**** BEGIN LOGGING AT Tue Feb 17 02:59:58 2009 Feb 17 03:44:22 make shr-lite-image ? Feb 17 04:00:13 gah, does anybody here know how to convert an int to a char* in c++? Feb 17 04:01:56 sprintf or itoa Feb 17 04:02:01 or itol Feb 17 04:03:12 cast it like it's hot Feb 17 04:03:20 those are c, it's for class so i can't use the c libs : ( Feb 17 04:03:37 * Dave stabs fenfir Feb 17 04:03:46 yay Feb 17 04:03:51 that's the third time today Feb 17 04:03:54 yeah, you deserved that Feb 17 04:11:27 http://pastebin.com/d24c523f7 Feb 17 04:11:29 ? Feb 17 04:13:27 fenfir: u are doing c./. and u can't use libc? Feb 17 04:13:41 c++ Feb 17 04:14:39 the professor is really picky about stuff, if i could i'd just use the c libs and do it in straight c Feb 17 04:14:41 exactly Feb 17 04:14:43 haha Feb 17 04:14:50 so u cant use printf? Feb 17 04:14:56 or malloc Feb 17 04:14:57 if you're using C++ you shouldn't be using (char *) Feb 17 04:14:57 or free Feb 17 04:15:00 or atoi Feb 17 04:15:03 etc. Feb 17 04:15:10 Professor Raster Feb 17 04:15:11 ie all the stanard most basic libc calls? Feb 17 04:15:21 is it that he wants you to use c++ stuff Feb 17 04:15:24 ie stl? Feb 17 04:15:28 instead of c Feb 17 04:15:29 yeah Feb 17 04:15:35 to force you to write "pure c++" Feb 17 04:15:41 exactly Feb 17 04:15:46 or is he forcing you to do it by hand so u learn the basics Feb 17 04:15:56 ie how to convert a number into a string Feb 17 04:16:03 you can set up a stringstream and "<<" the int into it Feb 17 04:16:05 std::string::Ftw Feb 17 04:16:09 huhu Feb 17 04:16:14 cast it like it's hot Feb 17 04:16:18 so u basically would have to implement your own code and the only functions u can use are write() and reader() for example Feb 17 04:16:19 regardless of what language you're in! Feb 17 04:16:42 bah Feb 17 04:16:55 your professor is... not living in the real world Feb 17 04:17:00 raster Feb 17 04:17:02 c++ has full access to all c constructs Feb 17 04:17:08 often its just easier to use c Feb 17 04:17:10 clearly you've never taken a college course Feb 17 04:17:17 he's teaching you incredibly bad habits Feb 17 04:17:21 i know Feb 17 04:17:21 they teach bullshit, backwards, and unrealistic stuff Feb 17 04:17:29 :D Feb 17 04:17:30 the "c++ way" almost always incurrs extra overhead for the abstraction Feb 17 04:17:38 heh go for a 0 and do it the *right* way Feb 17 04:17:38 raster, but it works! Feb 17 04:17:41 you should chose the simplest way to get what u want always Feb 17 04:17:44 and it looks prettier :P Feb 17 04:17:53 use perl; Feb 17 04:18:00 if u need a class, object and operator overloading Feb 17 04:18:00 then use c++ Feb 17 04:18:03 if that actually makes sense for you and what u are doing Feb 17 04:18:12 efficiency has little value in acedemia. Feb 17 04:18:20 always Feb 17 04:18:23 Just look at Openmoko! Feb 17 04:18:25 :D Feb 17 04:18:43 heh Feb 17 04:18:50 :D of course professors get paid to waste time ;D Feb 17 04:18:56 huhuhu Feb 17 04:19:02 and teach incomprehensible shit Feb 17 04:19:03 I came out of college to my first job: a quote from my boss after one of my customer projects: "Nice, but customers don't pay us for elegance, they pay us for results." Feb 17 04:19:19 buhuhuhu Feb 17 04:19:26 that's what my boss says to me Feb 17 04:20:24 but c++ is ticking me off , there was lots of stuff in c that was much easier to do and seemed much much more elegant Feb 17 04:21:24 mwester-laptop: aaah thats so true Feb 17 04:21:28 results count Feb 17 04:21:38 elegance is only good when it comes for free with the results Feb 17 04:21:39 :) Feb 17 04:23:39 i got the stringstream thing to work Feb 17 04:23:52 good. Feb 17 04:24:24 another problem i'm having with c++ is that visual studio has ruined me Feb 17 04:24:45 i can hit tab for twenty minutes and have a small program written with almost no actual input Feb 17 04:24:58 yeah Feb 17 04:25:07 should force u to be stuck with vi or emacs Feb 17 04:25:08 and nothing else Feb 17 04:25:18 u dont learn by becoming reliant on an IDE Feb 17 04:25:25 IDE's are nice.. once u know stuff already Feb 17 04:27:18 tab? Feb 17 04:27:30 oh, does it tab-complete keywords/instructions for you? Feb 17 04:28:54 function calls Feb 17 04:28:56 data types Feb 17 04:29:02 variables Feb 17 04:29:03 yes Feb 17 04:29:06 haha Feb 17 04:29:15 The old versions never did that. Feb 17 04:29:32 I do all my c and c++ with vi Feb 17 04:34:09 fenfir: good call Feb 17 04:34:22 fenfir: trust us - your professor is disconnected from the real world Feb 17 04:34:27 if u ask me Feb 17 04:34:35 raster: i know that Feb 17 04:34:52 best way to learn to code after getting a quick intro with basic or python or some high level thing just to get u an idea of hwat is posible Feb 17 04:34:54 is to go straigh to asm Feb 17 04:34:57 straight Feb 17 04:35:01 learn assembly Feb 17 04:35:07 get comfortable with it Feb 17 04:35:09 and the ideas Feb 17 04:35:23 is this computer SCIENCE or computer ENGINEERING? If it's CS, then C++ sounds like what it should be Feb 17 04:35:24 (registers, pointers, everything is a memory address, a number and all you do is logic and math ops) Feb 17 04:35:34 after u can get your head around all of that with comfort Feb 17 04:35:35 go to c Feb 17 04:35:40 c is macro-assembly Feb 17 04:35:40 budfive: CS Feb 17 04:36:01 just a faster way to write asm Feb 17 04:36:13 then after c - pick and choose - u've learnt how to do it now Feb 17 04:36:26 (before u do c++ i'd elarn how to do oo in c with structs, func pointers etc.) Feb 17 04:38:01 I've been doing C# for work for a couple of years, with a bit of everything else in my free time, haven't considered asm yet Feb 17 04:40:36 C# runs in a VM, right? Feb 17 04:41:06 raster: intro by basic spoils pupils for live, never will become real programmers. Feb 17 04:41:20 * docscrutinizer2 started on assembler Feb 17 04:41:28 budfive: yeah Feb 17 04:41:29 :) Feb 17 04:41:48 hex machine code, paper and pencil that is Feb 17 04:42:07 Punch cards! Feb 17 04:42:14 s360 \o/ Feb 17 04:42:20 EBCDIC! Feb 17 04:42:41 ebcdic yeeehaa Feb 17 04:42:45 :D Feb 17 04:44:39 docscrutinizer2: i started on basic Feb 17 04:44:47 so apparently i'm not a real programmer Feb 17 04:44:48 :) Feb 17 04:45:00 of course this was CBM basic Feb 17 04:45:02 on a vic20 Feb 17 04:45:05 with 4kb of ram Feb 17 04:45:22 and your basic had lots of PEEK and POKE calls to memory addresses Feb 17 04:45:22 :) Feb 17 04:49:39 I think computing lost something when they got rid of the WAIT Light Feb 17 04:59:21 HAHAHAH Feb 17 05:02:19 hit the road jack... Feb 17 05:05:58 what wait light? Feb 17 05:08:16 exactly Feb 17 05:08:34 good thing I recreated it several years back :D Feb 17 05:17:48 orzo! Feb 17 05:21:17 Dave Feb 17 05:21:20 i'm building a kernel Feb 17 05:21:24 for the gta01 Feb 17 05:21:58 that will allow the zd1211rw driver for 802.11 wifi from a usb adapter Feb 17 05:22:36 its a small patch to mwester's config Feb 17 05:27:50 orzo: evil wlan hacker ;) Feb 17 05:28:16 hahaha nice Feb 17 05:28:21 I have one of those :p Feb 17 05:28:34 really? Feb 17 05:28:36 yep Feb 17 05:28:38 you se that driver? Feb 17 05:28:45 on the moko? Feb 17 05:28:55 at all Feb 17 05:28:58 sure Feb 17 05:29:05 I've compiled it before, for linux 2.6 Feb 17 05:29:22 its included with the kernel Feb 17 05:29:28 Well now it is. Feb 17 05:29:36 This was back when there were no wireless drivers Feb 17 05:29:56 except the natives Feb 17 05:30:25 haha oh man, good times *rolls eyes* Feb 17 05:30:28 has anybody actually use 802.11 wifi on a gta01 ? Feb 17 05:30:46 one guy, I had a link to it... though he gave no instructions Feb 17 05:30:58 It was more of a resource about how he built his usb hub for it Feb 17 05:31:02 but you've already done that :) Feb 17 05:32:02 it's done Feb 17 05:34:06 yep yep Feb 17 05:34:14 hm, my uImage is 8440 bytes smaller than mwester's Feb 17 05:35:13 maybe you lack additional coolness! Feb 17 05:35:24 either that, or you're cooler than he is Feb 17 05:35:29 (which I doubt) :P Feb 17 05:35:46 no, its probably due to a compiler difference Feb 17 05:36:33 i didnt use the open embedded enviro Feb 17 05:37:07 i used a cross compiler supplied by apt-cross in debian used by the emdebian project Feb 17 05:37:33 gcc 4.3.2 Feb 17 05:38:46 arm-linux-gnueabi-gcc Feb 17 05:38:56 good shit Feb 17 05:40:30 yeah, apt-cross is nicer than oe. the whole cross environment is managable with apt and dpkg and such Feb 17 05:42:08 the problem is that debian as such is big and fat Feb 17 05:42:08 :( Feb 17 05:42:45 nah Feb 17 05:42:52 not for a pc anyway :p Feb 17 05:44:08 heh, ya, my cross enviro is on a pretty powerful laptop Feb 17 05:44:25 turion 64 x2 Feb 17 05:44:35 debian is fine Feb 17 05:44:59 but i run debian on teh phone too and it is a tad fat Feb 17 05:45:03 heh Feb 17 05:45:46 :D Feb 17 05:46:10 fso's framework is a tad fat Feb 17 05:46:12 heh Feb 17 05:46:20 memory usage wise Feb 17 05:49:34 i had problems where i was accidently trigguring firefox by accidently selecting help from a menu in a terminal program Feb 17 05:49:49 and firefox of course brought the