**** BEGIN LOGGING AT Tue Aug 18 02:59:56 2009 Aug 18 03:20:09 project B = Vacation Aug 18 03:20:30 i would bet on it Aug 18 04:23:03 hackable 1 coming up Aug 18 04:35:19 all i got was a login screen Aug 18 04:54:13 i need foe either doom or duke-nukem to work on shr-unstable Aug 18 05:31:08 is there a quiz to see if i qualify as an 'advanced user'? my cell phone is falling apart, i don't want to wait. Aug 18 05:31:35 i made a VB script once. just kidding, i've been writing C and using linux for 15 years. is that advanced enough? Aug 18 05:31:59 you'll probably manage Aug 18 07:09:51 hmmmmmmm Aug 18 07:09:55 maybe ill flash to 20090808 Aug 18 07:10:00 i havent flashed since 2008 Aug 18 07:13:50 if You upgraded, it shouldn't matter Aug 18 08:23:22 PaulFertser: ping ... Aug 18 08:24:08 BillK_: pong Aug 18 08:24:25 BillK_: i wanted to propose a simpler workaround for wlan problem Aug 18 08:24:27 You pinged me about 12 hrs ago ? Aug 18 08:24:34 yes: Aug 18 08:25:13 BillK_: it seems that just rmmod ar6000 && modprobe ar6000 after every "fsoraw + wpa_supplicant " invocation is enough. At least it worked reliably here. Aug 18 08:25:38 BillK_: (unfortunately fixing the driver properly is really hard i'm not sure i'll be able to do that :() Aug 18 08:25:45 PaulFertser: it will eventually fail ... what Aug 18 08:26:03 I found is that wpa_supplicant, Aug 18 08:26:17 avahi and all the other things running in the FR Aug 18 08:26:29 networking randomly crash & burn and Aug 18 08:26:41 need cleaning up - so even if you Aug 18 08:27:07 do the driver thing, it may fail on something else. Aug 18 08:27:31 Also, doesnt the request resources thing via dbus add/remove the module anyway? Aug 18 08:29:15 PaulFertser: searching my emails for a ref to the last comment ... Aug 18 08:29:17 BillK_: that's the trick, yes. FSO only binds/unbinds s3cmci driver, basically the result is the same as suspend/resume Aug 18 08:30:24 BillK_: and i doubt you need avahi on FR, at least on all desktops and laptops i used i never needed it and it even sometimes caused problems (e.g. with dns resolving). Aug 18 08:31:15 BillK_: it'd be nice to have at least reproducible testcases to fix in-driver bugs so even if something bad happenes in firmware the driver could be simply reloaded. Aug 18 08:32:01 PaulFertser: sorry, not avahi - udhcpc - shoulda checked first Aug 18 08:32:44 PaulFertser: I have had good reliability using those scripts I posted - I Aug 18 08:33:07 sometimes get a failure when signal is weak (the things DEAF :) but Aug 18 08:33:55 just running the script again when closer works - cant remember a failure in fact (against Aug 18 08:34:07 the 3 AP's I use regularly) Aug 18 08:34:51 s/failure/real failure/ Aug 18 08:35:46 If the ar6000 is crashing internally, its getting reset somehow ... Aug 18 08:35:54 BillK_: btw, why do you have to delete default route? As i see it if you establish connection and udhcpc adds another default route, it'll be prioritized. Aug 18 08:36:47 BillK_: ar6000 can crash internally but it doesn't do that all the time with all APs. There were some quite specific conditions, Werner was able to reproduce that reliably. Aug 18 08:36:56 PaulFertser: local network - I often use wifi when usb is connected, but when wifi is running, its usually doing backups/updates etc Aug 18 08:37:56 PaulFertser: I have not looked into prioritisation - Its a quick fix that works in my setup Aug 18 08:38:37 BillK_: i wonder why it's needed since i think that the route add later "takes priority" automatically. Aug 18 08:38:37 PaulFertser: Do you have a lits of the conditioons Werner found - would be interesting to see. Aug 18 08:38:57 Hi, is there an openmoko2009 build with kernel 2.6.30 that will work? Aug 18 08:39:38 PaulFertser: route didnt - I have usb setup to not be forwarded, whilst wifi gets full access to network/internet - when starting usb after Aug 18 08:39:56 BillK_: (sorry, nabble) http://lists.openmoko.org/nabble.html#nabble-td1581754 Aug 18 08:40:07 Hello. I would like to know what you would choose or which is knowen to be better QTExtended Improved or QTMoko? Aug 18 08:40:37 PaulFertser: tkx for link - lokking ... Aug 18 08:43:00 PaulFertser: have not seen the gwiscan bug since last year, the others "may" be Aug 18 08:43:18 there but I dont see them in my usage which is