**** BEGIN LOGGING AT Tue Jan 27 05:11:48 2009 **** BEGIN LOGGING AT Tue Jan 27 05:27:26 2009 **** BEGIN LOGGING AT Tue Jan 27 05:37:20 2009 Jan 27 06:12:03 i /thnik/ my dbus signals are going out, but my signal callbacks are seemingly never triggered Jan 27 06:23:04 AntonTakk, us mdbus -l to figure out if they are really going out Jan 27 06:23:24 thanks Jan 27 06:24:50 playya: it keeps crashing giving: ImportError: No module named gobject Jan 27 06:25:17 then python-gobject might be missing Jan 27 06:27:15 Failed to execute dbus-launch to autolaunch D-Bus session Jan 27 07:37:50 morning Jan 27 07:59:59 * spaetz looks for python-edbus examples that could help making use of asynchronous dbus calls in shr-setting Jan 27 08:00:56 spaetz: used edbus from a c app? Jan 27 08:04:11 <\marco> hi to all Jan 27 08:04:51 hi Jan 27 08:05:55 <\marco> what's happened to accelerometers? is the new framework? Jan 27 08:06:15 <\marco> apps like moko-eightball isn't working anymore.. Jan 27 08:06:23 <\marco> ( I'm using shr-unstable ) Jan 27 08:07:27 i did'nt think accelerometer support was part of the framework yet? Jan 27 08:08:30 <\marco> mhm.. I think you are right :).. but.. it's strange.. Jan 27 08:09:41 <\marco> due to last updates ( a lot of things ) aren't working anymore.. Jan 27 08:12:55 <\marco> I've tryied the example script in python on http://wiki.openmoko.org/wiki/Accelerometer_data_retrieval and they seem to work.. Jan 27 08:13:52 <\marco> ok.. Jan 27 08:14:35 <\marco> I seem to have found the problem Jan 27 08:14:52 <\marco> the new organization in /sys Jan 27 08:15:33 <\marco> there isn't a /sys/devices/platform/lis302dl.1/ anymore Jan 27 08:16:32 <\marco> now it is under /sys/class/i2c-adapter/i2c-0/0-0073/lis302dl.1/ Jan 27 08:18:42 what is /sys/class/i2c-adapter/i2c-0 a symlink to? Jan 27 08:22:23 <\marco> I'm looking inside moko-eightball source and.. no.. new /sys organization isn't the problem Jan 27 08:22:39 <\marco> moko-eightball use /dev/input/event3 Jan 27 08:22:51 <\marco> directly Jan 27 08:23:57 raster: did you see the "fix" for e consuming 30% cpu (in the current srcrev we are using) Jan 27 08:24:57 <\marco> I've to go Jan 27 08:25:12 <\marco> bye! :) Jan 27 08:45:48 mw|zZzZ: no? Jan 27 08:47:41 grrr, seems to be sending dbus stuff ok, but it's not triggering network registration like i think it should Jan 27 08:48:44 Anyone knows if the command 'dd' is also available for windows? :/ Jan 27 08:48:59 er Jan 27 08:49:05 maybe wih cygwin Jan 27 08:51:08 AntonTakk1: Thought about that as well, but I dunno if it would support block devices. The problem is I need to make a 1:1 copy of my SDHC card to another, bigger one, and as my card reader does not support SDHC I though I'd maybe use the EEEPC (windows version) Jan 27 08:54:10 as i understand it, this should be enough to trigger a gsm network registration where no pin is required: http://pastebin.com/m571db1bb or do i have it wrong? Jan 27 08:54:34 beniwtv: http://www.chrysocome.net/dd Jan 27 08:55:47 i'm pretty sure the first two dbus messages get sent, since when i run the app with dbus-monitor --system running it shows two messages every time the app runs Jan 27 08:55:57 PaulFertser: Seems exactly what I need, thanks :) Jan 27 08:56:35 beniwtv: don't forget to set bs= to something big enough or it will be slow :) Jan 27 08:57:25 PaulFertser: does it really matter with SD card? Jan 27 08:58:13 lindi-: i think it does. At least it matters with hard drives, why shouldn't it matter for SD? :) Jan 27 08:59:02 PaulFertser: they are just much slower Jan 27 09:02:21 lindi-: i'm still not convinced. ;) Let's ask beniwtv to perform some real-life testing and compare bs=1 with bs=1M. Jan 27 09:02:34 raster, i sometimes see enlightenment using ~20% cpu continously on the gta02 Jan 27 09:02:50 PaulFertser: bs=1 probably is slow yes Jan 27 09:02:58 PaulFertser: but that surely won't ever be default? Jan 27 09:03:01 it is reading the same data from the x server socket over and over again, it seems Jan 27 09:03:27 any idea what this could be? Jan 27 09:03:59 shoragan: xtrace it? Jan 27 09:04:14 oh, didn't know about xtrace Jan 27 09:04:48 PaulFertser: Will do, once I'm able to find the charger for it :P Jan 27 09:04:54 shoragan: ask mw|zZzZ Jan 27 09:04:58 i have never seen it Jan 27 09:05:31 beniwtv: or better, compare bs=1M to the default (no bs= parameter, i think it's 512 bytes). Jan 27 09:05:38 raster, ok, and i'll try to get some more details Jan 27 09:09:08 raster, also where do i configure what is called by the power button menu? Jan 27 09:09:35 hmmm Jan 27 09:09:45 that... u need to fiddle with xfg files by hand with eet Jan 27 09:09:49 no gui for it Jan 27 09:09:51 or ipc Jan 27 09:10:08 and even then.. u'd better configure it before egets going (ie while its not runing) Jan 27 09:10:12 or e will overwrite it Jan 27 09:10:16 what is it using for suspend and shutdown right now? Jan 27 09:10:29 e's sysactions Jan 27 09:11:04 i'm asking because for fso it should use the dbus calls Jan 27 09:11:10 check /etc/enlightenment/sysactions.conf Jan 27 09:11:16 its pretty self-explanatory Jan 27 09:11:19 and documented Jan 27 09:11:26 ok Jan 27 09:11:49 make the commands dbus-send calls then Jan 27 09:12:38 oh, very nice! thanks :) Jan 27 09:13:08 ....ok, now i really feel dumb Jan 27 09:13:40 been wondering all this time why i couldn't communicate with fso over dbus, the framework isn't installed :) Jan 27 09:13:52 HAHAHAHAHHAHA Jan 27 09:14:14 :) Jan 27 09:18:06 still not working thou Jan 27 09:22:51 AntonTakk, on system or session bus? Jan 27 09:23:48 system Jan 27 09:24:25 is the daemon already running Jan 27 09:24:34 it is now Jan 27 09:25:43 playya: i pastebinned my code earlier, if that would help? Jan 27 09:27:49 ehh you have a function and a var with the same name Jan 27 09:30:01 @ AntonTakk Jan 27 09:30:23 looked like that was what i needed from the code i was looking at Jan 27 09:33:06 cleared those up Jan 27 09:33:22 doesn't look like it is sending the dbus messages Jan 27 09:35:32 fso.Usage doesn't have a Status signal Jan 27 09:36:30 then that should simply never be triggered Jan 27 09:36:33 right? Jan 27 09:36:55 yes :) Jan 27 09:37:39 all i'm trying to do right now is prove i can talk to fso by making it try ro register Jan 27 09:38:07 shouldn't sending those three messages handle that? Jan 27 09:43:04 I'm not really familiar with edbus. but if the fact that E_DBus_Signal_Cb and the function have the same name might be a problem Jan 27 09:44:04 gcc didn't seem to care Jan 27 09:44:15 but i corrected that anyways Jan 27 09:53:57 PaulFertser, lindi-: bs=1M is waaaay quicker than the default.... I had to abort the default as it was taking ages Jan 27 09:54:39 beniwtv: what's the default? Jan 27 09:55:36 lindi-: 512 Jan 27 09:56:08 beniwtv: odd, that's on windows? Jan 27 09:56:20 lindi-: yep Jan 27 09:59:35 I told ya :p Jan 27 09:59:48 4 minutes for 2GB... not bad Jan 27 10:00:05 but the SDHC card has gotten a bit hot :P Jan 27 10:04:12 woot! got the gsm resource requested :) Jan 27 10:15:29 ok, think i almost got it, except that my test app acts like it has a maid, so it seems that limits it to one run, then reboot :-\ Jan 27 10:57:08 hi all, a while ago a bug in gsm0710muxd where a 4096 bytees buffer has silent overflows. i can;t seem to find that article again, or who is working on it. do one of you know? Jan 27 10:58:09 edje2, that is fixed in current gsm0710muxd Jan 27 10:58:23 (i think) Jan 27 10:59:16 brilliant, it did bother me