whole thing to a screeching halt Feb 17 05:50:07 hahaha Feb 17 05:50:11 though it probably would have been ok if i wasnt running gps and other things Feb 17 05:50:13 hellllllllll yeaaaaaaaaah Feb 17 05:51:16 anyway, emdebian is supposed to be leaner than debian proper Feb 17 05:51:36 it probably uses busybox Feb 17 05:51:50 eeek Feb 17 05:52:02 crap Feb 17 05:52:09 yep Feb 17 05:52:35 i havent run emdebian on anything Feb 17 05:52:50 but they're work on making debian cross-buildable is much appreciated Feb 17 05:52:53 their Feb 17 05:54:37 Dave: for living on 64,128 or 256m of flash.. it is Feb 17 05:54:37 :) Feb 17 05:54:46 haha well yeah :) Feb 17 05:55:00 orzo: yes. fso gobbles up ram Feb 17 05:55:26 orzo: i havent tried embedian Feb 17 05:55:36 but to me it seems like that'd be a better way to go Feb 17 06:01:55 my rootfs for debian is on an sd card Feb 17 06:01:55 i have qtopia on the built in flash Feb 17 06:01:55 but i hardly use it Feb 17 06:01:55 raster: is the openmoko os going to use fso? Feb 17 06:01:55 is fso going to be re-implemented in vala or something at some point? Feb 17 06:01:56 heh Feb 17 06:01:58 i assume vala would goble less ram than python, but i guess i dont really know Feb 17 06:02:05 hells yeah Feb 17 06:02:37 orzo: yes. fso as tyhe infra for a phone is what om plans to use Feb 17 06:02:40 thats why they are funding it Feb 17 06:02:50 beyodn that dont ask me - dont work for om :) i cant speak for them Feb 17 06:03:47 python's dbus bindigns seem oddly complicated to me Feb 17 06:04:37 i guess i haven't thought abuot it a great deal, but i would think such a flexible oo system would be able to hide more complexity Feb 17 06:05:35 firefox's xpcom itnerface is a lot uglier than apple's objective c interface to webkit Feb 17 06:06:01 im not sure what teh original c++ interface to webkit is like Feb 17 06:06:25 but i figured the dynamic oo in objective c was a factor Feb 17 06:07:12 dbus in c isnt that nice Feb 17 06:07:18 but its doable Feb 17 06:07:24 you just have to wrap it up anjd do footwork Feb 17 06:07:30 write helper funcs for common things Feb 17 06:12:36 i find matchbox-desktop eats ram Feb 17 06:12:40 on my debian anyway Feb 17 06:12:52 maybe there are too many .desktop files Feb 17 06:14:31 what does mb-desktop do? Feb 17 06:14:34 how much ram? Feb 17 06:16:10 loading .desktop file data isnt cheap Feb 17 06:16:13 there can be a lot of it Feb 17 06:16:18 depends how you intend to use it Feb 17 06:16:32 i know in e we are forced to pretty much load them all to kepe a ram copy of most common fields Feb 17 06:16:39 because we allow search by globs/fields Feb 17 06:16:51 and this needs to be fast (syncronous) Feb 17 06:16:54 as part of the ui Feb 17 06:18:56 well fat debian loads it's .desktop files with translations for all teh languages Feb 17 06:19:01 iduno if that matters Feb 17 06:22:21 mb-desktop is using 10.5% of physical mem, frameworkd is using 15.7%, xorg is using 12.9%, kapula is using 12.6%, everything else is single digits Feb 17 06:22:43 whats the RSS and SHR? Feb 17 06:22:46 i think kapula and frameworkd could be made to share a python instance heh Feb 17 06:22:58 measuring memory usage is a fine art Feb 17 06:23:07 never read those numbers and just assume them Feb 17 06:24:39 that % mem includes SHR Feb 17 06:24:49 u'll find if u add them up they will often be > 100% Feb 17 06:25:12 shared memory is things like mmaped files (whihc are paged to/from disk by the vm system anyway) Feb 17 06:25:19 as well as the executable itself whihc gets paged Feb 17 06:25:25 i have no swap Feb 17 06:25:28 shared libs (again paged.. AND shared) Feb 17 06:25:36 so if everyone uses libc Feb 17 06:25:46 everyone will show the same 1mb "shr" segment Feb 17 06:25:51 even tho only 1 copy of 1mb of glibc is in ram Feb 17 06:25:55 everyone shows to use 1mb Feb 17 06:26:03 u dont need swap to page Feb 17 06:26:04 mb-desktop is using 12M RES and 7768 SHR Feb 17 06:26:12 if tis mapped from disk Feb 17 06:26:15 it will page Feb 17 06:26:26 swap is a place for memory to go when its NOT mapped to/from disk Feb 17 06:26:35 (ie allocated ram) Feb 17 06:26:37 hmm Feb 17 06:26:42 thats not to obad Feb 17 06:26:51 7.8m is the xecutable and shared libs Feb 17 06:26:56 use pmap to check Feb 17 06:27:04 pmap PID-OF-PROC Feb 17 06:27:10 and it will show you all the memory mapings Feb 17 06:27:18 u'll see just how much is from each shared lib Feb 17 06:27:24 eg libgtk will be in SHR Feb 17 06:27:27 libc Feb 17 06:27:28 libm Feb 17 06:27:31 a tonne of libs Feb 17 06:27:38 as mb-desktop is gtk based Feb 17 06:27:48 the mapped ram from gtk will be the SAME mammped ram all other gtk using apps will use Feb 17 06:28:01 generally speaking its wise just to ignore the SHR segment Feb 17 06:28:11 worry about RES - SHR Feb 17 06:28:18 thats ram u would normally need to swap out Feb 17 06:28:21 if u have swap Feb 17 06:28:23 sicne u dont Feb 17 06:28:26 it has nowhere to go Feb 17 06:28:33 thats the stuff to care about Feb 17 06:28:35 :) Feb 17 06:28:48 does top have a RES - SHR column? Feb 17 06:28:49 heh Feb 17 06:28:54 no it doesnt Feb 17 06:29:01 u'll need to do the math Feb 17 06:29:02 :) Feb 17 06:29:21 well i wanted a live sort and all Feb 17 06:30:17 u'll have to do your own :) Feb 17 06:30:28 but really mb-desktop is using closer to 5-6m ram Feb 17 06:30:59 measuring the real memory usage of an app is near black magic Feb 17 06:31:02 thus u cant say for sure Feb 17 06:31:07 the ONLY way is this Feb 17 06:31:14 start a systsme cleanly from scratch Feb 17 06:31:16 system Feb 17 06:31:27 now start all apps EXCEPT the one u want to measure Feb 17 06:31:35 run free Feb 17 06:31:44 -/+ buffers/cache: 490780 1547112 Feb 17 06:31:48 that number first there Feb 17 06:31:55 the 490780 Feb 17 06:32:04 thats the real usage to worry about Feb 17 06:32:07 now - run app Feb 17 06:32:10 then run free again Feb 17 06:32:13 look at same number Feb 17 06:32:17 subtract 2nd one from first Feb 17 06:32:26 and u get an "idea" of its footprint Feb 17 06:32:31 relative to the system its on Feb 17 06:32:42 no app has an absolute footprint unless its 100% statically compiled Feb 17 06:32:49 and even then the shared pages are pagable even without swap Feb 17 06:33:42 (sorry - just trying to clear up myths on evaluating memory usage as very few people really know what all the numebrs mean that u get from top/ps/free etc. - and those that do know stopped looking at the numbers long ago as they know its just voodoo) Feb 17 06:33:42 :) Feb 17 06:34:12 :D Feb 17 06:38:34 (you just track all ram u malloc via toosl like memprof or valgrind etc. and the rest you specially worry about case-by-case) Feb 17 06:40:01 suppose i have an application Feb 17 06:40:19 its running now Feb 17 06:40:32 can i quit it and then run free and then start it again and compare? Feb 17 06:40:39 or do i need to reboot? Feb 17 06:40:48 and keep it from running on the next boot Feb 17 06:41:19 you can Feb 17 06:41:50 it works in reverse - to an extent Feb 17 06:41:56 some serives the app may use Feb 17 06:41:58 like x Feb 17 06:42:07 or other daemons may have 1-time setup costs for that app Feb 17 06:42:18 ie if that app starts for the first time and asks to allocate something Feb 17 06:42:25 lets say for example a wallpaper for your desktop Feb 17 06:42:33 that wallpaper normally should live in video ram Feb 17 06:42:38 but on some devices - or many Feb 17 06:42:42 video ram is in system ram Feb 17 06:42:44 its the same thing Feb 17 06:42:54 so the wallpaper is malloc()ed FOR that app BY x Feb 17 06:42:59 extreme example Feb 17 06:43:27 1920x1200 - a wallpaper is 9mb @ 32bpp Feb 17 06:43:30 so app starts Feb 17 06:43:41 allocs a wallpaper and sets root window from its plain blakc color to wallpaper Feb 17 06:43:46 u kill app Feb 17 06:43:50 wallpaper is still there Feb 17 06:43:56 if wallpaper pixmap lives in system ram for x Feb 17 06:44:12 this 9m will still be alloced even if the app is killed as x holds the resource until its not needed anymore Feb 17 06:44:16 (someone resets the wallpaper) Feb 17 06:44:22 so it can be misleading Feb 17 06:44:38 from a linux image is there an easy way to tell what uname -r will tell me without actually booting the image? Feb 17 06:44:51 hmmm Feb 17 06:45:02 i swear there used to be one Feb 17 06:45:08 i cant remember the tool now Feb 17 06:45:22 last i used it i think was 111 or 12 years ago Feb 17 06:45:25 err Feb 17 06:45:26 11 or 12 Feb 17 06:45:27 :) Feb 17 06:46:16 i was gona guess 12 Feb 17 06:46:22 given teh choice Feb 17 06:46:57 i'm reminded of bilbo's eleventy first birthday Feb 17 06:47:47 hahahahah Feb 17 06:49:07 :p Feb 17 06:49:09 orzo: u can try this Feb 17 06:49:33 cat uImage | strings | grep 2.6. Feb 17 06:49:40 heh Feb 17 06:49:43 ok Feb 17 06:49:48 5:49PM ~ > cat /boot/vmlinuz-2.6.24-23-generic | strings | grep 2.6. Feb 17 06:49:48 2.6.24-23-generic (buildd@vernadsky) #1 SMP Mon Jan 26 00:13:11 UTC 2009 Feb 17 06:52:19 well uIMage didnt work, but the vmlinux did Feb 17 06:52:53 the strings are probably compressed in the uImage version