either Aug 18 08:44:07 flat out streaming (backups is probably most traffic, but also intrnet radio streaming via mplayer) Aug 18 08:44:29 PaulFertser: and an our or two of inactivity between uses. Aug 18 08:44:48 BillK_: http://svn.openmoko.org/developers/werner/wlan/freeze/README Aug 18 08:45:58 PaulFertser: also tried "wmiconfig power maxperf" but didnt see an advantage, and it Aug 18 08:46:00 BillK_: and some other tests like that can reliably crash firmware Aug 18 08:46:09 mooo Aug 18 08:46:18 plays havoc with usb audio when streaming from the internet Aug 18 08:46:25 Morning DocScrutinizer :) Aug 18 08:49:14 ~nabble Aug 18 08:49:18 BillK_: You mentioned scripts for setting up wifi, do you have a link to them? I could never get my wifi working. Aug 18 08:49:57 PaulFertser: hmm - I am not roaming ... cant remember seeing any of the error messages Aug 18 08:50:12 ajvogel: wait 1 Aug 18 08:52:20 ajvogel: pasted email contents to: http://pastebin.com/m643d6275 Aug 18 08:53:15 BillK_: Thanx alot, I appreciate the help Aug 18 08:53:38 PaulFertser: tyop, its bluetooth audio, not usb audio that wmiconfig maxperf causes continual stutters. Aug 18 08:53:48 s/tyop/typo/ Aug 18 08:53:48 BillK_ meant: PaulFertser: typo, its bluetooth audio, not usb audio that wmiconfig maxperf causes continual stutters. Aug 18 08:54:12 BillK_: usb audio is quite a realtime application. On a weak CPU that is hard to get right when you need to do any other task meanwhile (like transfer over WiFi) Aug 18 08:54:45 PaulFertser: see typo above - meant bluetooth, not usb :) Aug 18 08:54:53 even worse Aug 18 08:55:21 well, depends Aug 18 08:55:38 PaulFertser: It looked like interferance between bluetooth an wifi at the wireless level, not cpu bound. Aug 18 08:55:52 gotta take son to soccer training, back later ... Aug 18 08:56:11 id you can use a IIS if at BT, it shouldn't be much diff to a normal audio-card Aug 18 08:57:09 BillK_: and BT and WiFi use same freq, they basically have to share it Aug 18 08:58:18 PaulFertser: might the 2-wire coexistence hw need some additional sw support for BT and WLAN to cooperate correctly? Aug 18 09:04:35 * DocScrutinizer wonders how many of the strange effects in BT and WLAN might be introduced by mutual influence over 2w-coex which afaik never been taken into account when checking either of those subsystems Aug 18 09:07:10 DocScrutinizer: i thought there's some documentation with guidelines. But i wouldn't surprise it was ignored just like many other more important issues. Aug 18 09:09:01 PaulFertser: exactly Aug 18 09:09:37 PaulFertser: we should pester werner about it. How much and what he knows about the topic. Aug 18 09:10:17 PaulFertser: are there any API functions dealing with coex, for BT or WLAN? Aug 18 09:10:27 DocScrutinizer: he says he's busy with gta02-core anyway. Advices me to learn network drivers programming by "fixing" ar6000... Aug 18 09:11:07 PaulFertser: he won't refuse to answer questions ;-) Aug 18 09:12:25 DocScrutinizer: i'm afraid it won't help much. "fixing" that driver would require real dedication and anyway it won't help with firmware problems. Aug 18 09:13:59 DocScrutinizer: can you possibly explain how can it happen that despite making a decent wifi chip is magnitude harder than writing a sane driver, atheros usually succeedes at first but can ultimately fail at second (as ar6000 is an example of really poor engineering). They're supposed to have QA etc... Aug 18 09:14:17 PaulFertser: a 'freeze' may as well be a wrong level on one of these 2 wires Aug 18 09:15:03 PaulFertser: no idea on atheros capabilities Aug 18 09:15:34 DocScrutinizer: but we all know the hardware it produces is "ok" and widely used. Aug 18 09:20:12 PaulFertser: any pointer to where 2W-coex is specified (operation principles etc)? Aug 18 09:23:26 hi Aug 18 09:24:06 someone here knowing a little help about qtmoko to sd-card installation? Aug 18 09:40:59 $ sudo tar -C /mnt/moko/ -xzvf qtmoko-debian-v6.tar.gz Aug 18 09:41:02 sry Aug 18 09:42:45 while trying to copy the qtmoko-fs to the sd-card i have the following error, which seems to be an error with the qtmoko-fs-archive: Aug 18 09:42:45 ./usr/lib/gcc/arm-linux-gnueabi/4.3/cc1 Aug 18 09:42:45 gzip: stdin: invalid compressed data--format violated Aug 18 09:42:45 Any idea? Aug 18 09:44:02 download it again and/or