**** BEGIN LOGGING AT Fri Feb 20 02:59:58 2009 Feb 20 04:58:04 i take it someone has noticed that the battery meter is all kinds of unpredictable on SHR on the GTA01? Feb 20 04:58:27 oh, does anyone know where a list of placed calls, or received calls would be? Feb 20 05:09:30 Yes, there seems to be some confusion about how the battery should be handled between the GTA01, FSO, and SHR. :( Feb 20 05:11:17 heh Feb 20 05:14:40 ok, ay idea what bug number it is? Feb 20 05:14:56 I doubt there's any bug on it. Feb 20 05:15:10 mind if i file one? Feb 20 05:15:19 heh Feb 20 05:15:20 Go ahead! Feb 20 05:15:39 Just set your expectations carefully. Feb 20 05:16:20 appropriately Feb 20 05:18:09 i expect it not to show charging when unplugged after three hours, and not to constantly show empty. Feb 20 05:18:12 ;) Feb 20 05:18:31 hehe! But that's not the expectations I was refering to, of course. Feb 20 05:18:46 yes :D Feb 20 05:19:49 i Feb 20 05:20:07 think expecting things to work when you're unwilling to file bugs.. well. ;) Feb 20 05:20:29 I don't file bugs. ;) I attempt to fix them, generally. :D Feb 20 05:20:48 juri has a bug fetish ;p Feb 20 05:21:02 I already fixed that one... but it seems that there is some disagreement between the parties about how it should work. Feb 20 05:21:40 And to make matters worse, the battery driver for the gta01 (kernel) was given to some Om dude with the promise that it would all just work... Feb 20 05:21:54 So nobody has any idea what/how it works for 2.6.28. Feb 20 05:22:47 Which is why I'm totally disgusted with that topic, and think we should lock the SHR, FSO, and Om kernel folks in a room together until they agree on how it should work. Feb 20 05:23:27 dave: the last bug i complained about on irc was fixed, after some debugging work and interaction (you were here.). as a result, i don't mind going through the trouble of filing bugs on what i find. Feb 20 05:23:42 you bribe me by fixing bugs, i'll reply by filing them. Feb 20 05:24:16 I know :P Feb 20 05:24:29 we send giftbasket. Feb 20 05:24:36 many thanks, yadda yadda :p Feb 20 05:24:54 juri_: Please, please, fill the bug. Feb 20 05:25:09 #328. Feb 20 05:25:19 juri_: And it is a kernel issue please let balaji know about it... Feb 20 05:25:29 how do i do that? Feb 20 05:27:13 juri_: "Balaji Rao" , or better send an email to the kernel list if you bug do not get noticed... Thanks ;-) Gotta go. BBL. Feb 20 06:07:07 aw Feb 20 06:09:34 im getting rpetty furstrated Feb 20 06:09:45 i have a python error Feb 20 06:10:04 it's a bug in frameworkd Feb 20 06:10:16 anybody around to help me? Feb 20 06:29:20 orzo: a bit more specific? Feb 20 06:29:21 mmm does one need to install MPFR in target system or does gcc automagically deals with the soft-fp? Feb 20 06:38:20 vininim: eabi? Feb 20 06:38:34 yes Feb 20 06:39:01 not sure what MPFR is, I think part of the point of eabi is to make soft float faster Feb 20 06:40:25 yes, eabi makes float operation into soft-fp without context change, but MPRF is a optimized fp implementation Feb 20 06:40:56 just wondering if the cross-compiler built with it will automagically make use or it will supposed a MPFR in the target system Feb 20 06:41:08 *suppose Feb 20 06:41:32 if it's built a dynamic you would have to have the library on the target Feb 20 06:41:36 as Feb 20 06:42:16 you would also have to use a build system in that case I think Feb 20 06:42:18 like OE Feb 20 06:54:46 mrmoko, unicode exceptiosn trying to dump sms Feb 20 06:55:03 the implicit ascii code is really annoying Feb 20 07:03:10 orzo: on ubuntu? Feb 20 07:03:25 or on the device Feb 20 07:03:31 on the device Feb 20 07:03:38 running debian Feb 20 07:03:42 I see those from every python application on ubuntu right now so wondering Feb 20 07:03:45 ah, yeah Feb 20 07:04:09 i managed to patch it Feb 20 07:04:12 it should be on stderr though, how are you trying to dump? Feb 20 07:05:06 where is the link on the hardware fix for the FR for power management? Feb 20 07:05:14 RetrieveMessagebook("all") Feb 20 07:05:35 on org.fresmartphone.GSM.SIM Feb 20 07:05:36 so interactively with python? Feb 20 07:05:48 with python ya Feb 20 07:05:54 you can also use msomething from mickey|zzZZzz Feb 20 07:05:54 interactively? iduno what you mean Feb 20 07:06:03 as in, you type python Feb 20 07:06:07 oh, no Feb 20 07:06:09 then > whatever Feb 20 07:06:13 i huave a little dump-sms script Feb 20 07:06:19 from lindi- Feb 20 07:06:53 what I'm saying is those messages should only be on stderr and shouldn't affect >sms.txt Feb 20 07:07:34 i dont know what you're talking about Feb 20 07:07:41 the messages are exceptions Feb 20 07:07:54 they are being routed to frameworkd.log Feb 20 07:08:18 teh exceptions are interfering with the dbus calls so that the dump-sms script failed to work Feb 20 07:08:39 tmzt: I read the assembly, and it's translated to stuff with 'bl __aeabi_dadd Feb 20 07:08:53 so I think libc is enough Feb 20 07:09:18 orzo: ok, sorry Feb 20 07:13:50 actually libgcc Feb 20 07:15:06 orzo: probably the easiest way is to file a bug at trac.freesmartphone.log and post your log there Feb 20 07:18:33 well it was caused by non ascii phone numbers Feb 20 07:18:43 i added code to convert to utf-8 instead Feb 20 07:18:52 and