Feb 17 07:01:11 can i use the dfu-util in debian on my x86_64 computer? Feb 17 07:01:28 i assuem so - if its the x86_64 build of it Feb 17 07:03:44 flashing kernel Feb 17 07:07:36 it works Feb 17 07:07:38 well Feb 17 07:07:49 its booted and modprobe -l shows zd1211rw.ko Feb 17 07:09:58 hot Feb 17 07:17:36 plugging in the adapter didnt make any dmesg statements Feb 17 07:17:41 :\ Feb 17 07:17:46 no hotplug, huh? Feb 17 07:18:01 loading teh module with the plugged in adapter has modprobe hang Feb 17 07:18:15 rut roh Feb 17 07:18:17 although im not 100% i copied the right firmware files Feb 17 07:18:36 debian unhelpfully has a firmware package that is not right Feb 17 07:18:49 get it yourself :P Feb 17 07:18:54 don't trust the repo shit :D Feb 17 07:20:20 after unplugging teh device and putting the neo back in device mode rather than host mode, modprobe unhung and loaded teh driver Feb 17 07:20:41 weird Feb 17 07:20:47 i'm going to try a usb keyboard to verify i havent broken my altoids thing by carying it around everywhere Feb 17 07:29:30 ok Feb 17 07:29:45 i changed a cord and got the usb keyboard to make output in dmesg Feb 17 07:31:57 it detected my device and created eth0 Feb 17 07:32:08 but im unprepared apparently i dont have wireless tools installed Feb 17 07:47:20 ok, ringing when suspended did not work. trying to flash gsm firmware now... Feb 17 07:47:23 morning... Feb 17 07:52:03 okay, i see my ssid in iwlist scan output :) Feb 17 07:52:14 i have wifi on a gta01 Feb 17 08:23:11 how long would flashing a 5MB image take? Feb 17 08:33:03 yay, it worked Feb 17 08:41:38 morning Feb 17 08:41:55 morning Feb 17 08:43:49 spaetz: so you now have resume from suspend on call ? Feb 17 08:44:45 guess i'll have to get around to flashing the new fw soon too Feb 17 08:45:45 new fw? u-boot,qi? kernel? rootfs? Feb 17 08:46:28 sicu: no really tested yet. I am just happy that flashing worked and I am still able to make phone calls :) Feb 17 08:46:45 admiral0: GSM Feb 17 08:47:07 ah nice Feb 17 08:47:28 confussled calypso Feb 17 08:47:45 i flashed moko10 beta some time ago, and i could finally use my vodafone SIM Feb 17 08:47:56 is there a new version? Feb 17 08:47:58 admiral0: same here =] Feb 17 08:48:17 yes, 11b2 or something along those lines Feb 17 08:48:43 the OpenMoko guys are giving to calypso a life cycle very long, with this firmware updates, you could use an ancient GSM modem with more or less pain :) Feb 17 08:48:43 admiral0: http://people.openmoko.org/joerg/calypso_moko_FW/moko11/ Feb 17 08:48:53 (i just love my freerunner, where in other cell phones you can flash firmware of GSM???) Feb 17 08:49:27 it's a step in the right direction =] Feb 17 08:50:12 there are also other ideas of open phone, cough*,raster,*couch* Feb 17 08:50:29 s/couch/cough/ Feb 17 08:50:30 admiral0 meant: there are also other ideas of open phone, cough*,raster,*cough* Feb 17 08:54:31 well, it doesnt work. I can scan networks but i've yet to be able to connect to anything Feb 17 09:01:22 orzo: if you can scan, you're almost there :) Feb 17 09:04:32 lindi-: AYT? Want to see a patch that allows oopsless recording with DAI 1 statefile? not that it allows VoIP, but still? I should understand WM8753 deeper to actually do something more meaningful. Feb 17 09:06:51 wpwrak: ya well, it doesnt seem like its going to work. I notice the iwconfig channel command fails also Feb 17 09:07:00 i have to sleep Feb 17 09:35:06 ptitjes: ping Feb 17 09:39:34 lindi-: https://paulfertser.is-a-geek.org/files/0001-Hack-to-temporarily-avoid-Oops-on-recording-with-DAI.patch Feb 17 09:57:32 intersting, after flasing the unstable from now, I don't have usb0 anymore Feb 17 10:00:37 blarg, first boot after flashing, of course it doesn't work Feb 17 10:00:47 reboot fixed it Feb 17 10:00:48 * spaetz hits head against wall Feb 17 10:02:08 * mrmoku passes spaetz some icepacks to cool the lump Feb 17 10:03:08 hello Feb 17 10:03:15 hi spaetz and mrmoku Feb 17 10:03:24 ptitjes: I'm having problems to build your stuff ;) Feb 17 10:03:30 arf Feb 17 10:03:44 like missing dependencies or... Feb 17 10:05:10 samples/Makefile.am:1: option `subdirs-object' not recognized Feb 17 10:05:16 (for libmodulo Feb 17 10:05:17 ) Feb 17 10:05:37 same for src/ subdir Feb 17 10:05:50 ok you can remove that I think Feb 17 10:06:08 you know my makefiles are mainly copy-and-paste Feb 17 10:06:11 :p Feb 17 10:06:25 ptitjes: :-) so remove and push... I don't have access :P Feb 17 10:06:41 * mrmoku could give himself access... but does not want to ;) Feb 17 10:06:51 btw I must tell you Feb 17 10:06:58 you might have a problem to run them Feb 17 10:07:31 because it relies on a vala's patch I hope will be in next release (I hope this week) Feb 17 10:07:37 so we'll have to bump vala Feb 17 10:07:42 for libmodulo to run Feb 17 10:07:44 :) Feb 17 10:10:13 * mrmoku moving to home office bbiab Feb 17 10:10:33 see u Feb 17 10:23:57 PaulFertser_: cool! Feb 17 10:24:41 lindi-: not that cool, read the commit message :-/ I have to learn a lot to understand this ASoC stuff. Feb 17 10:33:04 PauFertser: this is nice. Does it allow you to record with no problems? Feb 17 10:33:14 PaulFertser_, I mean. Feb 17 10:34:53 PaulFertser_: I think this patch will be liked. Feb 17 10:35:06 PaulFertser_: Will you send it? I think it helps. Feb 17 10:36:58 lindi-: for a long time I though I was not reproducing the workaround for recording but that's because my mic's volume is sooo low that I had not noticed I was actually recording... Feb 17 10:38:04 Arhuaco: It's not a solution, more like a quick hack. Can you test it? Feb 17 10:38:22 PaulFertser_: Sure!~ Feb 17 10:39:55 Arhuaco: ah, you're Nelson :) Are you looking in the alsa issues? ;) Feb 17 10:40:04 PaulFertser_: It helps us move from the problem of the .ops being NULL... I will test it now. I think Andy would take it. The problem when DAI=1 seems to be that somehow codec information is not there... I don't know, I'll debug again with your patch in. I'm planning to see what changing the DAI setting does.... Feb 17 10:40:20 PaulFertser_: I'm trying to... it's really hard :-) Feb 17 10:42:06 Arhuaco: I think you already saw my first humble steps ;) Feb 17 10:42:52 Arhuaco: i joined #alsa-soc and talking with Graeme atm, also broonie told me he can answer some questions later. Feb 17 10:43:17 PaulFertser_: Great! Can I snoop? :-) Feb 17 10:43:25 Arhuaco: why not? :) Feb 17 10:44:32 PaulFertser_: Not humble steps at all! It's great to see you're making progress on that issue. Feb 17 10:44:57 Arhuaco: at least now module reloading works Feb 17 10:45:56 PaulFertser_: Yes, it's a PITA to debug drivers when you cannot use modules... thus it helps a lot. Feb 17 10:46:11 * Arhuaco tests the patch Feb 17 10:47:59 Arhuaco: btw, how's supposed to ensure that stable/andy-tracking works on GTA01? Feb 17 10:49:43 PaulFertser_: AFAIK Balaji Raio got a GTA01 and he is testing (or will test) this. You might want to ping him via email if you want to be sure he is working on it. His email is balaji@open... Feb 17 10:50:31 Arhuaco: i hope he'll have time... GTA01 are _very_ sad because andy-tracking doesn't work for them for quite some time. Feb 17 10:51:11 PaulFertser_: Would you like me to ask him? If you want I can send him and email and CC you. Feb 17 10:51:47 mrmoku|away: corrected and pushed Feb 17 10:52:54 Arhuaco: I'm not a GTA01 owner, but i feel bad for all those who trusted OM to provide them with a decent kernel. Feb 17 10:53:16 PaulFertser_: Let's ask him... I'll send the email. Feb 17 10:55:03 Arhuaco: thanks a lot :) Feb 17 11:39:57 ptitjes: ok... will retry Feb 17 11:54:27 DOS , DOS1, DOOOS1 ! Feb 17 11:54:36 need you here Feb 17 11:58:40 spaetz: did he bork something? ;) Feb 17 12:02:43 ptitjes: http://pastebin.com/d4185a71e Feb 17 12:04:59 mrmoku: ok! this needs new vala :( Feb 17 12:05:20 heyho Feb 17 12:08:04 ptitjes: an not yet released version then... there is no tarball newer than 0.5.6... Feb 17 12:08:58 mrmoku: yeah I just wait for the new release... Feb 17 12:09:09 may come during the week Feb 17 12:09:23 ok Feb 17 12:30:48 mrmoku: no, I have a minor patch to shr-settings :) Feb 17 13:25:55 Hello. Guys Feb 17 13:26:12 Is there any decent guide for building FSO?? Feb 17 13:26:42 http://wiki.openmoko.org/wiki/Building_FSO. Is this link outdated? Feb 17 13:27:37 why is there no kernel-module-squashfs for 2.6.28? Feb 17 13:28:05 Zorkman, becouse squashfs wasn't in vanilla kernel Feb 17 13:28:41 squashfs included in kernel only in 2.6.29 Feb 17 13:34:46 KaraDog, http://wiki.openmoko.org/wiki/FSO Feb 17 13:42:05 KaraDog, http://wiki.openmoko.org/wiki/Building_FSO Feb 17 13:50:48 is there a bluez-utils-alsa - 3.33-r3 Feb 17 13:50:54 for shr unstable? Feb 17 13:51:53 i want to try out ad2p later on tonight, but the wiki http://wiki.openmoko.org/wiki/A2DP says i will need this bluez-utils-alsa Feb 17 13:57:47 Zorkman i have it downloaded from FSO http://downloads.freesmartphone.org/fso-testing/feeds/armv4t/bluez-utils-alsa_3.33-r1_armv4t.ipk Feb 17 13:59:34 but until now no success with a2dp on SHR :( i can