**** BEGIN LOGGING AT Wed Feb 18 02:59:57 2009 Feb 18 04:02:57 haha this is awesome... Feb 18 04:03:13 a system where charge rate is dependent upon cpu-usage Feb 18 04:03:23 * Dave wants to kill himself Feb 18 04:08:43 not surprising Feb 18 04:08:52 as you have limited input current Feb 18 04:09:03 as cpu usage goes up - it draws more of the input current Feb 18 04:09:10 less left to charge battery with Feb 18 04:17:05 raster - yes :( Feb 18 04:17:10 of course Feb 18 04:20:39 Actually IIRC Microsoft also wants a system where charge rate is dependent upon cpu usage -- although they mean a different kind of "charge". Feb 18 04:22:22 haha Feb 18 04:27:20 CHARGE!!! Feb 18 04:30:28 at least I won't look like an idiot running a pc off a DPS now :) Feb 18 07:56:11 morning Feb 18 08:06:41 Why is GTA03 using the same old WM8753 codec? I thought Wolfson has many new cool models and this one is oold. Feb 18 08:07:23 PaulFertser: low design risk :) Feb 18 08:07:44 wpwrak: low success chances ;) Feb 18 08:08:02 PaulFertser: bah, nobody cares what codec you use Feb 18 08:08:20 PaulFertser: with one exception, globally :) Feb 18 08:09:37 wpwrak: So, WM8753 is here to stay and it's worth spending time on improving the driver then. The thing is that broonie said it's too old and not very consistent and that new codecs have registers that can be mapped to user-accessible controls directly (for wm8753 alsamixer basically maps registers that doesn't make much sense without datasheet). And that affect every end-user. Feb 18 08:10:55 PaulFertser: well, but most of these controls are something you shouldn't touch anyway. besides, wehave o get this beast to work for GTA02 anyway. Feb 18 08:12:14 wpwrak: as now OM was so generous to send broonie a gta02, is he supposed to fix all the stuff? Do you know anything about the deal? ;) Feb 18 08:15:38 PaulFertser: nope, but i think he might be motivated to fix some things even without a great deal :) Feb 18 08:17:29 morning! Feb 18 08:17:40 wpwrak: you're doing great work on investigating wifi problems but have you heard from atheros anything lately? Looks like they didn't say anything for like 2 weeks now. Feb 18 08:44:52 PaulFertser: i just pinged them today :) Feb 18 08:44:59 Arhuaco: please take a look at my writeup on the ticket, i hope it will shed some light on this dark routing topic. BTW, do you have rights to add broonie to Cc of this ticket? Why can't i do it myself? Feb 18 08:45:36 wpwrak: and what was the response time? ;) Feb 18 08:47:42 PaulFertser: Sure! Thanks a lot for this update! It seems I have rights. I don't know how this ticket system works (admin stuff). wpwrak should know more about rights and such things.... I'll add him in the mean time to the CC list. What is his nick? (I'll look up broonie now to see if it exists in the Wiki as an user)... Feb 18 08:48:12 * Arhuaco tried to mean "trac" and not "wiki". Feb 18 08:48:45 PaulFertser: still counting :) but i sent it around 10pm their time, so i didn't quite count on an immediate reaction Feb 18 08:49:45 Arhuaco: err, what exactly do you need ? Feb 18 08:50:25 wpwrak: PaulFertser wanted to add someone to the CC list of a bug and he could not. I'll do it for him in the mean time... Is this the right thing to do? Feb 18 08:50:27 wpwrak: to add broonie to Cc of https://docs.openmoko.org/trac/ticket/2073 Feb 18 08:52:39 Arhuaco: i don't have any special rights in trac Feb 18 08:53:10 wpwrak, PaulFertser: Ok. I'll do it. In case I have some rights and you don Feb 18 08:54:25 wpwrak, PaulFertser: In case I can do something you cannot (I have no Idea, I'm adding him to the CC list now but it's slow) please send me email and I'll help you with that... Feb 18 08:55:00 Arhuaco: are you employed by OM-the-company? Feb 18 08:55:22 PaulFertser: Yes. Feb 18 08:57:57 Arhuaco: funny how Andy doesn't dare to close irrelevant tickets he obviously knows about. OM sucks :p Feb 18 09:00:31 PaulFerster: Hi this is Andy :-) I just stopped by to talk to Nelson and saw my name mentioned :-) Feb 18 09:00:45 Arhuaco: nice to meet you :) Feb 18 09:00:52 Actually I closed a bunch of tracs the other day, and I will close more in the next days Feb 18 09:01:06 Arhuaco: sure, i saw your efforts, just kidding :) Feb 18 09:01:14 What happened last time was quite interesting though and maybe it will shed some