ended up with symbols _@ and @ Feb 20 07:19:51 they have time stamps but no content Feb 20 07:20:04 im not sure what happened Feb 20 07:20:27 if those are sms messages that failed to store or if they are just blank spots on the sim that RetreiveMessagebook didn't know what to do with Feb 20 07:24:40 orzo: ogsmd debug log will help FSO devs to fix this issues Feb 20 07:25:14 well i'll file a bug eventually Feb 20 07:25:25 im just seeing what i can figure out first Feb 20 07:25:32 moo Feb 20 07:25:47 joerg_42: hey :) Feb 20 07:27:15 helau, alaaf Feb 20 07:27:33 * joerg_42 puts on the red nose ;-) Feb 20 07:27:35 joerg_42: btw, i've tried to use alsamixer yesterday and ended up with looking register numbers and then the driver sources to find which alsamixer control corresponds to what control on that routing diagram. Not exactly intuitive :) Feb 20 07:27:54 lol Feb 20 07:28:05 maximum PITA Feb 20 07:28:09 :D Feb 20 07:29:02 secret tip: use 'amixer controls' plus a few of the other options Feb 20 07:29:26 At least the datasheet and driver sources can be easily understood. Feb 20 07:29:34 you see a complex and a simple name, both associated to control.number Feb 20 07:29:59 easy understand driver sources? Feb 20 07:30:07 * joerg_42 feels stupid Feb 20 07:32:16 At least given enough time, yes, i think this driver is not too hard to understand. Not big, no funky threading, it's a thin interface to hardware, which is described in the datasheet. Feb 20 07:33:09 yup, ack so far. just the data structures are confusing like hell Feb 20 07:34:08 And, btw, the only really not user friendly control i found so far is 'Bass Volume'. It's weird in hardware and exposed as is to the user. Feb 20 07:34:24 mainly due to implicit numbering of array const index Feb 20 07:34:52 (bass) I've read your debate Feb 20 07:35:46 Looks like alsamixer does some weird name mangling on its own :( Feb 20 07:40:35 joerg_42: btw, i haven't found a way to feed both DAIs from the ADC at the same time, looks like it's not exposed in the driver. Feb 20 07:42:37 What's your opinion about recording GSM calls, should it be done? Feb 20 07:42:45 morning Feb 20 07:46:21 Another problem is that changing any controls (e.g. adding two virtual instead of one real for Bass Volume) will probably brake the existing statefiles and we don't have an alternative yet. Feb 20 07:46:36 PaulFertser: should be *possible* at least Feb 20 07:47:03 standard benchmark for new mixer handling concept ;) Feb 20 07:48:02 The problem with recording when using a BT headset is that either we route everything through HCI and it can add latency or we feed all the sound (including recorded from BT mic) back to the BT (and SoC at the same time). Feb 20 07:48:34 my alternative is near to start of first PoC Feb 20 07:49:18 hmm I see gsmBT & rec is a special case, yup Feb 20 07:49:40 err voipBT Feb 20 07:49:50 Are you still going to do that? I haven't seen any public messages that OM-the-company is interested. Feb 20 07:50:24 hmm I have to ask Mickey if he wants me to do it Feb 20 07:50:54 Wolf told it's up to Mickey to give order to me Feb 20 07:51:45 OM itself isn't interested in this kind of middleware any more Feb 20 07:52:18 But they need some middle and high level software to sell GTA03, don't they? Feb 20 07:53:46 yup but that's up to FSO if I got that right Feb 20 07:55:23 Yes, so they give Mickey freedom to decide what's needed and what's not on his own... Funny how Mickey gets so much flexibility while other devs (esp. Raster) were a lot more constrained. Feb 20 07:55:56 hmm, no opinion on that Feb 20 07:56:37 especially while I have my red nose put on Feb 20 07:57:07 No comments required, i just say what i see :) Feb 20 07:57:17 * joerg_42 invites PaulFertser for a last_workday beer Feb 20 08:02:04 PaulFertser: also your RFC suggestion isn't forgotten here Feb 20 08:04:22 if asoc internally is based on a graph, wouldn't it make sense to expose that somehow? Feb 20 08:05:26 tmzt: in my understanding asoc mostly maps registers to controls and sets plls to correspond to the rate. Feb 20 08:06:01 ah, where is the input and output function graph I've seen Feb 20 08:06:05 that must not be asoc then Feb 20 08:06:18 uvcvideo and v4l2 maybe Feb 20 08:06:54 err¿?¿ Feb 20 08:06:56 tmzt: though it'd be cool to visualize all the controls on the routing diagram to aid routing, it's not strictly required Feb 20 08:07:26 have a cloudy idea about that too Feb 20 08:08:13 sth around HTML, maybe flash... Feb 20 08:09:05 or ecmascript Feb 20 08:10:06 I consider that as abusing HTML and/or ecmascript :) Using XULRunner -- maybe... Feb 20 08:10:37 * PaulFertser will refrain from explecitives he thinks of when hearing about "flash" Feb 20 08:11:00 ;D Feb 20 08:11:06 ack Feb 20 08:11:09 tmzt: i think a graphical representation of the codec configuration, perhaps even with the possibility to change things, would be great Feb 20 08:11:32 we still have to deal with the codecs registers though Feb 20 08:11:44 if not as mixer elements then somehow Feb 20 08:11:45 tmzt: similar for the PMU and the current flows Feb 20 08:11:56 wpwrak: won't speed up gta03 ttm Feb 20 08:12:24 joerg_42: duh, maybe, when people can interactively explore the beast Feb 20 08:13:05 