pair i can connect but not play :( Feb 17 14:11:29 hmmm could i have to do with the FSO version of that bluez-utils-alsa von_fritz? Feb 17 14:11:36 and have you folowed the wiki? Feb 17 14:16:12 von_fritz: which bt device do you use? Feb 17 14:18:38 anybody else had succes with ad2p and shr? Feb 17 14:31:54 Zorkman b-speech headset, SHR use bluez4 and for pairing i have used simple-agent.py include in bluez4 package Feb 17 14:33:31 no success with "conventional" passkey-agent :( Feb 17 14:40:17 Zorkman, von_fritz: bluez-utils-alsa hit the SHR feed :-) Feb 17 14:49:13 lindi-: So, do you know of any sound problem that prevents real usage (as opposed to trying to make it oops on purpose)? Feb 17 14:54:13 mrmoku: thanks! you're great Feb 17 14:54:53 von_fritz: i only have bluez 3.33 Feb 17 14:55:09 and i'm using shr unstable from 02/09, where did you get bluez 4 from? Feb 17 14:57:22 is someone here who can help me with the shr toolchain? toggles_w ? Feb 17 14:57:28 PaulFertser_: not right now Feb 17 14:57:38 PaulFertser_: now that the audio recording from mic works Feb 17 14:57:57 lindi-: but it was working with voip state file before :) Feb 17 14:58:47 PaulFertser_: but i did not know that :) Feb 17 14:59:02 PaulFertser_: how about recording from mic, doing some filtering on cpu and sending it to gsm? Feb 17 14:59:43 Zorkman: got bluez4 to build... so it is in current SHR now Feb 17 14:59:59 lindi-: i'm still going to try to understand this stuff deeper, i guess this should be possible though doesn't make any sense to me. Feb 17 15:00:51 real 567m25.947s to fail to build shr-lite ;-) Feb 17 15:02:35 toggles_w: what package? Feb 17 15:04:24 mrmoku: the kernel, rerunning now for changes Feb 17 15:04:37 dir the PR getr bumped for the gsm fix? Feb 17 15:06:13 will shr unstable 02/09 break if i upgrade it? Feb 17 15:10:33 toggles_w: libframeworkd-glib? not needed... thought so. but is not for git :-) Feb 17 15:10:50 Zorkman: try ;) and don't spank me if it does not :P Feb 17 15:11:11 mickey|lunch: could you ping me after lunch please ? have some questions for you Feb 17 15:11:18 :D Feb 17 15:12:18 mrmoku: latest news for Vala's release says 1 week if no regression, 2 weeks else Feb 17 15:13:02 ptitjes: so... ologicd will hit SHR in about a month then ;) Feb 17 15:13:46 mrmoku: you mean ophonekitd, don't you ? Feb 17 15:14:24 ptitjes: all of them :-) Feb 17 15:14:38 mrmoku: anyway for ologicd, I just wait someone has the courage to do bb recipes for libiodbc, XSB and flora Feb 17 15:14:42 btw. if you want it earlier... we can always switch to vala svn :-) Feb 17 15:15:11 (Ainulindale told he would do them, but I don't think this will be these days, so...) Feb 17 15:15:41 mrmoku: In fact, my patches didn't yet go to svn... Feb 17 15:15:45 too bad Feb 17 15:15:47 ptitjes: bb's are ready :-) Feb 17 15:16:12 mrmoku: what bb's ? Feb 17 15:16:18 bitbake recipes Feb 17 15:17:02 new ophonekitd needs libmodulo, libfso-glib, vala-dbus-binding-tool Feb 17 15:17:31 and ologicd needs bb recipes for libiodbc, XSB and Flora Feb 17 15:17:54 but the ones for XSB and Flora may be really hardcore Feb 17 15:17:59 to do Feb 17 15:18:40 ahh, ok. So we might just wait for the vala release to happen :-) Feb 17 15:18:50 and get your stuff in svn :P Feb 17 15:19:25 yeah too bad to be on the bleeding edge :) Feb 17 15:24:23 ptitjes: pong Feb 17 15:24:32 hi mickey Feb 17 15:24:39 (vala) well, i have started to add patches into OE anyways due to them not being applied nor even commented on Feb 17 15:24:47 so we might as well add more Feb 17 15:24:55 oh nice! Feb 17 15:25:12 http://bugzilla.gnome.org/show_bug.cgi?id=571831 Feb 17 15:25:19 http://bugzilla.gnome.org/show_bug.cgi?id=572122 Feb 17 15:25:23 here are two I need Feb 17 15:25:25 :) Feb 17 15:26:02 mickey|osx: I have problems with the xml specs Feb 17 15:26:05 at least two SHR folks ahve commite access Feb 17 15:26:17 who are they ? Feb 17 15:26:36 does mrmoku has ? Feb 17 15:26:38 Ainulindale and quickdev iirc Feb 17 15:26:41 I am using openmoko daily unstable build 20090217, tangogps 0.93 -> config -> Save repository configuration does not really save the change value Feb 17 15:26:41 arf Feb 17 15:26:59 both are busy these days :) Feb 17 15:27:03 anyway Feb 17 15:27:08 since it's overdue that SHR appears upstream, more folks should apply for access Feb 17 15:27:21 ok, back to the specs. what's the problem? Feb 17 15:27:43 mickey|osx: My problem comes from Preferences and Phone Feb 17 15:28:30 Preferences is o.fs.Preferences and PreferencesService is o.fs.Preferences.Service Feb 17 15:28:37 the same occurs for Phone Feb 17 15:28:53 yeah Feb 17 15:28:59 Phone is o.fs.Phone and PhoneCall is o.fs.Phone.Call Feb 17 15:29:01 conincidentally two of the interfaces I didn't add :) Feb 17 15:29:10 héhéhé :) Feb 17 15:29:24 my problem is that when I bind to Vala Feb 17 15:29:42 I must chose whether Phone is a namespace or an interface Feb 17 15:29:58 so I bound Phone to Phone.Phone and Call to Phone.Call Feb 17 15:30:00 ptitjes, the answer: nested classes Feb 17 15:30:08 but I find it really ugly Feb 17 15:30:32 yeah but I can't do nested interfaces Feb 17 15:30:38 ... Feb 17 15:31:30 would you prefer I bind Phone to Phone and Phone.Call to PhoneCall ? Feb 17 15:31:56 then these two would be in the FreeSmartphone namespace Feb 17 15:32:18 sounds ok to me Feb 17 15:32:38 ok let's go to this Feb 17 15:32:39 then Feb 17 15:34:11 mickey|osx: BTW don't know if you figured this out but with gobject-introspection, it will be possible to invoke any glib-bindable language from any glib-bindable language Feb 17 15:35:18 but afaik there aren't that much tools out there which use the .gir files Feb 17 15:35:28 mickey|osx: which will means for us that we at some time will need to all use one typelib for FreeSmartphone Feb 17 15:35:49 playya_: python, vala, C are quite ok Feb 17 15:36:19 yeah, i've been watching this with interest, although I fail to be _that_ fascinated as many other language guys Feb 17 15:36:29 i just tried to use the python generator a few months ago and it didn't work :( Feb 17 15:36:32 since I already found my language of choice ;) Feb 17 15:36:40 or, language_s_ for that matter Feb 17 15:36:56 mickey|osx: and if it is possible mrmoku and I may apply for commiter access to oe Feb 17 15:37:24 ptitjes: feel free to apply for access. we need two coreteam votes to grant access though Feb 17 15:37:27 mickey|osx: yeah but that will mean you don't have to produce bindings specific to python Feb 17 15:37:58 ptitjes: true, although -- among the two of us -- i think we will move away from python sooner or later Feb 17 15:38:13 :D Feb 17 15:38:42 so should I write to the core team list ? Feb 17 15:38:54 ssh, don't tell anyone Feb 17 15:38:56 (for the commit access, I mean) Feb 17 15:38:58 Om will kill me Feb 17 15:39:01 ah, that Feb 17 15:39:04 no, write to oe-devel Feb 17 15:39:04 héhé Feb 17 15:39:08 ok Feb 17 15:39:46 so I'll first have to subscribe!! :p (oh no yet another list...) Feb 17 15:40:06 heh, yes, it's a requirement to be subscribed to the list for every contributor anyways :) Feb 17 15:42:57 mickey|osx: sooner would be better than later Feb 17 15:43:16 heh Feb 17 15:43:25 later is more likely though Feb 17 15:43:31 indeed Feb 17 15:43:44 but it's going to happen Feb 17 15:43:57 if vala develops in the right direction ... Feb 17 15:44:03 which we all hope Feb 17 15:47:17 Ainulindale: hey! you should just look at what Bobby sent me : http://www.dyers.org/blog/beards/beard-%20types/ Feb 17 15:47:36 :D Feb 17 15:48:29 ptitjes: haha. I randomly met this man last year. now that's impressive. http://www.beardteamusa.org/btusa/jackpassion.html Feb 17 15:59:27 updated the fso architecture boilerplate Feb 17 15:59:30 comments? Feb 17 15:59:43 http://trac.freesmartphone.org/raw-attachment/wiki/WikiStart/fso-architecture.png Feb 17 16:01:13 mickey|osx: small typo... comman :-) Feb 17 16:01:32 and maybe update the linux version... Feb 17 16:01:42 (connman) aargh Feb 17 16:01:46 linux >= 2.6.24 on purpose Feb 17 16:01:59 ahh, did not see > :-) Feb 17 16:02:00 this was when the power class and other important subsystems hit linux Feb 17 16:02:30 mickey|osx: btw... GetPowerStatus is supposed to work on gta01? Feb 17 16:02:46 yes, if you have the correct kernel module builtin Feb 17 16:02:54 there's a gta01-specific power class driver Feb 17 16:03:21 so we might have a borken gta01 kernel in SHR... Does it work in the FSO image? Feb 17 16:03:33 it doesn't Feb 17 16:03:51 * mrmoku feels less guilty then :-) Feb 17 16:03:54 i didn't have time to investigate throughly Feb 17 16:04:02 ptitjes: heh the federation standard Feb 17 16:04:08 the patch should be updated for 2.6.28 anyways, i'm not sure whether it's in andy-tracking or stable Feb 17 16:04:45 mrmoku: thanks for kickstarting the shr-lite diet Feb 17 16:04:51 (on the list) Feb 17 16:04:52 :-) Feb 17 16:05:10 * mrmoku checks if finally some gta01 defconfig arrived in andy-tracking... Feb 17 16:05:14 also sent a reworked timeout setting dialog to the list Feb 17 16:05:25 adapting