light on why I told you that it made trouble for me Feb 18 09:01:38 PaulFertser: now we no longer have a safe place to complain about Andy :) Feb 18 09:01:47 OMG Feb 18 09:01:53 Actually, many decisions to take action in OM cause trouble of one kind or another and it's that which is to blame for slow actions here sometimes Feb 18 09:02:28 Arhuaco: can you share what happened last time? ;) Feb 18 09:02:28 What happened was that after I closed a bunch of dead and rotting Bugzilla entries we will never attack as WONTFIX, Harald Welte complained on the "everyone" internal list here at Openmoko Feb 18 09:03:09 I defended myself (for closing these reports we will never touch) but there was such a flamethread, and Harald draws so much water, I found other things to do. Feb 18 09:03:49 So, please don't get the idea that it's timidity on my part, I make a lot of enemies trying to do what's right in OM and over time, the right thing does tend to get done. Feb 18 09:04:00 Interesting, i thought Harald has no reason to pay attention any more. Moreover the tickets you closed were obviously bogus by now. Feb 18 09:04:04 That's the end of the history lesson, off to help another colleague :-) Feb 18 09:04:11 Thanks a lot for your work on the audio problems. Feb 18 09:04:40 PaulFertser: I'm back now! Andy noticed his name when he came here. I didn't tell him, I swear! Feb 18 09:04:44 :-) Feb 18 09:04:54 I never blamed personally you for tickets. We (end-users) mostly blame OM-the-company all the time. Feb 18 09:05:11 PaulFertser: Andy left... He's helping someone else around.... Feb 18 09:06:03 Arhuaco: can you clarify why Harald cared about those tickets? Feb 18 09:06:21 PaulFertser: I'll go ask Andy... Feb 18 09:06:24 :) Feb 18 09:06:26 * Arhuaco will be right back... Feb 18 09:07:28 PaulFertser: i think he just thought we were closing tickets without solving the bugs Feb 18 09:08:03 wpwrak: but how it affects him personally? Does he care that much about health of OM-the-company? ;) Feb 18 09:08:12 PaulFertser: and indeed, some of the resolutions may have appeared a little rude. but i think harald made a bit too much of a stink anyway. Feb 18 09:08:39 PaulFertser: well, he's one of the founding fathers, so to speak :) Feb 18 09:09:15 wpwrak: imho it's better to close a ticket and wait for somebody to complain instead of leaving it dangling without any feedback from both sides. Feb 18 09:10:38 PaulFertser: yeah, probably Feb 18 09:10:42 PaulFertser: Andy seems to be busy with something... I guess wpwrak just told you what you wished to know anyway... Feb 18 09:10:55 wpwrak: yes, but as i see it OM-the-company is foobared anyway, to be honest ;) Like "defective by design", you know. But it's from a "hacker" POI, not from a "dumb end-user" poi, so ymmv. Feb 18 09:11:51 Arhuaco: thanks a lot for the clarifications. I thought Andy lives in a quiet house somewhere in UK and he's instead mad busy at the office... ;) Feb 18 09:13:40 PaulFertser: Yes, this world is just weird... Most of the time when you talk to me your text will show in some screen in Colombia... Feb 18 09:13:52 Andy doesn't like IRC very much but I think it's useful for development... I guess I'll like to hang out on IRC. Feb 18 09:14:10 I will get less done myself but I can collaborate with others and help them somehow... Feb 18 09:14:56 Arhuaco: are you very busy with kernel work for GTA03 now? Feb 18 09:15:07 PaulFertser: No, not at all. I'm working on GTA02! Feb 18 09:15:15 PaulFertser! I just got one for day-to-day use! Feb 18 09:15:20 Arhuaco: what's your focus ATM? Feb 18 09:15:34 PaulFertser: Help fix kernel bugs .... Feb 18 09:16:20 Arhuaco: are there many left? I don't experience any during day-to-day use, to be honest. Feb 18 09:19:52 PaulFertser: Sure. I really want to be able to know how it works for day-to-day use... I've seen complaints here in IRC that some of us are not really eating the dog's food thus I want start using the phone... Feb 18 09:20:59 PaulFertser: text-only rootfs is good for some tasks but I feel the phone is close to be very useful as a real phone, but I want to _know_ for sure... Feb 18 09:21:17 Arhuaco: one last question to you, i hope i don't distract too much. What's the status of efforts to get stuff upstream? Feb 18 09:22:42 PaulFertser: You are not distracting me. I was bouncing