quiet a bit. i use 2008.12 any pointer to where i can find the brand new gsm0710muxd ? Jan 27 11:02:44 2008.12 uses the qtopia muxer, i think Jan 27 11:03:08 otherwise OE builds the fixed version Jan 27 11:04:26 i'll check, thanks for the pointers Jan 27 11:25:34 helo Jan 27 11:42:33 yo Jan 27 11:51:12 freesmartphone.org: 03jluebbe 07framework * r027fd5d65d98 10/framework/subsystems/oeventsd/fso_actions.py: Jan 27 11:51:12 freesmartphone.org: oeventsd: support setting environment variables for the Command action (closes: #242) Jan 27 11:51:12 freesmartphone.org: Thanks to Peter Strapp for the idea and a patch. Jan 27 11:51:13 freesmartphone.org: 03jluebbe 07framework * reb4b6b398307 10/framework/subsystems/ousaged/generic.py: ousaged: add some more debug messages Jan 27 11:52:31 freesmartphone.org: 03mickey 07framework * r70a3c2b1128e 10/framework/subsystems/ogpsd/gpsdevice.py: Jan 27 11:52:31 freesmartphone.org: ogpsd: Check in all resource functions if a channel is None and warn Jan 27 11:52:31 freesmartphone.org: Under certain circumstances, i.e. with testdevices we don't have a Jan 27 11:52:31 freesmartphone.org: channel. Also remove stray shutdownChannel from commit Jan 27 11:52:31 freesmartphone.org: 0be860bbc94f497a877555fbb4c9c95107634ef3 Jan 27 12:01:57 any one knows how to fribid and / or pango works with e ? Jan 27 12:02:09 how to make Jan 27 12:02:51 no one ? Jan 27 12:02:59 any info or direction Jan 27 12:06:03 arhrhhh Jan 27 12:06:07 i get with SHR unstable Jan 27 12:06:19 An error ocurred, return value: -50512. Jan 27 12:06:31 when I want to install the locale :( Jan 27 12:07:14 libs/pbd/fpu.cc:35: Error: suffix or operands invalid for `push' <= ardour :-/ Jan 27 12:07:35 bad thing :) Jan 27 12:07:49 are you using the dev branch or the stable? Jan 27 12:07:57 i wish the stable... :) Jan 27 12:08:39 however i get -50512 when I want to install the locale Jan 27 12:15:48 Hire: 2.7.1 Jan 27 12:16:30 werid Jan 27 12:16:34 *weird Jan 27 12:16:46 here with debian I don't have problem Jan 27 12:17:32 it's at compilation Jan 27 12:17:45 sure :) Jan 27 12:18:01 i know it Jan 27 12:18:39 http://www.nabble.com/Compile-Ardour-2.7.1-on-Debian-sid-td21057075.html Jan 27 12:18:43 see here Jan 27 12:22:52 already saw it already tried it Jan 27 12:23:17 uh ... Jan 27 12:29:21 Heh! Managed to trick it :-) Jan 27 12:30:06 how? Jan 27 12:30:31 ( rebuild the package index :( ) Jan 27 12:35:13 Done Jan 27 12:35:29 I modified the SConstruct thingy to force the use of X86_64 ASM Jan 27 12:40:31 anyone knows why alarm app keeps going ? it wakes up the machine and rings/beeps every ~10 min =] Jan 27 12:40:53 it just refuses to accept that I'm awake :( Jan 27 13:12:48 Mononoke_: this is an innovative new approach. Our atempts to build a self-destruction system by exploiting battery LiIon properties failed completely, so... ;-) Jan 27 13:15:47 ("SHUT UP!!!"... baNNGgg %@) Jan 27 13:17:20 DocScrutinizer: o/ Jan 27 13:17:24 How's life ? Jan 27 13:18:06 Mononoke_: also - honestly - there have been considerations to detect awake-status of user by analyzing G-meters for some minutes of carrying around etc Jan 27 13:18:29 DocScrutinizer: detect awake-status or wanker-status? Jan 27 13:19:02 Ainulindale: hi! :-) Well how should I put it... Maybe: "in a hurry..." Jan 27 13:19:18 (wanker) lool XD Jan 27 13:19:33 DocScrutinizer: any good news recently? Jan 27 13:19:42 nope Jan 27 13:21:52 bbl, cya Jan 27 14:00:22 freesmartphone.org: 03mickey 07specs * re873af6cb4e4 10/ (3 files in 2 dirs): otime: document org.freesmartphone.Time.Alarm Jan 27 14:01:43 freesmartphone.org: 03mickey 07specs * ra1317f7256f4 10/ (2 files in 2 dirs): otime: add html and xml files Jan 27 14:21:28 freesmartphone.org: 03mickey 07specs * r833a43d963aa 