timeouts becomes much less painful with it Feb 17 16:06:10 timeouts? Feb 17 16:06:12 yup there is a gta01_moredrivers_defconfig... no packaging one though Feb 17 16:06:20 I need to try the GPSnofix stuff, just like Helge Hafting I haven't got a GPS fix sind end January Feb 17 16:06:31 mickey|osx: yes. The IdleNotifier timeouts in shr-settings Feb 17 16:06:37 ah, that Feb 17 16:07:01 spaetz: IMHO it should use elementary sliders... though I think the did not yet arrive in the python bindings :( Feb 17 16:07:29 mickey|osx: fso-abyss? Feb 17 16:07:30 Sliders would be nice. But -10 -1 +1 +10 buttons work well and fit on the screen. Feb 17 16:07:34 mrmoku: Balaji said he's going to test full functionality of .29 on GTA01 very soon. Feb 17 16:07:45 PaulFertser_: good news :-) Feb 17 16:07:50 it has more goodies like more responsive changing Feb 17 16:07:53 methril|away: fso-abyss is the successor to gsm0710muxd, which we will phase out Feb 17 16:08:42 well, this is important to me :) is where i need to look when other gsm modem doesn't work Feb 17 16:08:43 I'm getting "2009.02.17 12:03:23 ogsmd.channel DEBUG : sending 12 bytes: 'AT%Naec+nr\r\n'" on SHR-unstable... Feb 17 16:08:48 spaetz: yep... adjusting timeouts is (was) quite annoying... so did not want to offend your patch :-) Feb 17 16:09:06 methril|away: fso-abyss is something that will be actively maintained and has a sane upstream Feb 17 16:09:08 ...this should be AT%0187 Feb 17 16:09:15 methril|away: so yes, please lets co-work on that Feb 17 16:09:49 * methril|away nods with last mickey|osx sentence Feb 17 16:10:27 hmm, that reminds me Feb 17 16:10:36 * mickey|osx wants to write a huge blog post today Feb 17 16:10:51 cathing up with what i did in the last months and the roadmap for 2009 Feb 17 16:10:56 i should better start... Feb 17 16:11:12 * mickey|osx brews a coffee first Feb 17 16:11:17 please don't writ it in vala ;) otherwise I won't understand it.... Feb 17 16:11:21 +e Feb 17 16:11:39 * methril|away bbl to see the wonderfull mickey|osx blog post :) Feb 17 16:11:51 heh Feb 17 16:11:57 mrmoku: ;) Feb 17 16:15:56 mmh, midori desktop icon is still gone. Feb 17 16:18:05 interesting there is a file /usr/share/icons/shr/86x86/apps/midori.png but the midori.desktop uses Icon=web-browser Feb 17 16:18:30 * spaetz wonders what/who has changed things. Feb 17 16:20:35 yes, renaming Icon=web-browser to Icon=midori fixes things Feb 17 16:20:48 mmh, where does that .desktop file come from? Feb 17 16:21:03 spaetz: Koen Kooi :-) Feb 17 16:21:03 midori: update to 0.1.2, adjust to waf buildsystem, fix 'missing icon' problem Feb 17 16:23:02 bbiab Feb 17 16:24:52 mrmoku|away: ok, will try to bump version to 0.1.2 Feb 17 16:38:41 so if openembedded provides 0.0.17 and shr-overlay has 0.1.0 (and 0.1.2) why does it build 0.0.17? Feb 17 16:38:54 I thought bitbake chooses the highest version by default? Feb 17 16:40:32 it's supposed to Feb 17 16:40:37 check the files, perhaps there is DEFAULT_PREFERENCE somewhere Feb 17 16:40:53 or otherwise PREFERRED_VERSION settings Feb 17 16:41:10 actually, openembedded provides midori_0.1.2.bb too, so the shr-overlay one can be removed anyway. Feb 17 16:41:13 mickey|osx: grepping Feb 17 16:43:21 bahh: openembedded/conf/distro/preferred-om-2008-versions.inc:PREFERRED_VERSION_midor ?= "0.0.17" Feb 17 16:43:45 ok, need to override from SHR overlay somehow Feb 17 16:44:18 i'm happy to report that SHR on my openmoko has now been running, receiving and sending phone calls, for about a day. Feb 17 16:44:30 thats a nice improvement, from 20 seconds at a time. ;) Feb 17 16:45:09 hey, i cant find anything according to the kernel in the shr wiki, i want to compile the kernel with spaushfs support. or can i compile it like a regular kernel, and just add ".bin" at the end of the image name? Feb 17 16:45:37 juri, hope it lasts :p Feb 17 16:45:41 also, how is the quality? Feb 17 16:48:42 so far, no-one has reported any echoing. Feb 17 16:48:48 and the interface is *snappy*. Feb 17 16:49:42 juri_: which images are you running? Feb 17 16:52:55 the newest SHR unstable image. Feb 17 16:53:04 i need to upgrade the gsm firmwares... Feb 17 16:53:44 juri_: and which kernel? I'm using the newest 2.6.28 from the shr images, but suspend and resume is extremely wonky. Feb 17 16:55:53 oh, i'm still using the old 2.6.24. Feb 17 16:56:19 juri_: got it. thanks. Feb 17 16:58:27 someboydy has obexpush and obexftp for shr? Feb 17 17:11:08 blist Feb 17 17:11:18 sorry everybody, wrong window Feb 17 17:17:43 juri_: happy to hear you're happy :-) Feb 17 17:35:28 mrmoku: just wait until i find the next bug. ;) Feb 17 17:36:38 Beware! Feb 17 17:41:47 well, ok, the 'your sim doesn't exist' thing is annoying.. but workarroundable. the ophonekitd dieing every 20 seconds was not. ;) Feb 17 17:44:06 juri_: what is that 'your sim doesn't exist' thing? Sometimes on boot SIM gets not discovered and then you reboot? Feb 17 17:44:24 nope. Feb 17 17:45:02 since i've been running ophonekitd manually, sometimes it detects my sim, and does its job. other times, it splashes up the 'your sim is not installed' message, and tells me to go away. Feb 17 17:45:41 its NOT dependant on reboots. meerly waiting on ophonekitd to die (when it used to die on me), and re-launching it would work about half the time. Feb 17 17:46:27 * mrmoku is grepping for 'your sim is not installed' Feb 17 17:46:30 i have a second openmoko i can abuse, so if you like, we can start fighting this one. however, i think it would be prudent for me to update my GSM firmware first. Feb 17 17:46:36 er. thats not a quote. ;) Feb 17 17:46:48 i think it used 'sim is not present'.... Feb 17 17:46:54 hehe, ok Feb 17 17:47:13 anyways, my logs attached to #318 show one run with it working, another run with it not working. Feb 17 17:47:32 g_message("SIM card not present.") Feb 17 17:48:11 and it pops up an error dialog? Feb 17 17:48:15 oh, is it wrong that ophonekitd does not break when i hit control-c at a prompt, or is this just not real bash with job control. Feb 17 17:48:18 yes. Feb 17 17:49:04 juri_: I don't know why... but I have that control-c not working too... Feb 17 17:50:02 examining #318, the first log i present is it 'broken', and the second and third are it 'working'. Feb 17 17:53:32 mickey|osx: under what conditions does frameworkd emit org.freesmartphone.GSM.SIM.NotPresent? Feb 17 17:58:37 mrmoku, is there any calendar app in the SHR repos? Feb 17 17:58:57 Sharwin_F: no... don't think so :( Feb 17 18:01:30 Sharwin_F: do you know a good one? Feb 17 18:03:08 mrmoku, nop. There's one on the OM distro, would be difficult to port it? Feb 17 18:03:46 Sharwin_F: probably yes... don't know for sure though Feb 17 18:03:49 mrmoku, don't have to be a great app, just some nice app to look through the calendar for specific days to know the number, day of the week, etc. Feb 17 18:04:33 mrmoku, I'll look for it Feb 17 18:04:55 * Sharwin_F most wished app is a chess game with a GUI Feb 17 18:05:00 xD Feb 17 18:08:45 * mrmoku --> bitbake fltk-chess Feb 17 18:08:50 mrmoku, this one looks great, took package from angstrom, but there are lots of dependencies not covered on SHR Feb 17 18:08:52 http://gpe.handhelds.org/projects/GPE-calendar.shtml Feb 17 18:22:06 mrmoku: this should come for almost all operations when you don't have a SIM inserted Feb 17 18:22:26 mrmoku: it's a literal translation of the respective AT command error that occurs when you want to do anything that requires the SIM Feb 17 18:22:38 for sure it will come for setting the antenna power or sending the auth code Feb 17 18:22:59 blog posted. Feb 17 18:26:08 http://www.heise.de/newsticker/MWC-Hersteller-versprechen-einheitliche-Handy-Netzteile--/meldung/132700 Feb 17 18:26:22 Found "got 17 bytes from: '%CSTAT: EONS, 1\r\n'" in the log just after the phonebook retrieval Feb 17 18:27:25 mickey|osx: juri_ is experiencing it on gta01 when powering up the antenna... though retrying a bit later it works (without rebooting or fiddling with the SIM)... Feb 17 18:27:33 so, its just the gsm firmware in my phone being flaky. gotcha. Feb 17 18:28:09 Line 1726 of http://shr.bearstech.com/trac/attachment/ticket/318/frameworkd.log is mad strange, that EONS is funny and enexpected. Feb 17 18:28:14 bloody non-free.. Feb 17 18:29:00 PaulFertser: so you negotiated with broonie who's OM's next alsa-officer? Feb 17 18:31:23 what should I put for shr in DISTRO in local.conf of bitbake? DISTRO = "shr" ? Feb 17 18:31:28 docscrutinizer2: hm, i still don't know on what conditions he received his freerunner. It looks like he's going to fix all the known issues soon. My last talk with him was on public channel that you saw... Feb 17 18:31:39 Sharwin_F: openmoko Feb 17 18:31:44 mrmoku, ok Feb 17 18:32:46 Sharwin_F: ok... bitbaking package-index... after that you have fltk-chess and gpe-calendar to try :-) Feb 17 18:33:31 mrmoku, on feeds or do I have to use bitbake? (never used it, just looking at it right now) Feb 17 18:33:41 docscrutinizer2: is _upgrading_ gsm from pre-moko6 safe? Feb 17 18:34:13 yup _should_ be perfectly safe