**** BEGIN LOGGING AT Wed Aug 05 02:59:56 2009 Aug 05 03:01:33 * Sharwin_F trying to make a swapfile as freeciv needs more memory :S Aug 05 04:08:07 anyone know why opkg upgrade (in shr) has tonnes of "unsatisfied dependencies" for libasound-dev gio-2.0-dev etc.? Aug 05 04:08:12 seems VERY broken Aug 05 04:08:26 or some source just isn't in my repository Aug 05 04:08:49 it also complains about missing virtual-locale..* Aug 05 04:17:17 http://pastie.org/572030 Aug 05 07:40:00 morning Aug 05 07:42:20 Weiss: very quickly checked latest changes. took the new kernel drm changes in addition to libdrm. now I got colored noise for a longer time before a crash, is that "correct"? also I was able to make out from the noise it had properly launched zhone among else for me before crashing... Aug 05 07:42:39 just tracking if I still get same results etc. Aug 05 08:02:22 moinmoin Aug 05 08:38:18 hi. anyone here know how to specify window size in a gtk+ prog? Aug 05 08:38:38 there's always cutoff in the 'dates' program window Aug 05 08:40:11 Mirv: hmm, for me it doesn't crash for a good few minutes, and i see things like the illume bar forming Aug 05 08:43:04 mrmoku|away: hi Aug 05 08:43:18 I can't seem to find any good documentation on 'dates' Aug 05 08:43:28 mrmoku|away: q-master asks why is shr-config required by shr-config-dev is unavailable atm. Aug 05 08:43:36 I'm also trying to backup my calendar.ics file Aug 05 08:44:04 but using an old calendar file doesn't seem to work Aug 05 08:48:22 PaulFertser: o/ Aug 05 08:48:49 scratch that, I can reuse old ics files after rebooting Aug 05 08:49:00 but the window size is still to wide :( Aug 05 08:49:27 DocScrutinizer: Yo, nobody uses dumb battery support for FR :D Aug 05 08:50:17 Mirv: although, the coloured noise not getting overwritten is correct (there's something preventing accelerated operations from getting executed straight away) Aug 05 08:50:30 wait two years, until all gta02-bat are worn. Then everybody will *love* your work ;-) Aug 05 08:50:49 DocScrutinizer: haha, ok Aug 05 08:53:46 Weiss: I was busy doing some other stuff, so it may have been several minutes. I should've looked closer but just briefly tested it among other stuff. Aug 05 09:00:23 Mirv: no problem. here's a screenshot from me: http://www.bitwiz.org.uk/s/2009/08/kms-xorg-driver---almost.html - although only that "crashed" window is covering up the coloured noise below Aug 05 09:08:09 mrmoku|away ? Aug 05 09:08:13 ping when you're here :-) Aug 05 09:09:41 Weiss: all colors colored noise or only green and blue? Aug 05 09:11:19 larsc: all colours. it's some kind of Glamo test pattern or uninitialised memory patterned in some way. you see it because the driver tells KMS to display a buffer but the bits which write something sensible to that buffer don't work just yet Aug 05 09:16:31 ah pk Aug 05 09:40:51 hey mickey|patio o/ Aug 05 09:40:53 freesmartphone.org: 03mickey 07cornucopia * r2d37fcadbe5a 10/ (5 files in 4 dirs): gitignore++ Aug 05 09:41:08 mickey|patio: we need helper functions in libfso-glib Aug 05 09:41:11 it's annoying Aug 05 09:41:15 plus it seems we have a weird issue Aug 05 09:41:29 ogsmd is not up, we try to get a proxy object for GSM.Device Aug 05 09:41:32 And it returns something Aug 05 09:41:34 No error, no nothing Aug 05 09:42:10 that's ok i think Aug 05 09:42:19 since the actual object may be present later Aug 05 09:42:37 No it's not ok Aug 05 09:42:45 Because if you want to use it it'll fail with no reason Aug 05 09:42:53 it should fail with a reason Aug 05 09:42:56 A proxy object should be NULL imho Aug 05 09:43:11 (At least that's what I think is logical :-) ) Aug 05 09:43:23 no, really, proxy objects are just that Aug 05 09:43:30 Damn it's ugly Aug 05 09:43:40 well Aug 05 09:43:51 you have two methods to check whether proxies are valid Aug 05 09:43:55 1) ask dbus Aug 05 09:44:01 2) try a 'ping' method Aug 05 09:44:13 FIrst I'll try a ping Deubeuliou Aug 05 09:44:16 hehe Aug 05 09:44:17 As he's working on that =) Aug 05 09:44:20 good plan Aug 05 09:44:23 yo Aug 05 09:44:23 (he's in front of me, mind that) Aug 05 09:44:30 ah, right Aug 05 09:44:34 new job now Aug 05 09:44:47 Yeah :-) Aug 05 09:45:01 new trainee to harass Aug 05 09:45:05 hehe Aug 05 09:45:11 i should come over eventually Aug 05 09:45:16 to work a week with you guys in the office Aug 05 09:45:20 or so Aug 05 09:45:38 Well Aug 05 09:45:48 It could be a good thing :-) Aug 05 09:46:18 lets think about that once you got familiar with all Aug 05 09:46:18 I'll ask Louis if it'd be doable and if we can be all there (that is khorben, Marcus, Jérome...) Aug 05 09:46:22 Yeah :-) Aug 05 09:48:25 Ainulindale: are you working for Bearstech now? Aug 05 09:48:31 yep :-) Aug 05 09:49:13 Ainulindale: cool, i hope you'll enjoy it :D Aug 05 09:49:19 anyone remember what function is called once a library gets loaded / initialized? Aug 05 09:49:21 I hope too :-) Aug 05 09:49:38 dlload ? Aug 05 09:50:20 not necessarily Aug 05 09:50:39 on program startup with usual .sos Aug 05 09:50:46 mickey|patio: libraries should export routines using the __attribute__((constructor)) ? Aug 05 09:50:59 PaulFertser: ah, right, that was it Aug 05 09:51:08 hmm, lets see how i can do that from Vala Aug 05 09:52:05 mickey|patio: is there a sample client for libfso-glib? Aug 05 09:52:51 not really, although fsousaged uses it Aug 05 09:53:09 (to call the functions a resource needs to export) Aug 05 09:54:11 I'm lacking a bit behind w/ my dialer that should be a showcase Aug 05 09:54:18 (IOW... it's not even started) Aug 05 09:54:23 yet Aug 05 09:54:26 Well Aug 05 09:54:29 In C it's plain crappy Aug 05 09:54:39 Deubeuliou has been working on that for three days Aug 05 09:54:41 So is mrmoku|away Aug 05 09:54:59 yeah, i'm afraid for C usage there needs some additional convenience functions in libfso-glib Aug 05 09:55:10 it's kind of optimized for using from Vala right now Aug 05 09:55:13 (i think) Aug 05 09:55:18 well yeah that's just what I told you :-) Aug 05 09:55:28 For instance adding a signal Aug 05 10:14:23 mickey|patio: hi! Aug 05 10:14:58 morning DocScrutinizer Aug 05 10:21:14 freesmartphone.org: 03mickey 07cornucopia * rff8bb278cf38 10/libfsotransport/fsotransport/ (4 files): libfsotransport: convert to using the libfsobasics logger Aug 05 10:28:27 freesmartphone.org: 03mickey 07cornucopia * rbff7837fac95 10/ (README libfsobasics/README libfsoframework/README): README cleanup Aug 05 10:30:13 freesmartphone.org: 03mickey 07cornucopia * rad72961b3a8d 10/README: README: add build order Aug 05 10:56:38 hi, I am having problems with shr-messages Aug 05 10:56:48 I have a log of the debug here : http://shr.pastebin.com/m823b056 Aug 05 10:57:43 something has died on my phone over the last couple of days without me changing anything. I think it is somewhere in the dbus. ophonekitd frameworkd area Aug 05 10:58:11 calls etc are all down too Aug 05 10:59:54 is unstable in a comparitably safe place for upgrading at the moment? Aug 05 11:13:09 tig|, depends how you upgrade Aug 05 11:13:16 i recommned using a step-by-step script Aug 05 11:13:21 since "opkg upgrade" dies Aug 05 11:13:26 I take it, kill x add swap Aug 05 11:13:46 I am just baffled as to why this has suddenly died Aug 05 11:16:14 tig|, id still recomned a strp-by-step script, since some ppl complain that even with swap it dies Aug 05 11:16:44 F4t: cheers I will give that a go in a minute if I can't resurrect this Aug 05 11:17:18 tig|, the script: http://build.shr-project.org/tests/shr-step-by-step-upgrade.sh Aug 05 11:17:44 thanks Aug 05 11:18:08 hmm : Aug 5 12:15:39 om-gta02 user.warn kernel: [ 263.785000] Alignment trap: shr-messages (1565) PC=0x400fc104 Instr=0xe7920104 Address=0x00000001 FSR 0x013 Aug 05 11:18:56 no idea X_x Aug 05 11:19:19 possibly linked with : http://shr-project.org/trac/ticket/363 Aug 05 11:19:31 tig|: ask shr-messages developers to care about unaligned access (it should be avoided) Aug 05 11:22:10 Arhuaco3: hey, how are you? :) Aug 05 11:23:10 PaulFertser, whats unaligned access? Aug 05 11:23:48 F4t: basically is when you dereference a pointer to, say, word but the address is not 2*n. Aug 05 11:24:53 F4t: can happen if you disabled struct padding and have suboptimal fields order too. Aug 05 11:25:43 F4t: almost all architectures have penalty for analigned access, for some it's more serious, on some it doesn't work at all. Aug 05 11:26:34 mickey|patio: something weird Aug 05 11:26:44 There's a struct called FreeSmartphoneGSMDevice Aug 05 11:26:52 Which is the self parameter for the dbus calls Aug 05 11:26:58 And I can't find the definition of this struct Aug 05 11:27:50 PaulFertser, so its just a warning... Aug 05 11:27:57 F4t: on arm yes Aug 05 11:28:23 PaulFertser, which procesor simply fails? Aug 05 11:28:39 F4t: probably ia64, i'm not sure Aug 05 11:29:23 i think its easy to solve with a bit overhead in the processor Aug 05 11:29:42 so i doubt any new procesors have this as a fail... Aug 05 11:30:28 is connman-plugin-loopback still part of shr image?! Aug 05 11:30:28 F4t: what for to "solve" that instead of doing the right thing in the first place? Aug 05 11:30:57 PaulFertser, the right thing won't be to force all apps to use aligned memory access.... Aug 05 11:31:09 its right to do that bu u cant force it on ppl Aug 05 11:31:13 but* Aug 05 11:31:16 F4t: you doubt? Aug 05 11:31:38 guys cu soon Aug 05 11:31:40 doubt what? Aug 05 11:31:46 I don't know how much I will able to connect during those two weeks Aug 05 11:31:53 bye Aug 05 11:31:57 F4t: that new processors would somehow be more friendly towards aligned access Aug 05 11:32:03 F4t: unaligned Aug 05 11:32:29 ptitjes: by the way ophonekitd-vala won't compile Aug 05 11:33:07 lindi-, i dont disagree, aligned access is faster and better, i was just saying that unaligned shouldnt be a fatal situation, just an overhead processing Aug 05 11:33:28 F4t: i don't understand why it can't be fatal Aug 05 11:33:44 F4t: it's so low level stuff that you shouldn't need to care Aug 05 11:33:53 uhm, arm doesn't support unaligned access, does it? afaik the kernel sets up a trap and does the decoding in there Aug 05 11:34:43 * lindi- just reading the datasheet Aug 05 11:35:16 and thats why it's so slow, cause you have at least two context switches for each unaligned access Aug 05 11:35:53 lindi-, because some code segments can cause an unaligned access, for example tightly packed structures Aug 05 11:36:01 if you get an unaligned access error your program probably did something wrong. Aug 05 11:36:23 lindi-, its like isuing warning or failing when you use "mov eax,0" instead of "xor eax,eax" since thats not efficient Aug 05 11:36:27 F4t: you can always avoid those by generating code that does not do it Aug 05 11:37:03 lindi-, you can always write your code in BFL, but thats not as easy as packing ur structure in memory the same as its later stored/sent over the wire... Aug 05 11:37:07 F4t: it's the same thing if you want e.g. "13-bit integers" instead of 16-bit ones Aug 05 11:37:27 packed structures shouldn't be a problem, cause gcc should create proper accessors. Aug 05 11:38:44 F4t: yeah it seems that armv4 at least does not do unaligned word access Aug 05 11:38:55 lets assume a structure with { BYTE flags, WORD info,BYTE stuff,DWORD checksum} if its not-padded, how exactly can you avoid unaligned access here? Aug 05 11:39:22 { BYTE flags, WORD info,DWORD checksum}* Aug 05 11:39:30 F4t: the compiler will generate two instructions to access the "info" field Aug 05 11:39:40 F4t: when accessing the info field gcc does two byte accesses and composes the word value out of it Aug 05 11:40:01 and THATS more efficient then implementing unaligned access in cpu? Aug 05 11:40:08 yes Aug 05 11:40:19 maybe Aug 05 11:40:36 implementing new functionality to processor for rare cases is not a good idea anyway Aug 05 11:41:23 lindi-, i dont know, for me the "access any memory address, with any size" is kinda mandatory cpu feature Aug 05 11:41:24 "The ARM architecture traditionally expects all memory accesses to be suitably aligned." Aug 05 11:41:39 F4t: even if size is e.g. 123123 bytse? Aug 05 11:41:46 but thats not the point. when your programm triggers a alignment trap you are doing something like: int i; char *p; int *q; i = &p; ++p; q = p; *q; Aug 05 11:42:00 lindi-, i ment any of the standart sizes... Aug 05 11:42:21 F4t: you can, you just have to use the standard alignment :) Aug 05 11:42:36 and void the "any address" aswell Aug 05 11:43:12 "ARMv6 introduces unaligned word and halfword load and store data access support. When this is enabled, Aug 05 11:43:15 the processor uses one or more memory accesses to generate the required transfer of adjacent bytes Aug 05 11:43:18 transparently to the programmer, apart from a potential access time penalty where the transaction crosses an Aug 05 11:43:21 IMPLEMENTATION DEFINED cache-line, bus-width or page boundary condition. Aug 05 11:43:23 btw, if gcc avoid unaligned access, how did shr-messages manage to have a compiled code that does it? Aug 05 11:43:24 whoops, bit long paste, sorry Aug 05 11:43:38 F4t: as i said above Aug 05 11:44:12 int i; *((int *)((char *)(&i) + 1)); Aug 05 11:44:37 larsc, the horror.... Aug 05 11:44:48 hmm, what does C standard say about casting between pointer types Aug 05 11:45:06 lindi-, have fun with it? Aug 05 11:46:27 lindi-, also, the fact that ARMv6 introduces ualigned access, proves my point that its needed and newer cpu's support it with the overhead it costs... Aug 05 11:47:11 F4t: no it doesn't :) Aug 05 11:49:35 lindi-, the c standard not saying "have fun with it" about casting between pointers? or the armv6 paste above doesnt prove my point? Aug 05 11:49:51 F4t: the arm paste Aug 05 11:50:07 F4t: just because ARM does it does not mean it should be something that every processor architecture should support Aug 05 11:50:42 lindi-, more like arm finaly supports it, impretty sure intel's processors support it long ago Aug 05 11:51:16 F4t: x86 has a really large instruction set, too large :) Aug 05 11:52:49 "Some architectures like MIPS have special unaligned load and store instructions. One unaligned load instruction gets the bytes from the memory word with the lowest byte address and another gets the bytes from the memory word with the highest byte address. Similarly, store-high and store-low instructions store the appropriate bytes in the higher and lower memory words respectively." Aug 05 11:54:50 http://lwn.net/Articles/259732/ Documentation about unaligned memory access Aug 05 11:57:08 F4t: i'm having hard time reading the C standrad Aug 05 11:57:55 it could be 6.2.6.1.5 Aug 05 11:59:00 O_o Aug 05 11:59:25 larsc: "A pointer to an object or incomplete type may be converted to a pointer to a different Aug 05 11:59:28 object or incomplete type. If the resulting pointer is not correctly aligned59) for the Aug 05 11:59:31 pointed-to type, the behavior is undefined. Aug 05 11:59:46 larsc: maybe your example C programs hits this? Aug 05 12:00:47 F4t: so if this is what I think it is, only C programs that do undefined things can hit this problem :) Aug 05 12:00:58 F4t: don't do undefined things and you are safe ;) Aug 05 12:01:44 lindi-, i think i agree now, after i've been told that packing tightly a structure still won't generate unaligned access Aug 05 12:01:45 lindi-: yep. shr messages is probably doing something nasty Aug 05 12:01:52 that was my only legit example... Aug 05 12:01:58 http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1336.pdf is the link to standard draft Aug 05 12:02:55 F4t: in C it is very easy to do something that has undefined behavior Aug 05 12:03:20 lindi-, undefined and unwanted behaviour yes, but undefined yet wanted is quite hard to accomplish :) Aug 05 12:03:22 F4t: your armv4 processor is doing you a favor and letting you know :P Aug 05 12:03:56 lindi-, thats another thing i dont abide by, why wont it generate warning for every ineffcient assembly command? Aug 05 12:05:25 PaulFertser: shr-messages is just dumb wrapper around libframeworkd-phonegui, and libframeworkd-phonegui-efl is going to death Aug 05 12:05:25 "looks like you are using mov to zero a variable, use xor instead" Aug 05 12:05:28 F4t: the processor? ;) Aug 05 12:05:33 ye :D Aug 05 12:05:42 PaulFertser: so that aligment trap is going to be fixed automatically :P (if only new UI library won't have the same problem ;D) Aug 05 12:05:58 F4t: it's not about using a inefficent command, its about using a impossible command. Aug 05 12:06:08 imagine