**** BEGIN LOGGING AT Mon Aug 31 02:59:57 2009 Aug 31 03:28:54 openmoko: 03rehar * r5591 10/trunk/gta02-core/docs/ecn/ (7 files): Aug 31 03:28:54 openmoko: - added commit to ecn0010 and ecn0015 Aug 31 03:28:54 openmoko: - removed ecn0020 from open ecn list Aug 31 03:28:54 openmoko: - added new ecn0036 "Removal of decoupling capacitor C3020" Aug 31 03:28:54 openmoko: - added new ecn0037 "Route ROUT1/LOUT1 back to LINE1/LINE2 input" Aug 31 03:28:55 openmoko: - added new ecn0038 "Loudspeaker EMI filter" Aug 31 04:23:14 just get another bum freerunner an replace your screen an case Aug 31 04:24:55 Slaxxer: but most dead ones have dead screens Aug 31 04:26:51 quit the horror stories, i have open openmoko on my desk and i don't want to break it :) Aug 31 04:28:19 I have a case that is guaranteed to protect your FR from anything you as an unaided human can do to it. Aug 31 04:28:55 Collection only, and bring transport capable of lifting a 500Kg block of concrete. Aug 31 04:29:09 SpeedEvil: heh Aug 31 04:29:36 lets see it Aug 31 04:29:45 SpeedEvil: i only have a water-proof case which I use while in forest Aug 31 04:31:07 so these august 22 images have improvement? Aug 31 04:31:14 shr -u Aug 31 04:32:50 prolly doesnt take 5-6 lbs to crack a screen anyways especially if its a sharper object like a pencil Aug 31 04:33:35 or are they still 090808 Aug 31 04:34:54 there are some defective freerunners with gps bad instead of the screen Aug 31 04:35:07 i bought 1 Aug 31 04:35:18 75 bucks Aug 31 05:00:29 * blindcoder is using his freerunner as a gps and mobile gaming platform atm ;) Aug 31 05:45:11 Digital_Pioneer: going to leave us? :( Aug 31 06:28:15 what is it Aug 31 10:26:48 good morninig Aug 31 10:26:55 i Aug 31 10:27:04 good afternoon Aug 31 10:31:50 good nfdndkfn Aug 31 11:54:03 PaulFertser: ping :D Aug 31 11:54:52 naeg: pong Aug 31 11:55:15 PaulFertser: do you have any idea how i can get your script working with xev? Aug 31 11:55:20 just packaged it... Aug 31 11:55:53 naeg: btw, since then i research two alternatives: xmonobut Aug 31 11:56:26 naeg: or xkeylaunch + stuff Aug 31 11:57:42 PaulFertser: one is enough i think ;) Aug 31 11:57:53 xmodmap is ok, we already have that Aug 31 11:59:04 xev | grep "button 1" takes too much time for output Aug 31 12:01:43 brb, reboot Aug 31 12:10:26 PaulFertser: I'm afraid so. Unless I can find a really KILLER deal on a A7 or whatever we're up to now. Aug 31 12:11:27 Digital_Pioneer: what will you do instead of FR fun? Aug 31 12:12:12 PaulFertser: Submit to a life of boredom with an Android phone, I think. Heheh Aug 31 12:12:31 Digital_Pioneer: probably wait for N900? Aug 31 12:13:22 PaulFertser: xev | grep -q -m 1 button takes few secs...any other idea? Aug 31 12:13:56 naeg: well, let me try Aug 31 12:14:42 you'll get a thanks to in that wiki page ;) Aug 31 12:15:02 PaulFertser: I keep hearing about it, but IDK what it is... Aug 31 12:15:09 Or how much it is. :P Aug 31 12:18:16 naeg: cat /dev/input/event1 | head -c1 Aug 31 12:18:39 Digital_Pioneer: search the internet, enough info is available already. Aug 31 12:19:15 naeg: i mean "head -c 1 /dev/input/event1" Aug 31 12:20:00 hehe, no, the first was more correct Aug 31 12:21:14 PaulFertser: but this requires root access which is ugly Aug 31 12:21:26 we use X as user Aug 31 12:21:29 naeg: indeed Aug 31 12:22:37 PaulFertser: xinput sounds good :) Aug 31 12:27:23 PaulFertser: Maemo, ehh? That would work. But how much will it cost? :P Aug 31 12:27:48 Digital_Pioneer: awfully much, $600 or so :( Aug 31 12:27:52 but xinput get-feedback Touchscreen gives me always some failed request error :( Aug 31 12:28:11 Ouch. Aug 31 12:28:40 naeg: i lack xinput here Aug 31 12:31:51 PaulFertser: on host too? here it is in xorg-server-utils Aug 31 12:32:55 PaulFertser: but with xinput test Touchscreen | grep button it also takes few seconds Aug 31 12:33:07 xinput test gives me all events Aug 31 12:35:06 xinput ...