**** BEGIN LOGGING AT Thu Aug 19 02:59:57 2010 Aug 19 08:06:09 CC src/network.o Aug 19 08:06:09 cc1: warnings being treated as errors Aug 19 08:06:09 src/network.c: In function ‘reset_available’: Aug 19 08:06:09 src/network.c:1182: error: ‘return’ with no value, in function returning non-void Aug 19 13:01:07 AT^CPIN? Aug 19 13:01:08 ^CPIN: READY,,10,3,10,3 Aug 19 13:01:08 OK Aug 19 13:01:21 kvalo: This should work on your Huawei modem as well. Aug 19 13:01:39 We just have to quirk it. Aug 19 13:02:54 AT^USSDMODE=? Aug 19 13:02:54 ^USSDMODE:(0-1) Aug 19 13:02:54 OK Aug 19 13:03:07 denkenz: Are the new Huawei modems maybe supporting PDU mode for USSD? Aug 19 13:03:09 holtmann: thanks. I hope to try it tomorrow Aug 19 13:03:58 holtmann: is there a dbus interface for the retry count already? Aug 19 13:04:03 I didn't find anything Aug 19 13:04:30 Nope. It is in the TODO list, but you have to come up with this. Aug 19 13:04:52 It is all vendor specific non-sense. Aug 19 13:19:09 holtmann: yeah, but it's quite usefull for the user Aug 19 14:15:23 holtmann: I saw that ^USSDMODE thing, but doubt it does PDUs Aug 19 14:56:26 pekka, balrog-k2n: Can someone fill me in whats going on with the CPIN polling? Aug 19 14:56:52 I keep seeing patches flying that touch that same area Aug 19 17:20:42 denkenz: on my mbm modem +CPIN usually works at the first try Aug 19 17:21:08 i.e. returns OK when sim present and ERROR otherwise Aug 19 17:21:20 Is this the MD300? Aug 19 17:21:27 Or the one in your X200? Aug 19 17:21:42 no, it's the minipci plugged into a usb adapter Aug 19 17:21:54 f3607 Aug 19 17:22:55 and i think pessi's modem will sometimes return +CME ERROR: 10 before it either succeeds or errors Aug 19 17:23:27 and as far i understand he now found that sometimes it will also give other errors, no necessarily +CME ERROR: 10 Aug 19 17:24:04 hence the last patch Aug 19 17:24:38 Ok, so his firmware is probably crappier than yours Aug 19 17:24:42 He has the F3507 Aug 19 17:24:57 or i didn't pay attention to the AT debug log Aug 19 17:25:47 And we have no command / unsolicited notification to get the SIM state? Aug 19 17:27:54 unsolicited notification would be ideal, but i can't find any Aug 19 17:29:47 they may have assumed it would always be a pci modem so you have to open the machine to change sim Aug 19 17:31:20 Doubtful since the MD300 / MD400 are the same form factor Aug 19 17:31:57 What about EMRDY? Aug 19 17:32:15 If CPIN fails, can we wait for that? Aug 19 17:34:21 i think we already wait for EMRDY: 1 in the plugin... not sure, it's some complicated logic :) Aug 19 17:35:25 In theory, EMRDY should tell us that the sim and modem are ready Aug 19 17:35:36 MBM refuses to even CFUN=1 until then Aug 19 17:36:04 iirc CFUN=1 works without sim Aug 19 17:36:05 Which could mean that pessi is using an Online/Offline enabled MBM driver Aug 19 17:37:18 Dunno, in the docs it says "The radio and SIM are both turned on" Aug 19 17:37:22 who knows Aug 19 17:38:16 pessi: Proper annotated and the compiler warns you ;) Aug 19 17:38:16 CC src/error.o Aug 19 17:38:17 cc1: warnings being treated as errors Aug 19 17:38:17 src/error.c: In function ‘__connman_error_invalid_property’: Aug 19 17:38:17 src/error.c:184: error: too many arguments for format Aug 19 18:44:17 ofonod[23816]: Pcui:< \r\n^BOOT:30563115,0,0,0,20\r\n Aug 19 18:44:17 ofonod[23816]: Pcui:< \r\n^STIN:0,0,0\r\n Aug 19 18:44:18 ofonod[23816]: Pcui:< \r\n^STIN:99,1,0\r\n Aug 19 18:44:18 ofonod[23816]: Pcui:< \r\n^SIMST:1\r\n Aug 19 18:44:18 ofonod[23816]: Pcui:< \r\n^SRVST:4\r\n Aug 19 18:44:28 kvalo: Are