the popups, "looks like you are trying to use an inefficient sorting algorithm. I suggest you use XYZ instead, ok/cancel?" Aug 05 12:06:08 F4t: the kernel is so nice and fixes it for you Aug 05 12:06:34 F4t: you can turn it off if you want and your programm dies on a unaligned access Aug 05 12:06:38 lsarc, if the kernel can fix it for you, its no longer impossible.... Aug 05 12:06:52 F4t: yes it is Aug 05 12:06:55 if it dies, i agree there should be a msg, but a warning? Aug 05 12:08:32 "warning: ur using mul command with 2 instead of shifting left once" Aug 05 12:09:09 F4t: the kernel wont see that Aug 05 12:09:15 F4t: you can use valgrind to do that Aug 05 12:10:31 F4t: if you do an unaligned access the cpu will throw an exception. Aug 05 12:10:40 larsc, if the mul reaches the processor, then why not? Aug 05 12:10:43 lindi-, larsc, F4t: if you'll try to search, where is that aligment trap, then don't search in shr-messages ;) Aug 05 12:11:02 that nasty thing has to be in libframeworkd-phonegui-efl Aug 05 12:11:38 F4t: if you do a mul the cpu does not throw an exception Aug 05 12:14:10 larsc, but it can warn you in case its a very ineeficient mul :D Aug 05 12:14:45 F4t: not really ;) Aug 05 12:18:06 F4t: again, just run your program under valgrind and you can make it emit such warninsg Aug 05 12:18:51 lindi-, i dont wish for it to emit such warnings, i dont want it to emit any kernel level warnings regarding preformance as long as its running and doing what it should Aug 05 12:19:23 F4t: but it doesn't Aug 05 12:20:11 dos1: have you finished depeing ContactsDomain on GenericDomain? I don't think so, but i just want to ask :) Aug 05 12:20:13 work Aug 05 12:20:21 the unaligned warning, its running? doing what it should? then i dont wanna see it Aug 05 12:22:24 you can disable it Aug 05 12:23:16 larsc, without recompiling the kernel? Aug 05 12:23:37 not sure Aug 05 12:23:50 there is /proc/cpu/interrupt Aug 05 12:24:10 ah Aug 05 12:24:22 /proc/cpu/unalinged Aug 05 12:25:16 /proc/cpu/alignment Aug 05 12:25:23 yeah Aug 05 12:25:27 k :) Aug 05 12:27:52 Heinervdm: no, but i hope i'll finish it today :) Aug 05 12:28:02 dos1: great :) Aug 05 12:57:13 hi Aug 05 12:57:15 I've that: Aug 05 12:57:16 mkdir: cannot create directory '.e': Stale NFS file handle Aug 05 12:57:20 what could it be? Aug 05 12:57:26 should I change sdcard? Aug 05 12:57:40 the partition also often disapear on this card Aug 05 12:59:18 and the filesystem is ext2 Aug 05 12:59:22 but that's not on the openmoko Aug 05 12:59:26 it's on the G1 Aug 05 12:59:35 the distro is SHR Aug 05 12:59:36 Gnutoo: filesystem is broken Aug 05 12:59:40 ok thanks Aug 05 12:59:46 I'll remake the filesystem Aug 05 13:03:55 Gnutoo: how's SHR on G1 working? Aug 05 13:04:08 I just had usbnet working Aug 05 13:04:27 there is the need to write an alsa driver unless someone want to port all apps to sdcard Aug 05 13:04:31 s/sdcard/ Aug 05 13:04:35 why would someone want to replace a wokring android? Aug 05 13:04:39 the android audio system Aug 05 13:04:52 in order to have tons of free applications Aug 05 13:04:54 F4t: cause android is nonstandard crap? :P Aug 05 13:05:16 dos1, oh and shr is full of standard crap? Aug 05 13:05:36 Gnutoo, tons of applications? on the moko? Aug 05 13:06:00 no on the G1 Aug 05 13:06:07 F4t: SHR is standard Linux, with glibc and other libraries which you also have on desktop Aug 05 13:06:18 for instance you can have irc clients Aug 05 13:06:23 you can run everything you want, i was for instance using firefox on moko ;P Aug 05 13:06:29 wesnoth etc... Aug 05 13:06:44 crawling firefox... Aug 05 13:06:48 not running Aug 05 13:06:50 :P Aug 05 13:07:01 F4t: it doesn't matter ;p Aug 05 13:07:17 android, when it comes to developing, is almost the same like iPhone Aug 05 13:07:35 dos1, its open source ffs, how can it be like iphone? Aug 05 13:08:06 F4t: yup, it's open source, but not so free software Aug 05 13:08:21 dos1, why not? Aug 05 13:08:58 F4t: buy G1 and try to add some interesting function to android's api Aug 05 13:09:40 F4t: oh well... you can't without hacking bootloader! you don't even have root without hacking Aug 05 13:09:48 F4t: you could say "but after hacking it's possible!" Aug 05 13:10:00 F4t: but then there is no difference with Android and iPhone Aug 05 13:10:07 mmm Xserver needs ipv6 Aug 05 13:10:15 F4t: as you can also jail-break iPhone ;p Aug 05 13:10:30 F4t, there is a lot of proprietary things on the g1 Aug 05 13:10:50 F4t, such as ril,gps libs etc...and it spys on you according to some blogs Aug 05 13:11:01 such as strange network activity after a fix Aug 05 13:11:01 dos1, even if you do jailbreak the iphone u still dont have the source of the op system Aug 05 13:11:19 F4t: you're right... but it doesn't change anything :P Aug 05 13:11:27 im talking about developing applications for the system, not changing the system itself... Aug 05 13:11:42 F4t: oh, you can develop applications even for Windows Mobile Aug 05 13:11:50 without having sources... Aug 05 13:11:51 F4t, what if you want to develop something tricky like a phone firewall? Aug 05 13:12:02 F4t,that block certain numbers Aug 05 13:12:08 Gnutoo: ++ Aug 05 13:12:13 I'm not shure you can do it,and it won't be easy Aug 05 13:12:16 dos1, but you couldnt develop applications for any mobile phone opsys until symbian and ms) Aug 05 13:12:55 F4t: but that doesn't make android phones free... ;p Aug 05 13:12:59 F4t, yes but why not taking freedom one step ahread Aug 05 13:13:05 Gnutoo: ++ ;D Aug 05 13:13:28 Gnutoo, since youd also be easily able to break/brick your phone? Aug 05 13:13:43 F4t, if you don't touch the bootloader... Aug 05 13:13:54 F4t, for now I boot via fastboot Aug 05 13:13:54 im ptryy sure no company wants to deal with endless kiddies who bricked thier phone Aug 05 13:14:00 pretty* Aug 05 13:14:06 F4t: well, if i want to brick it, i can do that in many different ways, not only software Aug 05 13:14:52 F4t: and they don't have to deal with them... Aug 05 13:15:02 F4t, mmm...you can brick an iphone too it's a lot more easy: jailbreak it and wait for apple bricking it,if you have fully free system it's a lot harder Aug 05 13:15:07 F4t: if they bricked their phone, then it's their problems Aug 05 13:15:31 Indeed I want not to require the engineering bootloader Aug 05 13:15:46 that's why I didn't flash it(and because I was afraid) Aug 05 13:16:40 Android based phones aren't free as gta0x, they are rather in the same league as iPhone Aug 05 13:16:49 you can read sources, but you can't do anything with them Aug 05 13:17:17 if Android had on GPLv3, G1 in actual state would break licence Aug 05 13:17:24 that's why i don't think it's free Aug 05 13:17:33 s/had on/had be/ Aug 05 13:17:38 had be on* Aug 05 13:17:39 ... Aug 05 13:17:40 Xfbdev from kdrive needs ipv6? Aug 05 13:17:47 Gnutoo: WTF? Aug 05 13:17:52 Gnutoo: why for? Aug 05 13:18:23 I've a black screen+ a message about ipv6 Aug 05 13:18:40 which message? Aug 05 13:19:05 _XSERVTransSocketOpenCOTSServer: Unable to open socket for inet6 _XSERVTransOpen: transport open failed for inet6/om-gta02:0 _XSERVTransMakeAllCOTSServerListeners: failed to open listener for inet6 Aug 05 13:19:15 I'll check if I've already a server running Aug 05 13:19:41 no no server running Aug 05 13:24:13 dos1, do you have a workarround or should I add ipv6 Aug 05 13:24:37 Gnutoo: no, i don't know anything about ipv6 implementation... Aug 05 13:24:47 ok Aug 05 13:24:52 I'll add ipv6 then Aug 05 13:25:42 Gnutoo ? Aug 05 13:25:50 yes Aug 05 13:25:56 ok read your mail Aug 05 13:26:00 let me recheck it Aug 05 13:26:01 Ainulindale, ok Aug 05 13:26:20 part history of the bug Aug 05 13:26:24 don't tell your life :-) Aug 05 13:26:27 ok Aug 05 13:26:30 remove it then Aug 05 13:26:39 or change it Aug 05 13:26:42 nah Aug 05 13:26:50 is the rest ok? Aug 05 13:26:51 You'll do it I don't want to do it for you :-p Aug 05 13:26:54 Yep pretty good Aug 05 13:27:01 ok Aug 05 13:27:04 So create an account on the blog Aug 05 13:27:13 Tell me when you're done Aug 05 13:27:23 I hope my english is not too bad Aug 05 13:27:32 Ainulindale, I'm adding G1 support for oe/SHR Aug 05 13:27:45 are you interested in it Aug 05 13:27:45 ? Aug 05 13:28:19 Well Aug 05 13:28:23 Greatly. Aug 05 13:28:29 I thought audio