my head on this ALSA problem and I would have wasted a lot of hours had not I been on IRC... I don't really know anything about ALSA.... I got to the point where I searched for some configurations to check if the DAI=1 thing was indeed the culprit of the crash and I confirmed it was the case. Feb 18 09:25:06 Arhuaco: Is GTA01 dropped completely? Feb 18 09:25:12 PaulFertser: I am supposed to help with that also, but IHMO bugs like this one have priority.... We have to send stuff upstream also I guess... A lot of files need to be polished and in some drivers some changes need to be made in order to be able to send things upstream. Specially when kernel functionality is duplicated (IE: SPI bitbang in the accelerometers). Feb 18 09:25:12 Arhuaco: i don't know anything about ALSA as well. But it's extremely cool to have a sane driver source :) Do you have any questions left after reading my writeup at ticket? Feb 18 09:25:16 CM_work: not at all Feb 18 09:25:26 And why is 2.6.24 recommended instead of .28? :) Feb 18 09:25:32 CM_work: Balaji has recently received a GTA01 and he promises to do something very soon. Feb 18 09:25:45 Nice Feb 18 09:25:58 The latest nightly build of fso worked really well on my gta01 Feb 18 09:25:59 CM_work: because .29 doesn't work good because nobody from the community was interested and OM lost interest for some time. Feb 18 09:26:10 PauFertser: I don't have questions yet, I will read this ticket in detail soon. I'm trying a very small patch that at least avoids the crash... Feb 18 09:26:44 Arhuaco: do you understand that there is no need to avoid this crash at all as there is no need to use DAI 1 mode? Feb 18 09:26:48 PaulFertser: hmm ... what's the issue with .29 ? Feb 18 09:26:58 Arhuaco: and DAI 2 with voip state works good. Feb 18 09:27:50 PaulFertser: Yes, but we can make a quick check and do some printk there... It servers as documentation in case someone follows the old instructions / we can save someone some time... Feb 18 09:27:58 s/servers/serves/ Feb 18 09:27:59 Arhuaco meant: PaulFertser: Yes, but we can make a quick check and do some printk there... It serves as documentation in case someone follows the old instructions / we can save someone some time... Feb 18 09:28:54 wpwrak: all end-users complained that "nothing works". But you know "end-users"... Feb 18 09:30:14 Arhuaco: i think that's not beneficial especially given broonie will be active on weekend. I'd really like to see the support upstream instead ;) Feb 18 09:30:17 PaulFertser: heh ;-) love those specific bug reports Feb 18 09:31:55 wpwrak: but you saw two bugfixes recently. They were possible only because someone started the kernel in QEMU and captured the backtrace. So, the kernel obviously doesn't work on GTA01 and it takes at least an advanced end-user to solve. I wonder where all hackers with GTA01 gone. Feb 18 09:33:00 CM_work: I have a GTA01 with me. Balaji is doing some nice stuff with power management there and he is in a better position to tell if the GTA01 is working well with the latest kernel... He told me he would test soon but if he gets busy I can test myself and report any bad behavior I eventually find. He tested with the latest kernel some time ago but the has not tried the very last changes. Feb 18 09:34:18 Arhuaco: Ok, thanks for the update Feb 18 09:34:26 CM_work, no prob... Feb 18 09:34:35 I'll be happy to test and report stuff, but I'm no kernel hacker.. Feb 18 09:36:18 CM_work: well, all what we need to know if whether it boots and it charges :-) I guess... And it works.. Or what does not work. I don't know... I'll go out soon to get two SIM cards so that I can make tests... Feb 18 09:37:22 The battery meter doesn't seem to work properly, and with the latest fso there's no terminal so I couldn't check Feb 18 09:37:40 But, it seems to run well enough while plugged into usb Feb 18 09:39:08 CM_work: with the latest andy-tracking? Feb 18 09:40:23 Arhuaco: please tell Andy when he see him next time that i'm sorry he has a hard time closing tickets and that i appreciate his efforts to try to beat Trac into shape. Feb 18 09:41:23 PaulFertser: OM is interesting on that regards -- I mean on priorities.. I guess I could think of priorities and prefer some tasks over others... make small choices ... There are 2 extreme lines of thought. 