**** BEGIN LOGGING AT Fri Dec 03 03:00:00 2010 Dec 03 08:38:11 pessi:? Dec 03 08:41:14 ! Dec 03 08:41:50 i found a weird issue with MeeGo-oFono in N900. Dec 03 08:41:59 the use case is back-toback calls Dec 03 08:42:27 make a MO, reject it at other end, immediately make another MO to the same number Dec 03 08:42:53 after this the dial request always failed Dec 03 08:43:02 it is blacklist Dec 03 08:43:57 i just mailed you the log Dec 03 08:43:59 if calls made by a an application (as opposed to a user) get rejected too many times, modem will block them for a while Dec 03 08:44:08 oh ok Dec 03 08:44:12 mandatory 3gpp feature Dec 03 08:44:41 now the problem is how we could know if call is dialed by user and not by some application Dec 03 08:44:46 yeh Dec 03 08:45:19 ? Dec 03 08:46:59 also defining "rejected too many times", Dec 03 08:47:06 which spec defines? Dec 03 08:47:24 i could see this happening with 2nd attempt Dec 03 08:53:06 22.001 annex e Dec 03 09:03:52 did you get a chance to see the log i mailed? Dec 03 09:07:13 yep, it would be much more helpful with OFONO_ISI_DEBUG=all Dec 03 09:19:52 after first rejected dial you have wait for 5 seconds, after second 1 minute, 3rd, 4th, 1 minute, after 5th, 3 minutes Dec 03 09:23:11 hmm ok Dec 03 09:58:57 pessi: you had some comments for Read/write EFcfis patch. Is it clarified now? Dec 03 10:43:11 jep, my point is that it provides little value to read the cf settings from sim card if you cannot get them from the D-Bus interface without doing full check from network Dec 03 11:00:13 pessi: ForwardingFlagOnSim is the main information which will be available via DBus. Basically, this is used to just indicate the VoiceUnconditional active Dec 03 11:00:38 This will be used for showing some icon in the UI Dec 03 12:04:12 denkenz: ping Dec 03 12:14:30 Jeevaka: did you move ForwardingFlagOnSim to sim atom? if so, t would make sense to remove OnSim from the name ? and Flag is apparent from the fact it is boolean... Dec 03 12:31:28 its still in the call forwarding Dec 03 14:12:21 Does anyone use a wwan modem with the connman/ofono in MeeGo? Dec 03 14:12:36 I'm trying to figure out which mini PCI modem I should buy. Dec 03 14:14:36 Seems like Ericsson mini PCI wwan modems are at least supported in the linux kernel. Dec 03 14:14:43 No proprietary driver needed, it seems. Dec 03 14:15:31 …so I've thought about the Ericsson F3607gw one. Dec 03 14:18:27 that should be pretty good Dec 03 14:20:31 msvb: I have this one and it's pretty good indeed. Dec 03 14:21:56 msvb: who is selling it in Germany? Lenovo Finland does not confess they ever sold one, nor they want to sell it ;) Dec 03 14:22:45 Sorry, I'm back. Dec 03 14:23:03 pessi: I found a few places just looking for online shops here in DE. Dec 03 14:23:20 …but don't remember what they were. Dec 03 14:23:37 I think one sold a older F3507g. Dec 03 14:23:52 And one or two others sold the newer F3607gw. Dec 03 14:24:19 It seems that the F3607gw draws less power. Dec 03 14:24:35 Which makes it more suitable for mobile (battery based) use. Dec 03 14:24:51 But sadly, I cannot find the specs anywhere. Dec 03 14:25:05 So it's impossible to know just how power hungry these two models are. Dec 03 14:25:47 pessi: And it's hard to understand Lenovo's position on wwan devices. Dec 03 14:27:50 usb-id whitelisting? Dec 03 14:28:30 pessi: Yes, and I though about hacking the BIOS. Dec 03 14:28:50 pessi: At first it seemed all you need is a hex editor. Dec 03 14:29:07 pessi: But then I heard about people using some win32 program Phoenix BIOS editor... Dec 03 14:29:18 pessi: And it makes me wonder if you need to get that first. Dec 03 14:29:25 I got a bios for dell 5530 (3507) from some