wasn't supported Aug 05 13:28:31 Is it now? Aug 05 13:30:23 Ainulindale: AFAIK it is, but though some nonstandard node and set of ioctls in /dev Aug 05 13:30:31 Ainulindale: so alsa driver needs to be written Aug 05 13:30:41 well that's what I remember too Aug 05 13:31:16 Ainulindale: but even without propertiary shit in kernel they managed to play sound by that nonstandard interface Aug 05 13:32:37 hmmm Aug 05 13:33:42 Gnutoo: so, did you create an user? Aug 05 13:33:55 I'll do it in few minutes Aug 05 13:34:01 Ok Aug 05 13:34:02 I've to post documentation Aug 05 13:34:04 Then I'll give you access Aug 05 13:34:15 Then you'll post your thing and I'll edit it a bit :-) Aug 05 13:34:26 Just to remove a sentence about you asking ptitjes and then asking me and then asking mrmoku|away :-) Aug 05 13:34:46 ok thanks Aug 05 13:34:54 lol Aug 05 13:35:36 hm enotes from -unstable feed don't want to install.. Aug 05 13:36:30 opkg install enotes give this http://wklej.to/TyEx Aug 05 13:38:25 soltys: is it really from -unstable feed, or from my repo Aug 05 13:38:27 ? Aug 05 13:38:37 dos1: sec Aug 05 13:38:59 dos1: hm your repo ;P Aug 05 13:40:43 I'll do it now Aug 05 13:43:59 done waiting for confirmation mail Aug 05 13:44:28 Well too bad for you Aug 05 13:44:31 You're only the 41st Aug 05 13:44:37 You could have been the 42nd Aug 05 13:44:40 :-) Aug 05 13:44:53 I manually enabled you Gnutoo Aug 05 13:45:02 ok thanks Aug 05 13:45:05 SHR: 03seba.dos1 07opimd-utils * ref0acdadcbf7 10/opimd-contacts: opimd-contacts: load whole contact list asynchronously Aug 05 13:45:06 SHR: 03seba.dos1 07opimd-utils * rea345976426e 10/opimd-contacts: opimd-contacts: dispose queries when they should be disposed Aug 05 13:45:08 SHR: 03seba.dos1 07opimd-utils * rbcd0b5206fb5 10/TODO: Add TODO Aug 05 13:45:15 And you're now an author Aug 05 13:45:23 ok Aug 05 13:45:26 http://blog.shr-project.org/cgi-bin/mt/mt.cgi Aug 05 13:45:36 Create your entry, do not publish it, tell me when you're done -) Aug 05 13:49:02 ok Aug 05 13:55:28 freesmartphone.org: 03mickey 07cornucopia * r360781002ad8 10/libfsobasics/ (MAINTAINERS fsobasics/logger.vala): libfsobasics: fill in MAINTAINERS Aug 05 13:56:12 freesmartphone.org: 03mickey 07cornucopia * r95860e5cb97f 10/libfsobasics/configure.ac: libfsobasics: post-release version bump Aug 05 13:57:25 freesmartphone.org: 03mickey 07cornucopia * r151664622151 10/libfsotransport/configure.ac: libfsotransport: require libfsobasics 0.8.1.0 Aug 05 13:58:25 SHR: 03seba.dos1 07opimd-utils * rf44f03d6b83d 10/ (TODO opimd-contacts): opimd-contacts: support Query.ContactAdded signal Aug 05 13:58:59 freesmartphone.org: 03mickey 07cornucopia * r3c3e1b862cdd 10/libfsotransport/configure.ac: libfsotransport: post-release version bump Aug 05 14:00:50 freesmartphone.org: 03mickey 07libgsm0710mux * r47edc43c9bca 10/configure.ac: require libfsobasics 0.8.1.0 and libfsotransport 0.9.3 Aug 05 14:02:33 freesmartphone.org: 03mickey 07libgsm0710mux * r6ea2787d49f2 10/configure.ac: post-release version bump Aug 05 14:04:29 freesmartphone.org: 03mickey 07fso-abyss * r449f09a597e3 10/ (configure.ac src/consts.vala): require libgsm0710mux 0.3.5 Aug 05 14:05:20 freesmartphone.org: 03mickey 07fso-abyss * r008485b24ab0 10/configure.ac: post-release version bump Aug 05 14:05:21 mrmoku|away: I'd like to have ' eject' cmdline-tool Aug 05 14:07:43 Ainulindale, if I click on submit it won't publish it? Aug 05 14:10:27 Ainulindale, ping Aug 05 14:13:07 Gnutoo: howdy! Aug 05 14:13:22 DocScrutinizer, hi Aug 05 14:15:23 Ainulindale, oops Aug 05 14:15:38 is the entry published I've clicked on "save this entry" Aug 05 14:15:41 help!!! Aug 05 14:15:59 Ainulindale, ping Aug 05 14:16:36 http://planet.openmoko.org/ does not have the entry Aug 05 14:21:27 I got a dbus.exceptions.DBusException: org.freesmartphone.GSM.Network.Unauthorized: Aug 05 14:21:41 when entering my PIN Aug 05 14:22:09 (with gsmsim.SendAuthCode) before trying to power the Antenna up Aug 05 14:22:59 but after trying to powering it up, it works. Aug 05 14:24:50 mmm Aug 05 14:24:58 I've a _ on top-left of the screen Aug 05 14:25:04 and Xfbdev is not starting Aug 05 14:26:50 Gnutoo: wait Aug 05 14:27:03 ok Aug 05 14:27:07 Title "..." ? Aug 05 14:27:10 oops Aug 05 14:27:17 I'll add a title Aug 05 14:28:22 Nah Aug 05 14:28:23 I just did it Aug 05 14:28:26 Do not edit it :-) Aug 05 14:28:31 I just modified some stuff Aug 05 14:28:34 It'll appear on planet later Aug 05 14:28:37 ok Aug 05 14:28:48 thanks a lot Aug 05 14:29:06 Thanks to you for your piece :-) Aug 05 14:29:26 btw I should send a mail to the person that did the initial port of om200x.x to the om Aug 05 14:29:30 s/om/g1 Aug 05 14:32:29 freesmartphone.org: 03mickey 07libgsm0710mux * r8e3e7533b286 10/gsm0710mux/ (channel.vala consts.vala multiplexer.vala): better align hex debug output Aug 05 15:28:35 DocScrutinizer, for automatic SIM ejection? Aug 05 15:28:49 muhaha, no Aug 05 15:29:04 mrmoku: usb-umts-stick needs this Aug 05 15:30:14 Ainulindale, ping Aug 05 15:30:53 PaulFertser, shr-config-dev should not even be installed :( Aug 05 15:31:01 should be save to remove with -force-depends Aug 05 15:31:14 Ainulindale, image from scratch was bad :( Aug 05 15:31:55 mrmoku: the stick registers as vfat mass storage device 16MB, and under windoof it's supposed there's some crap to install/automount the modem part. if you eject the stick, it immediately re-registers as modem-device Aug 05 15:32:14 DocScrutinizer, he, ok Aug 05 15:32:30 * mrmoku looks into OE for some eject package... Aug 05 15:32:42 DocScrutinizer: isn't there kernel support for this? Aug 05 15:33:02 lindi-: no idea what to answer Aug 05 15:33:03 If I request a resource, and forget to release it (for example, my app crashes), the resource will remain on? Aug 05 15:33:12 could someone explain me why, in libfso-glib's freesmartphone.h, are several Error enums defined but not for all domains Aug 05 15:33:19 Oops, why I'm logged in two times Aug 05 15:33:25 for instance, there is no FreeSmartphoneGSMSIMError Aug 05 15:33:37 but there is a ...GSMCallError Aug 05 15:33:45 thrashold: use fsoraw, and resource gets freed automagically Aug 05 15:35:07 thrashold: PaulFertser has published a neat mini-howto wrt resources yesterday Aug 05 15:35:36 DocScrutinizer, thanks. I'll consider it, it might turn out to be a better idea than requesting resources manually. Aug 05 15:35:51 Deubeuliou, maybe the GSMSIM domain throws no errors? Aug 05 15:36:18 it does Aug 05 15:36:28 http://git.freesmartphone.org/?p=specs.git;a=blob_plain;f=html/org.freesmartphone.GSM.SIM.html;hb=HEAD Aug 05 15:36:57 But that leads me to another question. Did fso-gpsd stop requesting the GPS resource in recent versions? Since I upgraded it I need to activate my GPS manually. Aug 05 15:37:07 it's basically no diff. fsoraw does also request resource "manually". the point is fsousaged knows when your app exits. you just need to use resource_alloc and NOT set_resource_policy Aug 05 15:38:02 Deubeuliou, hmm, let me check my libfso-glib... Aug 05 15:39:15 I guess these are function from the C API, I would be using dbus calls directly, most likely. Aug 05 15:43:11 Deubeuliou, no idea Aug 05 15:45:49 ok mrmoku Aug 05 15:48:19 thrashold: use RequestResource and sleep in peace Aug 05 15:48:20 ;) Aug 05 15:48:34 that's what they want to say to you : Aug 05 15:48:36 :D Aug 05 15:49:03 mrmoku: http://2.bp.blogspot.com/_Hht7VZ29XgM/SnmYi2KjwkI/AAAAAAAAUEY/NOlCYnTJSY4/s1600-h/win7-kde.jpg Aug 05 15:49:26 ffs :( Aug 05 15:49:33 now my connection really died Aug 05 15:49:44 mrmoku: part of article in some german newspaper about Windows history Aug 05 15:49:45 modem can't see PPPoE server ;( Aug 05 15:49:46 mrmoku: ;D Aug 05 15:50:29 dos1, but I still should call ReleaseResource when I'm finished? Aug 05 15:50:29 Where can I read some docs about the API, the one at docs.freesmartphone.org seems outdated? Where can I view this mini-howto by PaulFertser ? :) Aug 05 15:50:44 thrashold: you don't have to Aug 05 15:50:49 dos1, hehe... probably ComputerBild ;) Aug 05 15:50:53 thrashold: docs.freesmartphone.org outdated? Aug 05 15:51:04 oops Aug 05 15:51:22 probably grepping the page for RequestResource was a bad idea Aug 05 15:51:22 :D Aug 05 15:51:44 my bad Aug 05 15:51:44 sorry