|head -c1 too Aug 31 12:35:26 naeg: and |cat gives lag too Aug 31 12:35:28 dunno why all this x event things take so much time, without grep/head it report immediatly Aug 31 12:35:46 naeg: it's pipe buffering Aug 31 12:36:32 can i disable this somehow PaulFertser ? Aug 31 12:49:11 PaulFertser: unbuffer! :D Aug 31 13:34:57 PaulFertser: one last thing...unbuffer xinput test Touchscreen | grep "button" -m 1 -q && kill $$ doesnt kill Aug 31 13:44:15 freesmartphone.org: 03mickey 07cornucopia * rb80c8c802689 10/fsodeviced/src/plugins/kernel_idle/plugin.vala: Aug 31 13:44:15 freesmartphone.org: fsodeviced: kernel_idle: resync input devices when input devices appear or disappear during runtime Aug 31 13:44:15 freesmartphone.org: NOTE: This fixes the infamous 'screen blanking bug' when using USB or BT keyboards plugged in after system start Aug 31 14:30:00 ~seen laf0rge Aug 31 14:30:02 laf0rge was last seen on IRC in channel #openmoko, 5d 15m 30s ago, saying: 'doc: yes, that's the case'. Aug 31 14:31:17 So, what does one do when it keeps dropping to a root maintenance shell because of an fsck problem on the sd card? Aug 31 14:31:30 I fscked the card in another machine already Aug 31 14:31:37 but the FR still drops to the shell Aug 31 14:35:09 freesmartphone.org: 03seba.dos1 07framework * r02ede11c3d92 10/framework/subsystems/opimd/ (docs/TODO pimd_generic.py): opimd: fix merging Aug 31 14:35:41 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * rcdb1f5cbeb18 10framework/framework/subsystems/opimd/ (docs/TODO pimd_generic.py): Merge branch 'master' of git@git.freesmartphone.org:framework into dos/opimd-tracking Aug 31 14:38:18 Hi every one. Aug 31 14:39:05 Is there someone that knows how to get a fix with GPS uner SHR unstable from image of 02 of August ? I red many public forums and nothing really work. Aug 31 14:39:13 Clock is set but that's it Aug 31 14:40:12 someone has a clue what to do ? Aug 31 14:40:28 robinson: are you reading nmea data from serial port? Aug 31 14:40:40 the thing is that if I flash the image from july I get a fix fairly quickly Aug 31 14:41:01 sorry to ask but what is "nmea" ? Aug 31 14:41:26 robinson: the protocol used by gps Aug 31 14:41:29 robinson: en.wikipedia.org/wiki/NMEA_0183 Aug 31 14:42:19 oki I used gpspipe -r 127.0.0.1 2947 Aug 31 14:42:26 should I pipe the output ? Aug 31 14:42:30 paste sorry Aug 31 14:42:37 I think I see no fix Aug 31 14:43:43 ooops Aug 31 14:43:52 robinson: tried "delete agps data"? will take w while for first fix after that, but may fix the general issue. also I thin shr-u0808 is recommended atm Aug 31 14:43:52 this morning I did opkg upgrade Aug 31 14:43:56 seems to work now Aug 31 14:44:05 sorry sorry sorry Aug 31 14:44:09 to disturb Aug 31 14:44:20 it was about a month without working Aug 31 14:44:22 np :-) Aug 31 14:44:37 another question Aug 31 14:44:49 is someone in dev for opimd here ? Aug 31 14:45:30 robinson: /join #openmoko-cdevel; ping