format your sd card Aug 18 09:46:03 dns53: thank you Aug 18 12:45:33 2 Aug 18 12:45:38 hmpf Aug 18 14:35:12 hi there Aug 18 14:36:43 trick question : what would be the simpler way, for a script, to know whether the usb networking is on or not? Aug 18 14:36:53 if possible on shr-u :) Aug 18 14:36:55 thanks alot Aug 18 14:37:12 check for a route? Aug 18 14:37:28 olivierm: cat /sys/class/net/usb0/carrier Aug 18 14:52:17 PaulFerster: thanks alot :) Aug 18 14:52:38 SpeedEvil : no, the route isn't destroyed even when the cable is unplugged Aug 18 14:57:12 it works great that way paul, thank you very much again Aug 18 14:57:36 olivierm: say hi to mwester, it's his code :) Aug 18 14:57:43 oh, great :) Aug 18 15:05:39 http://jokvsms.mihanblog.com Aug 18 15:05:40 http://jokvsms.mihanblog.com Aug 18 15:13:02 r3 kernel-paroli white screens after resume Aug 18 15:19:41 dos1: I estimated John might take up to ~3500 years. So I'll stop it now Aug 18 15:23:55 dos1: the fan became a nuissance ;-) Aug 18 15:24:57 doc is there something wrong with the latest paroli? Aug 18 15:37:15 DocScrutinizer: ;D Aug 18 15:40:12 DocScrutinizer: too short attention span to wait that little? sissy! Aug 18 16:21:26 doc is there something wrong with the latest paroli? Aug 18 16:43:49 hi Aug 18 17:25:33 OK, I can't stand Midori. What's a good (JS/AJAX friendly) browser for FR/ Aug 18 17:25:35 ? Aug 18 17:29:17 It's in my brain atm, it should come out soon ;-) Aug 18 17:39:50 hej everyone. I like to install QI-bootloader but for some reason it doesn't work (following the wiki http://wiki.openmoko.org/wiki/Qi#Installation ). And thene I have a question. How can I flash my neo once I installed qi? As far as I understand DFU mode and NOR won't work after the installation Aug 18 17:41:36 linuksamiko: wrong Aug 18 17:41:38 linuksamiko: reread the wiki page Aug 18 17:46:58 lindi-: what do you mean by wrong? That I didn't follow the wiki right or that I can't flash my neo anymore (using dfu-util) Aug 18 17:47:19 linuksamiko: you can't write to NOR so you can't make it "not work" Aug 18 17:48:55 * PaulFertser high-fives lindi- Aug 18 17:49:03 and what is meant by that exactly: no DFU-Mode - USB is not initialized at all (but you can always boot from NOR) Aug 18 17:49:44 linuksamiko: when you boot Qi Aug 18 17:50:47 lindi-: so with Qi I can boot into NOR and flash as usual if I want to? Aug 18 17:51:02 linuksamiko: it's the hardware that allows you to choose that Aug 18 17:51:35 PaulFertser: btw, why is it hardware? why not just have it always boot from NOR and code in NOR to read the status of AUX? Aug 18 17:51:47 PaulFertser: this would allow you to resume by aux Aug 18 17:52:35 lindi-: ok, I just wanted to know if I can still flash my moko with dfu-util afterwords because I wasn't so sure about that. And sticking forever with one system (besides SDcards) wasn't exactly what I was looking for Aug 18 17:52:59 lindi-: no idea, that was design decision of OM folks made long time ago. Probably there're some obstacles in the way you propose. Aug 18 17:53:10 PaulFertser: I will just do the whole thing all over again and hope it works this time Aug 18 17:53:19 linuksamiko: it doesn't make sense! Aug 18 17:53:27 PaulFertser: ok, this is what the AVR microcontroller does for me :) Aug 18 17:53:45 PaulFertser: what doesn't make sense Aug 18 17:54:02 lindi-: btw, you know that reliable resume by AUX is possible with a bit of hacking. Aug 18 17:54:13 linuksamiko: "doing the whole thing from scratch" Aug 18 17:55:05 PaulFertser: well that is what you suggested (kind of) Aug 18 17:55:24 linuksamiko: no, i suggested reading the page, not "doing" anything with hardware. Aug 18 17:55:56 lindi-: the way it's done now allows you to boot both ways. If you bork NOR you can boot from NAND and vice-versa. That has its benefits to what you propose. Aug 18 17:56:53 PaulFertser: yes Aug 18 17:57:09 PaulFertser: but if I bork NOR I need debug board anyway? Aug 18 17:57:14 PaulFertser: so what's the point? Aug 18 17:59:31 lindi-: no, NOR can be written from the userspace, if you can boot anyhow. Aug 18 17:59:49 PaulFertser: but only with hardware hack? Aug 18 18:00:13 lindi-: it's so minor that it can't be considered to be a real hardware hack. Aug 18 18:00:23 ok it's still one :) Aug 18 18:00:36 lindi-: think of it as of safety