1) Get code upstream first and let's not care whether the damn phone works. 2) Fix the bugs and let's not care a bit about upstream because users need to have phones that works. I guess there is some point in the middle... Feb 18 09:41:50 PaulFertser: Andy did not get upset at all.... Email him if you want :-) But I'll tell him anyway. Feb 18 09:42:52 PaulFertser: I have to go talk to someone... I'll be back later... Thanks for all your help. I guess I'll be on IRC more often... BBL... Feb 18 09:47:21 Arhuaco: i think what OM needs to stop dissapointing developers and community is to state the choosen way more publicly. I mean, if i.e. Andy decided that bugfixing .24 doesn't worth it, explain in detail when and why he thinks .28 will be in shape for everyday use and what advantages will it give. Even more transparency is probably needed, community wants to know who is doing what. Move more communication from internal to pub Feb 18 09:56:05 PaulFertser: oh, if you reverse the beast, you'd go softmac anyway. so probably get those goodies as well Feb 18 09:56:07 wpwrak: btw, would you personally want to have a freerunner-like device which is bigger, PDA-size? In my opinion it'd be optimum and will allow for more cool ICs inside and more power cells etc. Feb 18 09:56:28 PaulFertser: bigger only if it's x86 Feb 18 09:56:34 Arhuaco: i think what OM needs to stop dissapointing developers and community is to state the choosen way more publicly. I mean, if i.e. Andy decided that bugfixing .24 doesn't worth it, explain in detail when and why he thinks .28 will be in shape for everyday use and what advantages will it give. Even more transparency is probably needed, community wants to know who is doing what. Move more communication from internal to pub Feb 18 09:56:40 PaulFertser: but i can get that already from OQO Feb 18 09:56:52 wpwrak: who needs x86 anyway? Feb 18 09:57:24 PaulFertser: from openmoko, i just want a phone :) Feb 18 09:57:24 PaulFertser: I use the latest nightly experimental builds, but so far I've used the 2.6.24 since the note said I should for a GTA01 Feb 18 09:57:51 CM_work: so, your report is irrelevant as the talk was about the latest andy-tracking kernel, sorry ;) Feb 18 09:58:00 PaulFertser: x86 binary compatibilty makes life easier. even if try to avoid closed source stuff Feb 18 09:58:38 Ok, is there any builds for that one? Just use the .28 or .29? Feb 18 09:59:09 wpwrak: but i don't understand all those people who says that the phone should be small. Just can't get it. As to x86 -- imo it's a very bad idea, it makes much more sense to fix bugs in arm-incompatible software instead. imho. Feb 18 09:59:37 CM_work: if you're not a kernel dev, probably you won't want to try it yet. If you're, compile from the andy-tracking HEAD> Feb 18 10:00:02 Ok, I'll give it a shot. I'm not a total newbie Feb 18 10:05:22 anybody see spontaneous sleeping with the newest shr? Feb 18 10:13:20 budfive: no Feb 18 10:16:48 mrmoku: I'm going to dig into it right now, unless you know what the issue is. The moko seems to be falling asleep without me prompting it at times. One consistent way to replicate what I'm seeing is to push the power button. The little window with the suspend, shutdown, etc buttons comes up as it's supposed to, but the moko falls asleep a second later without me pushing any of the buttons. sometimes it spontaneously wakes up f Feb 18 10:16:48 ideas? Feb 18 10:17:41 thats a borken rules.yaml from frameworkd... we fixed it... might try to update Feb 18 10:17:56 budfive: do you happen to have a build that has not the 1st rules.yaml rule commented out? Feb 18 10:17:57 (though I'm not completely sure the fix got built ;) Feb 18 10:18:11 then you are not using the latest unstable :) Feb 18 10:18:37 mrmoku: i flashed the most recent unstable, but let me look. glad i asked. :) Feb 18 10:21:03 mrmoku: just downloaded the latest .ipk from framworkd-config-shr and the rule is still there, it seems :-( Feb 18 10:21:58 spaetz: btw... I saw you did not do it for gta01... any reason? Feb 18 10:22:06 * mrmoku is rebuilding frameworkd-config-shr Feb 18 10:22:39 I did not update omgta01 at all, as it seemed to use some different rules and I am unable to test Feb 18 10:22:39 mrmoku: so can you tell me what I should patch, if it's not too much trouble for you? Feb 18 10:23:09 budfive: comment