**** BEGIN LOGGING AT Tue Mar 22 02:59:57 2011 Mar 22 16:28:54 denkenz: ping Mar 22 16:43:13 fdanis: pong Mar 22 16:43:29 call indicator should be set to 1 during time there is at least 1 active or held call, is my understanding correct ? Mar 22 16:45:05 that is what I remember, however I've learned and unlearned the BT HFP state machine twice now Mar 22 16:45:41 So its best to grok the bt hfp 1.5 spec and the other spec I mentioned Mar 22 16:45:47 Things become pretty clear after that Mar 22 16:46:49 this is what I understand from both specs, for 3-way calls the call indicator is not changed Mar 22 16:54:39 Then that is likely the case Mar 22 16:54:59 But you might still want to run a few HFP client implementations against your understanding Mar 22 16:55:20 Namely the PTS tester and our HFP client implementation Mar 22 16:55:31 AFAIK we passed all PTS tests for our client one Mar 22 20:43:45 denkenz: did you get the vl600 dongle working under linux? the AT interface is so bad that it probably doesn't make sense supporting it in ofono Mar 22 20:45:28 no, I just received it 10 minutes ago Mar 22 20:45:47 And yes, I suspect we need the Pantech ones Mar 22 20:50:25 all in all i think there's no difference between 3g and 4g visible to ofono Mar 22 20:51:08 There is, but not a huge one for the data-only usecase, especially with no 3g fallback Mar 22 20:51:31 the dongle does all the job, the vl600 is just terribly non conforming to any standards it claims to use Mar 22 20:51:51 i think the panatech one does fall back to 3g Mar 22 20:52:36 Not on Verizon, at least not in that sense ;) Mar 22 20:57:06 So how do you get the vl600 into AT mode anyway? Mar 22 20:59:05 you need to add a header to every frame you send to its tty and strip it off when you receive packets, one sec.. Mar 22 20:59:42 jeez, and how the hell did you figure that out ;) Mar 22 21:00:34 Is there a way to force two ttys to be created? Seems there are two AT descriptors on lsusb Mar 22 21:03:04 http://openstreetmap.pl/balrog/vl600-com.py is a dummy terminal that opens /dev/ttyACM0 and lets you send and receive at data Mar 22 21:04:07 practically no at command works the way it should Mar 22 21:06:46 also don't execute AT%FRST, it bricks the dongle :) Mar 22 21:07:39 good to know Mar 22 21:10:52 I'm seriously scared at how you found out this header stuff Mar 22 21:12:15 someone in ubuntuforums posted a log from an usb analyzer from running the windows driver back in december and then gave up Mar 22 21:13:10 i think i also have the ethernet frames decoding and encoding working, but i couldn't figure out the gateway/dns ips Mar 22 21:14:21 Have you tried running dhcp? Mar 22 21:15:16 yeah, although the windows log doesn't show it running dhcp Mar 22 21:19:47 Have you tried +CGCONTRDP? Mar 22 21:19:59 Or maybe CGCONTRDP Mar 22 21:20:07 As its showing up on CLAC :P Mar 22 21:23:21 they both either return ERROR or +CME ERROR: phone failure when i try, maybe there's some circumstance when they actually work, i haven't tried very hard yet Mar 22 21:24:33 there are some funny commands that return "[ACM something] This command will be implemented..." Mar 22 21:25:24 yeah, the at interface is a PoS it seems Mar 22 21:26:29 Some commands also seem to disappear into the ether Mar 22 21:26:36 so your magic might not be 100% correct Mar 22 21:29:24 that's possible but it's also possible they just return nothing ;) Mar 22 21:30:33 They get echoed Mar 22 21:30:39 But then disappear *poof* Mar 22 21:30:54 Would be fun to know the protocol this thing is using at least Mar 22 21:36:12 I think something is definitely up with your magic ;) Mar 22 21:39:15 it should throw an exception when it doesn't understand something, though Mar 22 21:39:55 yeah, i suspect this is some known protocol, btu i just don't know what it is as i haven't seen it before Mar 22 21:40:22 it's similar to rndis in that it just encapsualtes everything in headers and otherwise works like cdc acm and cdc ethernet Mar 22 21:41:58 in one place in the usb log the string MSFT appears which i think is some set of microsoft "standards" including rndis Mar 22 21:43:29 yep, but if we knew what it was it might get easier Mar 22 21:45:39 either way even if starts working it'd be hell to try to support in ofono Mar 22 21:46:04 Perhaps Mar 22 21:46:13 But yeah, lets try to get the Pantech dongles Mar 22 21:46:25 However, to me it seems like the vl600 is running a muxer Mar 22 21:47:01 yeah, the ACM port can also receive and receive binary frames Mar 22 21:47:16 receive/return Mar 22 21:47:35 there's one byte which decides if the frame is text or binary Mar 22 21:48:29 the way i make the dongle connect is actually by sending a couple of those binary frames extracted from the usb log Mar 22 21:48:44 so the led strats blinking green Mar 22 21:50:02 i suppose you can do the same with one of those at commands Mar 22 21:53:23 I doubt they screwed up AT+CGACT, but you never know of course Mar 23 01:47:25 :o **** ENDING LOGGING AT Wed Mar 23 02:59:57 2011