dos1 Aug 31 14:45:38 thx Aug 31 14:57:23 hi im looking to build a package thats about 30+ megs, normally it takes like a hour on my machine so i think it will take like 10 years on the neo, what is the best way to go about compiling on the neo? Aug 31 14:57:44 what is your machine? Aug 31 14:57:56 easy. don't compile on the neo. Aug 31 14:58:12 The FR may be only 10-20 tims slower. Aug 31 14:58:20 yeah - you can crosscompile. Aug 31 14:58:24 And usually should. Aug 31 14:58:44 crosscompile how? Aug 31 14:58:46 or potentially infinitely slower (i.e, impossible) if the compile is C++ and takes vast amounts of memory Aug 31 14:59:07 i remember compiling C with another compiler i forgot the name but i could share 2 cpus to compile the code Aug 31 14:59:35 tae: make -j2 Aug 31 15:00:19 lindi-: how do i get it to run both jobs on 2 different cpus? Aug 31 15:00:20 make -j2 is effective even with one CPU. I suggest make -j4 Aug 31 15:00:35 how do i add another CPU to the pool of cpus? Aug 31 15:00:47 i have dualcpu so i guess i could do -j6 Aug 31 15:00:48 make -j Aug 31 15:01:17 if i run this on the my computer it wont compile it for arm6 Aug 31 15:01:19 even -j10 can make sense with something like the linux kernel which is many small files. Aug 31 15:01:27 if i run this on my neo it will only use one cpu Aug 31 15:01:37 tae: sigh. I think you have a lot to learn Aug 31 15:01:42 how do i make it use multiple cpus, it wont magically know that i have a computer connected Aug 31 15:01:42 full of assumptions. Aug 31 15:02:57 tae: i had success with distcc, in some cases it can considerably reduce compile time on FR. Aug 31 15:03:15 ah Aug 31 15:03:19 he is confusing the build process with his program. And a lot of other stuff too Aug 31 15:03:19 distcc thats what i used Aug 31 15:03:22 thanks PaulFertser : ) Aug 31 15:03:32 naeg: probably it would if started inside a script? Aug 31 15:04:04 tae: consider using the latest with "pump" (or whatever it called) support, not the one in debian repos. Aug 31 15:04:33 i dont know what "pump" is, is it a flag or a patch? Aug 31 15:04:48 tae: and btw if you disconnect your FR for some reason from your host, it'll happily continue to compile on FR until you connect it again. Aug 31 15:04:50 PaulFertser: working now: http://www.archmobile.org/trac/wiki/AFR/Configuration/Rightclick thanks again! Aug 31 15:04:53 distcc-pump Aug 31 15:04:54 ah found it Aug 31 15:04:57 its in the ubuntu repo Aug 31 15:05:08 PaulFertser: thats nice Aug 31 15:05:20 naeg: community work rules :D Aug 31 15:05:32 ah got it compiling on all cpus :) thanks PaulFertser Aug 31 15:05:42 i had forgotten the name Aug 31 15:05:53 PaulFertser: true, the unbuffer idea is from #archlinux.de :D Aug 31 15:06:43 naeg: i guess to have rightclick emulation with tslib you need a patched X tslib driver, Debian already has that. And some special parameter in xorg.conf Aug 31 15:07:27 PaulFertser: man page documents it Aug 31 15:07:56 PaulFertser: yeah, i read that. i updated the patch to our version and it's not working :( it compiles everything fine, but nothing happens when i make a touch-and-hold-gesture Aug 31 15:08:15 lindi-: no doubt Aug 31 15:08:27 naeg: did you set the parameter? Aug 31 15:08:52 PaulFertser: the parameter depends on the xorg, they built it in, we don't Aug 31 15:09:00 we have to xf86-input-tslib packages Aug 31 15:09:07 two* Aug 31 15:09:11 odin er kul liksom. Aug 31 15:22:35 