out the first rule in rules.yaml Feb 18 10:23:27 budfive: http://dpaste.com/122052/ Feb 18 10:23:35 this one Feb 18 10:23:48 thank you Feb 18 10:24:09 might need a restart of frameworkd or a reboot to take effect Feb 18 10:26:16 mrmoku: looking at it closer, it seems that only the audio settings command is different between gta01 and 02 Feb 18 10:26:30 I'll update omgta rules.yaml when I get to it Feb 18 10:26:37 ok Feb 18 10:34:53 done :) Feb 18 10:35:11 * mrmoku is rebuilding again :P Feb 18 11:02:36 :) Feb 18 11:11:26 dos1: btw, something should be done so most people know that SHR testing is not supported atm. It causes too much confusion judging by the community mailing list. Feb 18 11:12:38 topic too long :( Feb 18 11:18:24 dos1: hah, good you are here Feb 18 11:18:30 Interested in another patch? Feb 18 11:18:47 also concerning timeouts, just arrange them in a table rather than a long list of boxes Feb 18 11:19:10 spaetz: of course ;) Feb 18 11:19:19 makes buttons be aligned and also centers the text labels in the middle rather than at the top of the boxes. Feb 18 11:19:32 I am finished, just need to create the patch Feb 18 11:23:09 http://dpaste.com/122059/ Feb 18 11:23:20 dos1: here is the patch. Want it mailed too? Feb 18 11:23:39 Also, I would put the "power actions" below the "timeout settings". Feb 18 11:24:02 the actions are available by pressing the power button too, so they are probably going to be used very seldomly Feb 18 11:29:48 ahh, I see that the power actions and the power status is one module. ok Feb 18 11:37:52 spaetz: i don't think we need "power actions" at all Feb 18 11:38:43 spaetz: i will test and commit your patch now, and then remove "power actions" frame Feb 18 11:41:24 dos1: sounds fine to me. Less on the screen means its faster to reach Feb 18 11:42:07 BTW, I am looking into how to use the ecore.add_Idler stuff Feb 18 11:42:22 so we can do the expensive dbus calls after the main GUI has been set up Feb 18 11:42:47 good call Feb 18 11:42:59 my suggestion is to also NOT do calls syncronously Feb 18 11:43:17 ie do the call and wait for a reply from the other end Feb 18 11:43:21 most difficult thing is that I never used e/ecore/efl/whatnot before Feb 18 11:43:25 that is one way to make sure performance will be bad Feb 18 11:43:35 can't help u with python Feb 18 11:43:37 but adding an idler does not seem that difficult Feb 18 11:43:45 but in c it's relatively simple Feb 18 11:44:04 it's pretty much the same in python as in c, just different syntax :P Feb 18 11:44:55 raster: first the idler. async dbus is lower down on my TODO list :) Feb 18 11:45:55 raster, first question: I add an idler, when will it be called Feb 18 11:46:11 1st time ecore mailoop is idle, or everytime when it becomes idle? Feb 18 11:46:33 ie do I need to allow it to enter the idle loop several times? Feb 18 11:46:49 * spaetz guesses yes Feb 18 11:47:44 commited Feb 18 11:49:26 idlers are called during idle time Feb 18 11:49:28 in a loop Feb 18 11:49:38 so instead of the process sitting idle waiting for a timeout or inoput Feb 18 11:49:41 it spins running idlers Feb 18 11:50:14 idlers keep looping forever undil they return 0 (meaing stop runing me and delete me) Feb 18 11:50:17 or they are deleted Feb 18 11:50:28 they only run during idle Feb 18 11:50:31 oh Feb 18 11:50:32 if a timeout happens Feb 18 11:50:38 or events come in those are processed Feb 18 11:50:50 so 1. make idlers do very litle work if u can Feb 18 11:50:54 raster: thanks, i was playing with it some time ago Feb 18 11:51:04 as you want them to cycle very fast to avoid latency in dealing in a timeout or event Feb 18 11:51:08 ok, returning 0 to stop and delete me is good, thanks Feb 18 11:51:21 2. be careful of return value Feb 18 11:51:36 * spaetz checks different return values in python Feb 18 11:51:48 3. they wont wake up the event loop if they stop runnign - so a new evas render cycle wont happen unless u then also queue some timer event to "wake up" the event loop Feb 18 11:51:59 (or queue some input data on an fd select is listening to) Feb 18 11:53:02 ok, returning "None" makes it runs only once, returing 1 makes it loop forever. Feb 18 11:53:11 seems to work fine Feb 18 11:53:53 ok, that was helpful, thanks Feb 18 11:54:39 dos1: do you have some worst offenders for dbus calls? Feb 18 11:55:01 spaetz: i don't understand? Feb 18 11:55:54 some dbus calls seem to take a long time, making shr-settings slow. Do you know which of the calls take long Feb 18 11:56:10 ie in which module should I start looking at adding an idle handler? Feb 18 11:56:17 dbus calls should not take long Feb 18 11:56:20 UNLESS they are syncronous Feb 18 11:56:26 whihc means they send a dbus request Feb 18 11:56:28 \and sit and wait Feb 18 11:56:33 waiting for a reply from the other end Feb 18 11:56:42 meaning your mainloop stops running while waiting Feb 18 11:56:44 raster: right, so what I need to do is make dbus calls asyncronous. Feb 18 11:56:46 and thats all bad Feb 18 11:56:50 bingo Feb 18 11:56:56 then u can send off 1, 2 10 100 calls Feb 18 11:57:00 and not wait for replies Feb 18 11:57:05 replies will come in later over time Feb 18 11:57:09 handle them as they come back Feb 18 11:57:40 raster: yes, we have syncronous dbus cals, and we should not, i know ;) Feb 18 11:57:40 ok, need more time to think about this. Feb 18 11:57:46 lunch time Feb 18 11:57:55 that will always be a problem you keep working around Feb 18 12:16:15 'afternoon Feb 18 12:16:29 does anyone know where I can get the squashfs module for 2.6.28-rc4 kernel? Feb 18 12:19:50 gromgull, take patch from official site, or just use 2.6.29-* Feb 18 12:20:29 is it included with the 2.6.29 modules? Feb 18 12:21:46 i'll download and check Feb 18 12:23:49 it's not in http://downloads.openmoko.org/distro/experimental/NeoFreerunner/modules-2.6.29-rc3-oe1+gitr119575+53967cb8ccaa5c0230f62f4fcf96f950c921992a-r2-om-gta02.tgz either :( Feb 18 12:25:03 but it is in the modules for 2.6.24 Feb 18 12:42:36 indeed, for 2.6.24 it's http://shr.bearstech.com/shr-testing/ipk/om-gta02/kernel-module-squashfs_2.6.24-oe3+gitrfb42ce6724576fc173faf8abfb04aa2c36d213b7-r1.1_om-gta02.ipk Feb 18 12:50:30 nice Feb 18 12:50:32 root@om-gta02:~# reboot Feb 18 12:50:32 Segmentation fault Feb 18 12:50:49 lol ? Feb 18 12:50:50 halt too Feb 18 12:53:21 dos1: heh? doing reboot all the time... Feb 18 12:53:52 mrmoku: my system was in state, when halt was segfaulting too Feb 18 12:54:05 mrmoku: and that happened to me first time ;D Feb 18 12:56:43 what did you do to your poor neo? Feb 18 12:57:27 hmm... thinking about that... had that too when fiddling with init Feb 18 12:57:48 mrmoku: but i was only fiddling with dbus in python Feb 18 12:57:50 ;x Feb 18 12:58:02 hmm.. playing with async dbus calls to frameworkd is dangerous... Feb 18 12:58:58 now frameworkd is borken - i even can't kill it :x Feb 18 13:00:15 ok, i can reproduce it... Feb 18 13:00:33 i'm calling odeviced and turning on/off in async calls Feb 18 13:00:45 i'm toggling it very fast Feb 18 13:00:52 and then frameworkd is not reacting to kill Feb 18 13:01:07 and then "reboot" displays "broadcast message" Feb 18 13:01:27 but nothing else Feb 18 13:01:37 and calling reboot second time - segfault Feb 18 13:02:18 * mrmoku thinks dos1 is doing DOS ;) Feb 18 13:02:26 :D Feb 18 13:02:40 hmm Feb 18 13:02:41 what are you calling exactly? Feb 18 13:02:49 in htop i see interesting processes Feb 18 13:02:51 ifconfig Feb 18 13:02:56 grep -q ^eth0 Feb 18 13:02:59 ip addr list usb0 Feb 18 13:03:15 /etc/rc6.d/K10dropbear stop Feb 18 13:03:53 mrmoku: Feb 18 13:03:55 proxy = dbus_system.get_object('org.freesmartphone.odeviced', '/org/freesmartphone/Device/PowerControl/WiFi') Feb 18 13:03:55 proxy.SetPower(obj.state_get(), dbus_interface='org.freesmartphone.Device.PowerControl', reply_handler=handler, error_handler=error) Feb 18 13:04:36 and when i'm calling SetPower second time, before reply_handler, it goes to that ugly state Feb 18 13:04:47 and then answers don't come Feb 18 13:04:50 from dbus Feb 18 13:06:53 poor frameworkd :P Feb 18 13:07:16 mrmoku: i know, i should use ousaged, but it's only test for async calls Feb 18 13:15:07 guys, is there a git for the qemu-neo1973 ? or has it been rolled in main qemu maybe ? Feb 18 13:15:51 anyone know whats up with gta01 not able to boot SHR-unstable from NAND? Feb 18 13:22:56 PaulFertser: as far as I can see... xglamo does not use the sysfs node... Feb 18 13:23:12 torpor: gta01 NAND?... thought it has only NOR Feb 18 13:23:32 ahh... u-boot only in NOR... Feb 18 13:24:21 * mrmoku needs coffee... bbiab Feb 18 13:24:40 well id ont get it Feb 18 13:26:45 w00t! real 560m27.430s