10/otime/org.freesmartphone.Time.Alarm.xml.in: otime: clarify that we're limited to dbus system services Jan 27 14:22:22 freesmartphone.org: 03mickey 07specs * r283d50f0770b 10/html/org.freesmartphone.Time.Alarm.html: otime: regen html docs Jan 27 14:36:25 argh the locales doesn't work Jan 27 14:36:30 ( SHR unstable ) Jan 27 14:44:31 yeah Jan 27 14:44:45 hell yeah Jan 27 14:46:59 yes and I am getting mad with that :| Jan 27 14:57:39 hehe Jan 27 15:18:53 Hire, would you know if there's an easy way to migrate cubase files to ardour? Jan 27 15:22:33 Ainulindale: did you read my messages yesterday? accelges works on testing now, should work on unstable also. but its a bit slow at the beginning Jan 27 15:26:56 Nice :-) Jan 27 15:48:56 хфбфе Jan 27 15:51:57 haha :) my yphonekitd now work on sqlite not on sim :) full data migradion on on button :) Jan 27 15:59:14 freesmartphone.org: 03jluebbe 07framework * r0583d9be9119 10/framework/subsystems/ogpsd/om.py: ogpsd: reenable workaround for the kernel that will be release with MS5 Jan 27 16:26:45 Does anybody know, on what events GSM is supposed to issue an IRQ? Jan 27 16:27:36 iirc the IRQ is sent if RING gets high Jan 27 16:27:42 but check the kernel sources to be sure Jan 27 16:27:54 s/gets/goes/ Jan 27 16:28:35 mickeyl: the question is: when is it gets high? I can check the kernel sources but i don't know how the modem is supposed to behave. Jan 27 16:28:54 ah Jan 27 16:29:04 usually RING gets sent on every unsolicited answer Jan 27 16:29:07 err Jan 27 16:29:09 raised Jan 27 16:29:40 mickeyl: moreover i'm afraid i can't monitor that line, because it seems to be somewhere deep and goes from one BGA (s3c) to another (calypso). Jan 27 16:30:09 what do you want to do? Jan 27 16:30:10 The thing is: i very much suspect that sometimes S3C doesn't wake up when it should due to incoming call/message. Jan 27 16:30:14 ah Jan 27 16:30:52 you should ask laf0rge on that Jan 27 16:30:56 he's most familiar with that circuit Jan 27 16:31:29 I think there's no circuit, just a trace from one BGA to another. Jan 27 16:31:48 well whatever :) Jan 27 16:31:55 <-- softie Jan 27 16:32:14 bear with me imprecisely describing hardware Jan 27 16:32:16 :D Jan 27 16:33:30 If it was somehow possible to connect to it, i could assert (or raise) it externally many times to check that it works good. Or if i could give calypso a command to order to raise it in 5 sec, i.e., that'd be cool for automated testing. Jan 27 16:34:42 So, it seems that either the IRQ pin is sometimes improperly configured by the kernel, or calypso sometimes doesn't do its job. Jan 27 16:35:37 But it resumes every time i use power button or insert jack, i can't see how modem irq line can be improperly configured while others work ok. Jan 27 16:38:34 DocScrutinizer: any thoughts on wake-by-modem IRQ not working sometimes? Jan 27 16:39:59 i'm afraid i don't see another way than to call it or send sms Jan 27 16:40:07 unfortunately +CALA doesn't seem to be implemented Jan 27 16:41:39 hi padowi Jan 27 16:41:43 Even if i confirm that one in 50 calls, e.g., doesn't result in resume, what will it prove?.. Jan 27 16:44:50 mirko-paroli: hiya Jan 27 16:46:05 mickeyl: do you want an ogsmd debug log where ogsmd detects recamping bug but never sends SLEEP=2? Jan 27 17:01:45 PaulFertser_: sure Jan 27 17:01:59 which ti_calypso_deep_sleep setting? Jan 27 17:06:29 mickeyl: adaptive. Here's the log: http://pastebin.com/m7d8b835c (no MiscChannel transfers after 16:51:24) version is from Jan 23. Jan 27 17:07:53 d'oh Jan 27 17:07:59 it say it turns it off, but doesn't Jan 27 17:08:02 lazy ogsmd Jan 27 17:08:07 let me take a quick look Jan 27 17:08:58 hmm Jan 27 17:09:02 code looks good Jan 27 17:09:06 the log is too short Jan 27 17:09:11 MiscChannel