**** BEGIN LOGGING AT Fri Jun 26 02:59:57 2009 Jun 26 04:03:23 openmoko: 03Duncanla * r5216 10/trunk/gta02-core/ (io.sch lcm.sch): Jun 26 04:03:23 openmoko: Removed duplicate CPU instantiaions from IO and LCM sheets. Jun 26 04:03:23 openmoko: Updated LCM schematics: Jun 26 04:03:23 openmoko: - Fixed bug in VD connections for 16BPP operation Jun 26 04:03:23 openmoko: - Moved LCM SPI interface from SPI1 to GPIOs Jun 26 04:26:50 openmoko: 03Duncanla * r5217 10/trunk/gta02-core/ (docs/ecn/ecn0001.txt docs/ecn/ecn0016.txt io.sch): Jun 26 04:26:50 openmoko: Updated ECNs 1 & 16 to reflect changes that were made to the schematics. Jun 26 04:26:50 openmoko: X'ed out unused pins in io.sch. Jun 26 04:59:53 hi, is it possible to install apt on openmoko? Jun 26 05:00:00 via ipkg/opkg Jun 26 05:00:12 so that i can have both Jun 26 05:00:37 i would wnat this very much, either that or perhaps install ipkg/opkg via apt on debian Jun 26 05:00:43 which ever works Jun 26 05:02:48 i will look though the mailinglist Jun 26 05:06:27 tae, afaik no. but you can install debian Jun 26 05:07:53 hmm, its a shame, its only a matter of compiling apt and all its dependencies right? Jun 26 05:16:03 doit Jun 26 07:54:36 lindi-: i've just borrowed again that headset that worked. Will test it with GSM soon (need to charge it a little). Jun 26 08:10:20 openmoko: 03alvieboy * r5218 10/trunk/gta02-core/docs/ecn/ecn0008.txt: Update GPS ext. ant. removal info Jun 26 08:20:10 lindi-: ok, i've just tried to use that working headset and it worked. One drawback though, debian has an old version of one lib. Jun 26 08:21:13 PaulFertser: working headset + hciconfig patch + what alsa state? Jun 26 08:27:51 http://www.tomshardware.com/de/Tegra-Nvidia-Netbook-Computer-on-a-Chip,news-243028.html#xtor=RSS-121 - GTA05? Jun 26 08:36:49 Wonka: Well, as soon as they open up everything on it, sure :) Jun 26 08:37:07 freesmartphone.org: 03seba.dos1 07framework * r73a5f06b09ec 10/framework/subsystems/opimd/pimb_sqlite_messages.py: opimd: SQLite-Messages: fix updating MessageRead, MessageSent and Processing fields Jun 26 08:37:08 freesmartphone.org: 03seba.dos1 07framework * ra2a529777132 10/framework/subsystems/opimd/pimd_messages.py: opimd: Messages: allow disabling merging in config Jun 26 08:37:09 freesmartphone.org: 03seba.dos1 07framework * rbf8dfd26b0a1 10/framework/subsystems/opimd/pimb_sim_messages_fso.py: opimd: SIM-Messages-FSO: rename Text field to Content Jun 26 08:37:10 freesmartphone.org: 03seba.dos1 07framework * r4a69d2858965 10/framework/subsystems/opimd/ (pimb_sqlite_contacts.py pimb_sqlite_messages.py): opimd: SQLite-*: don't store special _ fields Jun 26 08:41:20 lindi-: i've tested it with GSM. Jun 26 08:41:28 lindi-: for that no hciconfig patch needed. Jun 26 08:41:46 lindi-: the statefile is the one linked from the wiki (the "proper" method). Jun 26 08:42:10 lindi-: the headset is noname "Jabra", scan showed it as "Bluetooth headset" Jun 26 08:42:37 lindi-: the problem with fso is that it uses python alsa bindings that are too old in debian to release the PCM properly, so only first call works. Jun 26 08:42:50 lindi-: how can i provide you more info about the headset? Jun 26 08:48:12 PaulFertser: http://dl.getdropbox.com/u/453116/gsmbluetooth.state ? Jun 26 08:57:41 freesmartphone.org: 03seba.dos1 07framework * r510847d779d7 10/framework/subsystems/opimd/ (pimb_sqlite_contacts.py pimb_sqlite_messages.py): opimd: SQLite-*: handle database errors Jun 26 08:58:51 freesmartphone.org: 03mickey 07framework * r8bcbdd47ea1f 10/framework/subsystems/oeventsd/fso_actions.py: oeventsd: fso-actions: don't store values acquired by oprofile as variants Jun 26 08:59:43 freesmartphone.org: 03seba.dos1 07framework * r86fd996dbbc8 10/framework/subsystems/opimd/pimb_sqlite_contacts.py: opimd: SQLite-Contacts: fix accidentaly wrapped line Jun 26 08:59:44 freesmartphone.org: 03seba.dos1 07framework * r22a3a15ac18d 10/framework/subsystems/oeventsd/fso_actions.py: Merge branch 'master' of git@git.freesmartphone.org:framework Jun 26 09:06:59 lindi-: exactly the same i use. Beware of the alsa bug that prevents BT mic from working if you don't employ a workaround. Jun 26 09:08:51 lindi-: do you need anything else? I haven't tried mic over GSM but i think it should work. Will try a little bit later. Jun 26 09:11:37 which alsa bug is that? Jun 26 09:23:38 hello o/ does anyone know what the approximate delivery time is for a buzzfixed phone to be delivered? Jun 26 09:24:30 as in I sent it off, they have sent if back but the courier tracking page is a bit sketchy over when I should expect it Jun 26 09:25:19 I ran the status message through a google translation (yes I know this is not the best translation) and got : The shipment is on its way to becoming a competent office. :) Jun 26 09:25:32 Blu3: you'll find everything about it in this thread: http://www.mail-archive.com/openmoko-kernel@lists.openmoko.org/msg08663.html Jun 26 09:25:56 tig|: ROFL Jun 26 09:26:46 PaulFertser: it is a classic status message :) Yet slightly unhelpful and it may have lost subtle nuances during the translation process :) Jun 26 09:33:13 good morning Jun 26 09:33:22 morning Jun 26 09:40:37 PaulFertser: which workaround? Jun 26 09:59:46 lindi-: please read the thread i just mentioned Jun 26 10:15:01 freesmartphone.org: 03mickey 07cornucopia * r03f09d594349 10/fsousaged/TODO: add TODO Jun 26 10:19:51 freesmartphone.org: 