we handling STIN? Do we know what it is for? Aug 19 18:46:47 no mention of STIN anywhere in my docs Aug 19 18:48:41 Sounds like its SIM Toolkit related Aug 19 19:04:48 Do the numbers make any sense to you? Aug 19 19:04:57 I just pushed support for ^USSDMODE. Aug 19 19:05:09 We need to switch it to 0 = text mode. Otherwise it errors out. Aug 19 19:05:24 So I assume 1 (which is default for some reason) means PDU mode. Aug 19 19:05:45 And this one needs the Qualcomm quirk. But this time, the modem doesn't cut it short I get the full message. Aug 19 19:06:05 State: idle Aug 19 19:06:05 USSD: Prepaid-Menü Aug 19 19:06:05 1 Guthabenkonto Aug 19 19:06:05 2 Guthaben-Verfügbarkeit Aug 19 19:06:06 3 Aufladung Guthaben/Pack-to-Go Aug 19 19:06:06 4 Pack Manager Aug 19 19:06:09 7 Tarifinfo Aug 19 19:06:09 8 Hilfe Aug 19 19:06:11 State: user-response Aug 19 19:06:13 Enter response: 1 Aug 19 19:06:15 Ihr Prepaid-Guthaben beträgt 2.35 EUR. Jetzt auch Ihre Guthabenverfügbarkeit abrufen: einfach *102# und die Hörertaste eingeben. Aug 19 19:08:06 denkenz: Works way better than the crappy ZTE hardware. Aug 19 19:08:30 Pretty much reliable actually. Aug 19 19:09:01 what PDU mode are you talking about? Aug 19 19:10:06 'The default configuration value is 1, UE will use the new USSD solution, and UE will transform USSD data transparently' Aug 19 19:10:25 so wtf that means is anyone's guess Aug 19 19:11:08 I just guessed. Neverless it needs to mode=0 to make it work. Otherwise it errors out. Aug 19 19:11:18 Maybe I should read the docs first ;) Aug 19 19:12:23 They probably tried to be clever and decode the USSD data into IRA, instead of transfering the GSM characters Aug 19 19:13:26 Also, I'd really like you to move the CSCS Qualcomm quirk over to the huawei plugin Aug 19 19:13:46 putting it into ussd driver is a bit evil Aug 19 19:15:27 It is only enabled for ZTE and Huawei right now. Aug 19 19:15:35 I have to play with it a bit more. Aug 19 19:15:45 Worst case I write a custom driver for Huawei. Aug 19 19:16:08 Do we have information about the new USSD solution they are talking about? Aug 19 19:16:12 And besides that: Aug 19 19:16:13 AT^STGI=? Aug 19 19:16:13 ^STGI:(0-12,99),(0-8) Aug 19 19:16:13 OK Aug 19 19:16:13 AT^STGR=? Aug 19 19:16:13 ^STGR:(0-12,99),(0-8) Aug 19 19:16:16 OK Aug 19 19:16:22 Could this be seriously STK? Aug 19 19:19:53 All I have on USSDMODE is what I pasted you Aug 19 19:20:05 e.g. there's the 'old way' and there's the 'new way' Aug 19 19:20:12 No info other than that Aug 19 19:20:27 STGI/STGR are not documented, but sounds like some Huawei firmware versions do support STK Aug 19 19:21:10 So I have the gprs API changes completed Aug 19 19:21:13 Wanna play with them? Aug 19 19:21:16 Sure. Aug 19 19:21:36 So essentially the new USSDMODE way is not compatible with the standard. So we have to switch back. Aug 19 19:21:59 Most likely, not enough information Aug 19 19:22:10 As I said, USSD is a disaster area Aug 19 19:22:25 They should enable PDU mode for it and be done with it Aug 19 19:22:28 Like ISI does Aug 19 19:24:22 AT^STGI=0,0 Aug 19 19:24:23 ^STGI:0,4,"534D53205365727669636573" Aug 19 19:24:23 ^STGI:1,4,"496E666F205365727669636573",0 Aug 19 19:24:23 ^STGI:2,4,"577C7274657262756368",0 Aug 19 19:24:23 ^STGI:3,4,"454D61696C",0 Aug 19 19:24:23 ^STGI:4,4,"466178",0 Aug 19 19:24:31 Oh crap. It is menu driven :( Aug 19 19:25:50 Heh, last one is 'Fax' Aug 19 19:25:54 hilarious :) Aug 19 19:27:12 anyway pushed the gprs changes Aug 19 19:27:17 just voicecall left Aug 19 19:27:43 So for ZTE and the newer Huawei, we need STK encoder :( Aug 19 19:29:33 yah, and those drivers get messy fast Aug 19 19:39:11 denkenz: another question about the sim_ready_notify task. does the modem firmware cache all files on the sim, or just specific ones? I am wondering if it's safe to touch any file at all until the modem indicates ready. Aug 19 19:39:46 Totally firmware dependent Aug 19 19:40:16 We do our best, if the firmware is screwy, so be it Aug 19 19:41:22 denkenz: it looks like maybe the thing to do is to delay calling sim_initialize() until the firmware indicates it is ready. Aug 19 19:42:15 that avoids doing any file reads. Aug 19 19:42:24 There are two problems here Aug 19 19:42:41 This task is actually about firmware being ready after the PIN has been entered Aug 19 19:43:03 Because phonebook, imsi, sms, cbs, all are initialized then Aug 19 19:43:28 The SIM files that we read during sim_initialize have to be read even if the PIN is not entered Aug 19 19:43:41 If the modem screws this up, bohoo on them Aug 19 19:44:00 None provide a proper indication of that case Aug 19 19:44:56 ok - so you will assume that it's ok to ready anything prior to entering the pin, and delay anything that gets initialized after that to after firmware says it's done with whatever it is doing. Aug 19 19:45:10 yep Aug 19 19:45:27 are there any modems which exhibit a problem that I can test with? Aug 19 19:45:42 All of them ;) Aug 19 19:45:51 for example, see the MBM quirk in src/sim.c Aug 19 19:46:16 err drivers/atmodem/sim.c Aug 19 19:52:32 kristenc: I was running your icon patches on real SIM cards. So far non of them had any icons on it :( Aug 19 19:54:09 holtmann: bummer. I tested with phonesim, although I had to modify default.xml to get some files over 256 bytes. Aug 19 19:54:46 (and the phonesim patch I sent to the list is required for > 256 byte files) Aug 19 19:56:25 holtmann: So if I load gprs settings and I don't see any contexts, I will automatically add one with "internet" and everything else empty Aug 19 19:56:30 Good with that? Aug 19 19:56:46 Yes. That is what we want. Aug 19 20:15:14 holtmann: Ok pushed that one as well Aug 19 20:15:58 Let me go and test this. Aug 19 20:28:09 ofonod[2663]: Modem: > AT+CGAUTO=0\r Aug 19 20:28:09 ofonod[2663]: Modem: < \r\nERROR\r\n Aug 19 20:28:24 I have this and then my MD-300 stalls at AT+CGATT=1 Aug 19 20:28:30 Never returns with OK or ERROR. Aug 19 20:30:32 ofonod[2663]: Data: < \r\n*ESTKSMS: ,0,0\r\n Aug 19 20:30:35 And then this ??? Aug 19 20:30:41 What is this sim card doing? Aug 19 20:37:08 should I see an org.ofono.RadioSettings interface on every modem? Aug 19 20:37:13 I don't see one with phonesim Aug 19 20:37:16 no Aug 19 20:37:20 irksome Aug 19 20:37:33 So send a patch ;) Aug 19 20:37:33 got dbus-introspection for one? Aug 19 20:38:06 I have phonesim, but need to make a ui control for setting the TechnologyPreference property Aug 19 20:38:45 denkenz: I can has dbus-introspection xml? Aug 19 20:39:05 no such beast, its auto-generated Aug 19 20:39:47 there is such a beast.. just use the qt tools to generate it from an instantiated dbus interface Aug 19 20:41:03 or call org.freedesktop.DBUS.Introspectable::Introspect() on it Aug 19 20:41:28 You assume I have qt4 dev tools ;) Aug 19 20:41:53 I can feed you the appropriate dbus-send command Aug 19 20:42:32 lol, I know how to do it, just messing with ya Aug 19 20:42:51 I would do it, but I have phonesim on one side and an aava on the other Aug 19 20:42:55 neither have a real modem Aug 19 20:43:01 or at least one that works with ofono Aug 19 20:47:09 Not like you can't do this yourself by