Aug 05 15:52:16 thrashold: on docs.fre... there is even described way how resources work Aug 05 15:54:19 dos1: http://media.ubuntuusers.de/forum/attachments/2095330/K.jpg better resolution, under the picture is written: In the preview you can see the new design of Windows7 ;) Aug 05 16:05:04 * tig| wonders how long this update is going to take Aug 05 16:06:04 four hours 15 mins so far Aug 05 16:27:16 Ainulindale, * ERROR: Cannot satisfy the following dependencies for task-shr-minimal-apps: * fso-abyss * fso-abyss * Aug 05 16:27:25 strange thing Aug 05 16:28:28 hmpf Aug 05 16:28:32 recompile this ugly thing Aug 05 16:28:36 then remove the stamps Aug 05 16:28:39 I'm sick & tired of that Aug 05 16:29:23 Ainulindale, I don't even understand how task-shr-minimal-apps depends on fso-abyss... Aug 05 16:30:43 Some app must Aug 05 17:10:00 FSCK, anybody ever tried to open and use opkg.org in midori? :-( Aug 05 17:10:27 DocScrutinizer, eject is in OE and builds even :) Aug 05 17:10:32 will add it to our feed Aug 05 17:10:42 cool, thanks Aug 05 17:48:14 dos1: don't you think it's worthwhile and easy to implement to stop e.g settings-power from multiple invocation (aka multiple screens)? Aug 05 17:48:50 DocScrutinizer: multiple screens? Aug 05 17:49:50 dos1: open settings, click power 3 times. then pulldown illumebar and open the screens-list. you find 3 power-screens Aug 05 17:50:03 DocScrutinizer: so why do you want to click power 3 times? ;D Aug 05 17:50:19 FSCK 3%@37/\#&<52 Aug 05 17:50:37 >:-( Aug 05 17:50:53 ;) Aug 05 17:52:36 does someone know if files in libfso-glib as freesmartphone-*.vala are written by hand or generated from the specs ? Aug 05 17:57:19 dos1: still no commitlog yet?? WHAT'S UP? Aug 05 17:57:50 dos1: getting old? Aug 05 17:58:02 ;-P Aug 05 17:58:18 DocScrutinizer: i'll fix it in shr-settings launcher rewrite which i'm planning, as with actual mess it would be quite hard to do... ;x Aug 05 17:58:38 k, fair enough Aug 05 17:58:49 :) Aug 05 18:01:23 dos1: rewrite in vala? :P Aug 05 18:01:42 hi ptitjes Aug 05 18:01:53 I was looking for you :) Aug 05 18:01:58 hi Deubeuliou Aug 05 18:01:59 I would like to continue with shr-config but installing the efl-development environment is such a hassle :( Aug 05 18:02:03 how are you ? Aug 05 18:02:16 spaetz: if i'll find nice way to load python modules from vala then i'll consider vala rewrite ;) Aug 05 18:02:19 fine and you ? Aug 05 18:02:26 finally in holidays Aug 05 18:02:29 great. i'm fine Aug 05 18:02:30 yay, a whole evening alone at home and no traveling... Aug 05 18:02:31 ah ) Aug 05 18:02:32 :) Aug 05 18:02:35 spaetz, libeflvala does not build :( Aug 05 18:02:35 :p Aug 05 18:02:39 dos1: bah :) Aug 05 18:03:20 mrmoku: mmh, yeah that's another reason. living on the edge of development is quite annoying Aug 05 18:03:21 I needed to know whether libfso-glib's freesmartphone-gsm.vala is written by hand or generated with vala-dbus-binding-tool ? Aug 05 18:03:32 * spaetz considers a GTK-based settings tool Aug 05 18:04:46 checking logo submissions. I like the snake S still, could be improved but nice Aug 05 18:06:04 dos1: how are you selecting the duration (timespan) shown in settings-power in first line after "charge: xx% (" ? Aug 05 18:06:19 Deubeuliou: every single bit of the libfso-glib service interfaces is generated through vdbt Aug 05 18:06:40 ok Aug 05 18:06:43 Deubeuliou: the freesmartphone-utils.vala is hand made though Aug 05 18:07:03 just a utility function for non-Vala users Aug 05 18:07:16 so, why is there only a few errordomains ? I'm currently using libfso-glib in C, but I lack them Aug 05 18:07:30 DocScrutinizer: sec, i'll check Aug 05 18:08:14 Deubeuliou: time_to_empty_now and time_to_full_now sysfs nodes Aug 05 18:09:23 I was going to patch it but if it's generated ... Aug 05 18:10:04 dos1: please do *not* care about "status:" (3. line). Rather from the sysfs_nodes time_to_full and time_to_empty use the one that is *not* showing the 65535 value which results in 1092h15m display Aug 05 18:11:29 DocScrutinizer: hmm... ok, i'll look at it Aug 05 18:11:29 are these instructions up to date ? http://wiki.openmoko.org/wiki/Manually_using_GSM I try to connect through cu, freezes after "Connected", tried turning the mode on and off, also turning on the flow control after starting cu Aug 05 18:11:30 status sysfs node is still reporting nonsense Aug 05 18:11:45 modem Aug 05 18:11:53 doxy2: if you have fso running, then instead of cu use mickeyterm Aug 05 18:12:12 PaulFertser: might we eventually get a kernel with this braindamaged bug fixed or at least disabled Aug 05 18:12:13 ? Aug 05 18:12:14 dos1: nope, not fso Aug 05 18:12:38 dos1: picocom didn't work either Aug 05 18:13:02 doxy2: check, if something else isn't using gsm Aug 05 18:13:09 maybe by lsof Aug 05 18:13:17 doxy2: you tried to enter AT two times quickly? Aug 05 18:13:31 I wonder if I'm here. Aug 05 18:13:48 TAsn: you are there Aug 05 18:13:51 TAsn: pong ;p Aug 05 18:13:53 cool :) Aug 05 18:14:00 I wonder for how long Aug 05 18:14:04 DocScrutinizer: well, no Aug 05 18:14:10 for some reason a power out killed my connection for the whole day Aug 05 18:14:13 and only now it came back Aug 05 18:14:15 modem goes to sleepmode after 10sec, and you have to send a first arbitrary cmd "to nirvana" to awake it Aug 05 18:14:17 without me doing anything. Aug 05 18:14:18 :) Aug 05 18:14:29 anyhow, food. Aug 05 18:16:33 TAsn: enjoy! same here... Aug 05 18:17:15 DocScrutinizer: status? Is it charging/not charging/discharging? Aug 05 18:17:47 the old nasty mbc-restarter in kernel Aug 05 18:18:19 PaulFertser: which makes sysfs node report charging despite MBC evidentally isn't Aug 05 18:19:13 thank god the bug has that bug so it actually fails to restart charge process Aug 05 18:19:56 I'm starting to suspect it has something to do with the kernel, or my system in general... Aug 05 18:21:07 DocScrutinizer: mbc restart is fixed Aug 05 18:21:30 DocScrutinizer: and status is almost fix (except that it uses charger status instead of battery status even if bq27k is functional). Aug 05 18:21:40 we discussed nasty of tty's in linux recently. There's most probably some flaw in kernel's serial uart driver that makes it incredibly hard to get communication to e.g. modem right via cu etc Aug 05 18:21:49 * spaetz sighs, it seems to be difficult to get kernel bugs out of the kernel even, how many years after device release? Aug 05 18:21:52 doxy2: ^ Aug 05 18:21:59 * spaetz knows he is blatantly simplifying Aug 05 18:22:40 i see Aug 05 18:22:45 PaulFertser: ooh, sorry. Still kernel of 09-07-03 here Aug 05 18:22:59 but i canẗ access it through frameworkd either Aug 05 18:23:17 DocScrutinizer: :D Aug 05 18:24:26 doxy2: mickeyterm works great Aug 05 18:25:08 Hey all - a couple of very high-level questions. I'm looking at moving from OM2009 TR5 to SHR-Unstable. There are a *lot* of images in http://build.shr-project.org/shr-unstable/images/om-gta02/ - which is the best one to use? shr-image-om-gta02.jffs2 ? shr-shr-image-glibc-ipk--20090721-om-gta02.rootfs.jffs2 ? Aug 05 18:25:12 even without gsm0710muxd, if you use the right device parameter on invocation Aug 05 18:25:20 doxy2: we had success using socat with proper parameters Aug 05 18:25:41 doxy2: socat - file:/dev/ttySAC0,crtscts,crnl,b115200 Aug 05 18:25:51 doxy2: at least that's what lindi- always uses for testing. Aug 05 18:25:55 wjbaird: I'd recommend 090703 ;ß) Aug 05 18:27:08 wjbaird: to me it seems it's been just regressions since then, others here might disagree Aug 05 18:27:44 DocScrutinizer: Thanks. I guess you mean: shr-shr-image-glibc-ipk--20090703-om-gta02.rootfs.jffs2 - and does that mean I shouldn't upgrade after installing until things are sorted out? Aug 05 18:28:00 yup, and yup Aug 05 18:28:12 DocScrutinizer: I don't particularly want to be bleeding edge --- would just be happy with a working phone... Aug 05 18:28:38 using 090703 here as a daily phone Aug 05 18:28:50 Ok - I'll try that image out... Aug 05 18:28:52 thanks! Aug 05 18:29:35 DocScrutinizer: guess I should get the kernel build from the same day? Aug 05 18:29:53 wjbaird: you might want to adjust some earpiece volume and mic sensitivity in /usr/share/shr/scenarii/gsmhandset.state to your own preferences Aug 05 18:30:05 (kernel) exactly Aug 05 18:30:27 thanks - I've got a gsmhandset.state on OM2009 that seems to work well - I might just copy it over... Aug 05 18:30:47 (gsmhandset) see wiki page for details Aug 05 18:31:25 you may try, but maybe you find it yields not the results you want Aug 05 18:31:32 wjbaird: of course i'd recommend to upgrade kernel and kernel modules ;) Aug 05 18:32:06 PaulFerster: that's what I love about this community - everyone always agrees on everything! ;-) Aug 05 18:32:23 PaulFertser: give us a 3liner script to download and start to accomplish this ;-) Aug 05 18:32:51 wjbaird: well, kernel update might actually be ok Aug 05 18:33:05 DocScrutinizer: no idea, i don't use opkg here. But i know guys upgrade the kernel (and corresponding modules) with it. Aug 05 18:33:17 so I'd need to get the more recent kernels and then install the newer modules? Aug 05 18:33:32 any substantial improvements to the kernel in the last month? Aug 05 18:33:47 duh, interesting. how should opkg even know where my kernel lives? Aug 05 18:34:14 wjbaird: not afaik Aug 05 18:34:55 DocScrutinizer: (kernel/opkg) - well, if the kernel is configured through the package mgmt system, it'd know. that's how I updated my desktop kernel on ubuntu... Aug 05 18:35:04 but ask paul, he's kernel maintainer ;-P Aug 05 18:35:17 wjbaird: dumb battery driver Aug 05 18:35:43 http://lists.usefulinc.com/pipermail/gnome- Aug 05 18:36:06 oops Aug 05 18:36:07 DocScrutinizer: i'm not Aug 05 18:36:13 wjbaird: Neo is a little different wrt this (kernel) Aug 05 18:36:29 PaulFertser: ;-P Aug 05 18:36:41 PaulFertser: nice - but I just got my second battery through my buzz fix, so I don't think I'll need to try out a 'dumb battery' for a while... I think I'll stick with the July 3rd kernel for now... Aug 05 18:37:08 wjbaird: some nasty bug with wm8753 was fixed but i've never seen it surfaced in RL. Aug 05 18:37:25 PaulFertser: ???? Aug 05 18:37:36 wjbaird: ah, july 3rd already has that Aug 05 18:38:00 still ???? Aug 05 18:38:05 DocScrutinizer: has that patch :) Aug 05 18:38:15 still ???? Aug 05 18:38:22 which one? Aug 05 18:38:28 DocScrutinizer: the bug is with register caching, one register was not cached properly. Let me find the link. Aug 05 18:38:36 aaaah Aug 05 18:38:39 yes Aug 05 18:38:50 no link needed Aug 05 18:39:13 doc knows everything, just can't recall most of it ;-P Aug 05 18:39:57 DocScrutinizer: larsc fixed it, but i'm not sure when it was comitted. Aug 05 18:40:18 * DocScrutinizer considers acting according to time of day... dinner Aug 05 18:40:26 bbl Aug 05 18:40:28 DocScrutinizer: http://lists.openmoko.org/pipermail/openmoko-kernel/2009-July/010310.html Aug 05 18:40:39 yo yo yo Aug 05 18:40:54 the off by one issue Aug 05 18:41:24 good for some OOOPSes Aug 05 18:41:26 Comitted Sun, 5 Jul 2009 14:02:47 Aug 05 18:41:32 eeeek Aug 05 18:41:47 seems I need a kernel update ;-D Aug 05 18:42:04 DocScrutinizer: probably not, i'm not sure it resulted in oopses. Aug 05 18:42:39 DocScrutinizer: go out and enjoy your meal finally! Aug 05 18:42:41 ;) Aug 05 18:42:48 cya Aug 05 18:46:14 how safe is an opkg upgrade Aug 05 18:46:15 mrmoku, ? Aug 05 18:46:17 dos1, ? Aug 05 18:48:22 just did one today Aug 05 18:48:33 TAsn, yep Aug 05 18:48:33 and things seem fine Aug 05 18:48:50 TAsn, at your own risk... as always :P Aug 05 18:48:53 so a green frome here :) Aug 05 18:49:03 mrmoku, of course... Aug 05 18:49:22 who had volunteered to implement the green/yellow/red voting system again? :-) Aug 05 18:49:34 spaetz, it was abandoned Aug 05 18:49:43 since we decided to just make an always gree Aug 05 18:49:44 n Aug 05 18:49:45 testing image Aug 05 18:49:46 :) Aug 05 18:50:18 hehe Aug 05 18:50:19 ;) Aug 05 18:50:28 sounds good to me :) Aug 05 18:51:00 I vote for always green unstable images too Aug 05 18:52:25 spaetz: i'm against always green unstable ;P Aug 05 18:52:42 unstable is unstable is unstable, and i want it that way! Aug 05 18:52:49 how about a slightly more updated testing? Aug 05 18:53:08 but now unstable unstable isn't acceptable, as we don't have even unstable testing ;P Aug 05 18:53:14 dos1, same here :) Aug 05 18:53:19 (I mean I agree) Aug 05 18:53:39 so i vote for always green testing, and unstable sometimes green, sometimes yellow and sometimes red ;P Aug 05 18:54:00 wjbaird: we're fighting with new testing image for ~month Aug 05 18:54:56 dos1: is it really that hard to understand why the image can't be build? You're all python coders after all and have quite some OE experience. I think some dedication and the problem will be solved. Aug 05 18:55:02 dos1: I understand - I work in the software industry and know how fun stabilizing code can be... Aug 05 18:55:28 PaulFertser: oh, bitbake is mystery for me :P Aug 05 18:55:39 Oh - the issue is that you *can't* build the testing image? Not that you can stabilize for it? Aug 05 18:55:48 wjbaird: both Aug 05 18:55:57 wjbaird: when one we can do, then other is broken ;P Aug 05 18:56:09 wjbaird: we just don't have luck Aug 05 18:56:36 dos1: :-) Sounds like fun. and I guess freezing unstable when it isn't until you can build the test images isn't an option? Aug 05 18:56:44 now we have weird problems with packages not getting into image Aug 05 18:56:58 that's why there is no fresh unstable images too :P Aug 05 18:57:14 wjbaird: no, as we can't build complete unstable image as well :-) Aug 05 18:57:36 there are always problems, when one is fixed, then other is noticed Aug 05 18:58:15 :-) It's always exciting... well, I'm an old hand at Python - but I'm afraid I don't know the first thing about bitbake, so I'm not sure I can be of assistance... Aug 05 18:58:33 well - ok, I know bitbake builds stuff, so I guess I don't know the second thing about bitbake... Aug 05 18:58:59 wjbaird: then you have the same situation as me ;) Aug 05 18:59:46 mrmoku: so what is image doing? what is actual problem? ;) Aug 05 18:59:52 i don't know if i'm up-to-date ;D Aug 05 19:00:49 ok, dos1 Aug 05 19:00:56 tell me what I can do to fix opimd Aug 05 19:01:01 and what you need to know Aug 05 19:01:03 and how to get it. Aug 05 19:06:36 this: http://xkcd.com/619/ summs up connman pretty well :D Aug 05 19:06:44 PaulFertser: i tried that socat method. it starts and spits out "OK" infinetely Aug 05 19:06:57 hey F4t Aug 05 19:07:02 hey TAsn Aug 05 19:07:05 doxy2: that's because some echo should be disabled first. Aug 05 19:07:14 doxy2: if only i remembered the cause for that... Aug 05 19:07:44 F4t, this actually sums up flash. Aug 05 19:07:54 I already managed to use this strip as a reference today :) Aug 05 19:08:38 PaulFertser: Reducing the kernel log level doesn't help Aug 05 19:08:58 doxy2: doesn't help communicating with the modem? Aug 05 19:09:27 TAsn, this actually sums up linux kernel Aug 05 19:09:31 doxy2: http://wiki.openmoko.org/wiki/Neo_1973_and_Neo_FreeRunner_gsm_modem#Avoiding_Infinite_Echos Aug 05 19:09:44 F4t, this just sums up open source Aug 05 19:09:51 as usually devs like non mainstream features Aug 05 19:09:59 i agree with that Aug 05 19:10:01 as we live in caves and don't talk to other humans ;) Aug 05 19:10:12 (caves with wifi reception, duh.) Aug 05 19:12:28 hehe Aug 05 19:12:38 PaulFertser: ah, splendid Aug 05 19:19:28 PaulFertser: well now i don't get anything when i start socat Aug 05 19:19:32 damn Aug 05 19:19:58 doxy2: you're doomed for eternity Aug 05 19:20:15 doxy2: what's your problem in the first place? Aug 05 19:21:40 PaulFertser: I can't access GSM in any way. RequestResource on frameworkd doesn't work, cu doesn't work, picocom doesn't work. and not apparently socat either Aug 05 19:22:10 PaulFertser: Still, the modem isn't broken since it works in prebuilt distros Aug 05 19:23:04 i dunno, maybe it's my kernel. no idea Aug 05 19:23:42 doxy2: "prebuilt distros"? Are you using some special distro? Aug 05 19:23:58 doxy2: pastebin s3c24xx-gpio output too Aug 05 19:26:04 my self-built system, LFS. where do i get that output? Aug 05 19:30:34 that's weird Aug 05 19:30:37 I just did an opkg upgrade Aug 05 19:30:40 and opkg didn't crash Aug 05 19:30:45 and it fucking installed a lot of packages. Aug 05 19:33:27 probably kernel Aug 05 19:33:34 spaetz, duh. Aug 05 19:33:49 ;) Aug 05 19:39:50 dos1, nvm Aug 05 19:39:56 the little fuck face fixed itself