is stuck. Jan 27 17:09:34 does it claim a timeout later? Jan 27 17:09:38 or before? Jan 27 17:09:47 oh it does Jan 27 17:09:48 first line Jan 27 17:09:56 TIMEOUT 'AT+CMGW=34' => ??? Jan 27 17:10:07 well, sleep logic is ok then Jan 27 17:10:20 but for some reason the parser got stuck for the misc channel Jan 27 17:10:27 that could be due to invalid data read Jan 27 17:10:32 (muxer bug) Jan 27 17:10:40 i have yet to apply one safety measure Jan 27 17:10:46 which would be resetting the parser state on timeout Jan 27 17:10:49 i plan to do this before ms5 Jan 27 17:11:02 that won't fix the underlying issue Jan 27 17:11:11 but enable to recover Jan 27 17:11:15 Cool, at least channels won't be "stuck". Jan 27 17:11:36 yep Jan 27 17:13:05 And there was no invalid data from Misc before it stuck. COMPLETED 'AT+CMGL=2' => ['+CMS ERROR: 321'], then _readyToSend: watch timeout = None, then sending 'AT+CMGW=34\r' and then TIMEOUT Jan 27 17:13:28 I should give you a log :) Jan 27 17:15:38 yeah Jan 27 17:15:44 mickeyl: http://pastebin.com/m779551db (MiscChannel-related messages) Jan 27 17:16:17 hmm Jan 27 17:16:23 that's fishy Jan 27 17:16:31 2009.01.27 16:51:14 ogsmd.channel DEBUG : sending 11 bytes: 'AT+CMGW=34\r' Jan 27 17:16:34 there's a \n missing Jan 27 17:17:04 how the hell can that happen? Jan 27 17:17:11 did you use DebugCommand? Jan 27 17:17:20 How the hell i didn't noticed that myself! :) Jan 27 17:17:25 hehe Jan 27 17:17:28 Nope, i didn't use any Debug* Jan 27 17:17:35 i become used to reading these logs now Jan 27 17:18:19 so somewhere a character has been swallowed :/ Jan 27 17:18:32 That's just as i spot wrong placed semicolons in my freshman students' programs. They just don't notice when ; becomes empty operator but it's so obvious to me :) Jan 27 17:19:41 hehe Jan 27 17:19:54 right Jan 27 17:20:53 mickeyl: just read you're going to take up opimd Jan 27 17:21:28 i wish you good luck (and good SPEED ;-)) Jan 27 17:24:14 thanks Jan 27 17:33:24 gah it's building 12 hours now Jan 27 17:33:29 hey i have an question..why are all the apps written in python? wouldnt it be faster to write them in C or something? Jan 27 17:33:51 but the good part is it is building Jan 27 17:35:19 root_at: it was a descision from above Jan 27 17:35:37 haha Jan 27 17:35:46 Zorkman: but just for my knowledge...C/C++ would be faster? Jan 27 17:35:55 * Mononoke rises his fist toward the sky Jan 27 17:36:46 root_at: question if how fast it would be written? there is no problem with moving parts to C after they are done Jan 27 17:36:48 root_at: I have almost no knowledge and I would also expect that... Jan 27 17:37:11 if you want more info about it, ask someone like mickeyl, I might say things that aren't entirely correct Jan 27 17:38:18 raster: fileman overlay in default.edc has loop that goes forever in our version.. "go2" program Jan 27 17:38:31 * mickeyl won't comment on the language decision. We're doing dbus middleware so everyone can play Jan 27 17:38:47 it's up to the app developers to chose their language Jan 27 17:39:15 ok, root_at: ask raster then, he's good at cursing at python and was against it IIRC ;-) Jan 27 17:39:50 i'm off to buy food and will be happy with every working app, python or C :) Jan 27 17:40:00 cu Jan 27 17:40:44 raster: but it seems like that we don't need fileman module or the stuff in default.edc at all with our setup Jan 27 17:41:16 raster: also, I think somebody started to put an effert into reducing the loading overlay when starting apps to not cover entire screen, since that uses 30-50% cpu Jan 27 17:41:27 *afk* Jan 27 18:00:10 guys any interest to do http://www.ubuntu.com/products/mobile on freerunner , would look and feel amazing on free runner Jan 27 18:00:48 how do i do dsuc in c++ :-? Jan 27 18:01:04 s/dsuc/dbus Jan 27 18:03:44 Kagee, easy, go to the dbus main project page and read the glib/other documentation Jan 27 18:05:13 main project => dbus.freedesktop.org ? Jan 27 18:05:21 gfather_: don't take it as some kind of attack but with ubuntu's code/upstream give back.. I don't think it will work =] Jan 27 18:05:35 but that's just my view <.