measure against clueless lusers. Aug 18 18:08:08 ok, too much fuzz about nothing. It seamed to work all the time. THX anyway and helping me out Aug 18 18:08:37 linuksamiko: thanks for reading the wiki :) Aug 18 18:16:43 PaulFertser: lindi-: iirc I suggested a hack that would stabilize the otherwise bouncing AUX. In the end that gpf6 is a gpiO line that aux is running to Aug 18 18:16:49 DocScrutinizer: exactly Aug 18 18:17:12 DocScrutinizer: i still remember that, except that i don't see much demand for actual implementation. Cool hack :) Aug 18 18:19:42 PaulFertser: DocScrutinizer: would be cool to be able to put the device to a mode "record audio when button FOO is held" Aug 18 18:20:06 PaulFertser: DocScrutinizer: you can't do that with AUX since the phone won't wake with AUX. You can't do that with POWER since the system will shutdown in 10 seconds Aug 18 18:20:10 PaulFertser: well that's the meaning of the "hardware and synergy" subline on my business card Aug 18 18:20:23 lindi-: hm, you found a usecase indeed ;) Aug 18 18:21:05 lindi-: you obviously are mistaken wrt to what I can do ;-D Aug 18 18:21:34 Is SHR unstable stable enough to use? Aug 18 18:21:52 lindi-: why should FR be unable to wake on AUX? Aug 18 18:22:46 lindi-: aah, sorry, I missed the "link" between those two posts Aug 18 18:22:57 DocScrutinizer: well this bouncing issue? Aug 18 18:23:02 lindi-: use HOLD button Aug 18 18:23:24 DocScrutinizer: i don't have headset :) Aug 18 18:23:42 they should have marketed it as an extra button :) Aug 18 18:25:31 lindi-: hack NOR uboot then, to fix the bouncing Aug 18 18:26:35 dos1: pong Aug 18 18:26:54 mickeyl: did you test that idlenotifier stuff in fsousaged? Aug 18 18:27:03 lindi-: insert two opcodes: set gpf6-mode output; set gpf6 0 Aug 18 18:27:08 mickeyl: eigher i'm doing something wrong or it doesn't work :( Aug 18 18:27:14 dos1: what does happen? Aug 18 18:27:22 s/eigher/either/ Aug 18 18:27:22 dos1 meant: mickeyl: either i'm doing something wrong or it doesn't work :( Aug 18 18:27:33 mickeyl: just nothing Aug 18 18:27:35 hmm Aug 18 18:27:38 that's not enoug Aug 18 18:27:49 mickeyl: it was segfaulting without correct conf options (as expected) Aug 18 18:27:54 DocScrutinizer: hmm Aug 18 18:28:07 mickeyl: but now i can't see any debug message nor dbus signal about idle notifier when resume Aug 18 18:28:19 dos1: are you sure you are on HEAD? Aug 18 18:28:22 mickeyl: either from gsm or from power button Aug 18 18:28:29 mickeyl: yup, checked Aug 18 18:28:35 ok, let me retry here Aug 18 18:30:03 mickeyl: speaking of signals, could you please consider http://trac.freesmartphone.org/ticket/385 or is there some existing solution? Aug 18 18:30:28 Where does Woosh save its bookmarks? Aug 18 18:33:33 dos1: ups, 2009-08-18T18:33:08.840307Z UsageController [ERROR] : DBus Error while talking to IdleNotifier: Method "SetState" with signature "s" on interface "org.freesmartphone.Device.IdleNotifier" doesn't exist Aug 18 18:33:33 why is the letters on the keyboards soo smalll??? Aug 18 18:33:56 is-are Aug 18 18:34:01 lindi-: we could do that, although I'd prefer to forcefully detach the context before going into suspend Aug 18 18:34:12 mickeyl: I'm ok with both Aug 18 18:34:29 mickeyl: i'd just want to be able to suspend and be sure that the phone is not woken up after 60 seconds :) Aug 18 18:34:40 mickeyl: am I the only one who is hitting this? Aug 18 18:34:59 I'd imagine "stop IRCing and put phone to pocket" would be quite common use case Aug 18 18:35:25 dos1: aargh, found it Aug 18 18:36:05 lindi-: i agree, more folks should be annoyed by that Aug 18 18:36:34 ooh wait Aug 18 18:36:38 there's a much better way Aug 18 18:36:46 * mickeyl slaps forehead Aug 18 18:37:07 why do we get +CGEV at all? Aug 18 18:37:27 don't we disable that URC on suspend? Aug 18 18:39:30 mickeyl: what's the command to disable it? (I'd read calypso source but I don't have it, the command is probably not in my copy of GSM specs either..) Aug 18 18:40:01 freesmartphone.org: 03mickey 07cornucopia * r9447282b921d 10/fsousaged/src/plugins/controller/plugin.vala: Aug 18 18:40:01 freesmartphone.org: fsousaged: controller: talking to the right dbus object helps Aug 18 18:40:01 freesmartphone.org: Thanks dos1 for spotting Aug 18 18:40:16 +CGEV is a standard 07.07 URC Aug 18 18:40:24 let me check