Feb 17 18:34:29 Cool Feb 17 18:34:57 just downgrading to or beyond moko5 (iirc) frontier is strongly deprecated Feb 17 18:35:15 Sharwin_F: on feed Feb 17 18:35:21 now Feb 17 18:35:32 bbl Feb 17 18:36:06 PaulFertser: you have a minute for a private chat later on? Feb 17 18:36:14 docscrutinizer2: sure Feb 17 18:36:21 k :) Feb 17 18:38:26 Patch long-long-thumb.patch does not apply ----- any fix? Feb 17 18:39:28 does not apply on what? Feb 17 18:40:10 mpfr-native Feb 17 18:40:32 * mrmoku does not even know what that is ;) Feb 17 18:40:55 lol Feb 17 18:41:16 A lib required to build host gcc, i guess. Feb 17 18:45:12 Sharwin_F: ok... forget about fltk-chess :P Feb 17 18:49:09 mrmoku: would it be possible to add obexpush and obexftp for shr? Feb 17 18:50:25 Zorkman: is in OE... trying to build Feb 17 18:51:39 man, when you are ever in BE, be sure to notify me that i can give you some beer :) Feb 17 18:52:22 :-) Feb 17 18:52:52 mrmoku, what's wrong with fltk-chess? Feb 17 18:53:18 Sharwin_F: it takes just one quarter of the screen... mini-chess so too say :-) Feb 17 18:53:24 -o Feb 17 18:53:41 mrmoku, so it lacks a patch? Feb 17 18:53:49 probably Feb 17 18:54:08 gpe-calendar is nice though... just the desktop file seems to be borken. no icon Feb 17 18:54:13 but you can start it from the console Feb 17 18:54:44 mrmoku, I have experiencie editing desktop files on /usr/share/applications hehe Feb 17 18:54:50 :-) Feb 17 18:57:05 mrmoku, well, fltk-chess is better than gnuchess at least :P Feb 17 18:58:44 mrmoku: we all know that messing with dpi setting of X is not nice, but is can it be possible that fltk-chess become usable after tweaking it? Or probably switching to QVGA (that is reported to work with an additional tweak). Feb 17 18:59:46 mrmoku: btw, i think you can patch Xglamo for the time being to fix that annoying issue with QVGA (probably just change the sysfs node name). Feb 17 19:00:18 Sharwin_F: depends... Ainulindale might prefer gnuchess ;) Feb 17 19:00:41 phalanx is a good engine Feb 17 19:01:19 mickey|osx: some nice gui? Feb 17 19:01:49 no, just an engine Feb 17 19:02:25 PaulFertser: yeah... might be tweakable. And it starts fast :-) F in FLTK has a reason... Feb 17 19:03:13 hmmm obexftp brings in libgsm... problem? Feb 17 19:07:47 Zorkman: ok... obex stuff is in the feed Feb 17 19:08:43 mrmoku, I want virtualbox on the feeds! xD Feb 17 19:09:02 a fix for qvga would be very cool Feb 17 19:11:44 Sharwin_F: hehe, and I want a cray emulator :P Feb 17 19:11:59 PaulFertser: you say it is just a wrong sysfs node in Xglamo? Feb 17 19:12:05 any patch going around? Feb 17 19:12:38 mrmoku: no patch, but folks had success echoing the right value to the node manually after issueing xrandr -s command. Feb 17 19:13:54 mrmoku: https://docs.openmoko.org/trac/ticket/2162#comment:6 Feb 17 19:14:51 mrmoku: grep the Xglamo source for "qvga-normal", probably it is as easy as that. Feb 17 19:15:39 But according to the ticket one will want to disable framebuffer blanking while using QVGA Feb 17 19:23:23 hello Feb 17 19:23:54 dos1 :o Feb 17 19:25:06 * mrmoku` notes to order the 'do-not-disconnect' option for his dsl connection :( Feb 17 19:25:34 spaetz: thanks for patch :) Feb 17 19:26:20 mrmoku: how much is it? Feb 17 19:26:28 1,50 Feb 17 19:26:46 mrmoku: they work like a charm! Feb 17 19:26:52 * ptitjes1 notes to order for a computer that does not get so hot it reboots Feb 17 19:27:04 * Zorkman recommends thermal paste Feb 17 19:27:04 Zorkman: what do you use them for? Feb 17 19:27:11 I'd pay that if my isp allowed it :-/ Feb 17 19:27:15 mrmoku: sending and receiving files over bluetooth Feb 17 19:27:54 nice :-) Feb 17 19:28:07 it's always nice to see something work on the fr :) Feb 17 19:28:57 mrmoku: (1.50) me too, me too... ;) Feb 17 19:32:03 mrmoku: i'm getting some errors with installing fltk-chess, are you aware of this?: http://pastebin.com/mf37eaae Feb 17 19:33:27 :o Feb 17 19:37:46 Zorkman: nah... you installed some fltk package from somewhere else, right? Feb 17 19:39:36 PaulFertser: its two months now that I want to switch contract... drop PRI and have VOIP only Feb 17 19:39:55 mrmoku`: not that i know of, not deliberatly at least Feb 17 19:39:57 enhance bandwith and get that damn option Feb 17 19:40:39 Zorkman: try opkg remove libfltk and see what it complains about dependencies Feb 17 19:41:08 mrmoku`: (voip only) strongly deprecated. roundabout 5% downtime Feb 17 19:41:37 docscrutinizer2: ouch Feb 17 19:41:50 mrmoku` yeah i was doing that, seems that gwaterpas has already installed it Feb 17 19:42:06 what is gwaterpas? Feb 17 19:42:27 a waterpass program; to see if that painting really is hanging straight Feb 17 19:42:39 http://www.opkg.org/package_102.html Feb 17 19:42:55 http://wiki.openmoko.org/wiki/Gwaterpas Feb 17 19:43:06 ahh.. hmm not in OE :( Feb 17 19:44:02 the alarm in gpe-calendar seems to work, but there is no sound notification; overall it seems a very nice calendar app! Feb 17 19:44:27 someone should port it to opimd then... someday :-) Feb 17 19:46:32 docscrutinizer2: personal experience? Feb 17 19:46:35 (voip) Feb 17 19:47:19 yup, one of my "customers" as well as my auxiliary voip account Feb 17 19:48:20 docscrutinizer2: might depend on the provider? Feb 17 19:48:50 mrmoku: yup. but also on mains poer ;) Feb 17 19:48:59 power Feb 17 19:49:23 mains power? Feb 17 19:50:02 AC220V Feb 17 19:50:22 and then... its for private use (though my wife will kill me if it wont work ;) Feb 17 19:50:40 no voip without power. isdn and 600ohm still np Feb 17 19:51:25 MAXIMUM OHMAGE! Feb 17 19:52:33 that should be no problem... power outages are quite seldom here (ok, there was one for half an hour the other day... but thats really rare) and if I'm without power the last thing I'm thinking about is doing phone calls Feb 17 19:52:47 I sit in an angle and cry... because bitbaking is no more ;) Feb 17 19:53:24 and then there is my nice mobile :D Feb 17 19:53:31 (which does not work either) Feb 17 19:54:32 mrmoku: when my mains fails, *first* thing I do is calling to get to know when it's supposed to be back Feb 17 19:55:51 docscrutinizer2: I would not even know whom to call :o Feb 17 19:58:15 well, most voip downtimes due to dns-, registrar-, or gateway-server outage Feb 17 19:58:36 no carrier-grade MTBF :-/ Feb 17 19:59:20 * mrmoku is depressed now Feb 17 20:00:40 not as depressed as you'd feel after a switch to voip-only ;D Feb 17 20:02:03 ok, ok... will rethink my strategy Feb 17 20:10:31 :D Feb 17 20:10:55 * mrmoku wonders if using a moko as gsm router would be an alternative ;) Feb 17 20:11:17 gah Feb 17 20:11:20 ultimate failure Feb 17 20:11:24 the universe would be crushed Feb 17 20:11:44 just because my wife killed me? Feb 17 20:13:52 like moko+gsm=reiserfs^-1? *g* Feb 17 20:14:12 something like that :D Feb 17 20:15:34 :D Feb 17 20:15:46 haha, that's exactly what I was thinking :p Feb 17 20:17:41 mrmoku: have figured what's up with xglamo? Feb 17 20:18:18 PaulFertser: did not look yet... Feb 17 20:23:13 dos1: you are welcome Feb 17 20:23:25 I was just annoyed by slow timeout settings :) Feb 17 20:24:37 dos1: is it in? shall I rebuild? Feb 17 20:25:15 mrmoku: no, it will be in few minutes Feb 17 20:38:10 dos1: and BTW, that window is too wide (power settings), so the +10 button is somewhat off the screen. Feb 17 20:38:24 I believe it is beacuse the battery status is a very long string Feb 17 20:38:43 and that forces the window to be rather wide. Feb 17 20:39:35 spaetz: yes, i know Feb 17 20:39:45 it is when battery is nearly 100% Feb 17 20:42:52 ok, I'll send some patch for it when I have some time left. Feb 17 20:44:16 freesmartphone.org: 03ptitjes 07vala-dbus-binding-tool * r5b8d0b57bee1 10/src/vala-dbus-binding-tool.vala: Feb 17 20:44:16 freesmartphone.org: Enable concatenation of inner interfaces names Feb 17 20:44:16 freesmartphone.org: Signed-off-by: Didier 'Ptitjes Feb 17 20:46:26 spaetz, mrmoku: commited Feb 17 20:46:43 hej anyone speaking swedish here? Feb 17 20:47:10 freesmartphone.org: 03ptitjes 07libfso-glib * rfb6a3f97ae2b 10/ (9 files in 2 dirs): Feb 17 20:47:10 freesmartphone.org: Regenerated according to inner interfaces renaming option of vala-dbus-binding-tool. Feb 17 20:47:10 freesmartphone.org: Signed-off-by: Didier 'Ptitjes Feb 17 20:51:11 dos1: cool, thanks Feb 17 20:53:30 re Feb 17 20:55:31 dos1, spaetz: built Feb 17 20:57:59 yay Feb 17 21:23:48 i think #174, #215, #223, #232, #244, #248, #258, and looks like #265 too, can be closed ;] Feb 17 21:23:58 please specify that it is SHR :) Feb 17 21:24:02 SHR Feb 17 21:24:03 :D Feb 17 21:27:30 dos1: did you try #215 Feb 17 21:28:47 mrmoku: some time ago (~ week?) i ran without sim and it showed message to me Feb 17 21:28:56 mrdos Feb 17 21:28:59 mrmoku: message in window Feb 17 21:29:04 mrmoku: but now i'm trying it again Feb 17 21:29:40 ok Feb 17 21:32:41 mrmoku: yes, it's not failing Feb 17 21:36:08 dos1: and with #258 you mean #259? Feb 17 21:36:51 mrmoku: no, counter is in shr-settings Feb 17 21:37:05 sms inbox counter? Feb 17 21:37:13 mrmoku: yep Feb 17 21:37:20 mrmoku: in SIM module, GSM category Feb 17 21:38:08 hmm, ok... will come back when we have folders for messages :-) Feb 17 21:39:18 dos1: ok... closed them. Feb 17 21:39:33 mrmoku: thanks, and next: Feb 17 21:39:34 #42, #75, #148, at it looks like #142 Feb 17 21:41:17 hmm.. don't agree on #42... I want to see/change the profile without opening shr-settings and go into profiles Feb 17 21:41:50 mrmoku: but where it should be placed? Feb 17 21:43:07 dos1: I know space is limited in the top shelf... though I would keep the ticket open as reminder... Feb 17 21:43:24 mrmoku: hmm, ok. but i suggest changing name of it Feb 17 21:43:25 maybe it finds its place in a nice home screen some day :_) Feb 17 21:43:55 mrmoku: i was thinking about home screen too :) Feb 17 21:48:08 3 days ago there was 150 tickets Feb 17 21:48:12 now there is 124 Feb 17 21:48:13 ;) Feb 17 21:48:28 progress :D Feb 17 21:50:02 without package requests there are 110 Feb 17 21:51:04 mhh... package requests are for me :-) Feb 17 21:51:20 #139 can be closed, as is about old shr-settings app Feb 17 21:51:24 dos1: how do you get them? Feb 17 21:51:43 mrmoku: what? :D Feb 17 21:52:05 mrmoku: and maybe close #282 as wontfix Feb 17 21:52:19 package of package-request tickets :P Feb 17 21:53:17 dos1: no way to figure the status of services? Feb 17 21:53:50 mrmoku: i don't know any Feb 17 21:53:56 mrmoku: http://shr-project.org/trac/query?status=accepted&status=assigned&status=new&status=reopened&component=package+requests&order=priority&col=id&col=summary&col=status&col=type&col=priority&col=milestone&col=component :) Feb 17 21:54:02 mmh is in GSM operators listing in SHR settings working? Feb 17 21:54:35 http://feeds.digg.com/~r/digg/container/technology/popular/~3/kbAbhzOdgCM/ARM_shows_possibly_iPhone_bound_multicore_mobile_processor Feb 17 21:54:41 is there a super moko planned? ;] Feb 17 21:55:32 68 tickets with type=defect Feb 17 21:55:38 dos1: ahh, custom query... thanks :-) Feb 17 21:55:47 enhancement: 47 Feb 17 21:55:57 task: 8 Feb 17 21:56:28 TAsn: we can't even handle one core ;) Feb 17 21:56:41 mrmoku, ;] Feb 17 21:56:45 more cores -> more failures :P Feb 17 21:56:49 nah. Feb 17 21:56:57 ok... fails faster though Feb 17 21:56:59 our trac should be more cleaned Feb 17 21:57:11 more cores -> more speed -> we can write in python -> less failures ;] Feb 17 21:57:27 just think about the race conditions Feb 17 21:57:50 Ok operators listing fails , seems frameworkd prob :( Feb 17 21:58:08 von_fritz: yes, i filled ticket in FSO trac for it Feb 17 21:58:20 OK ;) Feb 17 21:58:25 * mrmoku does not like python :P Feb 17 21:58:26 von_fritz: there was three bugs in FSO, which was blocking shr-settings functionality Feb 17 21:58:35 von_fritz: now two of them are fixed Feb 17 21:58:49 von_fritz: last one is about operators listing ;] Feb 17 21:58:58 * Dave hisses at mrmoku Feb 17 21:59:03 yay! linuxpmi.org is back online! Feb 17 22:00:13 ok, now i have to go :] Feb 17 22:00:15 juri_: so... client paid his bills :P Feb 17 22:00:18 dos1: cu Feb 17 22:00:33 good night ;) Feb 17 22:01:07 dos1: pussy Feb 17 22:02:00 now i can stop messing with side projects, and get back to work. ;) Feb 17 22:03:32 mrmoku, ! you'll like python eventually... Feb 17 22:04:06 eolson: why should I? Feb 17 22:05:27 juri_: come back if your phone fails :P Feb 17 22:09:57 PaulFertser: will look at xglamo tomorrow... (if you don't beat me - *hint*) Feb 17 22:09:58 mrmoku, less typing, less brackets, fewer nitpicky mistakes to make, big std library, easier to read ==> lots more time to get things done. After you get used to the few things (indents, no compile time errors, etc.) and use it a little, it's good. Feb 17 22:10:29 eolson: used it... I have very fast fingers... sometimes faster than my brain :P Feb 17 22:10:39 so typing is no issue... Feb 17 22:11:05 but I have nothing against it... just don't like to use it myself Feb 17 22:11:30 (using it for small helper scripts.. yeah have to admit that :-) Feb 17 22:11:41 so, c++ instead? Feb 17 22:11:46 yep :D Feb 17 22:12:51 I used to be c++ everything, too, but had to learn python for work. Feb 17 22:16:34 I suppose the type-checking and c++ performance are useful for some things, but personally I can develop projects faster in python. Feb 17 22:21:49 *hiss* Feb 17 22:22:31 Anyone remembers what problems are there with bluetooth audio on GTA01? Feb 17 22:23:41 mrmoku, but mostly, Dave will make you like python eventually Feb 17 22:24:16 he might try (without success)... but not today... tired and heading to bed now :-) Feb 17 22:24:45 gnight all Feb 17 22:34:51 use perl; Feb 17 22:34:55 mrmoku|perl Feb 17 22:38:53 mrmoku, later, no pressure :) Feb 17 22:43:40 mrmoku: don't worry, i will. ;) Feb 17 22:45:53 lindi-: If you're interested to understand audio routing and DAI modes a bit, take a look at http://wiki.openmoko.org/wiki/Neo_1973_audio_subsystem#ALSA_Channels, i've updated it with new diagrams. And btw Doc suggested that for VoIP with BT one doesn't neet do use BT DAI at all, and use USB interface to BT instead. Feb 17 22:46:30 PaulFertser: aha Feb 17 22:47:05 I was just too fascinated to see a way to abuse WM8753 in a weird way :) Feb 17 22:47:08 PaulFertser: unfortunately i do not have supported bt heaset Feb 17 22:47:32 I thought they all follow some standard. Feb 17 22:47:43 I don't have any bt headset, but i can borrow one for testing. Feb 17 22:48:53 uh oh Feb 17 23:03:15 Also, Joerg, I wish the FR bat was NiCd Feb 17 23:03:24 PaulFertser: i thought so too :) Feb 17 23:03:31 huh Feb 17 23:03:33 ? Feb 17 23:04:16 lindi-: So, your BT earpiece is compatible with all the dumb cellphones and yet is incompatible with bluez? Feb 17 23:04:32 Paul, that seems to be common. Feb 17 23:04:41 PaulFertser: so it seems Feb 17 23:04:52 We knew this six months ago :D Feb 17 23:05:00 lindi-: tried to debug that with bluez folks? Feb 17 23:05:05 PaulFertser: since bt traffic is encrypted it is hard to figure out how it differs Feb 17 23:05:23 PaulFertser: only with a few friends so far Feb 17 23:06:01 lindi-: encrypted yes, but how can it be that dumb cell phones work with it? Feb 17 23:06:13 bluez-alsa driver is kinda flakey Feb 17 23:06:16 PaulFertser: i have no idea Feb 17 23:06:23 i can't see their traffic Feb 17 23:06:28 freesmartphone.org: 03mickey 07fso-abyss * r8367d4eb6721 10/ (README src/server.vala): update README Feb 17 23:06:31 and compare against hcidump Feb 17 23:07:21 heard about btsco yielding better results Feb 17 23:07:58 lindi-: so, it doesn't connect at all? I thought that some initialization is a bit different but your attempt to connect with this earpiece can be compared to a successful attempt to connect to another (working) earpiece probably. Feb 17 23:08:36 PaulFertser: it works completely except that when it is supposed to send audio it only sends quite short packets always with the same contents Feb 17 23:08:44 PaulFertser: it beeps, the buttons work etc. Feb 17 23:09:45 lindi-: Funny. So, all your traces look the same as succesful traces with another device, but no audio? Feb 17 23:10:08 PaulFertser: i don't have other devices to compare with really Feb 17 23:10:14 PaulFertser: but yes, i should send bug reports Feb 17 23:10:15 lindi-: others have Feb 17 23:10:58 lindi-: i think i can borrow the simplest earpiece tomorrow and try it myself, can provide you logs of any detail level i hope. Feb 17 23:11:55 PaulFertser: i'm probably too busy with the usb stuff. i was supposed to look for usb security vulnerabilities for a university course and thus i have a deadline on that work Feb 17 23:11:59 check http://tech.groups.yahoo.com/group/twinklephone for a few reports about bluez/btsco for voip Feb 17 23:12:25 lindi-: oh Feb 17 23:12:54 PaulFertser: goal is to be able to execute arbitrary code on a PC when I plug my phone to it :) Feb 17 23:13:55 lindi-: i doubt that's possible at all with usb. If FR had firewire... Feb 17 23:14:21 amazing, virus coding is a university discipline now? ;D Feb 17 23:14:55 PaulFertser: maybe webcam drivers are of very low quality Feb 17 23:14:58 s/maybe/many/( Feb 17 23:15:13 PaulFertser: we are not targeting usb itself but the linux usb drivers for various pieces of hardware Feb 17 23:15:35 lindi-: Only UVC is included mainline, so how does it matter? Feb 17 23:15:45 PaulFertser: only UVC? there is stv680.c too Feb 17 23:15:54 PaulFertser: and lots of others Feb 17 23:16:04 lindi-: ah, i'm too far away in the past it seems :) Feb 17 23:16:05 PaulFertser: just look at the incredible code quality of stv680.c Feb 17 23:16:47 selected quote: *(output + AD (x, y, vw) + GREEN) = ((int) *(output + AD (x - 1, y, vw) + GREEN) + (int) *(output + AD (x + 1, y, vw) + GREEN) + (int) *\(output + AD (x, y - 1, vw) + GREEN) + (int) *(output + AD (x, y + 1, vw) + GREEN)) >> 2; Feb 17 23:17:36 Is it mainline? Cool :) Feb 17 23:17:58 PaulFertser: has been in mainline since 2001 or so Feb 17 23:18:08 judging from the comments Feb 17 23:18:47 PaulFertser: and the nice thing about modern desktop distros is that they automatically load the module