**** BEGIN LOGGING AT Sat Oct 24 02:59:56 2009
Oct 24 15:24:29 <holtmann> denkenz: Do you follow the command order we use inside ConnMan?
Oct 24 15:24:46 <denkenz> holtmann: More or less
Oct 24 15:24:49 <holtmann> It seems that ENAP and E2NAP have to be executed in the right order.
Oct 24 15:25:02 <denkenz> holtmann: I don't execute E2NAP
Oct 24 15:25:07 <denkenz> its only used as unsolicited
Oct 24 15:25:22 <holtmann> You might have to. Try to enable it.
Oct 24 15:25:40 <holtmann> E2NAP enables the notifications.
Oct 24 15:26:55 <denkenz> E2NAP is never executed then
Oct 24 15:28:05 <holtmann> You just need to send it to enable the notifications.
Oct 24 15:28:39 <denkenz> I doubt that this is causing it to crash though
Oct 24 15:28:43 <denkenz> But lemme try
Oct 24 15:29:49 <holtmann> Yeah. Also on the 35xx hardware it needs to be send after the ENAP for some stupid reason. At least there is commit inside ModemManager that indicates that.
Oct 24 15:30:42 <denkenz> Unpossible
Oct 24 15:30:47 <denkenz> it crashes on the ENAP
Oct 24 15:31:14 <denkenz> E.g. ENAP=1,1 -> OK, crash
Oct 24 15:32:10 <holtmann> But using ConnMan this works for you?
Oct 24 15:33:10 <denkenz> haven't tried connman actually
Oct 24 15:33:31 <holtmann> Can you do that first. If that also crashes then it is the card.
Oct 24 15:33:52 <denkenz> Not really a solution
Oct 24 15:33:53 <holtmann> I can't really try it here without causing roaming charges like crazy. I don't have a Japanese SIM card.
Oct 24 15:34:14 <denkenz> Connman can easily control the command order, ofono can't
Oct 24 15:34:30 <holtmann> I know, but then we at least know it can work and is not the firmware.
Oct 24 15:36:08 <holtmann> Also maybe waiting for +PACSP1 is required before calling ENAP.
Oct 24 15:36:16 <holtmann> There are some weird corner cases.
Oct 24 15:37:44 <denkenz> wtf is this pacsp1?
Oct 24 15:37:57 <denkenz> can you share the command set btw?
Oct 24 15:38:15 <holtmann> I only have what is public available and inside ModemManager.
Oct 24 15:39:01 <denkenz> well considering they don't do jack with those commands
Oct 24 15:39:12 <denkenz> And also considering they've no idea wtf they're doing...
Oct 24 15:39:32 <holtmann> The order is important for some of them. That is a problem.
Oct 24 15:41:17 <denkenz> Nah, they just have no idea wtf they're doing
Oct 24 15:41:38 <holtmann> I am serious. You have to wait for certain events before you can send the next command.
Oct 24 15:41:55 <denkenz> Not on the mbm
Oct 24 15:42:00 <denkenz> Works manually btw
Oct 24 15:42:31 <holtmann> That is because you are too slow manually. Seriously some stuff is just weird and black magic.
Oct 24 15:42:32 <denkenz> The MBM hardware does not register to GPRS until you attach
Oct 24 15:43:29 <denkenz> Actually no, it takes ages for me to get signal where I am
Oct 24 15:43:37 <denkenz> The modem has plenty of time either way :)
Oct 24 15:47:45 <denkenz> Still dies in oFono even with E2NAP=1
Oct 24 15:48:03 <denkenz> I get an OK for ENAP, immediate E2NAP=2 and boom
Oct 24 15:50:56 <holtmann> We don't store the context data yet?
Oct 24 15:52:32 <denkenz> no
Oct 24 15:55:48 <denkenz> still goes kaboom after I took out all atoms except sim / netreg / gprs
Oct 24 15:59:06 <holtmann> How do I activate a context?