< Jan 27 18:06:26 Mononoke every thing can work , if people interested Jan 27 18:12:55 nomeata: Hi :) Do you want to package gsm0710muxd now or after MS5 release? Jan 27 18:13:18 PaulFertser_: is it compatible with frameworkd 4.1? Jan 27 18:15:42 nomeata: to the best of my knowledge, yes :) Jan 27 18:16:14 PaulFertser_: then I can do it now, I guess Jan 27 18:16:26 PaulFertser_: does it bring any notable improvements? Jan 27 18:18:08 nomeata: 0.9.2.2 never closed any virtual channels and didn't return an error if no more channels (4 is maximum for calypso) can be available. Moroever, it returned the pty before the modem acknoledged channel opening, and if app is too fast to write to it, some data was lost. Jan 27 18:19:05 ok, sounds like an upgrade would be nice :-9 Jan 27 18:19:34 freesmartphone.org: 03mickey 07fso-gsm0710muxd * r53afc4c1428e 10/ (3 files in 2 dirs): binding at least compiles now Jan 27 18:19:34 freesmartphone.org: 03mickey 07fso-gsm0710muxd * r5ee7d6fafe10 10/src/multiplexer.vala: more Jan 27 18:19:36 freesmartphone.org: 03mickey 07fso-gsm0710muxd * r309b846dcd6a 10/src/ (main.vala multiplexer.vala): Jan 27 18:19:36 freesmartphone.org: compiles and runs Jan 27 18:19:37 freesmartphone.org: NOTE: don't name a vala-function 'read', this will lead to funny SIGSEGV Jan 27 18:19:39 freesmartphone.org: 03mickey 07fso-gsm0710muxd * rc42fab6914e7 10/ (5 files in 3 dirs): add posixextra vapi and start wrapping read and write Jan 27 18:19:42 freesmartphone.org: 03mickey 07fso-gsm0710muxd * rfff66a00055d 10/ (Makefile.am gsm0710/gsm0710.vapi src/multiplexer.vala): complete read/write cycle to/from 0710 core<-->device, turn int results to bool in 0710 core mapping Jan 27 18:20:05 nomeata: it be cool to setup some automated system to be able to package FSO from git HEAD. Jan 27 18:20:41 PaulFertser_: yes, a few people have called for that, but there are bigger issues IMHO, and itБ─≥s not very debianish to do so :-) Jan 27 18:21:21 re Jan 27 18:21:43 nomeata: if you're wondering, wtf fso-gsm0710muxd is, that's qtopia's muxer that mickeyl is working on to have an alternative. Jan 27 18:22:35 nomeata: Debian way works great for stable software, not so great for WiP, imho. Jan 27 18:23:40 PaulFertser_: at least not as quick as OE, I have to admit that Jan 27 18:24:26 mickeyl: if you're not going to investigate why \n was lost now, i think i should open a ticket and paste my log there, right? Jan 27 18:24:35 yes, pleas Jan 27 18:36:30 PaulFertser_: ok, uploadecd Jan 27 18:36:32 *uploaded Jan 27 18:37:16 mickeyl: you added the paroli bb to OE right? Jan 27 18:37:32 nomeata: cool, thanks! :) Jan 27 18:39:07 sigh, that Jan 27 20:39:18 ginger user.debug 2009.01.27 20:39:18 ogsmd.pdp DEBUG route status. old=usb0, last=ppp0, current=ppp0 debugging message spams my logs :/ Jan 27 18:39:43 mirko-paroli: iirc, yes Jan 27 18:39:53 iirc ? Jan 27 18:39:55 lindi-: upgrade Jan 27 18:40:02 mirko-paroli: if i recall correctly Jan 27 18:40:07 ah Jan 27 18:40:08 mickeyl: i am working on how to run git head yes Jan 27 18:40:09 k Jan 27 18:40:29 lindi-: i'm running git head on debian. Jan 27 18:40:29 mickeyl: but it is not build yet, correct? or in fso repo or likewise Jan 27 18:40:44 PaulFertser_: just from the source directory? Jan 27 18:40:55 lindi-: basically i did python setup.py install --prefix=/usr/local Jan 27 18:41:07 PaulFertser_: i'd like to avoid that Jan 