hand in 3 seconds Aug 19 20:47:28 http://pastebin.ca/1920312 Aug 19 20:47:32 except phonesim doesn't have the ()**)#$)(#()ing interface Aug 19 20:48:06 pastebin.ca is flagged by the intel network as a potentially harmful site Aug 19 20:49:16 yeah, and what pastebin site isn't? Aug 19 20:49:31 I hate computers Aug 19 20:49:43 Fine, I email it to you Aug 19 20:49:54 pastebin.com is A-OK according to IT Aug 19 20:52:55 I saw it.. then it vanished from my inbox... WTF Aug 19 20:54:34 denkenz: can resend to mikeleib@linux.intel.com? I appear to be impaired in every way possible today Aug 19 20:54:45 next time I see you, I beer you Aug 19 20:58:55 denkenz: n/m I have worked around my problem Aug 19 20:59:43 * mikeleib writes qt interface to said dbus interface Aug 19 21:01:22 kristenc: So far no luck with SIM cards with icons. Or your patch is broken ;) Aug 19 21:02:40 denkenz: I think our icon handling API is over-designed. No SIM card is using this. Aug 19 21:02:59 how are you using the API? Aug 19 21:03:05 Are you doing it from the menu? Aug 19 21:03:06 when you guys do development, what modems do you use? Are there usb modems that work well? Aug 19 21:03:32 mikeleib: 'well'? No, good-enough, sure Aug 19 21:03:45 * mikeleib got nothing Aug 19 21:03:56 denkenz: No. Hacking into the source code. Aug 19 21:04:16 * mikeleib wonders if majid has one laying around I could borrow Aug 19 21:04:55 none of the sims i tested have icons either Aug 19 21:05:39 but i need to re-check Aug 19 21:06:25 mikeleib: Get yourself an Option iCon 452 Aug 19 21:06:57 mikeleib: They are readily available on Amazon, also check with majid if he has a Huawei E160G Aug 19 21:07:22 The golden standard are mbm modems, but only the MD300 comes in USB form Aug 19 21:07:29 Rest are PCI Express Aug 19 21:07:47 * mikeleib has a notebook and an aava Aug 19 21:08:00 I do most of my development with the notebook Aug 19 21:08:17 Then check whether you can stick in a Dell 5530 in there Aug 19 21:08:30 was that? Aug 19 21:08:33 Or the 5540 Aug 19 21:08:57 http://accessories.us.dell.com/sna/products/wireless_wifi/productdetail.aspx?c=ca&l=en&s=bsd&cs=cabsdt1&sku=313-9288 Aug 19 21:08:57 the nice part of the usb is that if usb-otg works on the aava, then I can use it there too Aug 19 21:09:00 * balrog-k2n 's recent test modem is an MBM that majid had just "laying around" Aug 19 21:09:16 * mikeleib finds majid Aug 19 21:09:26 He doesn't have any more Aug 19 21:09:31 I need one myself Aug 19 21:09:34 gah Aug 19 21:09:38 * mikeleib aborts finding majid Aug 19 21:09:48 Of the mbm ones anyway Aug 19 21:09:51 He might have others Aug 19 21:10:04 * mikeleib uses email Aug 19 21:11:30 I don't even know if I have free miniPCI slots on my notebook Aug 19 21:12:20 holtmann: Ok walk me through how you do this icon stuff Aug 19 21:12:28 I've a few exotic sim cards Aug 19 21:12:39 luckily majid had it laying around together with a minipci to usb adapter Aug 19 21:12:44 I put a big DBG into the efimg_read function. Aug 19 21:12:58 EFIMG Aug 19 21:12:58 ok 0 length 0 record length 0 Aug 19 21:30:36 So in theory the EFimg is in a different DF Aug 19 21:30:42 we might need extra magic to access it Aug 19 21:41:57 can you tell if it has the DFgraphics ? Aug 19 21:43:04 it is 5F50 subdirectory Aug 19 21:51:10 My F3507 simply gives an error when I try to access the EFimg Aug 19 21:51:21 And gives proper sw1 & sw2 for other files Aug 19 21:53:40 And my Huawei actually accesses EFimg properly Aug 19 21:55:25 holtmann: Can you test on something other than mbm? Aug 19 21:57:18 I did, but none gave me