**** BEGIN LOGGING AT Thu May 20 03:00:01 2010
**** BEGIN LOGGING AT Thu May 20 03:51:34 2010
May 20 07:07:25 <holtmann> #2  0x000000000042f0b2 in creg_notify (result=0x7fff89efcec0, user_data=0x1a8bb20)
May 20 07:07:26 <holtmann>     at drivers/atmodem/network-registration.c:770
May 20 07:07:26 <holtmann> #3  0x0000003fe3a57a9c in IA__g_slist_foreach (list=<value optimized out>, func=
May 20 07:07:26 <holtmann>     0x41e660 <at_notify_call_callback>, user_data=0x7fff89efcec0) at gslist.c:743
May 20 07:07:26 <holtmann> #4  0x000000000041f831 in g_at_chat_match_notify (rbuf=<value optimized out>,
May 20 07:07:28 <holtmann>     user_data=<value optimized out>) at gatchat/gatchat.c:331
May 20 07:07:28 <holtmann> #5  have_line (rbuf=<value optimized out>, user_data=<value optimized out>) at gatchat/gatchat.c:508
May 20 07:07:30 <holtmann> #6  new_bytes (rbuf=<value optimized out>, user_data=<value optimized out>) at gatchat/gatchat.c:648
May 20 07:07:30 <holtmann> #7  0x00000000004215c8 in received_data (channel=0x1a89c20, cond=<value optimized out>, data=
May 20 07:07:33 <holtmann>     0x1a8b130) at gatchat/gatio.c:119
May 20 07:07:34 <holtmann> #8  0x0000003fe3a3923e in g_main_dispatch (context=0x1a7e000) at gmain.c:1960
May 20 07:07:37 <holtmann> #9  IA__g_main_context_dispatch (context=0x1a7e000) at gmain.c:2513
May 20 07:13:48 <denkenz> holtmann: When does this happen?
May 20 07:14:07 <holtmann> Driving in the city of Munich and me re-starting ofonod manually ;)
May 20 07:14:47 <holtmann> There is also an issue with MBM, when it sort of crashes in between and tries to recover.
May 20 07:14:58 <holtmann> I lost all atoms, but Powered=1. No idea on how to re-produce that easily.
May 20 07:16:27 <denkenz> There's nothing in creg notify that should cause this unless the netreg atom has gone away
May 20 07:18:48 <denkenz> what's in stack trace at #1 and #0?
May 20 07:20:26 <kvalo> denkenz: so you want to have at_util_convert_signal_strength()?
May 20 07:21:15 <denkenz> kvalo: might as well
May 20 07:21:51 <kvalo> denkenz: or would you prefer _report_signal_strength() which would also call ofono_netreg_strength_notify()?
May 20 07:27:00 <denkenz> nah, convert is probably better
May 20 07:27:48 <kvalo> I was thinking the same
May 20 08:25:03 <denkenz> kvalo: The preferred way of doing multi-port debugs is:
May 20 08:25:17 <denkenz> g_at_chat_set_debug(data->event, huawei_event_debug, "FooChannel:");
May 20 08:25:40 <denkenz> s/huawei_event_debug/huawei_debug
May 20 09:19:57 <kvalo> denkenz: ok, I'll use that
May 20 09:29:09 <singesang> hi
May 20 09:56:06 <kvalo> denkenz: why are you still awake?
May 20 11:41:02 <denkenz> kvalo: yep
May 20 11:43:44 <kvalo> denkenz: you must be in europe :)
May 20 11:44:40 <denkenz> yep
May 20 11:45:51 <kvalo> denkenz: if I use g_at_chat_suspend() in setup_ppp(), I can't get pass phase 1
May 20 11:46:43 <denkenz> Well, first thing to test is whether this works with gsmdial
May 20 11:47:17 <kvalo> oh, I assumed that gsmdial will work :(
May 20 11:47:40 <denkenz> gsmdial works on all modems we have
May 20 11:47:46 <denkenz> But no guarantee it works on yours
May 20 11:47:48 <kvalo> good
May 20 11:48:10 <denkenz> Also, gprs context uses +CGDATA
May 20 11:48:20 <denkenz> it might be that Huawei actually needs ATD*99
May 20 11:48:26 <denkenz> I'm just guessing here
May 20 11:48:51 <kvalo> huawei works for me with cgdata
May 20 11:49:12 <denkenz> so something's wrong with ppp setup then
May 20 11:49:14 <kvalo> and I can ping the dns server
May 20 11:49:44 <kvalo> but if I try to fix the deactivate crash with g_at_chat_suspend(), ppp doesnt' work
May 20 11:49:56 <kvalo> I'll try gsmdial now
May 20 11:50:10 <denkenz> Are you doing everything that gsmdial does?
May 20 11:50:20 <denkenz> e.g. using ppp_new_from_io?
May 20 11:50:34 <kvalo> no
May 20 11:50:39 <kvalo> I was actually wondering that
May 20 11:51:15 <denkenz> yeah the ppp stuff is still a bit raw, to make it work with g_at_chat_suspend you need to pass around the GAtIO object
May 20 11:51:35 <kvalo> ah, that explains. thanks a lot
May 20 11:52:01 <denkenz> I did the atmodem gprs context in about 10 minutes and never tested it
May 20 11:52:15 <denkenz> Its actually a miracle it works ;)
May 20 11:52:25 <kvalo> hehe
May 20 11:52:55 <kvalo> it works pretty good for me. I just want to deactivate it without crashing ofono :)
May 20 11:53:41 <denkenz> g_at_ppp_shutdown would do it, unfortunately you wouldn't be able to activate it again
May 20 11:53:54 <kvalo> yeah
May 20 11:54:21 <kvalo> gsmdial works for me and it did send CFUN after a TERM signal
May 20 11:54:36 <kvalo> no reply to CFUN, but I guess that's by design
May 20 11:57:04 <denkenz> excellent, so your modem doesn't hup
May 20 11:57:16 <denkenz> sounds like if you do all the steps from gsmdial it should just work
May 20 12:00:49 <kvalo> yeah, I'll do that
May 20 12:33:20 <denkenz> holtmann: Ok the crash in creg_notify should be fixed
May 20 12:33:45 <holtmann> Thanks. I let you know once I re-tested it.
May 20 12:43:39 <kvalo> denkenz: just like you said, g_at_ppp_new_from_io() works with chat_suspend()
May 20 12:44:05 <denkenz> cool
**** BEGIN LOGGING AT Thu May 20 16:18:43 2010
May 20 17:42:10 <kvalo> denkenz: how do I know if huawei hanged up the line after ppp disconnect?
May 20 17:42:52 <kvalo> denkenz: I don't get any response for commands after resume :(
May 20 18:37:58 <wagi> hi, I have a question conserning IMSI, MCC and MNC. When the length of IMSI is more then 3 then MCC and MNC is appended. Is the length of MCC always constant, 3? And the MNC is between 0 and 3?
May 20 18:39:56 <wagi> hmm, I think I have to read up on IMSI, MCC and MNC 
May 20 18:40:34 <wagi> oh, it's prepend
May 21 01:27:44 <zhenhua> denkenz: how will we use PPP server side extension? this part is not so clear to me
May 21 01:28:55 <padovan> Q
May 21 01:28:58 <padovan> oops
May 21 01:29:37 <zhenhua> he
**** ENDING LOGGING AT Fri May 21 02:59:57 2010