joerg_42: and it would be a cool gadget for gta02 as well Feb 20 08:13:11 lol - go ahead suggest it to Wolf. I'm eager to accept that task Feb 20 08:13:53 joerg_42: naw, got other things to sell to him. i was more thinking of enthusiastic volunteers in the community :) Feb 20 08:14:58 * joerg_42 considers looking for a nice cobol job to make his living Feb 20 08:15:48 For practical purposes having a routing diagram with alsa control names printed over it is enough. Feb 20 08:16:44 a pity wasting all that time learning about WM8753 though Feb 20 08:17:50 that's ac97 isn't it? Feb 20 08:18:04 joerg_42: isn't cobol us-only thing? :) I hope there exists a decent job for you in DE without economists around. Feb 20 08:19:13 joerg_42: i don't consider any time i spent learning (if it's not some stupid not-useful-anyway insane twist from microsoft or the like) as wasted. Almost all i learned helped, sooner or later. Feb 20 08:20:35 wpwrak: is that clue about assertion (that should have been obvious to them immediately from the sources anyway) the only reply you got from Atheros? Feb 20 08:21:40 PaulFertser: on that issue, yes. they're also looking into the frequency change problem. they suspect it may be the AP. Feb 20 08:22:49 Hm, not much for more than 2 weeks :( Why big companies are _so_ slow... Feb 20 08:23:05 PaulFertser: ah, and no lean firmware for 6001 planned at the moment Feb 20 08:23:21 PaulFertser: we're not exactly their biggest customer :) Feb 20 08:29:31 wpwrak: if i was an engineer working for atheros, i'd contact you in private and at least make the whole thing more transparent for you. I think any honest man should do that... Feb 20 08:30:08 PaulFertser: :) Feb 20 08:32:55 IMO being kind is much more important than being smart Feb 20 08:33:51 PaulFertser: also their time is limited ... but i've asked them of course if they have any solution or workaround Feb 20 08:35:26 wpwrak: (time limited) yes. Still it doesn't take much time to be a bit more transparent in private, i hope. Feb 20 08:35:40 * PaulFertser has to go. CU Feb 20 10:51:09 orzo: sorry... was busy with daywork. How can I reproduce your problem? Feb 20 11:09:03 win 1 Feb 20 11:25:36 hm tangogps still doesn't show up on the desktop, is it ok, if i fix this? Feb 20 11:26:26 Azog: did not yet come around to fix it... but it is ok if you do it by hand Feb 20 11:26:49 mrmoku: i would just drop a patch in patches/tangogps/ Feb 20 11:29:20 Azog: me too... just did not yet get to it :-) Feb 20 12:05:49 ah, git is available again Feb 20 12:05:53 mrmoku: patch pushed Feb 20 12:10:36 Azog: thanks :-) Feb 20 13:25:09 is it possible to deactivate suspend, when gps is running? Feb 20 13:27:37 heyho Feb 20 13:33:00 [odeviced.idlenotifier] Feb 20 13:33:00 suspend = 0 Feb 20 13:33:07 in /etc/frameworkd.conf Feb 20 13:33:17 (info was for Azog) Feb 20 13:34:27 Zorkman: that will disable suspend completely imo Feb 20 13:44:50 freesmartphone.org: 03jluebbe 07python-helpers * r63e1435165e0 10/pycd/ (TODO pycd.conf pycd.py): pycd: add a python preloaded Feb 20 13:52:58 Azog: you can always manually syspend Feb 20 13:53:10 with the illume config, the shr-settings, or with the powerbutton Feb 20 13:53:16 *suspend Feb 20 13:53:30 ZORKMAN Feb 20 13:54:47 DAVIDOFF MAXIMUS Feb 20 13:55:04 how's it hanging in obama land? Feb 20 13:55:13 did you offer a new deal for FR owners yet? Feb 20 13:55:28 it's pretty awful :D Feb 20 13:55:36 no such new deals Feb 20 13:55:38 mobilize the national guard to help boot hundreds of powerdrained freerunners? Feb 20 13:55:43 hahaha Feb 20 13:55:44 I wish Feb 20 13:55:56 last I recall, bush sent them all to their deaths in iraq Feb 20 13:56:06 or some jihadist country Feb 20 13:56:17 read an article about a cartoon obama <-->ape Feb 20 13:56:20 what a fuss, man.... Feb 20 13:56:29 there were like 10000 cartoons about bush <-->ape Feb 20 13:57:41 http://static.crooksandliars.com/files/uploads/2009/02/delonas_b441a.jpg Feb 20 14:01:35 rape? Feb 20 14:01:51 hahaha obamanal snowman Feb 20 14:02:27 Dave: Greetings, master Dave :) Feb 20 14:03:42 GREETINGS MINIONS Feb 20 14:07:32 piss off Dave Feb 20 14:07:38 just piss right off Feb 20 14:11:45 wow, i'm sensins some hostile vibes Feb 20 15:16:43 freesmartphone.org: 03mickey 07libgsm0710 * r59f32eb03ff7 10/ (ChangeLog configure.ac src/gsm0710.h): Feb 20 15:16:43 freesmartphone.org: rewrite virtual channel v24 status defines (non-code change) Feb 20 15:16:43 freesmartphone.org: release as 1.1.1 Feb 20 15:17:56 freesmartphone.org: 03mickey 07libgsm0710 * r6fdfbc2bb50c 10/src/gsm0710-1.0.vapi: sync Feb 20 15:22:07 Dave: NOT A MINION? Feb 20 15:24:00 * mwester prints a label that reads "MINION" and pastes it onto his forehead. Feb 20 15:27:04 lol Feb 20 15:27:20 * jc_ labels himself "KERNEL TEST DUMMY" Feb 20 16:26:30 freesmartphone.org: 03mickey 07fso-abyss * re31e0105a7da 10/ (ChangeLog configure.ac src/multiplexer.vala src/server.vala): Feb 20 16:26:30 freesmartphone.org: * Added sending v24 data bytes on virtual channel from host -> module. Feb 20 16:26:30 freesmartphone.org: * Changed dbus test command signature to 'ay' (array of bytes), Feb 20 16:26:30 freesmartphone.org: since test commands can embed null-bytes. Feb 20 16:26:58 I'm testing the