27 18:41:14 mirko-paroli: i only added it to oe, no idea who builds it and who not Jan 27 18:41:18 PaulFertser_: either create proper package or use it from /home Jan 27 18:41:24 mickeyl: ah ok thx Jan 27 18:41:29 mirko-paroli: we have a severe need for a distro guy Jan 27 18:41:37 we need to pester wolfgang to look for one Jan 27 18:41:45 fso worked tons of hours on distro last month Jan 27 18:41:47 that is not good :/ Jan 27 18:42:00 lindi-: I think i can dump /usr/local whenever i like, it's designed for purposes of having improper installed software, afaik :) Jan 27 18:43:11 lindi-: you can do --prefix=/home/lindi/blubb as well Jan 27 18:43:23 lindi-: i think you can run frameworkd from any dir you want, just modify /usr/lib/python2.5/site.py to include your path to frameworkd python modules. Jan 27 18:43:54 nomeata: yeah true Jan 27 18:44:21 mickeyl: was just hoping to get a package into the fso feed Jan 27 18:44:42 but I'll come back with that when the MS5 stress is over Jan 27 18:45:01 mirko-paroli: i see, in that case we can add it to openmoko-feed and hope that the autobuilder at downloads.freesmartphone.org will pick it up Jan 27 18:45:47 mickeyl: sounds good Jan 27 18:46:03 k, i'll ad it Jan 27 18:46:08 mickeyl: thx Jan 27 19:05:14 mickeyl, you know that Ainulindale once offered to work as a distro guy? Jan 27 19:09:45 yes Jan 27 19:10:28 Hello everybody, is OE the only way to compile the Xglamo server? Jan 27 19:11:17 of course not, with an external toolchain and all the dependencies in place it should work as well Jan 27 19:11:18 Alinermedia: surely not Jan 27 19:11:25 don't ask me for how though Jan 27 19:11:25 Alinermedia: apt-get --build source xserver-xglamo works Jan 27 19:11:28 i'm using OE Jan 27 19:18:22 freesmartphone.org: 03mickey 07framework * rfc11a940555d 10/framework/subsystems/opimd/__init__.py: opimd: do not mess with system bus registration, the controller is doing that for you Jan 27 19:20:09 quickdev? Jan 27 19:20:28 Ainulindale, ? Jan 27 19:20:40 quickdev: your highlight about me? Jan 27 19:21:05 haven't you once offered openmoko to work as a distro guy? Jan 27 19:21:09 yep Jan 27 19:21:45 PaulFertser_: "No such file or directory: '/etc/freesmartphone/ogsmd/networks.tab'" -- http://rafb.net/p/T7ec6H78.html Jan 27 19:23:30 PaulFertser_: can i just copy that from git to /etc? Jan 27 19:23:38 Ainulindale: any idea how we want to integrate requesting Display and CPU resources? Jan 27 19:23:44 lindi-: yeah, should work Jan 27 19:24:36 lindi-: i keep configs for git frameworkd in ~. That is, i have ~/.frameworkd.conf and ~/.freesmartphone with all the files copied from git. Jan 27 19:24:45 mrmoku: right now, my head hurts, I have to slow myself for a bit Jan 27 19:25:06 Ainulindale: worked too much? Jan 27 19:25:11 yep Jan 27 19:25:33 PaulFertser_: i only want to run ogsmd so is that necessary? Jan 27 19:25:35 sorry for that... Jan 27 19:26:23 lindi-: i think not, networks.tab won't hurt your existing ogsmd installation anyway, i think it's safe to copy it there. Jan 27 19:26:30 ok, then i better build gsm0710muxd git head as a package too since it is easy Jan 27 19:26:52 lindi-: It was built by nomeata about 1h ago :) Jan 27 19:27:13 PaulFertser_: but not with debugging symbols Jan 27 19:27:37 maybe i should build xglamo from git too? Jan 27 19:28:32 lindi-: Ah, you still suspect bugs in the muxer that could be spotted with gdb. :) Good luck then. Haven't touched xglamo yet, have no idea. I hope it will be obsoleted soon by Graeme's work, but he doesn't seem to be active anymore. :-/ Jan 27 19:29:00 PaulFertser_: well, there _is_ a bug in gsm0710muxd since it dies sometimes Jan 27 19:29:22 it just happens rarely Jan 27 19:29:34 Graeme stoped working on the Xorg driver for the