specs Aug 18 18:40:43 'Set command enables or disables sending of unsolicited result codes, +CGEV:' Aug 18 18:41:20 +CGREG? Aug 18 18:41:28 dos1: please retry Aug 18 18:41:29 mickeyl: +CGEREP Aug 18 18:41:34 oh Aug 18 18:41:35 mickeyl: but i don't know which value it should be set to Aug 18 18:41:42 mickeyl: ok, thanks Aug 18 18:42:11 no, +CGEREP is something different Aug 18 18:42:21 oh Aug 18 18:42:25 wait, you're right Aug 18 18:42:30 anyway, please take care not to bork solicited resume by GPRS inbound traffic. I'm using this a lot Aug 18 18:42:32 let me check our suspend command biunch Aug 18 18:42:39 mickeyl: i'll read my logs Aug 18 18:43:03 DocScrutinizer: no, disabling +CGEV should have no effect on inbound gprs traffic wakeup Aug 18 18:43:19 ok :-) Aug 18 18:44:07 lindi-: according to the source code we disable it Aug 18 18:44:08 c.append( "+CGEREP=0,0" ) Aug 18 18:44:16 can you verify whether it's really sent? Aug 18 18:44:35 mickeyl: just in case you ever wondered: the fifo in modem uart seems 64 byte Aug 18 18:45:17 I'm afraid there is no command to say "stop stuffing more into the fifo and let it drain' Aug 18 18:45:24 that would be helpful during suspend/resume Aug 18 18:46:53 hmm, anyway I never seen data loss on resuming inbound GPRS packets. So probably moko FW is handling the internal CTS/RTS correctly Aug 18 18:47:12 *nod* Aug 18 18:47:38 given the original dedication of the TI Calypso, I'm positively surprised it works at all with our AP/BP suspend/resume design *cough* Aug 18 18:47:49 (for various degrees of 'working') Aug 18 18:47:52 :) Aug 18 18:48:01 :-P Aug 18 18:48:25 * DocScrutinizer >dinner Aug 18 18:48:41 ~bon appetit Aug 18 18:48:42 hmm... bon appetit is smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Smaklig m�ltid!. Hyv�� ruokahalua. Bo Proveito Aug 18 18:49:33 mickeyl: $ zcat /home/scratch/neo/2009-08-02/home/lindi/bugs/creg-wakes-up1.log.gz | grep -E "(CGEREP|CGEV)" | cut --bytes=1-20,90- | nc paste.dyndns.org 1234 Aug 18 18:49:36 http://paste.debian.net/44524/ Aug 18 18:50:10 mickeyl: maybe you want to see the whole log? Aug 18 18:50:14 yeah, please Aug 18 18:50:32 i admire your cmdline skills though ;) Aug 18 18:54:14 ~seen ptitjes Aug 18 18:54:16 ptitjes was last seen on IRC in channel #htc-linux, 54d 38m 5s ago, saying: 'you here!!! 8-]'. Aug 18 18:54:22 ~seen ptitjes_ Aug 18 18:54:23 i haven't seen 'ptitjes_', mickeyl Aug 18 18:54:42 mickeyl: blah, that's different log Aug 18 18:54:56 mickeyl: that's about ": got 26 bytes from: '\r\n+CREG: 5,"2396","92F2"\r\n'" waking the phone up Aug 18 18:56:29 ? it wakes up by +CREG? that means the suspend commands are not properly sent Aug 18 18:56:41 but that's different bug Aug 18 18:56:56 today my Neo1973 became my primary phone! Now I need a pim suite and enhanced framework functionality! :) Aug 18 18:56:58 then I have one log where the phone wakes up due to "NO CARRIER" Aug 18 18:57:09 heh, that's fun Aug 18 18:57:15 we have no way to disable that :( Aug 18 18:57:23 any idea wy Windoooozzzz seemed to provide only 100mA over USB? Aug 18 18:57:24 mickeyl: I need to replicate the log for #385 Aug 18 18:57:34 lindi-: ok, please add it to the bug once you have it Aug 18 18:58:02 Kero: perhaps it only enumerates known hardware (e.g. for which it has .inf files or drivers) Aug 18 19:00:13 Kero: yes, you must install driver Aug 18 19:00:23 Kero: (not driver actually, just inf file) Aug 18 19:00:30 Kero: or force "fast charge" Aug 18 19:00:33 which driver, the one tha came on the CD with my phone? Aug 18 19:00:45 Kero: no idea, download the one linked at wiki. Aug 18 19:00:51 :) Aug 18 19:00:57 Kero: or just force fast charge on the phone in case you don't need networking. Aug 18 19:00:58 will dig it up, thx Aug 18 19:01:17 Kero: you might want to try opimd-utils for PIM functionality Aug 18 19:03:32 Kero: another possible trick: rmmod g_ether and modprobe g_filestorage (with some dummy file). Aug 18 19:03:56 isntallation of opimd-utils on 20090808 won't mess up existing apps? Aug 18 19:04:18 Kero: you might even want to create a special small vfat-containing file and stuff that damned inf there. Aug 18 19:04:57 Kero: (mess up) afaik, it's ok. Aug 18 19:14:11 mostly not implemented, yet... Aug 18 19:20:07 I'm trying to compile my project with the arm-g++ compiler, but the toolchain's