", ENV{OFONO_DRIVER}="calypso" Dec 28 00:30:58 Just before the renaming rule. Dec 28 00:31:05 Also the kernel name should be still valid. Dec 28 00:31:40 And actually maybe not rename the TTY actually. Dec 28 00:31:51 That renaming is pointless anyway. Dec 28 00:31:58 oFono does not care what's the name. Dec 28 00:32:21 That renaming is just for humans ;) Dec 28 00:32:50 i dont care as long as it works... :/ Dec 28 00:33:34 KERNEL Dec 28 00:33:36 Match the name of the event device. Dec 28 00:33:36 NAME Dec 28 00:33:37 Match the name of the node or network interface. It can be used once the NAME key has been set in one of the preceding rules. Dec 28 00:34:01 So the one holtmann quoted should work always Dec 28 00:34:10 regardless of the renaming rule Dec 28 00:34:13 hm, didn't work that way. will check tomorrow, too tired & hungry Dec 28 00:34:59 Could be that because of the rename the device node on disk is different now. Dec 28 00:35:11 NAME=="ttySAC0", ENV{OFONO_DRIVER}="calypso" should be fine. **** ENDING LOGGING AT Tue Dec 28 02:59:56 2010