the CRSM for it. Aug 19 21:58:49 You need the quirk Aug 19 21:58:58 QUALCOMM_MSM Aug 19 21:59:04 For MBM? Aug 19 21:59:27 For huawei Aug 19 21:59:37 And get-icon script just returns not implemented for me Aug 19 21:59:38 For CRSM? Aug 19 21:59:42 for the sim atom yes Aug 19 21:59:59 - data->sim = ofono_sim_create(modem, 0, "atmodem", data->pcui); Aug 19 21:59:59 + data->sim = ofono_sim_create(modem, OFONO_VENDOR_QUALCOMM_MSM, Aug 19 22:00:00 + "atmodem", data->pcui); Aug 19 22:00:29 denkenz: get-icon returns not implemented on the Huawei? Aug 19 22:00:35 Yep Aug 19 22:01:03 And I haven't looked at your patch at all Aug 19 22:02:34 ofonod[32091]: Requested file structure differs from SIM: 4f20 Aug 19 22:03:05 You need to update the oFono sim file database Aug 19 22:03:11 similar to how you did the phonesim one Aug 19 22:03:22 hum, why doesn't that break for me? Aug 19 22:04:31 as far as I can tell, ofono doesn't store expected structure. Aug 19 22:04:37 it gets that when it first reads the fileinfo. Aug 19 22:05:07 Not on 3G sims Aug 19 22:05:22 3G sims have a different file info, one that is way too complicated Aug 19 22:05:32 And most modems don't even bother giving access to the necessary files Aug 19 22:05:33 where is the ofono sim file database? Aug 19 22:05:35 so we fudge it Aug 19 22:05:40 src/simutil.c Aug 19 22:06:50 oh, so the reason it works fine for me is because phonesim give the fileinfo, and this real modem does not, so it goes to the database instead. Aug 19 22:06:59 that makes sense. Aug 19 22:10:52 yep Aug 19 22:11:04 I tried making a 'native' 3G sim file info parser Aug 19 22:11:22 but gave up when MBM refused to give me access to the needed sim files with the access descriptors Aug 19 22:13:07 At it looks like on the '3 UK' sim that I have, EFimg isn't linear fixed Aug 19 22:15:37 ofonod[32577]: Pcui:> AT+CRSM=192,20256,0,0,255\r Aug 19 22:15:38 ofonod[32577]: Pcui:< \r\n+CRSM: 144,0,"621F8202412183024F20A5068001719201808A01058B036F060A80020009880108"\r\n\r\nOK\r\n Aug 19 22:19:17 what is it? Aug 19 22:22:36 That's the file info structure according to etsi 102.221 Aug 19 22:24:26 For some reason the code is breaking, so need to figure out why Aug 19 22:25:14 It is reporting the file as transparent Aug 19 22:26:12 byte 4 is the structure - which is 2 in the string above, correct? Aug 19 22:28:00 No, format is different, refer to sim_parse_3g_get_response Aug 19 22:30:15 Hah, this SIM is truly reporting it as transparent Aug 19 22:30:17 Bizarre Aug 19 22:31:09 kristen: FYI Aug 19 22:31:12 Program received signal SIGFPE, Arithmetic exception. Aug 19 22:31:12 q0x0000000000486124 in sim_efimg_read_cb (ok=0, length=0, record=0, data=0x0, Aug 19 22:31:12 record_length=0, userdata=0x6f96b0) at src/sim.c:1071 Aug 19 22:31:12 1071 int num_records = length / record_length; Aug 19 22:31:51 looks like we need to do more sanity checking. Aug 19 22:32:27 you can't assume the data is sane if ok == false Aug 19 22:32:35 do you have another sim which has efimg on it? I wonder if it is widespread to report EFimg as transparent. Aug 19 22:32:53 or if this modem is short-bus special. Aug 19 22:33:09 Lemme try Aug 19 22:33:56 Same thing on my AT&T sim Aug 19 22:34:45 are we just reading an old version of the spec and it is supposed to be transparent? It seems odd that *everyone* would violate the spec like that. Aug 19 22:35:27 Could be Aug 19 22:35:42 Maybe there's a bug in the get_response parser Aug 19 22:35:49 But then again, it works everywhere else Aug 19 22:36:10 let me make sure I've got the most recent spec. Aug 19 22:36:20 