03mickey 07framework * rc4200cc8a465 10/framework/subsystems/odeviced/kernel26.py: odeviced: kernel26: remove APM power class supply. We do not support APM. Jun 26 10:22:34 freesmartphone.org: 03Frederik.Sdun 07dbus-hlid * rb2a72f1a9153 10fso-monitord/src/monitor.vala: Jun 26 10:22:34 freesmartphone.org: Add debug output for NameOwnerChanged Jun 26 10:22:34 freesmartphone.org: skip new unowned busnames Jun 26 10:22:34 freesmartphone.org: Move restart info to CTOR Jun 26 10:22:34 freesmartphone.org: 03Frederik.Sdun 07dbus-hlid * rb50032f74ff3 10fso-monitord/src/monitor.vala: Add count for running monitors Jun 26 10:22:38 freesmartphone.org: 03Frederik.Sdun 07dbus-hlid * r20668753301b 10fso-monitord/src/monitor.vala: Using new vala syntax for dbus objects **** BEGIN LOGGING AT Fri Jun 26 10:29:06 2009 Jun 26 10:36:14 lindi-: do you want me to try to find a way to make a photo of that noname headset? Jun 26 10:48:03 lindi-: ok, just tested gsm call in both direction. It works. Probably requires tweaking "Mono" a bit otherwise other side can have difficulties hearing me (it might also depend on the headset). Jun 26 11:00:21 Guys, i updated instructions on using bluetooth headsets with my success report, please see http://wiki.openmoko.org/wiki/Manually_using_Bluetooth#Once_Again.2C_Bluetooth_Headset_on_Freerunner Jun 26 11:21:34 PaulFertser: i'm bit afraid to upgrade frameworkd -- I wonder what breaks in the process :) Jun 26 11:29:27 lindi-: i know. Just wanted to let you know that it works. OTOH it seems almost nobody needs bt headsets. Jun 26 11:31:04 There is a big disconnect between 'needs' - and 'needs, and has the skills to try to get it working, and doesn't give up at the 'it doesn't work stage'. Jun 26 11:33:42 SpeedEvil: the wiki instructions are for end-users, not programmers. Jun 26 11:33:55 SpeedEvil: Jan wrote the support, i wrote the instructions. Jun 26 11:36:22 PaulFertser: i'd appreciate a stand-alone program that does this without having to upgrade frameworkd Jun 26 11:37:26 lindi-: does what? Jun 26 11:37:51 lindi-: loading alsa state file? Tweaking one control to workaround kernel bug? Activating audio over PCM? Jun 26 11:37:59 lindi-: you need a 3-line script? ;) Jun 26 11:38:17 PaulFertser: activating audio over pcm i guess since i can alsactl the statefile Jun 26 11:40:55 lindi-: please see http://git.freesmartphone.org/?p=framework.git;a=blob;f=framework/subsystems/ophoned/headset.py;h=7a33bf20ca4146e5bb7baffc7928ebbef8979dc4;hb=HEAD and feel free to ask questions. I think you'll find it trivial to reimplement the way you like. Jun 26 11:42:24 lindi-: only that debian's alsa bindings lack close() call. Jun 26 11:44:27 lindi-: also you'll need to do subprocess.call( [ "amixer", "sset", "Capture Left Mixer", "Analogue Mix Right" ] ) and subprocess.call( [ "amixer", "sset", "Capture Left Mixer", "Analogue Mix Left" ] ) after loading the state file. Jun 26 11:52:47 PaulFertser: can't the state file just be fixed? Jun 26 11:53:48 lindi-: this sequence is essentially a noop but it workarounds a kernel bug. Jun 26 11:55:56 I'm somehow very sure there're at least 2 gta01 in moscow that are lying somewhere unused. And i can't find them :( Jun 26 11:59:17 PaulFertser: lindi-: (subproc.call) I suggest to call a script instead of multiple amixer calls. See ACI. Advantages: easily editable, extensible, configurable. Possibly also faster (again see ACI which loads amixer only once and pipes multiple cmds to input) Jun 26 12:00:33 DocScrutinizer: sure, good advice :) But you see how enthusiastic are people about BT support? Fuckers, first they complain for months that it doesn't work, then Jan finally implements the support and nobody uses it. Jun 26 12:00:57 heh, true Jun 26 12:02:11 I don't say that the support is excellent and complete. But it already works, there's already something to try. The reason the remaining bugs and inconveniencies (like ignoring headset buttons) are not fixed is exactly lack of end-user feedback. Jun 26 12:02:55 PaulFertser: we know it works with some headsets :) Jun 26 12:03:22 lindi-: we knew it in february already, when Jan implemented it. Jun 26 12:03:57 PaulFertser: do you understand my problems with a picture of me published without any copyright notice? Would you have done in a better way than me after 3 weeks of watching it spreading all over the net? Jun 26 12:04:15 PaulFertser: what does debian need to make multiple calls? restart of bluetoothd? Jun 26 12:05:55 s/witout any/with false GPL/ Jun 26 12:09:12 * SpeedEvil cannot find his bluetooth headset. Jun 26 12:14:00 PaulFertser: do you think we should get the files liked to on bt-wiki to a more centralized place? Jun 26 12:14:12 liNked Jun 26 12:31:07 DocScrutinizer: yes, i understand. Jun 26 12:32:00 DocScrutinizer: probably you could have complained to other members of SHR team more explicitly so they understand that it's not a joke and i'm sure they'd remove it very fast. Jun 26 12:32:16 DocScrutinizer: at least that's what i'd done. I'm not sure if it's better or not. Jun 26 12:32:27 mrmoku did in 5min Jun 26 12:35:30 lindi-: if frameworkd is used? Jun 26 12:35:45 lindi-: Debian needs updated bindings. Jun 26 12:37:28 DocScrutinizer: (picture) i understand that this story is quite uncomfortable for you. :-/ I think Ainu wouldn't have done it if he understood that you're talking really seriously. So many nasty arguments happen because