compiler can't seem to find my system libraries (like iostream and queue, etc) and I don't remember how to make it use them. any suggestions? Aug 18 19:20:47 chocotuar: are you using OE? Or emdebian? Or what? Aug 18 19:21:53 PaulFertser: I'm fairly certain it's the OE one, whichever one this is http://wiki.openmoko.org/wiki/Toolchain Aug 18 19:22:52 chocotuar: hm, i can tell you that most devs here use plain OE. Or OE installed with the help of SHRMakefile. With that you just need to specify proper dependencies in your recipe. Aug 18 19:24:33 PaulFertser: at this time I'm just trying to make a binary file to test on my phone itself... what dependency is queue and iostream and algorithm? they're supposed to be basic c/c++ include files Aug 18 19:26:20 chocotuar: no, OE is like gentoo for embedded. Aug 18 19:26:42 chocotuar: if you want to try something simple you can install all dev tools on FR itself but it'll be a bit too slow. Aug 18 19:27:01 PaulFertser: how do I go about getting this SDK? or the include files on the FR? Aug 18 19:27:25 chocotuar: opkg install g++ (to have dev environment on FR) should do it. Aug 18 19:27:46 chocotuar: as to OE, i'd recommend using "Building SHR" instructions from shr-project.org wiki. Aug 18 19:29:16 PaulFertser: for that when I build a package and I using a makefile? or something like a BB recipe Aug 18 19:29:17 chocotuar: do you know Gentoo? OE is very similiar but aimed at cross-compiling distros for embedded targets. To build some software with it you write a recipe (in gentoo they're called ebuilds) to specify what are compile and runtime dependencies of that package, where to get the sources and how to build it. Aug 18 19:29:20 does the openmoko have a EPEAT ranking? Aug 18 19:29:33 kardan__i: wtf is that? Aug 18 19:30:45 http://www.epeat.net/ Aug 18 19:30:57 PaulFertser: I've definitely heard of gentoo... are you suggesting that I install it on my own computer? or am I just using some SDK environment? where do I get it, either way? Aug 18 19:31:38 EPEAT is a system that helps purchasers evaluate, compare and select electronic products based on their environmental attributes. The system currently covers desktop and laptop computers, thin clients, workstations and computer monitors. Aug 18 19:31:56 mobile phones are not yet included oviously Aug 18 19:32:41 chocotuar: i suggest you to use OE by following "building SHR" instructions. Aug 18 19:33:15 kardan: i don't think there's any sense in comparing freerunner with anything. Most free software folks already know what that is. Aug 18 19:33:39 yes, they should :) Aug 18 19:34:21 additionally it would be great to see promote the om as a green mobile phone on epeat Aug 18 19:35:05 there also is a Environmental Benefits calculator: http://www.epeat.net/FastBenefits.aspx Aug 18 19:36:34 there are instructions to be listed by epeat: http://www.epeat.net/Manufacturers.aspx Aug 18 19:40:41 maybe the best is to write a proposal to the list Aug 18 19:45:38 the confusing part is that there are 3 databases in /etc/freesmartphone/opim/ but they are empty except for calls. Aug 18 19:46:19 * Kero thinks he should read up on the opimd dbus API Aug 18 19:48:40 Kero: you need to switch contacts and sms backends to opimd but then you'll have to use only opimd-enabled apps (at least for smses). Aug 18 19:49:14 Kero: dos1 has exlained something like that on community ML recently. Aug 18 19:53:25 ~seen rms Aug 18 19:53:29 pieterc: i haven't seen 'rms' Aug 18 19:53:30 *gr* now I find the basic API, but no clue on which bus, destination or interface to pass to mdbus. Aug 18 19:53:35 pfff, buu to rms Aug 18 19:54:22 mdbus -s is useful, though. Aug 18 19:57:42 Kero: want to know a faster way than mdbus? Aug 18 19:58:00 Kero: dbus-send ;) Aug 18 20:05:34 lindi-: do you by any chance want to try my patch that should improve stability of my testcase and (i hope) shouldn't affect others? Aug 18 20:06:06 lindi-: my testcase is "bind the driver, start wpa_supplicant, use wifi, kill wpa_supplicant; unbind the driver" multiple times. Aug 18 20:07:19 lindi-: i want to check if there're any regressions but lack decent skills. Aug 18 20:11:44 lindi-: http://paste.debian.net/44536/ Aug 18 20:11:49 lindi-: please Aug 18 20:16:43 PaulFertser: now way too busy right now :( Aug 18 20:16:51 lindi-: not right