**** BEGIN LOGGING AT Tue Apr 20 03:00:04 2010 Apr 20 07:24:01 jeua guys Apr 20 07:24:05 rahter heya Apr 20 07:24:28 having a problem flashing my moko- battery seems to die after about 3 minutes Apr 20 07:24:36 but then loads up fine a little later Apr 20 07:27:07 andylockran: do you mean the device turns off during flashing via dfu? Apr 20 07:27:13 yup Apr 20 07:29:08 andylockran: hm, it shouldn't be a problem with battery because if dfu-util finds the device, it means it's enumerated on bus and that means it's allowed to draw 500mA, so it needs no battery. Apr 20 07:29:40 that's kinda the conclusion I came to Apr 20 07:29:49 andylockran: but if you're using NOR u-boot, it has an idle timeout of 30 sec. Apr 20 07:29:52 as then cu'ing in shows that it's drawing the higher charge Apr 20 07:30:09 it's longer than 30 second Apr 20 07:30:15 this is a GTA01 btw.. Apr 20 07:30:30 andylockran: heh, that makes a completely different story. Apr 20 07:30:36 bought on release day - finally got round to having a play :) Apr 20 07:31:08 I have dfu'd a new kernel fine, but rootfs no luck. Apr 20 07:31:51 andylockran: i'm afraid for gta01 you'll need a fully charged battery to flash. You can use either any charged nokia BL-4/5/6C or charge your moko battery in a cheap (that's important) nokia-compatible charger. Apr 20 07:32:16 andylockran: and if your battery looks "pregnant" it means you have to replace it. Apr 20 07:32:22 balls Apr 20 07:32:43 it's only 6 weeks preggers.. still a problem? Apr 20 07:32:55 andylockran: should be flat... Apr 20 07:33:28 andylockran: do you know how to use a soldering iron? Because charging a nokia battery on FR requires a little hardware mod. Apr 20 07:33:34 * andylockran puts a chair on the battery and jumps up and down Apr 20 07:33:36 s/FR/GTA01/ Apr 20 07:33:36 PaulFertser meant: andylockran: do you know how to use a soldering iron? Because charging a nokia battery on GTA01 requires a little hardware mod. Apr 20 07:34:09 I know how to use a soldering iron.. I'm guessing that's the issue - this battery is a nokia BL-5C rather than my moko battery Apr 20 07:34:21 (which is 3,000 miles away) Apr 20 07:34:31 andylockran: then you can't charge it in the device, but you can charge it in any nokia phone or charger. Apr 20 07:34:48 what's the hardware mod? Apr 20 07:35:10 andylockran: to be able to charge it in the device you need to add a resistor of 10-15kOhm from the middle pin (ID) to the ground (-). Apr 20 07:35:59 andylockran: i disassembled one battery and decided that it's too damn hard to attach a resistor to it, so you'll have better luck modifying the device itself (soldering to the backside of the battery connector probably). Apr 20 07:38:02 andylockran: another option would be to use some foil between the battery and the neo connector and attach the resistor to that. Apr 20 07:38:36 hmm.. do FIC still sell the batteries? Apr 20 07:39:12 andylockran: with software tweak you can charge the battery with ~200mA once the device is booted but those 200 are going both to power the system and to charge the battery, so if the device is not suspended, the battery almost doesn't charge. Apr 20 07:39:35 andylockran: i think some of the distris has some in stock... Apr 20 07:40:05 andylockran: finding a way to attach a resistor to the device itself would be an ultimate solution. Apr 20 07:41:28 andylockran: just find a small enough tip for you iron (or make one yourself from a copper wire) and solder it to the backside of the connector. I admit i haven't tried it yet but it looks doable. Apr 20 07:44:46 yeah Apr 20 07:44:58 I'll take a look.. thanks for that Paul.. Apr 20 07:45:22 will the battery charge normally once I get software on there, or is this purely a hardware thing. Apr 20 07:46:29 11:39 < PaulFertser> andylockran: with software tweak you can charge the battery with ~200mA once the device is booted Apr 20 07:46:40 but blablabla Apr 20 07:47:31 andylockran: so, unfortunately, no. There's just no way to convince