Oct 24 15:59:15 <denkenz> CreateContext
Oct 24 15:59:25 <holtmann> Got that part. Set the APN etc.
Oct 24 15:59:28 <denkenz> on context: SetProperty("AccessPointName", "foo")
Oct 24 15:59:33 <holtmann> Just to Active=true?
Oct 24 15:59:34 <denkenz> SetProperty("Active", True)
Oct 24 15:59:47 <denkenz> Oh and SetProperty("Powered", True) on DCM
Oct 24 16:04:37 <holtmann> So mine doesn't crash.
Oct 24 16:04:56 <denkenz> Yah, sounds like a broken device
Oct 24 16:05:07 <holtmann> It also never connects either.
Oct 24 16:05:14 <denkenz> Heh
Oct 24 16:05:36 <denkenz> This is why I'm harping on andrew, the logic for registering is fucked
Oct 24 16:05:56 <denkenz> I've already had a few infinite CGATT=0 / CGATT=1 loops happen
Oct 24 16:06:37 <holtmann> Do we get a signal when we are attached?
Oct 24 16:08:04 <denkenz> The way my MBM device works is that it doesn't send CGREG until we attach
Oct 24 16:08:14 <denkenz> However, I'm working with shitty signal right now
Oct 24 16:08:15 <holtmann> Also we might wanna disable the scanning when connecting to GPRS. Most modems can't handle it.
Oct 24 16:08:28 <denkenz> yeah I know ;)
Oct 24 16:08:31 <holtmann> I don't see the CGREG at all.
Oct 24 16:09:17 <denkenz> SetProperty("Powered", True) on DataConnectionManager
Oct 24 16:09:27 <holtmann> Yes. Did that.
Oct 24 16:12:25 <denkenz> roamingallowed = true?
Oct 24 16:13:00 <holtmann> Crap. That might be it.
Oct 24 16:16:51 <holtmann> And you have to read ENAP on startup to see if we are already have been connected.
Oct 24 16:17:25 <holtmann> Also Powered succeeds and Active succeeds even if RoamingAllowed=false.
Oct 24 16:17:31 <holtmann> It just doesn't work.
Oct 24 16:18:29 <denkenz> powered should
Oct 24 16:19:09 <denkenz> active should just check attached and fail if not
Oct 24 16:20:52 <holtmann> I think your problem is that CDGCONT and ENAP are send to quickly after each other.
Oct 24 16:21:05 <holtmann> But ConnMan indeed does the same.
Oct 24 16:21:21 <holtmann> ofonod[6442]: > AT+CGDCONT=1,"IP","internet.com"\r
Oct 24 16:21:21 <holtmann> ofonod[6442]: < \r\nOK\r\n
Oct 24 16:21:21 <holtmann> ofonod[6442]: > AT*ENAP=1,1\r
Oct 24 16:21:21 <holtmann> ofonod[6442]: < \r\nOK\r\n
Oct 24 16:21:30 <holtmann> It see this, but usb0 never gets LOWER_UP.
Oct 24 16:22:03 <denkenz> send e2nap=1 at init and see?
Oct 24 16:26:34 <holtmann> Crap. I forgot that you have to manually do ifup usb0.
Oct 24 16:26:51 <holtmann> Try to ifup usb0 first and see if that makes a difference.
Oct 24 16:28:56 <denkenz> it works when i do all of this manually
Oct 24 16:29:08 <denkenz> no ifup required
Oct 24 16:31:06 <holtmann> Please try it with ConnMan. I am curious if it works ther.
Oct 24 16:41:13 <denkenz> Ok, I give up on this for now
Oct 24 16:41:20 <denkenz> My card shoots itself randomly now
Oct 24 17:17:17 <denkenz> So taking out CGATT=1 and ifconfig usb0 up it seems to work now
**** ENDING LOGGING AT Sun Oct 25 02:59:56 2009