of misunderstanding :-/ Jun 26 12:39:39 I downloaded MokoMakefile and tried to run 'make qemu' but there is no rule for qemu Jun 26 12:39:46 what's the matter? Jun 26 12:40:00 PaulFertser: I'm just concerned about now a lot of ppl seem to feel pissed, while nobody cared the multiple times before when I used a more polite wording Jun 26 12:40:50 DocScrutinizer: probably it was so polite they thought you're joking. Sometimes politeness is counter-productive. Jun 26 12:41:25 Proton23: almost nobody is interested in qemu emulation of our devices. Jun 26 12:44:49 PaulFertser: Has this option been thrown out? Jun 26 12:46:26 PaulFertser: I guess the pages about QEMU emulation can be deleted so Jun 26 12:46:40 Proton23: quite possible so. Are you sure you need it? There's a reason why almost nobody is interested in virtual FR. You can have exactly the same software on host, why do you want emulation? Jun 26 12:48:10 I have no FR but would like to compile a program and test it Jun 26 12:48:33 PaulFertser: what kind of program? Jun 26 12:48:37 blah, Proton23 Jun 26 12:49:13 a little self-written qt program Jun 26 12:49:23 Proton23: why not run it on your PC? Jun 26 12:49:40 For essentially everything - you can just run it on your PC Jun 26 12:49:46 if you've got a GSM modem connected Jun 26 12:50:17 Hurrah my freerunner has returned (hopefully buzzfixed) :) Jun 26 12:50:39 it runs on my desktop, now I would like to give it a try on the FR Jun 26 12:51:11 tig|: why shouldn't it be buzzfixed? Jun 26 12:51:17 Proton23: qemu is not FR Jun 26 12:51:35 Proton23: what is this app? Jun 26 12:51:39 DocScrutinizer well I sent it to be buzzfixed but have not tested it yet :) Jun 26 12:51:40 lindi-: but emulats it? Jun 26 12:51:49 I am sure it will be fine :) Jun 26 12:52:09 Proton23: yes - but in almost every way - a 640*480 window on your screen, and you backed off several feet - is exactly the same Jun 26 12:52:29 Admittedly touchscreen interfacing is a bit of a problem - but that's not solved by qemu Jun 26 12:52:47 nor is working ouyt the speed, as qemu can be much slower than the real hardware Jun 26 12:53:13 lindi-: "qemu is not FR" :D Jun 26 12:53:26 + Jun 26 12:54:35 How to test if the cross compiling worked fine - without a FR? Jun 26 12:55:08 Proton23: you're supposed to release the source, not the binary ;) Jun 26 12:55:29 PaulFertser: :D Jun 26 12:55:34 I know Jun 26 12:55:40 Proton23: what's this app? Jun 26 12:55:52 (Though I'd be quite happy with a binary for some things) Jun 26 12:56:01 it's just a test case Jun 26 12:56:34 if compiling a qt program works Jun 26 12:56:45 before getting a FR Jun 26 12:56:57 Proton23: weird reasoning tbh Jun 26 12:58:26 PaulFertser: Maybe I AM weird ;) Jun 26 12:58:36 Proton23: obviously it's more a testcase for qemu than for croscompiling ;-) Jun 26 12:59:46 DocScrutinizer: Running on QEMU doesn't mean that it will run on the FR? Jun 26 12:59:54 Proton23: especially since crosscompiling works flawlessly for entire distributions :-D Jun 26 13:01:26 Proton23: more the other way round: it's obviously much more likely a qemu based app will fail, rather than crosscompiling had any issues Jun 26 13:04:04 DocScrutinizer: OK, but how to test without a FR? Jun 26 13:04:22 test WHAT? Jun 26 13:04:47 some programm Jun 26 13:06:05 test for your app is coded correctly? run on X86. -- test that it will run on FR? Run on FR, there's no better way (not even a worse way) Jun 26 13:08:01 "test that it will run on FR? Run on FR, there's no better way (not even a worse way)" but I don't have a FR :) Jun 26 13:08:43 so why do you care about any app actually running on FR then? Jun 26 13:09:59 Proton23: let me put it this way: your desktop PC with native X and all the OM-libs is the better qemu Jun 26 13:11:14 DocScrutinizer: Ok, that's something I can work with :) Jun 26 13:16:11 hi, does anyone know if a vanilla fso-console image is aeroplane safe i.e. gsm and wifi start up powered down? Jun 26 13:18:36 solar_george: any cellphone is airplane safe. Jun 26 13:19:26 i mean to keep the stewards happy, i assume that it is as i'm only running omgps Jun 26 13:19:35 solar_george: they can't check anyway. Jun 26 13:19:45 good point Jun 26 13:20:20 solar_george: basically you can just kill ophonekitd and it should release all resources and everything will be turned off. Jun 26 13:20:25 solar_george: (that's for SHR) Jun 26 13:21:11 i'm just using a plain fso-console image + X + omgps to keep power use to a minimum Jun 26 13:21:28 * PaulFertser imagines solar_george talking to stewards like "But you don't understand, that's fso-console image, it has everything turned off by default. Let me show you frameworkd log" Jun 26 13:21:37 :) Jun 26 13:22:22 imagine if it just booted into the console how that could scare people Jun 26 13:22:47 they'd think i was trying to hijack the plane by computer Jun 26 13:23:24 via wall-sockets :) Jun 26 13:25:53 :/ Jun 26 13:26:01 solar_george: I thought that only worked with macbook Jun 26 13:26:15 Don't even joke about that stuff around the American TSA security folks. Jun 26 13:26:29 i wonder how much battery life could be got while doing a gps track if i don't use X Jun 26 13:26:44 I've gotten maybe ... Jun 26 13:27:01 You won't end up in jail, or even charged with anything, but they're just a bunch of ***es looking for an opportunity to give trouble to someone... Jun 26 13:27:02 solar_george: i don't think using or not using X is much of a difference. Jun 26 13:27:07 same with the british airports Jun 26 13:27:22 mwester: yeah, quite stupid assholes afaik Jun 26 13:27:53 mwester: if only the usa general public would be smart enough to understand that Jun 26 13:28:29 PaulFertser yes i guess so, i'm just trying to get the maximum battery life to get a gps track while trekking away from power sources Jun 26 13:29:03 I'd not be looking at FR personally Jun 26 13:29:19 solar_george: why do you think using X has any effect? Jun 26 13:30:09 just a guess along the lines of greater cpu usage maybe Jun 26 13:31:13 Does the glamo competely powersave if the screen is off? Jun 26 13:31:21 If so, it shouldn't make a real difference Jun 26 13:31:23 probably just been given wrong impression by using a laptop - at the console you tend to use less power-intensive programs but that's different on fr Jun 26 13:33:30 SpeedEvil: hm, glamo can't completely powersave because it's used to access uSD. Jun 26 13:34:05 oh Jun 26 13:34:16 in that case would running from NAND make any appreciable difference do you think Jun 26 13:34:53 Extrapolating other stuff I've done, I think 12h on GTA01 is reasonable - with some hacks. Jun 26 13:34:57 solar_george: no good idea. Jun 26 13:35:05 With GSM off Jun 26 13:36:08 thats about what i got on GTA02 with console-image + omgps sitting on a windowsill (just to test battery life) Jun 26 13:36:38 hi Jun 26 13:51:38 does Qi check for bootable partitions based on the position on the disk (start>end) or in an order like the kernel? Jun 26 13:53:20 by like the kernel i meant the order in /dev/ Jun 26 13:53:58 ignore me i'm typing nonsense will test and findout myself Jun 26 14:03:59 solar_george yes, a kernel on the first partition, then the second, then the third Jun 26 14:08:45 solar_george: in order of partition table Jun 26 14:09:24 solar_george: /dev has no explicit ordering. It's just 'ls' that creates a collation Jun 26 14:09:57 does that mean even if /dev/sdc1 is at the end of the disk Qi will try it before /dev/sdc2 Jun 26 14:10:25 probably Jun 26 14:10:26 solar_george: nope. Jun 26 14:10:33 thats good Jun 26 14:10:36 ignore me Jun 26 14:11:06 qi is using the sequence in partition table Jun 26 14:11:34 partition table order isn't always that on disk Jun 26 14:11:49 sdc4 canbe at teh beginning of a disk, and sdc1 at the end Jun 26 14:12:07 it's not usually - as most tools don't do that - but there isn't any explicit reason why not Jun 26 14:12:08 i'll just try and see what happens in my case Jun 26 14:12:21 sdc is a kernel /dev thing depending on udev rules etc Jun 26 14:12:55 i know that sdc1 is after sdc2 as i've resized it Jun 26 14:13:02 the partition table sequence is unambiguous Jun 26 14:13:24 check with cfdisk or the like Jun 26 14:15:48 I'm not entirely sure about how Qi handles extended partitions, but AIUI it will "climb up" the sequence in MBR/part-tab by physical byte adress of the particular partition entry, while skipping fs-types not appropriate for boot Jun 26 14:16:52 PaulFertser for sure can tell more precisely Jun 26 14:17:35 DocScrutinizer: i don't think it supports extended partitions at all. Jun 26 14:17:49 * DocScrutinizer thinks the same Jun 26 14:18:10 * solar_george has seen that on the wiki Jun 26 14:20:42 SpeedEvil: sure part-tab first entry can define a part at end of hd, while second entry defines one at start of hd. I don't think that has any impact on the way Qi works though Jun 26 14:21:29 damm qi booted sdc1 first even though its at the end of the disk Jun 26 14:21:43 i.e. I'd assume Qi still checking first *entry* first Jun 26 14:21:53 yeah - that was what I was trying to get to. Jun 26 14:22:00 sorry for confuzzling. Jun 26 15:13:36 hi, i'm installing android koolu beta7 on my freerunner. I put all the files on my SD card, but it doesn't install Jun 26 15:14:48 how exactly doesn't it install, how far do you get Jun 26 15:17:05 i get a white screen now, before I was getting an error message Jun 26 15:17:50 try pulling the battery out and rebooting Jun 26 15:18:38 i get the messaeg I was not able to find installation files, sorry Jun 26 15:19:01 i created a single FAT partition on my SD card and untarred all the files there Jun 26 15:20:19 this may sound silly but did you remember to unmount the sd instead of just pulling it out Jun 26 15:20:33 have you tried redoing the copy Jun 26 15:21:12 i copied the files when the SD was on my freerunner Jun 26 15:21:21 and shut it down cleanly Jun 26 15:21:53 i try redoing the copy and try again Jun 26 15:22:07 afraid i really don't have a clue - i've never had any luck with the koolu android images Jun 26 15:22:28 i've never got beyond the koolu splash screen Jun 26 15:22:29 you've tried doing it before ? Jun 26 15:22:38 ok Jun 26 15:23:00 only with the older ones not beta7 Jun 26 15:23:06 not tried that yet Jun 26 15:23:09 okie.. Jun 26 15:23:22 any idea which of the images is most suitable for a daily phone Jun 26 15:23:40 no, i tend to use SHR Jun 26 15:24:17 okie i did use SHR testing until now, not as a daily phone though, thought I'll try android Jun 26 15:25:06 i've heard people recommend these images in the past http://panicking.kicks-ass.org/download/ Jun 26 16:01:14 openmoko: 03rehar * r5219 10/trunk/gta02-core/ (5 files in 2 dirs): Jun 26 16:01:14 openmoko: swaped pin LRC and BCLK in symbol WM8753LGEFL Jun 26 16:01:14 