xf86-video-glamo driver Feb 20 16:27:08 it gives me dlopen: /usr/lib/xorg/modules/drivers//glamo_drv.so: undefined symbol: exaDriverAlloc Feb 20 16:27:14 just letting you know :) Feb 20 16:28:17 jc_: I don't work on it, but which source tree are you using? Feb 20 16:28:27 cjb, git Feb 20 16:28:45 which git repo? Feb 20 16:29:06 git.openmoko.org/xglamo? Feb 20 16:29:25 cjb, git://git.openmoko.org/git/xf86-video-glamo.git Feb 20 16:29:45 cool, okay Feb 20 16:30:10 I'll test it from time to time :) We got ebuild for it in gentoo now, thanks to dagger :) Feb 20 16:30:25 xorg-x11 with fbdev and tslib works perfectly btw Feb 20 16:31:07 although it misses xrandr mode switching for anything besides 480x640 Feb 20 16:32:17 jc can you try adding AccelMethod EXA ? Feb 20 16:32:32 dagger, which section to ? Feb 20 16:32:45 checking Feb 20 16:32:46 the video dev ? Feb 20 16:33:30 card Feb 20 16:33:45 Option "AccelMethod" "EXA" Feb 20 16:35:10 I wonder if you need any extra library for EXA Feb 20 16:35:10 dagger, you mean the same section where Driver "glamo" is ? Feb 20 16:35:20 jc_: yeah Feb 20 16:35:25 ok did that, same problem Feb 20 16:39:04 dagger, I'll try compiling it with -g and debugging it later :) Feb 20 16:39:20 now we at least see it doesn't work 'out of the git' for now... Feb 20 17:16:52 freesmartphone.org: 03mickey 07fso-abyss * r2505336c7886 10/src/multiplexer.vala: send '!SHUTDOWN' message to higher layers and close logical channel, if we got instructed to do so from the modem side Feb 20 17:21:15 reports of massive breakup, and an almost 'ac like humm'. Feb 20 17:21:26 i *really* need to update my gsm firmware. Feb 20 17:22:48 juri_: highly recommended ;-) Feb 20 17:30:25 though probably it won't help for fixing "ac like humm" Feb 20 17:41:17 joerg: one day, i'll actually get documentation on how to do that. i don't have time to sit around and pester 'those who know' at the moment. Feb 20 17:41:58 huh? don't get you Feb 20 17:42:25 know about what? Feb 20 17:42:54 juri_ isn't confident that she can get the firmware flashed on the GTA01 using the instructions on the wiki as they are. Feb 20 17:43:01 doc on updating fw is in wiki Feb 20 17:43:29 juri_: should work Feb 20 17:43:47 mwester did Feb 20 17:44:34 lots of others did on gta02 Feb 20 17:45:20 no reports on severe fails so far Feb 20 17:46:18 um. Feb 20 17:46:22 juri_: don't be scared by the warnings! ;-) I placed them there to protect myself and OM Feb 20 17:46:27 here, let me start, even tho i don't have time. Feb 20 17:46:42 its not the warnings. its the imprecise nature of the instructions. Feb 20 17:46:46 allow me to *QUOTE*. Feb 20 17:47:13 no need to quote, I wrote the quoted text probably Feb 20 17:59:06 * juri_ starts at the beginning. Feb 20 17:59:14 at least someone *fixed* the broken links. Feb 20 18:02:04 what is this 'service data' dialog that keeps popping up in SHR unstable? Feb 20 18:04:51 Grah. Feb 20 18:04:52 ok. Feb 20 18:05:09 first problem: the directions specify for you to install the *GTA02* version of fso! Feb 20 18:06:08 and there is no document on flashing the GTA01. Feb 20 18:06:48 juri_: yup, you should get away by carefully stopping *all* unnecessary background processes on whatever distro U use Feb 20 18:07:11 flashing GTA01 is same procedure as 02 then Feb 20 18:07:19 ok, no how *exactly* do i do that on SHR? Feb 20 18:07:26 oh wait, there is no document on that. Feb 20 18:07:29 ;) Feb 20 18:07:40 ok I see what you mean Feb 20 18:08:29 i'm perfectly willing to update documentation myself, if someone wants to direct. Feb 20 18:14:43 i'm trying to build shr and getting parser errors due to "Could not resolve host name", is this a known issue? Feb 20 18:16:19 vininim: which hostname? Feb 20 18:16:27 (no known issue...) Feb 20 18:17:35 'shr.bearstech.com' and 'svn.openmoko.org' (which actually resolve with ping here...) Feb 20 18:20:09 heh, it was possibly just my connection being lousy Feb 20 18:20:13 :-) Feb 20 18:53:26 and then ftp.rpm.org is down u_u Feb 20 18:55:45 TAsn: went down in the basement today.... found some cable... :-) works great :D Feb 20 18:55:58 mrmoku, hehe I'm glad you like it ;] Feb 20 18:56:11 do you use it with your pointing finger or thumb? Feb 20 18:56:37 TAsn: I wrap it around the pointing finger Feb 20 18:57:02 use it as 'pointing-finger-extension-stylus' :-) Feb 20 18:57:03 mrmoku, and put the end exactly on the tip of the finger? Feb 20 18:57:10 hehe yeah ;] Feb 20 18:57:17 that's the recommended using method ;] Feb 20 18:57:50 and works really nice Feb 20 18:58:49 mrmoku, cool :} I'm glad you like it. I use it a lot, it fits into my wallet and it's very comfortable, that's why I upped the "tutor" Feb 20 18:59:06 and a "real" stylus costs a bunch. Feb 20 18:59:15 and is big and heavy ;) Feb 20 18:59:19 yeah. Feb 20 19:04:18 any glib / serial experts in da house for a small chat? Feb 20 19:05:30 i wonder whether it's really a good idea to use glib mainloop instrument for serial programming. the timing is not really deterministic with just one mainloop. I'm seriously considering using a dedicated thread for that. Feb 20 19:05:36 any opinions there? Feb 20 19:08:07 has anyone a kernel for shr unstable with squahfs support? or can someone give me a