thinkpad forum Dec 03 14:29:29 pessi: Which of course is a pain, because you first need to install Win32. Dec 03 14:29:46 pessi: Then find a place to copy the Phoenix BIOS editor. Dec 03 14:30:04 pessi: You install a BIOS that another person manipulated? Dec 03 14:30:12 pessi: I'd rather not do that. Dec 03 14:30:25 pessi: So in the end, it's probably all possible. Dec 03 14:30:32 pessi: But the hassle outweighs the benefit. Dec 03 14:30:46 pessi: …as well as the danger of bricking the Lenovo computer. Dec 03 14:31:24 I wonder if Lenovo has engineered other failures as well. Dec 03 14:31:49 …as well as the policy to limit the user's ability to work with good hardware. Dec 03 14:33:02 msvb: I waited until Aki got his laptop bios flashed first, and then installed on mine. always have a scratch monkey ;) Dec 03 14:34:14 pessi: That's good, and good that Aki got lucky as well. Dec 03 14:35:20 Bye bye. Dec 03 14:35:34 yep, it is embarrassing to go and tell, hey, I just bricked this 2000 € laptop, I need a new one ;-) Dec 03 14:45:41 As far as I know the Lenovo whitelist includes the Ericsson F3607gw. Dec 03 14:45:59 So that no BIOS hack is needed. Dec 03 14:46:15 I mean Lenovo Ideapad S10-3t, and probably others. Dec 03 14:47:01 So I wonder why pessi and Aki needed new BIOS for F3507g devices? Dec 03 14:48:18 because they had dell usb-id Dec 03 14:48:35 which was not on the whitelist Dec 03 14:57:07 so do not buy dell 5540 Dec 03 18:12:18 denkenz: about the race condition with pending online/offline. From what I checked that won't happen. Dec 03 18:13:07 when offline_cb or online_cb are called theu check the current state of the modem, and then return error if it wasn't the expected one. Dec 03 18:13:27 That should be enough to avoid the race condition Dec 03 18:14:09 that is not what I meant actually Dec 03 18:14:31 when you call set_property(online, blah) you set the pending var Dec 03 18:14:49 when you call ofono_modem_set_powered, that assumes that pending is for powered, not online Dec 03 18:15:04 So maybe this works by accident, but certainly not what is expected Dec 03 18:15:21 Ah, I see. Let me check it. Dec 03 18:16:35 denkenz: oFono sends NetworkRegistration.Strength as a uint16 although the documentation says it's a byte in the [0..100] range. What's the actual range ? Dec 03 18:17:24 0 100 Dec 03 18:17:32 probably a bug Dec 03 18:20:44 I got 220 and 240 for example. Dec 03 18:22:53 so definitely a bug. Dec 03 18:23:34 220? On what driver Dec 03 18:23:40 anyway push fixed Dec 03 18:23:43 err fix pushed Dec 03 18:25:15 on the mbm driver. Dec 03 18:25:39 hm, I've never observed that Dec 03 18:25:44 always worked fine for me Dec 03 18:25:53 do you have a log handy? Dec 03 18:31:10 denkenz: hmm, that's probably a bug on my side. It seems oFono is reporting a correct strength. sorry for the noise. Dec 03 18:36:16 denkenz: that might fix the pending problem http://pastebin.com/gbLw6vHL What do you think? Dec 03 18:38:54 what about simply sending failed error if pending is set Dec 03 18:39:07 and resetting pending? Dec 03 18:39:53 inside modem_reset that is Dec 03 18:42:45 however, that pastebin should be submitted as well Dec 03 18:42:48 to be extra paranoid Dec 03 18:43:45 denkenz: you idea is better, I'll do that. Dec 03 22:09:57 denkenz: Gah, the logic behind lockdown and powered is pretty nasty. I hope it right, at least my tests show that. :) Dec 03 22:44:00 denkenz: where does this numbers come from? Dec 03 22:44:01 __ofono_ussd_ssc_register(cs->ussd, "30", clip_colp_colr_ss, cs, NULL); Dec 03 22:44:02 __ofono_ussd_ssc_register(cs->ussd, "31", clir_ss_control, cs, NULL); Dec 04 00:20:51 padovan: 22.030 **** ENDING LOGGING AT Sat Dec 04 02:59:59 2010