now Aug 18 20:17:06 lindi-: but if you could do that in several days (test the patch for regressions) it'd awesome. Aug 18 20:17:48 not sure, i'm terribly busy at work nowadays Aug 18 20:18:11 PaulFertser: but if you send it to timo.lindfors@iki.fi i can keep it flagged as "important" and i'll hopefully eventually test it Aug 18 20:18:33 PaulFertser: note that i have not used wpa_supplicant manually before, always relied on ifupdown Aug 18 20:19:14 lindi-: heh, did it break for you too? I know only one definete way to break driver currently: to touch any eth0 property while the iface is down. Aug 18 20:19:43 PaulFertser: no i just use ifupdown on all systems so i don't see why i'd use something else on neo Aug 18 20:19:51 i think it calls wpa_supplicant Aug 18 20:20:43 lindi-: sure. But ar6000 is mad buggy, it can fail to do something others have no problems with. That's why details matter. Aug 18 20:26:48 latest SHR wont let me use wifi Aug 18 20:31:06 henkie_: no, it will Aug 18 20:31:13 henkie_: please read FSO_Resources wikipage Aug 18 20:31:28 s/read/read and understand/ Aug 18 20:31:29 PaulFertser meant: henkie_: please read and understand FSO_Resources wikipage Aug 18 20:36:03 PaulFertser, i see Aug 18 20:37:12 too bad it seems to involve a cli, mokonnect used to work for me Aug 18 20:37:27 will do fine for my home network tho Aug 18 20:37:38 (now trying hackable1) Aug 18 20:38:12 btw, removing the sd-card always makes me nervous :) Aug 18 20:38:40 the brackets seems a little fragile Aug 18 20:43:49 hi guys, can anybody tell me how to switch usb-hostmode with kernel 2.6.30.5/openwrt? Aug 18 20:44:05 :c Aug 18 20:44:32 grmlgrml: the same as with .29? Aug 18 20:45:00 doesn#t work Aug 18 20:51:18 error: can't open /sys/class/i2c-adapter/i2c-0/0-0073/gta02-pm-host.0/hostmode no such file... Aug 18 20:52:13 openmoko: 03werner * r5480 10/trunk/eda/fped/ (TODO gui_tool.c inst.c inst.h meas.c meas.h): Aug 18 20:52:13 openmoko: Finally found a nice way to draw arcs in the GUI. Aug 18 20:52:13 openmoko: - we freed a package's samples list after recalculating the number of samples, Aug 18 20:52:13 openmoko: which caused a crash after adding a new vector. We now record the original Aug 18 20:52:13 openmoko: list length in the package structure. Aug 18 20:52:16 openmoko: - when dragging a circle, offer the end point first, so that it becomes an arc Aug 18 20:54:11 grmlgrml: find /sys -name \*host\* Aug 18 20:55:15 mopls Aug 18 20:57:12 openmoko: 03werner * r5481 10/trunk/eda/fped/inst.c: - stupid typo made us crash when setting the package name Aug 18 21:00:45 PaulFertser: ok, found /sys/class/i2c-adapter/i2c-0/0-0073/gta02-pm-host.0/hostmode, gonna try this Aug 18 21:01:06 anyone tried gentoo on the freerunner? Aug 18 21:04:53 how do I link my computer's library files to the toolchain when I compile? Aug 18 21:05:01 include files* Aug 18 21:06:37 PaulFertser: well, seems to be charching current. dmesg: pcf50633 0-0073: usb curlim to 500 mA Aug 18 21:07:53 henkie_: http://people.openmoko.org/joerg/sdcard-handle/ to easy your pain Aug 18 21:08:46 cya... ->HP Aug 18 21:09:03 DocMobilizer, ah, seems a nice idea indeed Aug 18 21:11:41 all my ideas are nice (or mega-nasty) :-P Aug 18 21:12:10 hehe Aug 18 21:12:34 bbl Aug 18 21:13:11 generally, how can i find /sys/... parameters/states? greped in kernel source, don't know where/what to search for Aug 18 21:30:57 HP finally Aug 18 21:31:50 <|Marco|> wee, got my Freerunner A6 today :) Aug 18 21:32:16 <|Marco|> I've still got my A5 from last summer, still working fine :) Aug 18 21:32:18 congrats |Marco| Aug 18 21:32:22 grmlgrml: the path is build from the device hierarchy, so you won't find the the complete path in the kernel sources. the sysfs files are normally declared with DEVICE_ATTR(...). Aug 18 21:32:35 what comes preinstalled on the A6? Aug 18 21:32:44 <|Marco|> om2008.8 Aug 18 21:33:01 <|Marco|> stable build from "barbie" Aug 18 21:33:27 Barbie Bancock Aug 18 21:33:34 <|Marco|> possibly Aug 18 21:33:38 ;) Aug 18 21:33:42 <|Marco|> yeah yeah Aug 18 21:33:48 <|Marco|> pun intended ? Aug 18 21:33:50 <|Marco|> :p Aug 18 21:34:06 yes indeed ;-) Aug 18 21:34:59 <|Marco|> hehe, well.. I't operating fine, haven't tested with a simcard yet Aug 18 21:36:02 :) what's going to be your first flasg? Aug 18 21:36:04 flash Aug 18 