Aug 05 19:40:07 (after erasing the db and installing newest frameworkd) Aug 05 19:40:25 btw Aug 05 19:40:35 the last sent contained my last swear word ever on this chan Aug 05 19:40:36 :) Aug 05 19:40:43 (and hopefully in general) Aug 05 19:46:05 TAsn! I am appalled :-) Aug 05 19:47:18 spaetz, because of my swearing or my new lack of swearing? :) Aug 05 19:47:47 PaulFertser, ^ it's all your fault, I never said "fuck" (here) before you gave the russian mat link :) Aug 05 19:48:34 TAsn: hehe, russians always do that :D Aug 05 19:48:40 TAsn: blame canada Aug 05 19:49:19 PaulFertser, please, DO NOT remind me of southpark, that'll only increase my cursing needs :) Aug 05 19:49:31 :D Aug 05 19:52:36 mkkay. Aug 05 19:54:53 because of your swearing of course ;-) Aug 05 19:55:56 spaetz, I thought you may be russian, as PaulFertser is surprised when people don't swear. Aug 05 19:55:56 ;) Aug 05 19:56:40 hehe, nah I am not. I come close from the former west-german/east-german border though :) Aug 05 19:56:43 Hehe, i do. Aug 05 19:57:13 Ainulindale, http://www.thesun.co.uk/sol/homepage/showbiz/bizarre/2570875/Bar-Refaeli-is-considering-posing-for-Playboy.html <--- I bet you'll like israel even more after you'll see that. Aug 05 19:57:39 she's hot. Aug 05 19:58:30 she is Aug 05 20:03:14 TAsn: model business is sexists Aug 05 20:05:31 PaulFertser, yeah, they only take girls that look good. Aug 05 20:05:33 :| Aug 05 20:05:55 I think feminists should fight so ugly girls will also get to be models! Aug 05 20:06:43 TAsn: i think intelligent people (both girls and boys) should just boycott modelling. Aug 05 20:06:55 PaulFertser, I disagree Aug 05 20:07:04 dos1|away, you are crazy! Aug 05 20:07:21 wth is this opimd-utils progress?! when did you have time to do all of that?! Aug 05 20:07:28 TAsn: using a person as a thing is immoral imho. Aug 05 20:07:31 PaulFertser, I think modeling as art Aug 05 20:07:38 PaulFertser, again, I disagree. Aug 05 20:07:53 a model Aug 05 20:07:55 TAsn: i'd agree if that "art" wouldn't ruin lifes. Aug 05 20:07:57 is a beautiful object Aug 05 20:08:23 PaulFertser, you mean life of little girls who look up to the thin models? Aug 05 20:09:34 TAsn: both that girls, and girls who is grown up and suffer and the models themselves. Also it can produce a stereotype in men that someone who wants to be successful needs to "have" a live "doll". Aug 05 20:09:34 PaulFertser, I look at models as beautiful beings, models are the same as sculptures (in my pov) when they perform Aug 05 20:09:49 I think it's ok to adore beauty. Aug 05 20:10:17 PaulFertser, that's just poor education and weak character Aug 05 20:10:22 you can't always defend the weak. Aug 05 20:10:43 sometimes the weak have to get stronger themselves Aug 05 20:11:20 PaulFertser, btw, there are also male models Aug 05 20:11:25 what do you think about that? Aug 05 20:11:25 TAsn: businessmen are wolves doing everything to make money. Not all people are strong enough to defend themselves. Also society is weak and sick, otherwise it would oppose that thing. Aug 05 20:12:05 TAsn: i don't care if a model is male or female. The whole business is wrong and harmful. Aug 05 20:12:05 PaulFertser, society is weak and sick, with that I can agree. Aug 05 20:12:14 also about your statement about the businessmen Aug 05 20:12:36 PaulFertser, what's wrong about encouraging good looks? Aug 05 20:13:24 it's the same as buying a nice car Aug 05 20:13:24 or a nice sofa Aug 05 20:13:24 we are animals (humans) Aug 05 20:13:24 and as animals we have the basic instinct of attraction Aug 05 20:13:26 and therefore those who are more attractive get more attention Aug 05 20:13:26 TAsn: it's not good looks, it's looks that those business people can sell. Aug 05 20:13:33 power/fame/money Aug 05 20:13:44 PaulFertser, that's another issue Aug 05 20:13:49 TAsn: nice sofa? i'd prefer sleeping on the floor! Aug 05 20:14:13 * mrmoku has actually seen PaulFertser sleeping on the floor :P Aug 05 20:14:17 TAsn: nice car? brand-new-fancy-uber-comfortable? What i am, a disabled? Aug 05 20:14:32 PaulFertser, I see good looks as a "feature" Aug 05 20:14:40 one can be smart Aug 05 20:14:46 nice Aug 05 20:14:58 or good looking Aug 05 20:15:08 I mean Aug 05 20:15:27 all of those (and more) are attributes that a human can posses Aug 05 20:15:34 I don't think it's fair to prefer one over the other Aug 05 20:15:40 mrmoku: sleeping on the ground (the other night) was also nice (nicer in fact). Aug 05 20:16:07 PaulFertser, sleeping in the car wasn't bad either :) Aug 05 20:17:09 mrmoku: we were going to sleep in a car with my friend the saturday night (for no apparent reason) but his wife called him and ordered to go home :D Aug 05 20:17:27 yeah wifes :P Aug 05 20:17:49 PaulFertser, anyhow, ppl should not be embarrassed by their good looks or "big brain" Aug 05 20:17:59 we as a society encourage perfection Aug 05 20:18:01 in every feild Aug 05 20:18:03 field Aug 05 20:18:08 PaulFertser, btw. I did a new build from scratch overnight Aug 05 20:18:16 which had the same missing packages... Aug 05 20:18:17 sports/mind/beauty/etc Aug 05 20:18:31 but an error in rootfs log that fso-abyss was missing Aug 05 20:18:35 mrmoku: can't help much here... Aug 05 20:18:43 mrmoku: probably tomorrow, now i'm too sleepy already. Aug 05 20:18:53 PaulFertser, don't want you to help now Aug 05 20:18:58 just want to keep you updated :) Aug 05 20:19:06 I then manually built fso-abyss Aug 05 20:19:08 TAsn: i don't see why look encouraged by model business can be considered a road to perfection. Aug 05 20:19:12 cleaned the image Aug 05 20:19:18 removed all task stamps Aug 05 20:19:22 and rebuilt the image Aug 05 20:19:26 which was complete then Aug 05 20:19:41 PaulFertser, by definition Aug 05 20:19:48 looking good > looking bad Aug 05 20:19:53 (as good > bad) Aug 05 20:20:32 the modeling business just helps keeping this in mind Aug 05 20:20:35 :) Aug 05 20:20:38 TAsn: "looking good" is highly subjective. Aug 05 20:21:01 PaulFertser, only on the middle section of the scale. Aug 05 20:21:09 as the ends are clear. Aug 05 20:21:23 TAsn: ever been to a modern art museum? Aug 05 20:22:01 TAsn: when i see a girl in the underground i first look at what book she reads. I rarely consider attractive the one reading sidney sheldon. Aug 05 20:22:55 I don't think there's a doubt that Bar Rafaeli is hot. Aug 05 20:22:55 PaulFertser, I'm off taking a (beauty) shower Aug 05 20:22:55 *wink* Aug 05 20:23:13 and now for real, I don't always agree with the general perception of beauty Aug 05 20:23:26 for instance I think Angelina Jolie doesn't look that good. Aug 05 20:23:34 TAsn: when it comes to a UI, there's definitely no objective standard of 'beauty'... Aug 05 20:23:44 TAsn: hot==you want to have sex with her? Aug 05 20:24:10 * wjbaird assumes that this convo started at UIs... if it's purely about attractiveness of females, I'll butt out... Aug 05 20:24:32 wjbaird, I have. Aug 05 20:24:32 PaulFertser, I agree. Aug 05 20:25:02 PaulFertser, physical attraction, yeah. Aug 05 20:25:04 wjbaird: it's about society, stereotypes, sexism etc. Aug 05 20:25:13 Heya Aug 05 20:25:22 PaulFertser, though I don't pick my gf's based on physical attraction only Aug 05 20:25:26 Ainulindale, /lastlog yourself Aug 05 20:25:27 :) Aug 05 20:25:39 PaulFertser: ah - ok... that's what I get from diving in after being off at a meeting for a while... Aug 05 20:25:54 TAsn: yeah saw it Aug 05 20:25:59 She's not really that attractive Aug 05 20:26:15 Ainulindale, bah. Aug 05 20:26:16 TAsn: and FYI Angelina Jolie is ugly Aug 05 20:26:46 As for Bar Refaeli I just think she's too skinny Aug 05 20:26:51 you suck. ;) Aug 05 20:26:51 (suck isn't a real swear!) Aug 05 20:27:31 Ainulindale, I agree! (about Jolie) Aug 05 20:27:33 Ainulindale, I like skinny girls :) Aug 05 20:27:54 TAsn: ever heard about a smart model? Esp. female? Aug 05 20:28:08 shower, cya. Aug 05 20:28:41 PaulFertser: well a robot model Aug 05 20:28:49 PaulFertser, no, but there's one reason for that Aug 05 20:28:55 education Aug 05 20:28:58 for some reason here Aug 05 20:29:11 PaulFertser, btw, Marlyn Monroe was intelligent. Aug 05 20:29:14 anyhow, for some reason here Aug 05 20:29:17 smart = ugly Aug 05 20:29:18 ;\ Aug 05 20:29:26 so Aug 05 