I'm using v9.3.0 Aug 19 22:38:44 So far: O2 German, Vodafone Germany have no EFimg Aug 19 22:38:53 hum, yep, that is most recent. And I checked an older version, it has always been linear fixed. Aug 19 22:39:01 AT&T US, 3 UK report as transparent Aug 19 22:39:25 maybe they all have variations of the same firmware. Aug 19 22:39:41 Optus australia reports as transparent Aug 19 22:39:48 Or it might be the Huawei firmware is screwy Aug 19 22:39:59 well - I guess it doesn't matter, clearly we need to support both what the spec says and what reality dictates. Aug 19 22:41:03 Well, when you figure out what reality is, let me know ;) Aug 19 22:41:34 aren't you telling me that reality is that there are modems which report it as transparent? Aug 19 22:42:14 so it seems like we need to request the file info and see what the sim thinks it has, and then change how we read based on that. Aug 19 22:42:28 Nope, I'm just giving you anecdotal evidence of two modems Aug 19 22:42:35 One doesn't recognize there is EFimg at all Aug 19 22:42:42 The other gives something against the spec ;) Aug 19 22:43:00 ok, well then - you'll have to let me know what reality is, since I have zero modems :). Aug 19 22:43:19 Time to hit up majid for some ;) Aug 19 22:43:34 majid just made me give him all the ones I had including the sim card. Aug 19 22:43:47 Hah, then ask for his credit card Aug 19 22:45:30 He should also have at least 3-4 Sierra modems that I ordered Aug 19 22:46:58 I had to give him all mine so he could give them to someone else, so I am thinking he doesn't have any that are not spoken for. but I can check. Aug 19 22:50:39 Shrug, you're one of our core developers Aug 19 22:50:49 If you don't have 5 modems sitting on your desk, he's not doing his job Aug 19 23:08:08 Well that one wasn't happy, crashed my VM :P Aug 19 23:18:04 Funny, so MBM modems don't give me the icons and all Qualcomm based modems return structure as transparent Aug 19 23:18:34 However, the way we do things is definitely correct according to other AT command manuals Aug 19 23:18:55 So either Qualcomm firmware is wrong, or there's a bug in the parse_3g_get_response code Aug 19 23:37:45 * kristenc finds a bug I forgot to commit the fix for Aug 19 23:38:12 but it shouldn't affect your testing as it happens after efimg is read and only during the error path. Aug 19 23:40:24 No, its definitely a firmware bug Aug 19 23:40:42 I tried reading that file as binary and get different results every time I read Aug 19 23:43:40 ATI Aug 19 23:43:40 CS-10 R2.15.0-0634074 Aug 19 23:43:40 CS-10 R2.15 Aug 19 23:43:40 17-12-2009 Aug 19 23:43:42 (c) Nokia Aug 19 23:43:45 denkenz: Maybe this modem is better. Aug 19 23:43:55 Try it Aug 19 23:44:14 *TSTCR Aug 19 23:44:14 *TSTEV Aug 19 23:44:15 *TSTGC Aug 19 23:44:15 *TSTMODE Aug 19 23:44:16 *TSTMS Aug 19 23:44:16 *TSTPD Aug 19 23:44:18 *TSTRT Aug 19 23:44:19 *TSTRF Aug 19 23:44:21 *TSTTONE Aug 19 23:44:25 SIM Toolkit, yeah ;) Aug 19 23:44:34 Looks way too close to Qualcomm stuff :( Aug 19 23:45:00 I bet you all the manufacturers lease the same SIM toolkit firmware Aug 19 23:45:12 Because I've seen the same style everywhere Aug 19 23:45:36 I would assume so. Aug 19 23:45:47 Anyway, have to write the modem plugin for this one first. Aug 19 23:46:13 Nobody is writing original software anymore these days :( Aug 19 23:47:37 What is that modem anyway? Aug 19 23:49:01 And I give up on the icon stuff for now Aug 19 23:49:10 Probably nobody actually tested it Aug 19 23:49:31 However, google definitely tells me we're doing it right Aug 19 23:49:38 kristenc: http://www.wszystko.aplus.pl/io/Remote_SAT_Guide.pdf, grep for 20256 Aug 19 23:51:45 yep, we are right. we probably need to test with a modem we care about and not worry about the broken ones. Aug 19 23:52:29 yeah, so both sets consist of 'all' ;) Aug 19 23:53:28 I wasn't sure I understood your comment about trying to read as binary. Are you saying that when you tried treating it as a transparent file, it didn't work either? Aug 19 23:53:50 yeah, I just did a read_binary on it with length 9 Aug 19 23:54:01 every time I sent the command, it gave me a different content Aug 19 23:54:22 nice. well, that's going to be hard to write a workaround for :). Aug 19 23:55:09 So the only other thing I can think of is that we must manually select the graphics DF first Aug 19 23:57:16 * kristenc goes for family time for a while Aug 20 00:15:03 ofonod[7789]: Control:< \r\n+CSCS: ("GSM","HEX","IRA","PCCP","PCDN","UCS2","8859-1")\r\n\r\nOK\r\n Aug 20 00:29:02 ofonod[9175]: Control:> AT+COPS?\r Aug 20 00:29:02 ofonod[9175]: Control:< \r\n+COPS: 0,2,"26207,0\r\n\r\nOK\r\n Aug 20 00:29:11 denkenz: Unbelievable stupid modem :( Aug 20 00:30:38 How do I quirk that. Beside the open window option. Aug 20 00:34:05 ofonod[9330]: Control:> AT+CMGL=4\r Aug 20 00:34:06 ofonod[9330]: Control:< \r\n+CMGL: 1,3,,76\r\n0001FF008100004ECD72DA5D06B9CBF532885A6697CD6FB7BBDE6E97E5A0F49C0E82C56EB6D9CD868BD970381728394683CCF272BD0C6AA7C768905F5C9683CA697719E40C8FD1F2F4184D7701\r\n Aug 20 00:34:06 ofonod[9330]: Control:< \r\nOK\r\n Aug 20 00:34:10 And what is wrong here? Aug 20 00:42:56 * mikeleib flees the scene Aug 20 01:07:02 denkenz: Most likely you gonna kill me for the netreg hack for the Nokia cards. Aug 20 01:14:14 Entering new phase: 0 Aug 20 01:14:14 ofonod[11052]: Modem:< \r\nNO CARRIER\r\n Aug 20 01:14:17 So Qualcomm based ;) Aug 20 01:19:10 holtmann: COPS part is simply the modem too old to know what the carrier is Aug 20 01:19:15 and the NITZ hasn't come in yet Aug 20 01:19:33 It asks for the numeric one. Aug 20 01:19:50 And it should give us "26207",0 and not omit the ". Aug 20 01:20:11 Ah ok, fair enough Aug 20 01:20:17 My hack is nasty. Aug 20 01:20:34 Anyway, I am getting tired. It is 3am. Aug 20 01:20:45 I hoped the Nokia modem would be better, but it is just pure Qualcomm crap again. Aug 20 01:21:12 I think only STE modems get this right Aug 20 01:21:18 And maybe TI Aug 20 01:21:30 e.g. for 'real' phones Aug 20 01:21:47 So I have PPP part working, but I need to fix the NO CARRIER part like I did for the other ones. Aug 20 01:21:55 Same for ZTE, but not tonight anymore. Aug 20 01:52:55 So I'd like to get a very basic GUI client for Sim Toolkit Aug 20 01:53:14 So we can get Icons and rich text attributes tested Aug 20 01:53:25 Anyone wanna volunteer? Aug 20 01:54:35 me? but i know nothing there Aug 20 01:54:37 hehe Aug 20 01:54:48 Didn't you take Qt training? :) Aug 20 01:54:55 yeah. i have that one Aug 20 01:55:08 Qt is full with C++.;-) Aug 20 01:55:18 Well there's PyQt Aug 20 01:55:34 I wanna see something basic that can render HTML text + images Aug 20 01:55:35 of course i know that one Aug 20 01:56:00 which interface my test app should talk to Aug 20 01:56:10 Just look at test/simple-agent Aug 20 01:56:16 I just want a graphical version of that one Aug 20 01:56:49 i remember andrew draw a html based phonesim...no relationship? Aug 20 01:57:12 forgive my wrong guess Aug 20 01:57:20 will try then Aug 20 01:59:49 Yeah, but that needed a browser and stuff Aug 20 01:59:55 I just want simple GUI Aug 20 02:00:13 i see **** ENDING LOGGING AT Fri Aug 20 02:59:56 2010