21:36:35 <|Marco|> think I'll test om2008 for now, haven't touched it since some time last year Aug 18 21:36:51 <|Marco|> I've got Hackable1 on the A5 running well sofar Aug 18 21:36:52 hmmm... It's pretty much outdated Aug 18 21:37:10 imo, you're not doing any good testing om2008.8 :p Aug 18 21:37:15 <|Marco|> I could fetch a newer image of om200x.x Aug 18 21:37:23 sure Aug 18 21:37:30 2009 is pretty cool with paroli Aug 18 21:37:52 and at least you'll be able to install some recent packages Aug 18 21:37:56 like mokomaze ;) Aug 18 21:38:04 <|Marco|> aha :P Aug 18 21:39:14 <|Marco|> does the A6 have the Moko11 GSM fix ? Aug 18 21:39:41 |Marco|: it's something you should tell us ;) From whom did you bought it? Aug 18 21:40:10 <|Marco|> hrm, a buddy bought a bunch.. not quiet sure how or when :/ Aug 18 21:40:41 <|Marco|> can I get an output somewhere or is it only the stamp under the battery ? Aug 18 21:42:11 |Marco|: of GSM firmware version? Aug 18 21:42:29 |Marco|: there's AT+CGMR=? command Aug 18 21:42:47 |Marco|: no idea about how to use it in qtopia. Aug 18 21:42:57 <|Marco|> qtopia ? Aug 18 21:43:13 qtopia? Aug 18 21:43:15 <|Marco|> well, I'm hoocked up with it on usb and ssh shell Aug 18 21:43:31 <|Marco|> om2008.8 != qtopia Aug 18 21:43:50 <|Marco|> PaulFertser: yeah, the GSM firmware Aug 18 21:44:33 |Marco|: om2008.8 == qtopia/X Aug 18 21:44:48 <|Marco|> oh ? sorry, I didn't know that Aug 18 21:45:43 |Marco|: you better flash SHR unstable anyway, om2008 was deadborn and nobody's interested in it at all currently. Aug 18 21:46:16 <|Marco|> PaulFertser: no, not yet Aug 18 21:46:23 |Marco|: and if you want to try paroli, you can easily run it on SHR too. Aug 18 21:47:05 <|Marco|> we'll see Aug 18 21:47:42 <|Marco|> I'm not that keen on SHR, so we'll see what comes out of playing with om200x.x on uSD first for a while Aug 18 21:48:23 <|Marco|> I'll leave the om2008.8 on flash as backup if the uSD distro get's corrupted (as I've experienced so many times before in mainly SHR) Aug 18 21:48:29 |Marco|: I4m not keen on SHR, bad thing to say on this channel ;) Aug 18 21:49:28 Not at all Aug 18 21:50:02 <|Marco|> what I don't like is the people trying to "sell" it :/ If I find it comfertable, I'll use it, but I want to test other distro's first Aug 18 21:50:36 right Aug 18 21:50:53 I just wanted to notice there are many shr devs in here Aug 18 21:51:23 personally, I've found H:1 to be most desirable in terms of phone AND app usage. that is, if you have access to a 2gb uSD Aug 18 21:51:42 never tried h:1 yet Aug 18 21:51:48 maybe I should Aug 18 21:52:23 <|Marco|> chocotuar: have you tested with a uSD larger than 2GB and filling it with whatever over 2GB while in H:1 ? Aug 18 21:52:36 it's very responsive and most things work out-of-the-box Aug 18 21:52:50 larsc: thanks, found:ohci-s3c2410.c:static DEVICE_ATTR(usb_mode, S_IRUGO | S_IWUSR, show_usb_mode, set_usb_mode), seems to be boolean, but thats quiet cryptic to me. well i'm learning Aug 18 21:53:01 |Marco|: nope, I have about 6 2gb uSDs, but none bigger than that Aug 18 21:53:19 <|Marco|> on all distro's so far I've hit this magical limit and I'm puzled.. I'm guessing it has todo with the SDHC protocol ? Aug 18 21:53:24 <|Marco|> *on some cards Aug 18 21:54:03 <|Marco|> chocotuar: okay, I've got a 2GB and a 4GB Aug 18 21:54:10 <|Marco|> both are kingston Aug 18 21:54:43 |Marco|: no, sdhc should work as well. Just to boot from those you need to use a modern bootloader. Aug 18 21:54:44 |Marco|: the one I use is a kingston and it works fine, I just suggest ext3 instead of ext2 like the wiki calls for... less filesystem checks Aug 18 21:56:03 <|Marco|> chocotuar, PaulFertser I've got ext3 and qi on it, and it boot's just fine, just can't write more than 2GB data in total, get device is full errors at ~2GB Aug 18 21:56:23 |Marco|: are you using vfat there? ;) Aug 18 21:56:27 <|Marco|> no Aug 18 21:57:47 |Marco|: never heard of that happening... I remember I had a problem where my computer couldn't recognize the size of the uSD if it was bigger than 1gb, but that had nothing to do with the phone. might be the same issue Aug 18 21:58:44 <|Marco|> chocotuar: I've got no idéa Aug 18 22:00:13 <|Marco|> hrmf.. the theme and desktop changed to E .. I liked the black desktop that was originally in om2008.8 Aug 18 22:03:54