pcf50606 to start charging a battery with full current if it thinks the temperature is out of range. Apr 20 07:48:20 Unfortunately, the butt-heads who were responsible for this circuit just copied the example from the datasheet (which uses 10k reference) instead of using ~70k to make it compatible with nokia batteries. :( Apr 20 07:54:28 PaulFertser: sorry - by that comment, I thought you meant booted into uBoot Apr 20 07:55:48 andylockran: well, you need to set "pre-charge current" to the maximum to get those 200mA but it won't make sense if u-boot is running because it all be consumed by the device itself. Apr 20 07:56:12 andylockran: do not you have a nokia phone/charge nearby? Charge using that, you'll have enough energy to flash rootfs and to play with the device for some time. Apr 20 08:20:24 nope.. Apr 20 10:20:43 anyone know what the 'change globally' tick box does when you're editing your moko list subscription Apr 20 10:22:55 I guess all the moko sublists... Apr 20 10:23:00 eg users/announce/dev .etc Apr 20 10:23:14 Psi: then it will change those things for all your subscripĆ¼tions on that server Apr 20 10:24:19 ah, good, thats what i want Apr 20 10:59:42 hello Apr 20 11:00:58 I have touble cloning the git repo. Does anyone else got problems with this? Apr 20 11:01:50 which repo? Apr 20 11:03:19 git.openmoko.org/git/kernel.git Apr 20 11:04:03 siebz0r: http://lists.openmoko.org/nabble.html#nabble-td4894050 Apr 20 11:05:17 siebz0r: and to skip problems they reported, you should probably git checkout some branch first Apr 20 11:05:51 yeah, I've red that one (first hit @ google). I've downloaded the snapshot, but when I 'git pull' it repotrs a 'bad pack header' Apr 20 11:06:07 branch from openmoko? Apr 20 11:06:32 git checkout origin/om-gta02-2.6.32 -b om-gta02-2.6.32 Apr 20 11:06:35 or whatever you want Apr 20 11:07:12 error: Entry 'MAINTAINERS' would be overwritten by merge. Cannot merge. Apr 20 11:07:23 while checking out andy-tracking Apr 20 11:08:54 nvm Apr 20 11:09:11 -f parameter did the trick Apr 20 11:10:53 but the pull still fails Apr 20 11:11:39 what about git fetch; git rebase -i the_branch_you_checkouted? Apr 20 11:12:21 because bitbake updates from this tarball everytime and successfully.. so you should be able to repeat those steps by hand to Apr 20 11:14:13 git rebase -i andy-tracking opens vim? Apr 20 11:14:28 whatever is in $EDITOR Apr 20 11:14:53 and you want origin/andy-tracking Apr 20 11:15:48 so replace noop with origin/andy-tracking? Apr 20 11:16:12 no git rebase -i origin/andy-tracking Apr 20 11:17:14 and if our checkout is exactly the same as origin/andy-tracking it will say noop, check "git status" for unstaged files Apr 20 11:17:33 allright, I've done that. But then it opens $EDITOR, what am I supposed to do here? Apr 20 11:17:57 I'm sorry I find this very confusing :P (new to git) Apr 20 11:18:01 what is shown? Apr 20 11:18:15 noop again? (as it was with -i andy-tracking)? Apr 20 11:18:23 yes Apr 20 11:18:38 our checkout is exactly the same as origin/andy-tracking Apr 20 11:18:40 git rebase -i origin/andy-tracking Apr 20 11:18:49 and what is in git state? Apr 20 11:19:23 sorry status Apr 20 11:20:09 #On branch andy-tracking \ nothing to commit (working directory clean) Apr 20 11:20:25 git fetch works? Apr 20 11:21:23 no Apr 20 11:21:27 remote: fatal: Out of memory? mmap failed: Cannot allocate memory Apr 20 11:21:34 remote: aborting due to possible repository corruption on the remote side. Apr 20 11:21:39 fatal: protocol error: bad pack header Apr 20 11:22:10 siebz0r: are you in that untared snapshot? Apr 20 11:22:15 aloah Apr 20 11:22:22 yes Apr 20 11:22:53 I unpacked it Apr 20 11:23:18 and there was 1 dir ".git/" and the snapshot Apr 20 11:24:16 there are multiple directories/files now (looks like a kernel src) Apr 20 11:24:52 i noticed a possible bug in dfu-util: after "usb_claim_interface()" a "usb_set_altinerface(_rt_dif.dev_handle, 0)" call seems to be necessary to actually switch to DFU-Mode when the