openmoko: added CPU parts to USB and Audio sheet Jun 26 16:42:00 openmoko: 03daveb * r5220 10/trunk/gta02-core/usb.sch: use USB_DN / USB_DP labels for Bluetooth usb nets. Jun 26 16:55:13 openmoko: 03daveb * r5221 10/trunk/gta02-core/usb.sch: revert unintended elements of the last commit. Jun 26 17:39:52 TAsn: :){ Jun 26 17:40:38 ? Jun 26 17:40:56 PaulFertser, ? Jun 26 17:41:23 TAsn: that's a smiling guy with traditional jewish beard Jun 26 17:41:33 lol. Jun 26 17:41:53 n1 Jun 26 17:42:09 TAsn: n1? Jun 26 17:42:18 :)} <--- also looks like a jewish orthodox Jun 26 17:42:21 n1 = nice one ;) Jun 26 17:42:29 Yep :D Jun 26 18:09:23 hi Jun 26 18:09:51 I got a small "problem" with my fr (doesn't matter which distro) Jun 26 18:10:22 when I rotate X to landscape mode, my fr makes a strange beep noise Jun 26 18:10:29 Darkas: same for me. Jun 26 18:10:31 is this normal or expected? Jun 26 18:10:39 Darkas: is it kernel-dependent? Jun 26 18:10:42 Darkas: just don't rotate it :-) Jun 26 18:10:48 :) Jun 26 18:10:49 Darkas: i'll try to reproduce it with an older kernel. Jun 26 18:11:03 I haven't tried it with QTopia yet Jun 26 18:11:05 Darkas: seriously, it's not stable Jun 26 18:11:14 Darkas: no, really. I didn't have it in January. Now i do have it. I wonder if some kernel change made it emit that hum. Jun 26 18:11:35 Darkas: i'm talking about kernels, please answer if you tried rotating with some old kernel. Jun 26 18:11:39 lindi-, I like that feature for reading pdfs in the bus ;) Jun 26 18:11:49 PaulFertser, no Jun 26 18:12:11 I got 2.6.29-rc3-de84609-09314-gde84609 Jun 26 18:12:14 Darkas: i like it too but i value stability more :) Jun 26 18:12:21 BTW, i now know for sure that at least 2 gta01s were somewhere near moscow. So i still have chances to obtain it soon. Jun 26 18:12:36 Darkas: can you try some older kernel please? Like from before the NY? Jun 26 18:13:09 lindi-: do you have that high-pitch hum too? Jun 26 18:13:36 PaulFertser, I'll try (but wtf is NY?) Jun 26 18:13:45 lindi-: i for sure didn't have it. Need to boot some older kernel to verify. Would be an easy one to bisect :D) Jun 26 18:13:48 Darkas: new year Jun 26 18:13:54 ahh :D Jun 26 18:15:32 PaulFertser: no but I think Mirv had Jun 26 18:17:26 PaulFertser: have you managed to get the headset work with your laptop btw? currently bluetooth-wizard does not show the device at all but hcitool scan does Jun 26 18:18:19 lindi-: which headset? The one i borrowed from my friend (that didn't emit sound with FR) or the one i borrowed from my co-worker? Jun 26 18:18:50 PaulFertser: BH-203 Jun 26 18:18:56 PaulFertser: ah, well, any headset :) Jun 26 18:19:47 lindi-: the one from my friend works without any issues with my laptop (broadcom bt chip). The one from co-worker i tested only with my FR and it worked. Jun 26 18:20:22 PaulFertser: (laptop & usb-bt dongle & BH-203 & bluez 4 & simple-agent) => arecord outputs nothing, hcidump shows the audio data as SCO frames Jun 26 18:20:40 PaulFertser: kernel logs hci_scodata_packet: hci0 SCO packet for unknown connection handle 6 Jun 26 18:21:05 lindi-: are you sure you want to have alsa in the mess? why not try directly? Jun 26 18:21:35 PaulFertser: what was the direct way? Jun 26 18:22:12 lindi-: hstest Jun 26 18:22:20 lindi-: from test directory in bluez4 sources Jun 26 18:22:23 PaulFertser: no such binary in debian Jun 26 18:22:32 lindi-: no such debian without compiler Jun 26 18:22:40 PaulFertser: sigh. i wonder why these tools are not included in the package :( Jun 26 18:22:45 lindi-: me too Jun 26 18:22:55 and why I need to use some "code example" to send the PIN Jun 26 18:23:06 "code example" that has no manual page or comments Jun 26 18:23:18 (talking about /usr/share/doc/bluez/examples/simple-agent) Jun 26 18:23:24 lindi-: ask holtmann :-/ Jun 26 18:23:34 lindi-: but i think i know the answer Jun 26 18:23:52 gnome? Jun 26 18:24:27 lindi-: it's supposed that any guy who's going to make a good UI for bluez knows how to understand code examples. And others should use already available (gui) tools. Jun 26 18:24:27 lindi-: don't even get me started Jun 26 18:24:47 lindi-: I also love the commentary in the docs there Jun 26 18:24:48 PaulFertser: aiee Jun 26 18:24:58 lindi-: If you find that you ABSOLUTELY MUST pair your device WITHOUT X Jun 26 18:25:01 lindi-: and that's it's a pity that there's no good cli tool for that but bluez devs are busy enough to not bother about it. Jun 26 18:25:03 ridiculous Jun 26 18:25:18 python -c '....' Jun 26 18:25:20 PaulFertser: That's not true, it used to work just fine, and they broke it Jun 26 18:25:22 johnsu01: well, python is available almost everywhere. And their example script works. Jun 26 18:25:30 johnsu01: what used to work fine? Jun 26 18:25:31 but the example script is not in PATH Jun 26 18:25:39 you can't call it from an ifup script Jun 26 18:25:43 PaulFertser: You used to be able to pair without X with no trouble Jun 26 18:25:59 and no sample script or random .c file in /usr/share/doc to compile Jun 26 18:26:16 johnsu01: with bluez3? Bluez4 is complete api redesign. So it's ok to loose something during transition. Jun 26 18:26:37 PaulFertser: Actually this change happened during bluez3 Jun 26 18:26:55 johnsu01: we can stop this argument by 1) figuring out what examples we need to work with BT 2) rewriting these as proper tools 3) packaging these tools :-) Jun 26 18:27:09 just using getopt will