**** BEGIN LOGGING AT Wed May 11 02:59:58 2011
May 11 14:42:21 <naquad> hi
May 11 14:44:22 <naquad> i've got zte modem with vendorid=19d2 and productid=ffed, problem is i can't find out how to make ofono see it. any ideas?
May 11 14:48:24 <holtmann> You need to change /etc/udev/rules.d/*-ofono.rules file to include that id.
May 11 14:49:30 <holtmann> ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0031", ENV{OFONO_IFACE_NUM}=="03", ENV{OFONO_ZTE_TYPE}="modem"
May 11 14:49:30 <holtmann> ATTRS{idVendor}=="19d2", ATTRS{idProduct}=="0031", ENV{OFONO_IFACE_NUM}=="01", ENV{OFONO_ZTE_TYPE}="aux"
May 11 14:49:58 <holtmann> Something like this is an example. However you do need to figure out which ports of your modem are the main TTYs for PPP and aux data.
May 11 14:50:23 <holtmann> Every product is different. So this is something you need to do by yourself and for your modem.
May 11 14:50:33 <holtmann> Things like /proc/bus/usb/devices might help here.
May 11 14:50:42 <holtmann> Once you have that working, submit a patch for it.
May 11 15:05:25 <BertrandAygon> Hi Marcel. Looking at cppcheck, I have found a leak, but would like to be sure to propose the correct patch.
May 11 15:05:34 <BertrandAygon> In plugins/nokia-gpio.c, static int gpio_probe_links(void)
May 11 15:05:54 <BertrandAygon> the funtionc opendir but never closedir. Seems not to be good.
May 11 15:06:34 <BertrandAygon> You confirm that there is no 'magic' stuff there, and that we have to close the dir upon return
May 11 15:15:42 <holtmann> That is real Nokia magic stuff. And you are most likely right there. They forgot the closedir.
May 11 15:16:29 <BertrandAygon> Ok, I will propose something
May 11 15:16:30 <holtmann> That just misses the closedir(). I had a quick look.
May 11 15:17:29 <BertrandAygon> And what about the cmtdir?
May 11 15:18:02 <BertrandAygon> It's created, but then, if the opendir failed, we return 
May 11 15:18:20 <BertrandAygon> what is this dir used for? Do we have to clean it up?
May 11 15:19:04 <BertrandAygon> This dir is created during the probe, but never removed. Is it OK?
May 11 15:20:02 <BertrandAygon> Ok, I will forget about cmtdir, to not break anything, and just add the closedir
May 11 15:24:05 <BertrandAygon> Do I have to check the result of closedir? From my point of view, no.
May 11 15:27:19 <holtmann> No idea. Assumption is cmtdir is the cmtspeech stuff.
May 11 15:27:36 <holtmann> At this point you need to ask the Nokia guys.
May 11 15:27:54 <holtmann> git blame plugins/nokia-gpio.c is your friend to figure out who wrote that code.
May 11 15:28:27 <BertrandAygon> Yep I tried to talk yesterday, but failed :)
May 11 15:28:38 <BertrandAygon> I will retry tomorrow
May 11 15:47:48 <crevetor> Hi,
May 11 15:48:32 <crevetor> ever since I've started using ofono I've had this "problem" : the operator name always shows as "Rogers" but I'm using "Fido" which is an MVNO of Rogers.
May 11 15:49:18 <denkenz> Your modem does not implement low level EF reading or EFspn is not present on your sim card
May 11 15:53:14 <crevetor> denkenz: it's an N900 
May 11 15:53:52 <denkenz> N900 should implement EFspn, but none of the low level EF reading
May 11 15:54:04 <crevetor> denkenz: should ofono report the name correctly if it is on the sim or does the OS side have to do that ?
May 11 15:54:24 <denkenz> oFono reports the name according to EFspn if available
May 11 15:54:36 <denkenz> Otherwise the COPS name is all it can give
May 11 15:55:34 <crevetor> denkenz: I remember seeing the name of my operator in the ofono logs, I 'd have to find ou exactly when and how..
May 11 15:58:52 <crevetor> denkenz: what's the env variable to see ISI protocol messages ?
May 11 15:59:15 <denkenz> no idea
May 11 15:59:28 <crevetor> oh ok
May 11 15:59:29 <denkenz> OFONO_ISI_DEBUG?
May 11 15:59:34 <crevetor> I think so
May 11 16:04:21 <crevetor> it's actually OFONO_ISI_TRACE that I was looking for
May 11 16:11:28 <crevetor> denkenz: the isimodem specialist is akiniemi right ?
May 11 16:11:37 <denkenz> nod
May 11 16:12:49 <holtmann> Any of the Nokia guys might be able to help here.
May 11 16:13:03 <holtmann> Neither Denis or I have any insights into ISI actually.
May 11 16:13:43 <crevetor> ok
May 11 16:13:46 <crevetor> akiniemi: there ?
May 11 16:13:55 <crevetor> who are the other nokia guys &
May 11 16:13:58 <crevetor> ?
May 11 17:25:43 <denkenz> ==20475== Syscall param capget(data) points to unaddressable byte(s)
May 11 17:25:44 <denkenz> ==20475==    at 0x60D2CB7: capget (in /lib64/libc-2.11.3.so)
May 11 17:25:44 <denkenz> ==20475==    by 0x5BF9C58: ??? (in /usr/lib64/libcap-ng.so.0.0.0)
May 11 17:25:45 <denkenz> ==20475==    by 0x5BFA8AD: capng_clear (in /usr/lib64/libcap-ng.so.0.0.0)
May 11 17:25:45 <denkenz> ==20475==    by 0x46EECE: main (main.c:141)
May 11 17:25:45 <denkenz> ==20475==  Address 0x0 is not stack'd, malloc'd or (recently) free'd
May 11 17:27:11 <holtmann> Yeah.
May 11 17:27:41 <denkenz> Actual error or ignore?
May 11 17:32:56 <courmisch> Does strace show EFAULT ?
May 11 17:34:34 <denkenz> capget(0x20080522, 0, NULL)             = 0^M
May 11 17:34:44 <denkenz> capget(0x20080522, 21117, {0, 0, 0})    = 0^M
May 11 17:34:47 <courmisch> so ignore (valgrind bug)
May 11 17:35:22 <denkenz> sounds good
**** ENDING LOGGING AT Thu May 12 02:59:58 2011