USB-Device still uses its default interface. usb_claim_interface() doesn't send anything over the bus, but the device needs to know when to serve DFU-Requests (like the following dfu_get_status() request) Apr 20 11:25:20 it works without it on devices where the DFU-Interface == 0 (or the only interface available) Apr 20 11:29:24 oh sorry, just found #openmoko-cdevel Apr 20 12:05:43 hi from greece. i just unpacked my neo. Apr 20 12:06:15 cool :D Apr 20 12:06:34 in the first 10 minutes the first conclusion is that i must install a new OS!! Apr 20 12:06:37 :-) Apr 20 12:06:46 design is very cool Apr 20 12:07:28 in OM2009 it asks for pin os my sim card but the keyboard has no numbers!! Apr 20 12:07:36 os->of Apr 20 12:09:20 i deactivated my pin code for ages to play with neo... Apr 20 12:09:51 you can switch the keyboard by moving you finger up or downwards on the keyboard Apr 20 12:09:57 don't remeber exactly Apr 20 12:14:48 Heinervdm: that's it. thanks for the keyboard tip. anyway my first task is a new OS! Apr 20 13:05:38 JaMa: have you got any idea what my git problem might be? Apr 20 13:06:13 siebz0r: sorry, no Apr 20 13:06:53 siebz0r: but only .git first is right (bitbake fetches only git objects to .git and checkouts directly to builddir when needed) Apr 20 13:07:28 siebz0r: btw what revision do you have in your andy-tracking checkout (first item in git log) Apr 20 13:07:42 atm? Apr 20 13:07:49 y Apr 20 13:08:14 wait Apr 20 13:08:25 I just removed the files (starting over) Apr 20 13:08:39 and I just unpacked the git tar.gz Apr 20 13:08:58 and when I do a rebase on origin/andy-tracking Apr 20 13:09:12 it says "working tree is dirty" Apr 20 13:09:36 and git status? Apr 20 13:10:35 # On branch master Apr 20 13:10:36 * JaMa unpacking too Apr 20 13:10:58 and then a whole bunch of lines: # deleted: *.c Apr 20 13:11:02 git checkout -b andy-tracking origin/andy-tracking Apr 20 13:11:24 and if it still complains "git reset --hard" first Apr 20 13:11:49 git checkout requires -f I think Apr 20 13:12:07 complains about overwrite Apr 20 13:13:23 git checkout is finished Apr 20 13:14:02 git rebase -i origin/andy-tracking > quit editor > successfully rebased and updated refs/heads/andy-tracking Apr 20 13:16:38 siebz0r: replied to that email Apr 20 13:16:43 is this the revision?: checkout: moving from andy-tracking to ee4642aa62413667969261b94b5457f97d49f06e Apr 20 13:18:50 that's really old Mon, 24 Nov 2008 08:13:36 +0000 (08:13 +0000) Apr 20 13:19:27 :' Apr 20 13:19:29 :') Apr 20 13:20:24 but with instructions in my e-mail reply git pull works ok (http://paste.pocoo.org/show/204046/) Apr 20 13:21:13 page not found :P Apr 20 13:21:50 nvm Apr 20 13:21:52 got it Apr 20 13:22:49 your pulling from the unpacked tar.gz? Apr 20 13:23:33 http://paste.pocoo.org/show/204048/ Apr 20 13:23:39 is what I just sent to ML Apr 20 13:24:27 k, thnx, I'll try ;-) Apr 20 13:28:02 with git checkout is says that andy-tracking already exists Apr 20 13:28:35 and git branch -D master cannot delete branch master 'wich you are currenty on' Apr 20 13:28:37 it's expected if you run it twice with same .git :) Apr 20 13:29:11 and if it exists you can "git checkout andy-tracking" Apr 20 13:29:18 to switch branch from master to andy Apr 20 13:31:53 done, and deleted the master Apr 20 13:32:17 some hidden files got in the way I suppose ;-) Apr 20 13:33:53 what revision should it be? Apr 20 13:34:21 b460dd0c3712d12a0c15745fe74b5603630bc39d ? Apr 20 13:36:57 siebz0r: latest is 973a41fce60e5f6edfcf0eaf94056a57eee65e16 Apr 20 13:38:28 how could I update it? Apr 20 13:42:22 siebz0r: git pull should be enough.. Apr 20 13:42:26 * JaMa leaving, bye Apr 20 16:55:51 hi again. in my new neo device i encountered the problem that having it usb connected to my main pc , and having checked from dmesg that is recognized as usb0 nevertheless trying to ifconfig it gives 'Cannot find device "usb0" Apr 20 19:32:40 i have problems configuring the usb0 as network interface in debian squeeze Apr 20 19:55:12 