when they see the right vendor/product id Feb 17 23:18:47 lindi-: now i understand why you persue usb issues on FR so actively :) Feb 17 23:18:59 Yep, i got the idea. Feb 17 23:24:07 freesmartphone.org: 03Frederik.Sdun 07fso-monitord * r290bafc5cd76 10/src/ (device.vala monitor.vala phone.vala preferences.vala): Feb 17 23:24:07 freesmartphone.org: fixed a typo in object paths Feb 17 23:24:07 freesmartphone.org: added some debug output Feb 17 23:25:19 DocScrutinizer: that's called "security research". Wish i could do a paper somewhere on some funny topic like this :( Feb 17 23:26:47 i'm bit worried at the moment. we are not yet behind schedule but having to tackle with gadgetfs bugs is not something we planned for Feb 17 23:27:22 lindi-: is the bug FR-specific? Feb 17 23:27:31 PaulFertser: probably not Feb 17 23:27:41 maybe i should send it to bugs.kernel.org too Feb 17 23:28:17 kind of hard to say since i don't have other usb device hardware that'd run linux Feb 17 23:28:50 lindi-: no iPAQs around? Feb 17 23:28:58 nope Feb 17 23:29:54 lindi-: i can try to reproduce your bug on friend's hx4700, i think. It uses 2.6.21-hh20 or smth like that, nothing more recent supports it. Feb 17 23:30:17 lindi-: get a N810! Feb 17 23:30:32 PaulFertser: i'd be great if you have time. https://docs.openmoko.org/trac/ticket/2232 should have exact steps to reproduce it Feb 17 23:30:36 docscrutinizer2: no thanks :) Feb 17 23:30:41 s/i/it/ Feb 17 23:30:42 lindi- meant: docscrutitnizer2: no thanks :) Feb 17 23:30:47 bzzbot: grr Feb 17 23:30:48 lindi-: sure, i saw that Feb 17 23:31:18 :D Feb 17 23:31:22 PaulFertser: if gadgetfs starts to work i'm also planning to write usb keyboard device emulation Feb 17 23:31:31 PaulFertser: so that freerunner can be used as a usb keyboard for a PC Feb 17 23:32:16 lindi-: will not you be too busy with creating a decent spl0it? ;) Feb 17 23:32:36 PaulFertser: depends :) Feb 17 23:40:09 lindi-: btw, i think pop noise on resuming can be fixed. WM8753 with minimising all that noises in mind. Feb 17 23:40:21 PaulFertser: yep i fixed pop noise on resume Feb 17 23:40:42 lindi-: By disabling the speaker completely? ;) Feb 17 23:40:44 PaulFertser: my suspend script does amixer -q -d sset "Amp Spk" mute Feb 17 23:40:55 PaulFertser: and then amixer -q -d sset "Amp Spk" unmute on resume Feb 17 23:41:23 lindi-: i guess that's not exactly right solution as you still can hear click in headset. Feb 17 23:42:05 no headset here :) Feb 17 23:42:21 Cheater :p Feb 17 23:44:30 a2dp is working! Feb 17 23:45:28 Zorkman: great, which headset is that? Feb 17 23:46:11 some generic cheap-o from e-bay, i'll give you a link Feb 17 23:46:27 Zorkman: can you save full hcidump somewhere online? Feb 17 23:47:26 lindi-: http://cgi.benl.ebay.be/Slim-A2DP-Stereo-Bluetooth-Headphone-Headset-Black-UK4_W0QQitemZ270345968236QQcmdZViewItemQQptZUK_Mobile_Home_Phones_Bluetooth_Acc_ET?hash=item270345968236&_trksid=p3286.c0.m14&_trkparms=66%3A2|65%3A15|39%3A1|240%3A1318 Feb 17 23:47:27 lindi-: (mute) cool. should be done by alsadriver, but hey as long as it works... Feb 17 23:47:45 it costed about 12 euro for me (shipping inculed) Feb 17 23:48:09 lindi-: do i have to start hcidump before i pair it? Feb 17 23:48:47 docscrutinizer2: now if we could figure out why changing alsa states wakes up the LCD backlight suspend/resume would be totally undetectable :) "alsactl restore wakes up X" -- http://docs.openmoko.org/trac/ticket//2209 Feb 17 23:49:04 Zorkman: would be cool Feb 17 23:49:11 what?? Feb 17 23:49:25 docscrutinizer2: yes, exactly that :) see the bug report for steps to reproduce Feb 17 23:49:35 Zorkman: in the ideal world we would have a global database of hardware and each one would include a traffic capture Feb 17 23:49:43 wtf is relation of statefiles and backlight??? Feb 17 23:49:46 ok, i'll get on it Feb 17 23:49:57 docscrutinizer2: i don't understand it either. just try the steps Feb 17 23:50:11 which ones? Feb 17 23:50:28 docscrutinizer2: steps 1-4 in http://docs.openmoko.org/trac/ticket//2209 Feb 17 23:50:34 k Feb 17 23:51:54 lindi-: -sh: hcidump: not found Feb 17 23:52:02 Zorkman: apt-get install ... Feb 17 23:52:18 Zorkman: sorry i don't know if om2008.12 has it Feb 17 23:52:30 in debian it is 'usr/sbin/hcidump admin/bluez-hcidump' Feb 17 23:52:35 i have shr unstable Feb 17 23:53:02 lindi@ginger:~$ zgrep hci cache/opkg.gz Feb 17 23:53:02 bluez-hcidump - 1.42-r0 - Linux Bluetooth Stack HCI Debugger Tool. Feb 17 23:53:06 Zorkman: that one? Feb 17 23:53:46 don't have that one Feb 17 23:53:52 maybe if i install bluez-utils-dbg Feb 17 23:53:52 #define GTA02_GPIO_AMP_SHUT S3C2440_GPJ1 /* v2 + v3 + v4 only */ ?? Feb 17 23:54:03 Zorkman: ah so it is only in om2008.12 Feb 17 23:55:07 PaulFertser: no idea Feb 17 23:55:18 bogus noise Feb 17 23:55:35 prolly means: not in v1 Feb 17 23:55:43 docscrutinizer2: yep Feb 17 23:56:11 numb coder Feb 17 23:57:13 mokopatches is his name :) Feb 17 23:57:36 :D Feb 17 23:57:54 get me a JM Feb 17 23:57:56 freesmartphone.org: 03mickey 07framework * r5355d64b43c0 10/framework/subsystems/odeviced/idlenotifier.py: Feb 17 23:57:56 freesmartphone.org: odeviced: fix bug in Display resource preventing the display to be lit when Feb 17 23:57:56 freesmartphone.org: you request the resource while it's already past the state it's trying to Feb 17 23:57:56 freesmartphone.org: prevent falling into. This should fix FSO bugs #336, #343, and #352. Feb 17 23:58:12 (werner's medicine aganst flu ;) Feb 17 23:58:20 bbl Feb 18 00:00:00 lindi-: i confirm that it's not that AMP that produces the click. Feb 18 00:03:11 PaulFertser: GSM_DL :( :p Feb 18 00:06:16 docscrutinizer2: s3c24xx-gpio J6=0; s3c24xx-gpio J6=1 produces sound somewhat louder and longer than wakeup pop sound. Feb 18 00:06:18 lindi-: (#2209) sorry, sw-domain. X is beyond my scope. Feb 18 00:06:31 lindi-: but 2209 is reproducible, funny :) Feb 18 00:09:06 PaulFertser: if you run (and load) 2 shellcommands, and compare to a kernel machinecode sequence - sure there'll be a diff Feb 18 00:09:09 docscrutinizer2: it also prints "w" to console if X is not running i think Feb 18 00:09:34 who? alsactl? Feb 18 00:10:30 who is printing to which console Feb 18 00:10:45 system console? Feb 18 00:11:20 docscrutinizer2: system console yes, to my login prompt Feb 18 00:11:24 docscrutinizer2: sure, i understand that. Will rip out that chip soon anyway, that'll be enough to check :) Feb 18 00:11:30 so seems there's some debug print left over in alsa drivers Feb 18 00:11:34 docscrutinizer2: try "sudo chvt 1" and run steps 1-4 Feb 18 00:11:45 docscrutinizer2: seems to print "P" this time Feb 18 00:13:02 "[P" to be exact Feb 18 00:13:09 though I still dont get it why X resumes on console output. Anyway its odd Feb 18 00:13:29 docscrutinizer2: surely X resume display when it detects keyboard input? Feb 18 00:13:42 yup Feb 18 00:14:03 but what can cause that input to be sent to console? Feb 18 00:14:05 but what is kbd device for FR X? Feb 18 00:14:07 it can't be a printk Feb 18 00:14:38 thats what I told before yea Feb 18 00:15:23 print TO console != input FROM console/kbd though! Feb 18 00:16:17 read(0, "\33"..., 1) = 1 Feb 18 00:16:17 read(0, "["..., 1) = 1 Feb 18 00:16:17 write(1, "["..., 1) = 1 Feb 18 00:16:17 read(0, "P"..., 1) = 1 Feb 18 00:16:17 write(1, "P"..., 1) = 1 Feb 18 00:16:26 is what getty sees Feb 18 00:16:57 looks like an ansi escape sequence Feb 18 00:17:13 yup Feb 18 00:17:43 check, prolly to screensave-control console. origin:X Feb 18 00:18:37 you mean X is sending these? Feb 18 00:19:01 yup, I wouldn't be surprised Feb 18 00:19:10 freesmartphone.org: 03mickey 07fso-abyss * r6028a98c0cd5 10/src/server.vala: dbus naming policy++ Feb 18 00:20:07 or whatever controls screenblank in the end Feb 18 00:20:10 docscrutinizer2: strace should reveal that, right? Feb 18 00:20:26 maybe sorta screensaver deamon Feb 18 00:20:47 i don't run any such daemon :) Feb 18 00:21:00 fsobla? Feb 18 00:21:14 i'm completely clueless Feb 18 00:21:23 i can stop X if it makes you believe it is not X issue :) Feb 18 00:21:52 a silly question: if i would like mokoko/pythm (mp3 players) to output the sound through bluetooth, what is the best way to make them do it? manually launching mplayer works with mplayer -ao alsa:device=bluetooth song.mp3 Feb 18 00:21:56 gitgrep for [P Feb 18 00:22:11 docscrutinizer2: i stopped X and getty still sees [P Feb 18 00:22:19 me away now Feb 18 00:25:51 man 4 console_codes : esc-[ =CSI, P="delete # of chars" Feb 18 00:26:15 * DocScrutinizer waves Feb 18 00:30:09 (me jackass, came back from downstairs ;P ) maybe ogus read frm touchscreen AD-converter? Feb 18 00:30:37 ogus? Feb 18 00:35:45 freesmartphone.org: 03mickey 07framework * rb67edfb3e2d8 10/framework/subsystems/ogsmd/modems/ti_calypso/modem.py: ogsmd: [TI CALYPSO] fso-abyss features autoconfiguration now, take this into account. Feb 18 00:41:05 doc means 'B'ogus Feb 18 00:52:34 :D Feb 18 00:55:14 freesmartphone.org: 03mickey 07fso-abyss * rf1e8c8d32075 10/README: todo++ Feb 18 00:55:29 night **** ENDING LOGGING AT Wed Feb 18 02:59:57 2009