help a lot Jun 26 18:27:15 lindi-: Yeah, all true Jun 26 18:27:51 lindi-: for me simple-agent and hstest.c are enough to not bother. Jun 26 18:28:03 for me not :) Jun 26 18:28:17 lindi-: and in fact i'm _very_ frustrated now when i hear about BT headsets. Jun 26 18:28:23 when I rotate X to landscape mode, my fr makes a strange beep noise Jun 26 18:28:23 Darkas: same for me. Jun 26 18:28:33 PaulFertser, Darkas, just checked, nothing like that here. Jun 26 18:28:34 shr latest Jun 26 18:28:37 (unstable) Jun 26 18:28:39 * mwester thinks part of Linux died when bluez and connman were "released". Jun 26 18:28:49 lindi-: those fuckers asked for support for months. And now they ignore working support! Jun 26 18:28:55 mwester, lol. Jun 26 18:29:04 TAsn: that's not beep noise, that's high-pitch hum. Jun 26 18:29:19 I don't have that either Jun 26 18:29:28 Yep, it's become Windows now: No GUI == Not useful. Jun 26 18:30:10 PaulFertser: what working support? where? ;) Jun 26 18:30:20 mwester: that's not very true for a guy who knows a bit of python. Jun 26 18:30:33 johnsu01: I managed to get my BH-203 work with friend's laptop that has integrated bt chip Jun 26 18:30:35 lindi-: i tested today and bt headset worked for me as advertised. Jun 26 18:30:37 But I do agree with lindi- that we need to just create a simple C command line set of tools that we can use to debug and test without a GUI, after all it's darned hard to do automated regression testing against a GUI,a nd we really need to start doing that. Jun 26 18:30:47 johnsu01: but I can't get it to work with external usb bt dongle Jun 26 18:31:06 PaulFertser, when Linux replaces /bin/sh with /bin/python, then I will agree with you on that point. Jun 26 18:31:09 lindi-: and i used teh instructions i wrote. Jun 26 18:31:25 PaulFertser: yep but I'm still talking about my laptop :) Jun 26 18:31:33 mwester: if you think i like the fact that bluez4 lacks proper hacker-friendly ui, you're wrong. Jun 26 18:31:35 sorry for the confusion Jun 26 18:31:39 lindi-: hm, I actually have bluetooth working on my laptop again (new laptop), so I can get back to trying that Jun 26 18:31:58 PaulFertser: should I use patched hciconfig on my laptop to get audio data over usb? Jun 26 18:32:08 lindi-: and i'm talking about endlessly whining users who "wanted" support, got it months ago and do nothing, not even bugreporting to improve it. Jun 26 18:32:18 lindi-: is it CSR-based device? Jun 26 18:32:20 PaulFertser, I just think that you like to take up a contrary position to me always! ;-) :D Jun 26 18:32:43 PaulFertser: the usb bt dongle? hciconfig -a does not mention CSR at least Jun 26 18:32:45 lindi-: then check output of hcitool hciX revision | grep SCO Jun 26 18:33:07 PaulFertser: does not print anything Jun 26 18:33:12 lindi-: that trick is only for CSR chips. And some laptops/usb dongles really need tweaking for SCO to work. Jun 26 18:33:17 mwester: :D Jun 26 18:34:36 mwester: but you know, i spent quite some time digging these bluetooth issues, tricks about SCO routing, proprietary CSR oddities etc. Tried a lot to make my friend's headset work just for the sake of testing/bugreporting. And i feel dissatisfied that this work that seemed to be useful is not because end-users lied to me that they needed it. Jun 26 18:35:04 johnsu01: PaulFertser: talk about robust tools: http://paste.debian.net/40354/ Jun 26 18:36:52 lindi-: haha Jun 26 18:37:04 PaulFertser, I am as puzzled as you about that -- after all, it is bluetooth that will solve the echo and buzz issues, for those that cannot (or will not) apply the hardware fixes. And it is the only way to get the buzzfix if you wish a headset. I need bluetooth to make the devices useful; I cannot understand why nobody else seems to care? Jun 26 18:38:05 mwester: i did report that hcid crashes ages ago with bluez3 Jun 26 18:38:31 I know that there have been some who have cared... Jun 26 18:39:59 lindi-: haha, i think the syntax is wrong Jun 26 18:40:10 lindi-: hcitool -i hci0 revision ? Jun 26 18:40:10 PaulFertser: "hciconfig hci0 revision | grep SCO" prints "BD Address: 00:1A:9F:92:97:3C ACL MTU: 1021:8 SCO MTU: 64:1" Jun 26 18:40:33 "hcitool -i hci0 revision" does not print anything Jun 26 18:40:40 lindi-: no, hciconfig is what matters Jun 26 18:41:59 PaulFertser: so can you figure out if this means it is CSR chip or not? Jun 26 18:42:28 lindi-: can you pastebin the whole hciconfig output? Jun 26 18:42:39 PaulFertser: hciconfig -a? Jun 26 18:42:55 $ sudo hciconfig -a | nc paste.dyndns.org 1234 Jun 26 18:42:56 http://paste.debian.net/40356/ Jun 26 18:43:19 lindi-: Manufacturer: Broadcom Corporation Jun 26 18:43:45 PaulFertser: ok so this needs a different set of tricks? :/ Jun 26 18:44:13 lindi-: probably none, at least i think i remember only about CSR tricks from my googling. Other chips didn't have that kind of problems. Jun 26 18:44:17 PaulFertser: or I'll just write a perl script that parses the audio data from the output of "hcidump -X -V" Jun 26 18:44:44 since hcidump can see it, kernel just complains all the time and does not send it to arecord Jun 26 18:44:47 lindi-: better use hstest.c Jun 26 18:45:00 we really need a package for that Jun 26 18:45:18 lindi-: i'm very bad at packaging. Jun 26 18:47:39 johnsu01: you wouldn't happen to know the debian bluetooth maintainers? ;) Jun 26 18:48:05 lindi-: no, sorry :( Jun 26 18:52:38 PaulFertser: hstest does not seem