so hey Apr 20 19:55:19 PaulFertser, you here? Apr 20 20:24:44 rooly: yep Apr 20 20:24:49 Hi Apr 20 20:39:54 does anyone know how to get the andy-tracking kernel from git? Apr 20 20:41:24 I'm getting nothing but errors Apr 20 20:51:18 PaulFertser, sorry about that, stupid internet Apr 20 20:51:36 PaulFertser, remember back in the day when om devs were looking into psuedo-multitouch for the neo line? Apr 20 20:58:15 hola Apr 20 20:59:40 rooly: yes, i do Apr 20 21:00:09 would you believe that a windows mobile developer has only just started looking into the same technology? Apr 20 21:00:30 rooly: emulating multi-touch with a resistive touch-screen? Apr 20 21:00:35 yes Apr 20 21:01:39 hmm, seems that its an xda-dev guy doing it Apr 20 21:01:40 http://www.engadget.com/2010/04/20/resistive-htc-devices-can-have-pseudo-multitouch-too/ Apr 20 21:01:41 rooly: what do you mean by your question? Apr 20 21:02:49 rooly: i mean "a windows mobile developer" sounds too abstract Apr 20 21:02:51 ;) Apr 20 21:03:43 multi-touch is so interesting feature Apr 20 21:03:52 its required only for zoom Apr 20 21:04:01 max_posedon: emacs is an interesting feature. Multi-touch isn't ;) Apr 20 21:04:04 and as is only in web browser Apr 20 21:04:24 meh, like the old joke goes, emacs is a decent os...but it lacks a good text editor Apr 20 21:04:31 and openmoko haven't web browser which even could support it in a nice way Apr 20 21:04:51 so, why people so often speak about this useless feature? Apr 20 21:05:15 I played with iphone and htc hero with multitouch, found it useless Apr 20 21:05:21 rooly: i tried editing config files with it and it was cool. I tried ircing and it was cool. Emacs is the coolest environment for a smartphone. Apr 20 21:05:22 I just don't want zoom anything Apr 20 21:05:28 max_posedon, the fr has a resistive screen which doesn't support traditional multitouch anyway. any multitouch on the device will be emulated Apr 20 21:05:52 Q is, what you will do with multitouch? Apr 20 21:06:05 emulated it or not. Apr 20 21:06:26 hi. in my new freerunner i connect it to a usb port but i get no network interface , (debian squeeze) Apr 20 21:06:41 chomwitt, check dmesg, it might be ethN Apr 20 21:06:47 max_posedon: ok Apr 20 21:08:11 max_posedon: i unlplug and plug it having 'tail kernel.log' and i get http://debian.pastebin.com/kAmKuywR Apr 20 21:09:16 1 00:02:48 localhost kernel: [19963.226907] usbcore: registered new interface driver cdc_ether Apr 20 21:09:27 after 30 secs it was unregistered Apr 20 21:09:32 I don't know why Apr 20 21:09:39 may be your phone suspended Apr 20 21:09:50 anyway, I think you should update to smth more recent Apr 20 21:09:54 (on your neo) Apr 20 21:10:26 usb0 gone a year ago Apr 20 21:10:33 max_posedon: it a new phone wiht om2009. and now its dark but i though it was just screen saver or sth Apr 20 21:10:48 looks like suspended Apr 20 21:11:34 yes. i push the power button and it was recognized from the kernel again as usb0 Apr 20 21:12:49 but even with that should i get in 'ifconfig -a' a usb0 entry? (i have netusb module loaded) Apr 20 21:13:31 ok i turned suspend off Apr 20 21:15:37 om2009 a 1-year-old project Apr 20 21:15:44 I still suggest smth more recent Apr 20 21:16:34 max_posedon: as i get it in order to install a more recent dist i need to have established a usb connection Apr 20 21:17:11 nope, you need flash smth new via dfu-util Apr 20 21:17:20 rooly: anyway, thanks for letting me know. Apr 20 21:17:31 np Apr 20 21:17:47 max_posedon: thanks very much. i'll look again about flushing. Apr 20 22:42:59 hi Apr 20 22:43:50 i can't boot my GTA02... after installing Qi it just reboots every second (USB powered) or every 10 seconds (battery powered) while the screen stays completely black Apr 20 22:44:01 does anyone know this issue & a way to fix it? Apr 20 22:44:09 or how'd i debug this without a debug board? Apr 20 23:59:01 how can i delete in SHR keyboard!? Apr 21 02:08:01 wassup **** ENDING LOGGING AT Wed Apr 21 02:59:56 2010