shr-lite yay! Feb 18 13:26:55 wahts that? Feb 18 13:27:11 time take to build shr-lite image Feb 18 13:27:18 now to install and see if it works.. Feb 18 13:27:33 let me know Feb 18 13:27:36 i wanna run it on my gta01 Feb 18 13:27:39 but it just don't work. Feb 18 13:28:22 mrmoku: gta01 has NOR? i thought it has only NAND :P Feb 18 13:28:51 mrmoku: gta02 has bigger NAND than gta01 and NOR for backup bootloader :P Feb 18 13:33:29 dos1: and gta01 has no backup boot loader in NOR :P Feb 18 13:34:13 torpor: though... it works on gta01... where does it hang? Feb 18 13:35:15 mrmoku: because it doesn't have NOR :D Feb 18 13:35:46 dos1: ehmm... no. because it does not have u-boot in NAND :D ??? Feb 18 13:37:41 mrmoku: i don't think so Feb 18 13:37:53 mrmoku: gta01 doesn't have NOR flash; only NAND Feb 18 13:38:03 mrmoku: and it have everything in NAND Feb 18 13:38:20 mrmoku: gta02 has everything in NAND, plus backup bootloader in NOR Feb 18 13:38:21 * mrmoku thinks dos1 is right and me is stupid :-) Feb 18 13:38:49 dos1: yep... only NAND... no NOR Feb 18 13:38:56 * dos1 doesn't have gta01, i'm repeating what i read Feb 18 13:39:47 anybody tried AVRCP with bluetooth? (next and previous song) Feb 18 13:41:58 and are there plans to port a torrent client to shr? Feb 18 13:44:58 mrmoku: i get 'kernel panic: unable to load root filesystem on unknown-device(34,0)' Feb 18 13:45:32 but it doesnt matter because i just put together the filesystem for root-on-sd and will use that .. Feb 18 13:46:20 mickey|zzZZzz: please take a look on #329 at FSO trac - it's blocking shr-settings functionality Feb 18 13:46:52 i implemented async call with operators list, but FSO isn't returning reply - only timeout :x Feb 18 13:49:33 okay, i'm booting SHR on my gta01 finally .. looks nice! Feb 18 15:46:43 http://ramon.blogli.co.il/files/2009/02/myscreenshot_001.png <--- very good idea! using transparent icons so the bg image will be visible! Feb 18 15:56:57 TAsn: you could just use no icon... Feb 18 15:57:46 right. ;] I guess that's what he did ;] Feb 18 15:57:57 hehe, thanks for the idea ;] Feb 18 15:59:58 i'm a bit annoyed by my clock always changing to the wrong time Feb 18 16:00:06 what's going on there? Feb 18 16:01:04 orzo: what distro? Feb 18 16:02:56 debian with fso frameworkd Feb 18 16:05:42 current frameworkd automagically set the clock... maybe not to what you want? Feb 18 16:06:45 Hi. Is fso-abyss ready to replace gsm0710muxd for a MS5 frameworkd already? Feb 18 16:07:39 mrmoku: ya, not really waht i want, but even so, it's wrong. (It interferes with ntpdate which says the socket is in use) Feb 18 16:08:21 does frameworkd try to get the time from the gsm network when there's no internet available? Feb 18 16:08:33 and set it wrong Feb 18 16:09:26 the minutes are right Feb 18 16:09:34 so maybe its some incorrect time zone handling Feb 18 16:10:13 my system is in EST Feb 18 16:10:59 /etc/localtime is a symlink to /usr/share/zoneinfo/EST5EDT Feb 18 16:11:34 it has current time as 16:11 EST Feb 18 16:11:46 orzo: [otimed] disable=1 should disable that me thinks... Feb 18 16:11:47 when it is really 11:11 EST Feb 18 16:12:27 it must be assuming i'm in UTC Feb 18 16:15:39 mrmoku: where do i put that? Feb 18 16:26:10 any one here have problems with phonelog (shr-unstable) Feb 18 16:26:11 dos1: operator list should work with the latest FSO and muxer, especially with the moko11 firmware. If you don't get a reply, probably that's dbus timeout that is firing too early. Feb 18 16:26:12 ? Feb 18 16:26:49 mrmoku: well, you can add a facility to write to a sysfs node when xrandr is requested to change resolution then :) I'll take a look at the code, though i don't promise a patch. Feb 18 16:27:19 becouse it don't want to start on my FR ;/ Feb 18 16:28:10 PaulFertser: with old frameworkd (milestone4) it worked; with HEAD, it timeouts Feb 18 16:28:30 awww... Feb 18 16:28:43 why does qi not boot partition 4? dammit.. Feb 18 16:28:45 soltys, can you try running it from terminal? Feb 18 16:28:51 DISPLAY=:0.0 phonelog Feb 18 16:29:38 TAsn: http://wklej.org/id/54417/ Feb 18 16:30:07 opkg list | grep phonelog Feb 18 16:30:44 pyphonelog - 0.16.2-r0 - PyPhonelog is a phonelog gui that connects to the shr daemon/a custom daemon Feb 18 16:30:45 pyphonelog-dbg - 0.16.2-r0 - PyPhonelog is a phonelog gui that connects to the shr