glamo Jan 27 19:29:50 freesmartphone.org: 03mickey 07framework * r05bbd7f9cdc6 10/framework/subsystems/otimed/alarm.py: Jan 27 19:29:50 freesmartphone.org: otimed: alarm trigger is now calling org.freesmartphone.Notification.Alarm() Jan 27 19:29:50 freesmartphone.org: on the root object Jan 27 19:30:09 lindi-: ah, you've talked about that, true. Coredump will be handy :) Jan 27 19:30:15 Alinermedia: why? Jan 27 19:30:44 Lars-Peter Clausen has some code Jan 27 19:30:51 But I haven't seen it yet Jan 27 19:31:31 Alinermedia: do you have any idea why Graeme started the work and why he stopped it? Jan 27 19:32:00 PaulFertser_: xglamo is probably a much more complex thing to debug Jan 27 19:32:11 PaulFertser_: but just getting gdb backtrace on the case where it hangs would be useful Jan 27 19:32:34 [PaulFertser] : Check http://lists.openmoko.org/nabble.html#nabble-td2192154|a2192187 Jan 27 19:32:48 lindi-: sure, but if it will be obsoleted soon by x.org, there's no much sense to do that. But it might be an easy bug once you have the trace, so it can worth the effort. Jan 27 19:33:07 PaulFertser_: we could still spot errors that are present in the xorg port too Jan 27 19:33:10 The plan is to use xorg and build a driver for the glamo chip Jan 27 19:33:33 My concern is that, I can't use the toolchain Jan 27 19:33:53 I've seen the code, I need to test some stuff on it Jan 27 19:34:05 But I'm not familiar with OE/bitbake system Jan 27 19:34:12 xorg drivers fortunately can be built separately Jan 27 19:34:20 Alinermedia: The link you gave me doesn't give a slightest idea why he started the work and what's happened that he stopped it. I've seen those messages. Do you have any more info, please? ;) Jan 27 19:34:25 so i don't need to compile the phone xorg source code, i only need headers Jan 27 19:34:32 s/phone/whole/ (?) Jan 27 19:34:59 Ainulindale: do you know what's a cheap possibility to get prepaid mobile Internet in France? Jan 27 19:35:33 PaulFertser_: btw, wouldn't it be useful if frameworkd did not log timestamp when it logs to syslog? Jan 27 19:35:35 I might go there for a week in June and I and my colleagues need Internet near St.Tropez Jan 27 19:35:46 PaulFertser_: lines like Jan 27 21:36:06 ginger user.debug 2009.01.27 21:36:06 ogsmd.channel DEBUG : sending 19 bytes: 'AT+CNMI=2,1,2,1,1\r\n' are quite long... Jan 27 19:36:03 PaulFertser_: since it has the date twice there Jan 27 19:36:13 PaulFertser_: i can write a patch :) Jan 27 19:36:57 spaetz: nope Jan 27 19:37:09 lindi-: i fully agree. I log to the separate file though. :) I think mickeyl will be interested in your patch, but remember that if you don't write to syslog, the timestamps are still necessary. :) Jan 27 19:37:24 [PaulFertser] http://n2.nabble.com/Xglamo-moving-to-Xorg-architecture-%28was%3A-The-forbidden-topic%3A-Glamo-OpenGL%29-tp1518701p1518701.html Jan 27 19:38:23 PaulFertser_: my only problem is that there are two places that set logging up Jan 27 19:38:39 PaulFertser_: SetDebugDestination() and config.py Jan 27 19:38:48 PaulFertser_: so the same code is already duplicated to two places Jan 27 19:39:28 Alinermedia: i've read that also, but it doesn't gives any insights on why the work was started. Did anybody pay Graeme to port the driver? Or it was his own spare-time initiative? Jan 27 19:40:01 Ainulindale: ok, thanks. Jan 27 19:40:02 lindi-: bug mickeyl about it :), i'm not the one in charge (and even not the one who can write python) here :) Jan 27 19:40:14 mickeyl: ping? Jan 27 19:40:35 what'sup? Jan 27 19:41:08 mickeyl: when frameworkd logs to syslog it is useless to include timestamp since syslog does it already Jan 27 19:41:26 mickeyl: Jan 27 21:36:06 ginger user.debug 2009.01.27 21:36:06 ogsmd.channel DEBUG : Jan 27 19:41:30