reference how to compile a kernel for the shr? Feb 20 19:10:37 new squashfs or old? Feb 20 19:11:09 docscrutenizer: how goes? :) Feb 20 19:11:39 tmzt, well. what do you have? i need it fore evopedia(wikipedia offline), dont really know Feb 20 19:12:02 squashfs 4.0 is upstream now, it's not compatible with older ones Feb 20 19:12:25 my question is basically do you have an existing squashfs you are trying to mount or do you intend to create a new one? Feb 20 19:12:58 tmzt, i have a squashfs file which i need to mount. it contents all this wikipedia articles Feb 20 19:13:17 juri_: Hey, btw, do you respect Harald Welte. Feb 20 19:13:20 I dont really have to mount it, the client does it Feb 20 19:13:21 then you need to find the proper patch and patch the kernel Feb 20 19:13:39 tmzt, i did this already, but how can i compile this kernel now for arm? Feb 20 19:13:56 but i took squashfs 3.4 Feb 20 19:13:58 not sure for OE Feb 20 19:14:15 so i need to bitbake the kernel? Feb 20 19:14:29 you might have to add your patch to the .bb first Feb 20 19:14:47 ok i have no idea about OE : / Feb 20 19:41:50 ophonekitd is segfaulting for me after manually updating libframeworkd-glib and libframeworkd-phonegui{,-efl} Feb 20 19:41:56 what's the prefered way of debugging that? Feb 20 19:51:51 first, install gdb. Feb 20 19:52:42 then create a ticket, run GDB against it, set a breakpoint at abort(), exit(), and _exit() (yes, one of these will have to 'wait' and gdb will warn you). Feb 20 19:53:42 run the program inside of gdb, wait for it to break. when it does, do a backtrace (bt full), and paste the results of all of this into your newly created ticket. Feb 20 19:53:57 paul: never heard of him. ;) Feb 20 20:10:06 juri_: is it a joke or you really don't know about Harald Welte? Feb 20 20:10:54 tilman: there's no need to set a breakpoints if an app is really segfaulting Feb 20 20:11:04 tilman: just start it from gdb and wait until segfault Feb 20 20:11:10 tilman: then do "bt full" Feb 20 20:13:35 juri_: Guess what software he said that about: "It's the only proprietary software that I've been willing to license (and thus pay for) in something like a decade. Feb 20 20:16:13 PaulFertser_: the answer i was after is 'install gdb on the fr' then :) Feb 20 20:16:52 tilman: have you upgraded ophonekitd, btw? Or only libs? Feb 20 20:17:14 i built and upgraded ophonekitd, too, but that didn't help Feb 20 20:17:48 tilman: then yes, install gdb :) Feb 20 20:45:18 freesmartphone.org: 03mickey 07fso-abyss * r46cbb7d677c7 10/src/ (channel.vala multiplexer.vala): Feb 20 20:45:18 freesmartphone.org: stop sending data to a virtual channel when the FC bit is set Feb 20 20:45:18 freesmartphone.org: let's see whether we should Feb 20 20:45:18 freesmartphone.org: a) not read from the input buffer; waiting for FC to be cleared again Feb 20 20:45:18 freesmartphone.org: b) read from the input buffer and throw data away Feb 20 20:45:22 freesmartphone.org: c) read from the input buffer and stuff in a dedicated buffer. Feb 20 20:45:24 freesmartphone.org: First try: a) Feb 20 20:50:52 hm, in today's unstable shr-settings is broken: "ImportError: /usr/lib/python2.6/site-packages/elementary/c_elementary.so: undefined symbol: elm_list_item_show" Feb 20 20:52:49 which file(s) define what gets built with make image in shr? Feb 20 20:57:04 oh, task-shr-feed Feb 20 20:57:07 .bb Feb 20 21:01:29 vininim: look in the images dir Feb 20 21:04:13 may i paste a three line bt in here? Feb 20 21:08:35 :D Feb 20 21:09:36 oh, it seems, that python-elemntary is broken, not shr-settings Feb 20 21:16:00 tilman: three lines should be ok Feb 20 21:16:09 * mrmoku` is interested in ophonekitd segfaults Feb 20 21:17:26 sorry, too late Feb 20 21:17:39 first i installed libx11-dbg, then i updated libx11 Feb 20 21:17:49 the bt still didn't made sense, so i rebooted Feb 20 21:17:53 and now ophonekitd works Feb 20 21:18:00 ahh, ok :-) Feb 20 21:18:09 moku moku moku Feb 20 21:18:26 at least the apps work... if i run ophonekitd in gdb it still segfaults ._o Feb 20 21:18:52 pity, we can't bump e Feb 20 21:18:54 http://61c0ac3626b867ac.paste.se/ Feb 20 21:18:57 mrmoku`: ^ fwiw Feb 20 21:19:11 sliders are no implemented Feb 20 21:19:19 PaulFertser_: ping Feb 20 21:19:24 DocScrutinizer: pong :) Feb 20 21:19:50 Dave: !! sup? Feb 20 21:20:15 Toggles! Feb 20 21:20:21 spaetz: yep, because raster also reverted the 'dict is slow' patch and current e works quite nice Feb 20 21:20:21 Joerg :) Feb 20 21:20:33 Dave: whaazzzzabi? Feb 20 21:20:48 tilman: looks like you would have to rebuild more stuff... Feb 20 21:21:15 spaetz: but I'm looking into the segfault... Feb 20 21:21:33 cool Feb 20 21:21:46 coll will be when I found it ;) Feb 20 21:21:49 -l+o Feb 20 21:23:40 mrmoku`: ththen again... i didn't set $DISPLAY when i ssh'd in. so maybe the connectino to the x server failed as well Feb 20 21:24:07 tilman: uhh, yeah... ophonekitd needs DISPLAY Feb 20 21:24:38 * toggles_w wants to get xmessage working but can't finger out what XAW_CHECK_XPRINT_SUPPORT is all about Feb 20 21:25:06 xaw can be built with optional xprint support Feb 20 21:25:32 that macro probably checks whether your xaw build supports xprnint ;) Feb 20 21:29:31 * Dave boggles toggles Feb 20 21:41:28 