morning Aug 31 15:56:33 hi Aug 31 16:00:30 Is there any spefic order to the call ID numbers given by frameworkd? Aug 31 16:01:43 openmoko: 03rehar * r5592 10/trunk/gta02-core/components/mkloe: Added antenna WTE06016 to mkloe script Aug 31 16:32:49 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r622056a4ebe0 10framework/framework/subsystems/opimd/ (3 files): opimd: Messages: move to GenericDomain Aug 31 16:53:08 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r42c6e5c54490 10framework/framework/subsystems/opimd/pimd_messages.py: opimd: Messages: fix copy'n'paste issues Aug 31 16:53:21 Suddenly i found out that i almost literally followed these advices (http://xkcd.com/306/) when i went roller-skating with one of my female students :D Aug 31 16:53:36 Except that it wasn't a date. Aug 31 17:06:32 can anybody explain me ..what is "Deep Sleep" in GSM Aug 31 17:14:09 Rategain: well, the most efficient mode wrt powersaving. Aug 31 17:14:21 Rategain: the modem's still available to receive calls/smses in it. Aug 31 17:25:26 wasssup Aug 31 17:25:38 ^ Aug 31 17:25:38 | Aug 31 17:32:20 freesmartphone.org: 03seba.dos1 07framework * r64c827a46d42 10/framework/subsystems/opimd/pimd_generic.py: opimd: fix returning multiple fields with the same data after merging Aug 31 17:32:25 are there any v8 openmokoss? Aug 31 17:32:35 ' Aug 31 17:40:07 Slaxxed: no Aug 31 17:40:18 v7 is the last revision made Aug 31 17:40:24 (probably forever) Aug 31 17:41:49 mickeyl: probably v8 was made but not MPed? Aug 31 17:42:43 PaulFertser: well, that could be Aug 31 17:55:57 yup, i saw something about preparing v8 Aug 31 18:26:00 http://www.artecgroup.com/thincan Aug 31 18:28:03 wow Aug 31 18:28:20 they make some cool stuff Aug 31 18:51:24 freesmartphone.org: 03seba.dos1 07dos/opimd-tracking * r8ed2b2861a44 10framework/framework/subsystems/opimd/ (docs/TODO pimd_generic.py): opimd: fix possible incorrect query results when using _at_least_one mode Aug 31 20:08:54 openmoko: 03werner * r5593 10/trunk/gta02-core/modules/Makefile: - oops, forgot to commit. Thanks, Dave. Aug 31 21:10:15 hi Aug 31 22:21:04 hi, I am having problems compiling glibc-2.9 for ARM, ../sysdeps/unix/sysv/linux/umount.S:8: Error: cannot represent SWI relocation in this object file format Aug 31 22:40:34 ferar: does this help? http://www.sourceware.org/ml/crossgcc/2005-03/msg00198.html Aug 31 22:43:04 thanks Psi, the following link is more descriptive : http://www.mail-archive.com/linux-arm@vger.rutgers.edu/msg04405.html Aug 31 22:57:53 I found A solution by adding : Aug 31 22:58:12 #include Aug 31 22:58:22 #define __NR_syscall (__NR_SYSCALL_BASE+113) Aug 31 22:58:32 #define __NR_umount (__NR_SYSCALL_BASE+ 22) Aug 31 22:59:46 to /glibc-2.9/glibc-ports-2.9/sysdeps/unix/sysv/linux/arm/sysdep.h Aug 31 23:37:29 openmoko: 03daveb * r5594 10/trunk/gta02-core/ (5 files in 3 dirs): further modem corrections. Aug 31 23:58:04 openmoko: 03daveb * r5595 10/trunk/gta02-core/modem.sch: add some notes on remaining tasks Sep 01 01:47:27 good night locos :) Sep 01 01:47:57 * fredrin freerunner is going strong :) Sep 01 01:54:11 openmoko: 03werner * r5596 10/trunk/gta02-core/docs/ecn/ecn0030/telit-docs: GE865 documentation overview. Sep 01 02:12:16 openmoko: 03daveb * r5597 10/trunk/gta02-core/ (docs/ecn/ecn0030.txt memory.sch modem.sch sd-sim.sch): more modem updates, yes we do need DTR (AT cmd reference, pg139) **** ENDING LOGGING AT Tue Sep 01 02:59:57 2009