to have any docs either :) Jun 26 18:52:57 lindi-: sources are self-explaining Jun 26 18:53:21 lindi-: i know all that lack of documentation and tools is bad. But semi-motivated hacker can get without... Jun 26 18:53:31 PaulFertser: it prints "hstest record [channel]" has usage if I try "hstest record a.wav" Jun 26 18:54:01 lindi-: sure, you need the third argument, HS address. Jun 26 18:55:10 PaulFertser: it prints "Voice setting: 0x0060" and does not exit Jun 26 18:55:35 lindi-: probably that's an intended behaviour even Jun 26 18:57:15 seems bluetoothd has crashed Jun 26 18:58:00 hmm, now the whole laptop froze Jun 26 19:00:30 PaulFertser: btw, do you run HAL? Jun 26 19:00:42 lindi-: on FR yes Jun 26 19:01:10 PaulFertser: if you strace it, do you see that it wakes up every time you touch the screen? Jun 26 19:02:41 lindi-: unfortunately i don't have apm emulation disabled so for me it wakes about every second to read /dev/apm Jun 26 19:16:01 I just noticed that my NAND menu only has "Boot"... there are no other options. what gives? :-) Jun 26 19:16:17 I flashed it with gta02v5_and_up-u-boot.bin at one point Jun 26 19:16:32 SM2k: a clear hint you should start using Qi ;) Jun 26 19:16:37 eh? Jun 26 19:16:52 SM2k: Qi is fast and reliable and supported. Jun 26 19:17:05 but the uBoot would be the same, no? Jun 26 19:17:10 it's only that one phone. :-) Jun 26 19:17:12 SM2k: u-boot on FR is slower, depends on enviromnet and is less supported. Jun 26 19:17:30 this Qi.... is then a new device? Jun 26 19:17:42 SM2k: what? Jun 26 19:22:41 sorry. phone call. :-) so Qi is a uBoot replacement. does it still have the USB console thingy? Jun 26 19:22:56 SM2k: no, you're not supposed to need it. Jun 26 19:23:13 I broke my spash screen. I kinda need it. :-) Jun 26 19:23:20 SM2k: no Jun 26 19:23:41 SM2k: Qi boots kernel in 2 seconds. Then the kernel and userspace is responsible to show you anything you want. Jun 26 19:24:45 I assume GTA02-stable-mumble would be a good choice? Jun 26 19:25:07 SM2k: you mean Qi? I think experimental is the best as it's built from the latest sources. Jun 26 19:25:25 ah my bad those others are uImage Jun 26 19:25:26 SM2k: read the wiki page before flashing Qi! Jun 26 19:25:31 sigh Jun 26 19:26:05 PaulFertser: with older kernel hstest prints "Can't connect RFCOMM channel: Connection refused" Jun 26 19:26:06 :-) very well. I'll do that. the moko tends to get cranky with me when I play fast and loose 8-) Jun 26 19:26:33 lindi-: i've heard bluetooth kernel parts are in active development. Jun 26 19:34:34 PaulFertser: and now it magically works. I can't explain this Jun 26 19:34:58 lindi-: hstest works? Jun 26 19:35:02 lindi-: and you got sound? Jun 26 19:35:25 PaulFertser: arecord works, hstest fails Jun 26 19:35:36 lindi-: funny Jun 26 19:35:58 lindi-: anyway, how is it relevant, what is your target? Jun 26 19:36:32 PaulFertser: be able to use bluetooth headset with laptop and with freerunner with GSM and mplayer Jun 26 19:37:14 lindi-: i see. BTW, holtmann recommended me to use very latest kernel for good SCO support. Jun 26 19:44:48 PaulFertser: whoops sorry, it did not work. I had wrong alsa configuration and was recording from real mic... Jun 26 19:46:25 lindi-: it looks like something is wrong with you today ;) Jun 26 19:47:33 i blame the heat and lack of time :) Jun 26 19:51:06 hmm, now it really works (yes I spoke to the headset in another room) Jun 26 20:21:13 openmoko: 03daveb * r5222 10/trunk/gta02-core/ (cpu.sch io.sch): prep cpu sheet, move cpu timer block to CPU sheet. Jun 26 20:22:37 is there a fix for h Jun 26 20:22:47 the non-working apps in SHR? Jun 26 20:23:25 ie openmoko-messages, openmoko-contacts, and phonelog? Jun 26 20:23:41 im on a fresh install, but these applications dont work Jun 26 20:24:02 try a more recent SHR unstable Jun 26 20:24:29 the last testing I tried was kinda broken, but have been using a fairly recent unstable with good results Jun 26 20:24:40 shr-testing is not the recommended version? Jun 26 20:24:43 okidoki Jun 26 20:24:49 ill get shr-unstable then Jun 26 20:24:56 it varies as to what is recommendde Jun 26 20:24:59 recommended Jun 26 20:25:06 could be one thing one day, another the next Jun 26 20:25:15 just need to keep up with the developers Jun 26 20:25:51 how? :-))) I only check in to the wiki every now and then to see if there is anything new Jun 26 20:26:17 hang around here for a while Jun 26 20:26:22 is what I do Jun 26 20:29:37 http://build.shr-project.org/shr-unstable/images/om-gta02/shr-image-om-gta02.jffs2 and http://build.shr-project.org/shr-unstable/images/om-gta02/uImage-om-gta02-latest.bin Jun 26 20:29:47 are those the latest unstable-versions? Jun 26 20:30:00 ndnihil, xqo: if you want to listen what SHR developers say, I suggest #openmoko-cdevel ;) Jun 26 20:30:25 alright, ill try that. thanks Jun 26 20:30:37 ndnihil, xqo: but don't bother them too much or else they'll become as angry and rude as me Jun 26 20:30:54 :) Jun 26 20:31:05 how about those images? Jun 26 20:32:30 xqo: those should be fine, but double check the dates Jun 26 20:41:46 alright, thanks for all the halp Jun 26 20:42:24 Hi! :) Jun 26 20:42:39 I just received a neo1973 and I have some questions... Jun 26 20:42:55 sabrod: really? Jun 26 20:42:59 sabrod: where did you get one? Jun 26 20:43:12 I'd like to know if there actually IS any accelerometer in it... Jun 26 20:43:40 Heh, yes, ebay!! ( I'm