tilman: smartarse... ;-) Feb 20 21:42:10 toggles_w: sorry, i didn't know how much you knew about that stuff :P Feb 20 21:44:28 tilman: heh, i was joking, thanks for the info, I know 0 about it... the macro appears to be missing, causing a syntax error, i don't really want to have to patch the configure script but i guess i may have to Feb 20 21:44:40 i suppose i will learn how to add patches to .bb's Feb 20 21:45:51 i wonder whether oe will build the xserver with xprint support for the FR Feb 20 21:46:18 we have no use for it (and it's deprecated, too) Feb 20 21:46:31 ahh.. Feb 20 21:48:04 Pimp My Freerunner! Feb 20 22:12:59 my .emacs file is gone :/ Feb 20 22:13:52 Ainulindale: ping Feb 20 22:25:41 hmm.. xmessage is too much xaw for me, i give up Feb 20 22:43:52 toggles_w: hmm? i have been using xmessage on fr for at least 4 months Feb 20 22:58:14 lindi-: shr? the shr build wants a lot of xaw in makeform.c which doesn't seem to exist Feb 20 23:02:49 hi all Feb 20 23:03:13 anyone around that is familiar with the 2.6.28 kernel changes ? Feb 20 23:09:26 in older kernel /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/chgstate could be used to detect usb cable presence, but find /sys -name chgstate doesn't show anything on 2.6.28 , and I do not see any obviuos candidate for replacing it Feb 20 23:10:44 wtf has usb to deal with i2c? Feb 20 23:11:24 probably was a very generic way to call some PMU state Feb 20 23:11:53 that's how QtExtended detected usb cable presence in the past. if /sys/devices/platform/s3c2440-i2c/i2c-adapter/i2c-0/0-0073/chgstate = enabled , cable connected Feb 20 23:12:21 looks rather odd Feb 20 23:13:06 well, I'm trying to find a 'proper' way to detect usb cable presence, but can't find any decent info on the /sys changes Feb 20 23:13:09 probably that's the reason it vanished Feb 20 23:13:25 joerg_42: i2c-connected PMU should know it's charging mode, that's why it was there probably. Feb 20 23:13:54 that's what i said (or at least meant) Feb 20 23:14:53 joerg_42: you're the hardware expert if i'm not wrong. What would be the recomended way to check if the usb cable is connected ? Feb 20 23:15:24 basically you can't detect USB *cable* reliably Feb 20 23:15:31 FilipLinuxBE: you can judge by usb_curlim if USB is available probably Feb 20 23:15:45 of it's a simple extension cable e.g. Feb 20 23:15:58 (reliable detection of cable) lol Feb 20 23:16:02 s/of/if/ Feb 20 23:16:02 joerg_42 meant: if it's a simple extension cable e.g. Feb 20 23:16:09 and /sys/class/power_supply/usb/online ? Feb 20 23:16:43 would detect power stage, but not a connected device e.g. Feb 20 23:17:37 I'd say you're asking the wrong question Feb 20 23:18:07 probably it's not exactly the cable you are interested in Feb 20 23:19:17 more like "do I get power" or "is a device connected" or "am I hooked up to a host" etc Feb 20 23:19:18 goal: to prevent going into suspend if it is connected with a usb cable Feb 20 23:19:43 duh, connected to *what*? Feb 20 23:20:49 :))) Feb 20 23:21:00 as I told you there's no single bit indicating "connected" Feb 20 23:21:12 QtE powermanagment does it this way , to determine if the USB cable is present Feb 20 23:21:29 :-S Feb 20 23:21:56 it CAN'T!! Feb 20 23:22:44 insert a plain cable and it simply won't detect that. impossible by hw Feb 20 23:22:47 well, it doesn't work in 2.6.28, but did work in pre 2.6.28 Feb 20 23:23:00 FilipLinuxBE: you're not paying attention Feb 20 23:23:20 FilipLinuxBE: plug USB cable in FR. But don't plug the other side anywhere. Feb 20 23:23:34 * joerg_42 away now because of that Feb 20 23:23:50 ah, ok ; sorry, I'm not native english. Feb 20 23:24:02 I mean usb connected to anything else Feb 20 23:24:24 FilipLinuxBE: the same. Connect it to a turned off PC i.e. Feb 20 23:25:05 ok, call it an 'active' other usb device Feb 20 23:25:32 aanything supplying 5V , and potentially being used for networking or other purpose Feb 20 23:26:12 FilipLinuxBE: then your question should be: "how do i know if my FR is powered over USB atm?" Feb 20 23:27:01 yes, correct. sorry for the confusion Feb 20 23:27:07 FilipLinuxBE: i think usb_curlim is the answer Feb 20 23:27:30 and what does /sys/class/power_supply/usb/online does ? Feb 20 23:27:48 I expected it to be useable, but did not react as i expected. Feb 20 23:27:57 I don't know Feb 20 23:28:38 so anything >0 in usb_curlim should mean powered, correct ? Feb 20 23:28:57 root@om-gta02:~# ls /sys/class/power_supply/ Feb 20 23:28:59 bat Feb 20 23:29:00 root@om-gta02:~# Feb 20 23:30:17 DocScrutinizer: yo mean cat /sys/class/power_supply/battery/status ? Feb 20 23:31:09 FilipLinuxBE: no Feb 20 23:31:32 FilipLinuxBE: he meant that there's only bat present in /sys/class/power_supply/ when FR is powered only by bat. Feb 20 23:31:52 that's old kernel though Feb 20 23:32:01 ah, ok Feb 20 23:32:08 but me away now Feb 20 23:32:31 anyone know if /sys is documented somewhere (besides the source code) ? Feb 20 23:35:57 FilipLinuxBE: on the new kernel i still seem to have usb present there even when unplugged. So, don't count on it. Feb 20 23:36:18 I think I'm going to go create a MacGuyver jumpstarter kit for the freerunner :P Feb 20 23:36:21 FilipLinuxBE: no, gta-specific sysfs is only documented by