daemon/a custom daemon Feb 18 16:30:47 pyphonelog-dev - 0.16.2-r0 - PyPhonelog is a phonelog gui that connects to the shr daemon/a custom daemon Feb 18 16:30:59 thanks. Feb 18 16:31:45 this was hapeningo also with older version Feb 18 16:32:02 weird, would you mind sending me the phonelogging db? Feb 18 16:32:08 /var/db/phonelog.db iirc Feb 18 16:32:21 sec Feb 18 16:32:55 dos1: an ogsmd debug log is needed to further investigate Feb 18 16:33:17 toggles_w: Qi only supports 3 partitions on SD, sorry Feb 18 16:33:19 hmm. 48 hours of uptime, complete with improved battery life. Feb 18 16:33:25 good job, people. ;) Feb 18 16:34:06 TAsn: szelka.net/~soltys/phonelog.db Feb 18 16:34:13 thanks. Feb 18 16:34:31 np ;) Feb 18 16:34:39 i hope it helps Feb 18 16:36:49 I can't see why it happened, though you have a corrupted row in the db. Feb 18 16:36:59 or so it seems. Feb 18 16:37:15 if you don't care about the db, removing the db and restarting ophonekitd should solve it. Feb 18 16:37:17 i try to delete the db Feb 18 16:38:01 please tell me how it goes. Feb 18 16:38:21 the problem seems worst than I thought, I'm looking into it. Feb 18 16:40:13 TAsn: deleting the db helped Feb 18 16:40:23 will see for how long ;) Feb 18 16:40:56 anyhow, I'm adding extra checks, debug info and how to solve ;] Feb 18 16:41:17 ok ;) Feb 18 16:48:14 orzo: /etc/frameworkd.conf Feb 18 16:48:59 juri_: :-) Feb 18 16:49:37 Zorkman: this orrery thingie is giving me a hard time... but I'm getting closer :-) Feb 18 16:51:12 nice Feb 18 16:51:32 it was one of the first really good Fr programs (that worked!) Feb 18 16:51:36 TAsn: finally "built" a stylus using your method. It's almost as handy as a regular one and i'm no longer afraid to lose it :) Thanks a lot! (btw, i even got used to use fingernails for typing in matchbox keyboard while after i lost my stylus ;) ) Feb 18 16:51:54 ;] Feb 18 16:52:22 * mrmoku notes to go down into the basement to look for some old european style cables... Feb 18 16:52:59 it's actually very convenient, I just wrap it around my pointing finger and that "makes" my finger pad more sharp and therefore more accurate. Feb 18 16:53:00 ;] Feb 18 16:53:11 I'll up a pic of me using it soon enough ;] Feb 18 16:53:50 BTW, wanna hear a joke? How the "for" command implemented in windows' shell (cmd.exe). Feb 18 16:54:03 how? Feb 18 16:55:02 soltys, the checks + debug info will be available in the next auto generated package. Feb 18 16:55:17 mhm Feb 18 16:55:19 thanks a lot for the report. Feb 18 16:55:24 np Feb 18 16:55:30 TAsn: comitted? Feb 18 16:55:31 something is really going wrong with the ophonekitd ;] Feb 18 16:55:33 mrmoku, aye. Feb 18 16:55:38 * mrmoku goes building Feb 18 16:55:40 ;] Feb 18 16:55:51 there isn't a major change Feb 18 16:56:10 just made it more crash proof, if when the ophonekitd does stuff wrong ;] Feb 18 16:56:34 I should have done this from the start, I guess I was too naive the db won't get corrupted ;] Feb 18 16:57:04 They have a special syntax for some common tasks. So intead of using "for i in `seq 50`; do smth %i; done" a windows user has to type "for /L %i in (1, 1, 50) do only_one_command_allowed %i" Feb 18 16:57:20 anyhow, now the user gets some kind of an "alert" when the db is corrupted, so it'll be able to report, but still use the daemon. Feb 18 16:57:36 And if he uses it from a shell script, he must use %%i instead :) Feb 18 16:57:44 See how consistent and elegant? Feb 18 16:58:00 PaulFertser, yeah, it's also very intuitive. Feb 18 16:58:07 same as in every other language I know. ;] Feb 18 16:58:17 give or take. Feb 18 16:58:42 PaulFertser, I wonder what made them do that. Feb 18 16:58:55 TAsn: that "for" command even has a primitive parser for text files or a special arcane syntax for using "backtick" command. Feb 18 17:00:12 ;[ Feb 18 17:00:26 why are you writing windows shell scripts anyway?! Feb 18 17:00:34 just install python/perl/bash ;] Feb 18 17:01:32 I don't. I just went to an old writer who wanted to have 100 word files appropriately named, so i created one and used the command that to copy it to Chapter_%i.doc Feb 18 17:01:42 oh. Feb 18 17:01:46 Empty files Feb 18 17:02:58 TAsn: he was born in 1922, so it's easier to create empty files with a correct "text style" instead of explaining him how to do that when he needs it. Feb 18 17:03:33