always a little late ;) ) Jun 26 20:44:06 It is niiice!! :p Jun 26 20:45:06 sabrod: i don't think so. gta02 has two but gta01 seem to lack it. Jun 26 20:45:10 no accelerometer in the neo1973 Jun 26 20:45:12 I've seen a blog post showing some use of a neo1973 with "its" accelerometers... Jun 26 20:45:46 sabrod: btw are you capable of some hardware hacking? Jun 26 20:45:47 A guy that play Duke Nukem!? ( well he prob. have a neo case on a FR... :/ ) Jun 26 20:46:06 Hehe!! No I don't think so... Jun 26 20:46:07 It would be clearer if you use the "gta0x" terminology instead of "neo" Jun 26 20:46:14 neo1973 == gta01 Jun 26 20:46:19 neo freerunner == gta02 Jun 26 20:46:22 very different devices. Jun 26 20:46:23 Ok. sorry. Jun 26 20:46:36 Yep! Jun 26 20:47:05 If you know which you have, the wiki should be accurate, if you search/read using the "gta0x" term instead. :) Jun 26 20:48:28 And sooner today... guess what? I have found in my an old case in my room, one of these first kind of acceleromers sold by nokia as a 3220 option for 2 games! :D Jun 26 20:48:48 in a case in my room, sorry! :) Jun 26 20:48:50 sabrod: which interface do they use? Jun 26 20:49:05 Hehe, 4 contacts! :) Jun 26 20:49:21 sabrod: probably spi? Jun 26 20:49:25 Everything is shielded. Jun 26 20:49:46 most of the board is shielded, there is 2 big shields in fact. Jun 26 20:49:50 sabrod: gta01 has spi testpoints near the debugboard connector. Jun 26 20:50:04 Wow! ;) Jun 26 20:50:12 Shielded accelerometers? strange Jun 26 20:50:39 Yes... Jun 26 20:51:15 Maybe because there is not much space on the case ( it was a special case, a lot bigger... :) ) Jun 26 20:51:50 So in prevision the user will sit on the case / phone / accel. board ?? Jun 26 20:53:43 Ha, now I have it in front of me, it's 5 pins... Jun 26 21:12:02 openmoko: 03daveb * r5223 10/trunk/gta02-core/io.sch: labels for nand usage selection Jun 26 22:11:25 openmoko: 03daveb * r5224 10/trunk/gta02-core/ (8 files in 2 dirs): Jun 26 22:11:25 openmoko: New componant, ht110 (single LED) Jun 26 22:11:25 openmoko: initial work for CPU sheet Jun 26 22:24:59 is it possible to run fedora, ubuntu on free runner? Jun 26 22:25:21 fedora or ubuntu* Jun 26 22:29:49 possible yes, it just depends on how much work you want to do Jun 26 22:30:11 there is a debian dist packaged for it, that might be a bit quicker for you Jun 26 22:32:25 ermm.. okay, thanks Jun 26 22:35:54 yeah, the debian is prebuilt and has installation scripts Jun 26 22:36:04 for others you'd have to put in nontrivial effort Jun 26 22:36:31 what's the preferred OS for OM? Jun 26 22:37:24 and there is any installations/updates manager..? Jun 26 22:38:46 well, debian upgrades with apt-get in principle. For the more OMish things there's opkg Jun 26 22:39:21 I'm probably going for debian again when I can bother; last time it didn't function as a phone too well Jun 26 22:40:35 Hi! Jun 26 22:40:54 what's the preferred OS for OM? the mose stable one Jun 26 22:41:04 most* Jun 26 22:41:25 Ddorda, SHR Jun 26 22:41:44 So i got the most recent SHR-unstable images, and again an application crashes. openmoko-messages crashes with no error. Is there a fix for this somewhere? Jun 26 23:07:11 do you think OM is stable enough to be used as a main cell phone? Jun 26 23:07:50 shr is Jun 26 23:07:54 is it stable enough? how long can it work without charging? Jun 26 23:08:06 20090617 unstable Jun 26 23:08:30 and how long can it work without charging? Jun 26 23:08:45 well usage depends i go 2 days with some usage Jun 26 23:09:03 i have Qi Jun 26 23:09:08 shouldn't expect over a day not to be disappointed Jun 26 23:09:17 uboot works as well Jun 26 23:09:33 unsuspend times are a lot quicker now Jun 26 23:10:16 neovento worked ok Jun 26 23:10:48 i thought the battweery was used too fast but that was some time ago Jun 26 23:11:46 so i need to charge it everyday then Jun 26 23:11:49 fine with me Jun 26 23:11:52 yup Jun 26 23:12:05 just like the iphone:) Jun 26 23:12:09 and how long does it takes to load it? Jun 26 23:12:22 sdoftware? Jun 26 23:12:26 battery? Jun 26 23:12:31 my gun? Jun 26 23:12:48 to open it, to start it, to run it, to boot it, w/e Jun 26 23:13:05 u have 1? Jun 26 23:13:42 i don't. i think about buying one Jun 26 23:13:52 Qi+shr is 2 minuits to boot unsuspend time is 2 sec Jun 26 23:14:20 android was longer but may not be now Jun 26 23:14:35 and i guess i can make it boot faster if i edit the kernel Jun 26 23:14:52 debian im sure someone else can tell u Jun 26 23:15:42 paroli is in dev right now Jun 26 23:16:11 slackware no one has touched yet Jun 26 23:16:38 choices ,, lots of choices Jun 26 23:17:26 wouldnt suprise me if 2007.1 makes a complete comeback Jun 26 23:17:54 simplicity is devine Jun 26 23:18:11 an faster than shr Jun 26 23:24:22 How can I check which Koolu release I'm currently running? Jun 26 23:27:57 Hi folks! Jun 26 23:28:31 I've got audio problems on my freerunner using SHR unstable. When I use aplay to play a file it hangs, and 'dmesg' shows "BUG: stream = 0, pos = 0x2588000, buffer size = 0x4000, period size = 0x400" repeatedly. Jun 26 23:34:43 ok, killing some apparently alsa-hogging process seemed to have worked ;-) Jun 26 23:40:43 what's the OM screen size? Jun 27 00:12:47 2.8" Jun 27 00:36:23 openmoko: 03daveb * r5225 10/trunk/gta02-core/ (6 files in 2 dirs): one additional componant, and CPU is done. ish. **** ENDING LOGGING AT Sat Jun 27 02:59:56 2009