the source itself Feb 20 23:36:49 Dave: what will it include except sticky tape and a switch-army knife? Feb 20 23:37:03 :P Feb 20 23:37:12 I thought I just needed silly putty and a paperclip :p Feb 20 23:38:00 Nah, you better use the tape to attach FR to your laptop, then you'll always have laptop to connect with ssh handy ;) Feb 20 23:38:43 PaulFertser_: yep,I had the same here. Could not find a way to reliably tell if it was powered or not. Feb 20 23:38:53 FilipLinuxBE: but i told you the node Feb 20 23:39:33 didn't work on my kernel Feb 20 23:40:06 is it supposed to go back to 0 when nplugging ? Feb 20 23:40:32 toggles_w: i just used apt-get Feb 20 23:42:26 FilipLinuxBE: it does on mine Feb 20 23:43:48 PaulFertser_: immediately, or have to wait several seconds ? testing again ... Feb 20 23:44:46 FilipLinuxBE: hm, this time it didn't work for me Feb 20 23:45:59 is there any distro already using this kernel ? how do they prevent to resume when powered over usb ? Feb 20 23:46:21 s/resume/suspend Feb 20 23:46:44 FilipLinuxBE: no, looks like usb_curlim is working Feb 20 23:48:21 PaulFertser_: indeed, worked this time for me as well. Feb 20 23:49:11 possible values are: 0 / 100 / 500 ? Feb 20 23:50:05 FilipLinuxBE: i think it depends. If connected to a hub, it can be less than that. Also wall charger can be used that provides 1A. Feb 20 23:50:51 and 1000 it seems Feb 20 23:50:52 FilipLinuxBE: FSO uses a netlink socket to NETLINK_KOBJECT_UEVENT Feb 20 23:50:59 FilipLinuxBE: 1000 == 1A Feb 20 23:51:32 FilipLinuxBE: so, FSO gets notifications about plenty of thing kernel knows about through kobject Feb 20 23:52:00 but 1000 can only be forced, correct ? only 100 and 500 can be negociated ? Feb 20 23:53:55 i prefere sysfs, it is supposed to be a safe interface between kernel and userspace Feb 20 23:55:42 FilipLinuxBE: so is kobject. Feb 20 23:56:05 hmm, judging from the name it will only b valid if there is actully a network running over the usb. allowing it to suspend on a wallcharger Feb 20 23:56:28 but staying alive on usb networking Feb 20 23:57:10 FilipLinuxBE: no, kobject is "network" between kernel and userspace. Feb 20 23:58:29 PaulFertser_: but I mean the NETLINK_ part of the name Feb 20 23:59:05 FilipLinuxBE: yes, it is a special kind of sockets. To communicate with the kernel. Feb 20 23:59:38 ah ok, I see. Feb 21 00:07:53 PaulFertser_: thanks, looks like curlim will do the job. Feb 21 00:08:19 FilipLinuxBE: i don't like qtopia, but hth Feb 21 00:08:29 FilipLinuxBE: i see no reason to use it at all. Feb 21 00:08:46 FilipLinuxBE: if i needed a phone, i'd buy a cheap 2nd hand one. Feb 21 00:08:49 I'm running qtopia on top of debian (H1) . Feb 21 00:09:23 to me it is still the best environment for having a phone, while i can fall-back on deb for all the other stuff Feb 21 00:09:45 and I can always launch matchbox without having to reboot Feb 21 00:10:22 FilipLinuxBE: why do you prefer H1 to Debian with FSO? Feb 21 00:10:52 notreally prefer, it was just the quickest way to get a micro-sd running with a debian base Feb 21 00:11:47 FilipLinuxBE: probably you'll like FSO more, as it's easy to write your own programs that use it. Even shell-scripts. Or python. Works quite stable. Feb 21 00:12:28 is FSO already using 2.6.28 ? Feb 21 00:12:56 Sure Feb 21 00:13:13 For like 2 months at least Feb 21 00:14:44 i've got some spare µ-SD's , might give it a try. Feb 21 00:18:29 anyone knoe where I can get a gpe-edit package for shr? angstrons is all ive found and thats armv5t Feb 21 00:18:44 otherwise its vi for the shopping list :) Feb 21 00:38:41 freesmartphone.org: 03mickey 07framework * r13a3069efe96 10/framework/subsystems/oeventsd/fso_actions.py: oeventsd: use QueuedDBusAction only for triggers that really need it; RingTone for now Feb 21 00:38:42 freesmartphone.org: 03mickey 07framework * rf6ca17af5e34 10/framework/subsystems/ogsmd/modems/cinterion_mc75/ (modem.py unsolicited.py): ogsmd: [CINTERON MC75] catch up with changes in ogsmd Feb 21 00:39:22 freesmartphone.org: 03mickey 07framework * rb652f9cc4efb 10/ (docs/ogsmd/lessons tools/cli-framework): tools: add cell broadcast to cli-framework Feb 21 01:00:18 according to lsof, python is listening on the ntp socket (on SHR) Feb 21 01:00:29 what app could be doing this? Feb 21 01:00:36 frameworkd Feb 21 01:00:40 prevents running ntp, ntpdate etc Feb 21 01:00:41 more concrete Feb 21 01:00:44 the ogsmd subsystem Feb 21 01:00:47 err Feb 21 01:00:48 otimed Feb 21 01:00:58 otimed is aggregating time sources Feb 21 01:01:03 ideally you don't need ntp any more Feb 21 01:01:12 if there's something not working, please open a bug Feb 21 01:01:17 on 2008.12 I got a fair bit Feb 21 01:01:26 of time drift which I solved by Feb 21 01:01:33 btp over wifi Feb 21 01:01:42 here its drifting off and Feb 21 01:01:48 I cant correct it Feb 21 01:02:01 is otimed supposed to sync to an outside Feb 21 01:02:01 that's because we don't have networking done yet Feb 21 01:02:03 source? Feb 21 01:02:08 yes Feb 21 01:02:13 it will try to sync with an ntp server Feb 21 01:02:20 regularly Feb 21 01:02:21 billk: you can selectively disable various ntp time sources Feb 21 01:02:49 ok, I'll look into it. Feb 21 01:03:22 PaulFertser_: by the way, Ive been able to get a working gsm0710muxd on SHR Feb 21 01:03:24 billk: timesources in [otimed] Feb 21 01:03:36 billk: what did you do for that? Feb 21 01:03:44 by basicly ripping out the internal initialisation Feb 21 01:04:07 and doing it externally (serial port power on/off and reset) Feb 21 01:04:10 Hm, interesting. Feb 21 01:04:23 there are two lines in the mux code Feb 21 01:04:34 that read the port setiings Feb 21 01:04:41 and write them straight back Feb 21 01:04:49 they were key - should Feb 21 01:04:56 not have had that effect. Feb 21 01:05:09 need to look further at why of course Feb 21 01:05:22 in the meantime, its the same Feb 21 01:05:28 reliability as 2008.12 Feb 21 01:05:49 billk: do you enjoy SHR? ;) Feb 21 01:06:24 yes, always liked it but could change permanently until I ot a working phone Feb 21 01:06:46 moko11 beta firmware - no lost SMS messages anymore - yippee!!!! Feb 21 01:06:51 :) Feb 21 01:07:00 Now I am going through the pin of getting Feb 21 01:07:06 it to work the way I want Feb 21 01:07:19 and tripping over things like ntp :) Feb 21 01:07:47 s/pin/pain Feb 21 01:09:26 anyone tried tremor decoding on freerunner? Feb 21 01:10:07 i guess we do, yeah Feb 21 01:10:14 iirc FSO ships tremor Feb 21 01:10:18 as is SHR Feb 21 01:11:02 vininim: i did, thanks to lindi- Feb 21 01:11:14 through mplayer? Feb 21 01:12:21 vininim: or no, i didn't. Only lindi- did. Using example program from the lib sources. Feb 21 01:13:05 oh, just noticed it being discussed today on the list heh Feb 21 01:14:36 hmm Feb 21 01:14:44 who wanted to work on that gsm 07.10 simulator? Feb 21 01:14:47 Hi mickey :) Feb 21 01:14:49 i need one... Feb 21 01:14:50 yo dave Feb 21 01:15:51 there's a gsm simulator!? Feb 21 01:17:45 no, someone wanted to work on 07.10 (multiplexing protocol) Feb 21 01:17:55 to create a dummy modem Feb 21 01:17:55 sweet! Feb 21 01:18:01 that can speak muxing protocol Feb 21 01:18:03 I hope they know what they're getting into! :P Feb 21 01:18:04 would ease our tests Feb 21 01:18:17 Significantly! Feb 21 01:18:22 mickeyl: it was lindi- :) Feb 21 01:20:53 righto Feb 21 01:21:19 Aha! Feb 21 01:21:22 Oh, Lindi :P Feb 21 01:21:34 Such a nice guy... too interested in debuggery Feb 21 01:21:38 hehe Feb 21 01:21:45 ~praise lindi Feb 21 01:21:50 In automated testing i'd say Feb 21 01:22:03 hmm Feb 21 01:22:09 i don't want to work with qtopiasim Feb 21 01:22:15 but it's said it can do it Feb 21 01:23:30 my issue from last night was because unicode delta was occuring in my number field in ssms Feb 21 01:23:33 sms Feb 21 01:24:00 Hey orzo Feb 21 01:24:04 i had _@ and /\@ in my last two smses Feb 21 01:24:06 hi dave Feb 21 01:24:15 where /\ is intended to represent delta Feb 21 01:24:27 o.O Feb 21 01:24:32 those must be some special place holders Feb 21 01:24:44 they occur in the origin field, as the phone number Feb 21 01:24:44 ooOoooOooOooh Feb 21 01:24:49 wtf :D Feb 21 01:26:51 unicode u"\u0394" Feb 21 01:28:50 http://rafb.net/p/TNBlXs54.html Feb 21 01:29:07 that's what i used to make my copy of frameworkd work Feb 21 01:29:43 it allows PDUAddress to have a unicode encoded string in the number feild Feb 21 01:30:15 ok fellas. Seems I'm entitled to continue on development of a decent alternative to statefiles and relatives Feb 21 01:30:27 \o/ Feb 21 01:30:56 what do you have in mind Feb 21 01:31:11 awesoe Feb 21 01:31:15 *awesome Feb 21 01:31:26 Will Joerg be our salvation? :) Feb 21 01:31:26 for us, don't know if it's for you ;) Feb 21 01:31:45 an entirely different concept avoiding some of the nightmares of alsa.state Feb 21 01:32:07 orzo: can you please attach this to a bugtracker including the said PDU? Feb 21 01:33:45 Joerg, Yes! It sounds a great alternative Feb 21 01:35:11 how do i format the pdu Feb 21 01:35:30 i have a dump-sms script that works after my patch Feb 21 01:35:50 i can paste the source and the output Feb 21 01:35:54 well, after wrapping your head around it for >2months it's not that scary anymore and idea seems to reach state where I actaully could write down some specs and implement a proof of concept Feb 21 01:37:06 Wow Feb 21 01:37:13 That sounds great, Joerg! Feb 21 01:37:38 decent resource management instead of brute force reconfiguring the whole mixer Feb 21 01:37:56 orzo: run mickeyterm and get the PDU with AT+CMGL= err Feb 21 01:38:02 AT+CMGR Feb 21 01:38:22 :) Feb 21 01:38:23 we need the PDU in encoded format so we can analyse it correctly Feb 21 01:38:32 separation of "code" and variables like volume settings Feb 21 01:38:38 THat sounds like a logical plan :) Feb 21 01:40:33 at least a "echo 110 >speakervolume" sounds better to me than a "vi gsmspeakerphone.state" Feb 21 01:41:12 :) Feb 21 01:44:37 mickeyl: hanks :-) Feb 21 01:44:42 Thanks Feb 21 01:44:51 (damn keys) Feb 21 01:44:56 welcome on board Feb 21 01:44:58 (again) Feb 21 01:45:01 lol Feb 21 01:46:02 haha Feb 21 01:53:35 how do i make a new ticket on trac? Feb 21 02:13:19 there you are, mickeyl, bug 362 Feb 21 02:13:29 http://trac.freesmartphone.org/ticket/362 Feb 21 02:13:54 i certainly hope i put the right SMS pdu output in there and not something embarassing Feb 21 02:14:06 hehe Feb 21 02:14:09 excellent, thanks Feb 21 02:14:26 g'night **** ENDING LOGGING AT Sat Feb 21 02:59:57 2009