20:29:34 so good looking girls try to look dumb Aug 05 20:29:41 anyhow Aug 05 20:29:46 SHOWER! Aug 05 20:30:32 hey All Aug 05 20:30:47 is there anyone working on location aware configuration changes? Aug 05 20:30:55 That proves my point that model business is harmful to society. Aug 05 20:31:01 as in: the phone operating differently based on where it is? Aug 05 20:32:22 PaulFertser: I don't know - the modeling business provides (slightly) gainful employment for the attractive but dumb men and women... Is it *that* harmful to society? Aug 05 20:33:38 wjbaird: imho yes. And one can be a good person even if (s)he's dumb. But it's much harder being a doll. Aug 05 20:35:48 PaulFertser: Unfortunately I missed the begining of this discussion - is your argument mainly around the unrealistic expectations that the whole modeling biz puts on the rest of society? Aug 05 20:37:37 intone sucks Aug 05 20:39:18 wjbaird: i somehow don't like the idea to use humans as dolls in general. I see many reasons why this is bad. Especially when business wants to make that a profession of a person and if it wants to convince everybody to "follow the trend". I think modelling is so bad that should be boycotted. Aug 05 20:44:00 PaulFertser: Hmm. I can see your point, and I think that there are huge problems with the modelling industry taking advantage of models and encouraging young women to have extremely unrealistic expectations about their bodies. but the fact is that clothes look very different on a moving 'real' human than they do on a mannequin. I don't think it's realistic to expect them to stop... Aug 05 20:44:41 wjbaird: i'm in conflict with reality anyway Aug 05 20:45:26 PaulFertser: :-) Aren't we all? Aug 05 20:45:57 wjbaird: And _i_ would appreciate if a girl designs and makes some cool/funny/creative clothes from rags herself rather than following some new fashion trend. Aug 05 20:46:49 PaulFertser: couldn't agree more - so much so in fact, that my wife has a certificate in fashion design. :-) Aug 05 20:49:06 wjbaird: if there was a way for the models to show that they're in fact true living individual creatures, probably i would consider it's ok to use them to present some artistically designed clothing. Aug 05 20:54:09 PaulFertser: I certainly agree that would be a big improvement... Aug 05 20:57:14 bringing things back on topic slightly :-) I'm just booting SHR for the first time... Is the OM wiki up-to-date about importing contacts? That the only way to do it is a script available via the gopher protocol? Aug 05 21:01:32 TAsn: COLD shower? :-P Aug 05 21:02:14 spaetz, haha duh. (after Bar's announcement there's no other option...) Aug 05 21:02:24 grarg_: we'd be happy if it works in some places already Aug 05 21:02:52 or better, we'd be happy if it worked independent of its location :) Aug 05 21:04:25 How long should a first boot of SHR take? I've been sitting at a mostly blank screen with a few messages like "Unknown boot option 'g_ether.host_addr....' ignoring" at the top of the screen for about 6 or 7 minutes now... Aug 05 21:05:32 wjbaird, I've had to kill and restart a first boot before Aug 05 21:05:41 unstable build I think Aug 05 21:05:58 spaetz, I can use it as a phone Aug 05 21:06:39 wjbaird: probably it's generating locales. Aug 05 21:06:42 grarg_: well - the screen just went blank. I'll give it another minute or two, and then try killing it... Aug 05 21:06:45 wjbaird: you know it's never fast. Aug 05 21:07:16 yeah, I think I gave it like 30mins to an hour before giving up Aug 05 21:07:45 PaulFertser: True - although I don't think any of the other distros I've tried took quite *this* long on first boot. 4 or 5 minutes wouldn't have surprised me at all... but I'm at the 10m mark now... Aug 05 21:08:18 wjbaird: can you ssh in? Aug 05 21:11:15 PaulFertser: when I pulled and re-plugged the usb cable, I did see a usb0 device created - but I get a no route to host after I configured the usb0 device and try to ssh... Aug 05 21:12:38 wjbaird: are you using an old bootloader? Aug 05 21:12:49 wjbaird: with new you're supposed to get eth? on host. Aug 05 21:13:28 PaulFertser: I'm using Qi. Not sure which version - it was installed by SDG systems when they buzz-fixed me... Aug 05 21:13:34 wjbaird: ah, you see g_ether.host_addr etc, so you're uptodate. I wonder where that usb0 can come from then. Probably g_ether module is not loaded properly on the fist boot. Aug 05 21:16:11 PaulFertser: I'm coming up on 20 mins since I booted... should I continue to let it run, or kick it in the head? Aug 05 21:17:06 wjbaird: kick Aug 05 21:18:31 PaulFertser: ok - boot attempt 2... I'm still seeing the Unknown boot option - g_ether messages... is that something to worry about? Aug 05 21:18:45 wjbaird: 20 min is definitely too long Aug 05 21:19:03 it takes something like 5 minutes max after a reflash for me Aug 05 21:19:08 wjbaird: nothing to worry. Means that g_ether is not built_in which is a good thing in case you want to use other gadget drivers. Aug 05 21:19:20 the unknown boot option warning is harmless Aug 05 21:19:21 wjbaird: those settings will be picked up by modprobe on module load. Aug 05 21:19:36 PaulFertser: ok - thanks. Aug 05 21:19:45 spaetz: yeah, 5m was about what I was expecting... Aug 05 21:20:19 first boot takes about a minute longer than subsequent boots for me Aug 05 21:20:35 ouch - 4 minute boots? that's kinda painful... Aug 05 21:23:18 yep, you don't reboot often, though Aug 05 21:24:39 spaetz: I hope not... but with most of the other distros I tried I needed to reboot every day or two if I wanted things to be stable... Hope I won't have to do that with SHR... Aug 05 21:25:02 btw - coming up on 7 minutes after the second boot... still sitting at a screen with the g_ether messages... Aug 05 21:26:20 still seeing a usb0 device created, but no route when I ssh... Aug 05 21:26:22 :-( Aug 05 21:28:08 Hum. tried rebooting from the u-boot in nor, and now I see the "SHR" splashscreen. Aug 05 21:28:16 does SHR not work with Qi? Aug 05 21:29:40 It works Aug 05 21:30:32 strange... The two boots with Qi went no where, but it looks like it's starting up fine from u-boot... Aug 05 21:31:05 yup - enlightenment just started, so I think I'm good... Aug 05 21:35:26 well, I must admit that the first impressions are a lot better than OM2009... :-) one question though - I was prompted to select the things I wanted on the 'quick launch' during the startup, but once I was done I got into a launcher that seemed to show everything, not just what I selected... Aug 05 21:36:12 Hehe, surprise Aug 05 21:38:03 hehe Aug 05 21:38:11 the quicklaunch is useless Aug 05 21:38:14 ignore it Aug 05 21:38:28 we should remove that Aug 05 21:38:40 the launcher shows whatever the .desktop files state Aug 05 21:40:18 also - I was given the choice between an "Illume" theme or an "Illume-SHR" theme, without any clear indication of the difference between the two... Aug 05 21:40:35 there is indication in the about part :-) Aug 05 21:40:47 and I think we should remove illume as well :-) Aug 05 21:40:53 Well, yeah, but "customized for SHR" doesn't tell me much... Aug 05 21:41:03 as far as I could see, that's the only difference in the description... Aug 05 21:41:15 I picked that one, but it was just a guess... Aug 05 21:41:26 I understand, and you're right anyway :-) Aug 05 21:41:38 and I like the idea of a language selector - but if there's only one option - should we bother showing it? Aug 05 21:41:40 :-) Aug 05 21:41:48 wjbaird: well you can have several Aug 05 21:41:51 if you install another one Aug 05 21:41:55 then remove your conf Aug 05 21:41:59 you'll be able to change it Aug 05 21:42:05 if you build with several languages Aug 05 21:42:08 you'll be able as well Aug 05 21:42:11 I see.. That makes sense... Aug 05 21:42:29 any suggestions on the Qi thing? Or should I just re-install a new u-boot? Aug 05 21:42:44 qi boots on µSD as a default Aug 05 21:43:00 is your SHR on µSD ? Aug 05 21:43:22 no - it's on the flash, but I don't have my uSD configured so Qi recognizes it... Aug 05 21:43:48 I previously had android in flash, and Qi skiped the uSD (I feel the vibration when it does), and booted flash... Aug 05 21:44:05 Well I have three phones Aug 05 21:44:11 I still feel the little vibration, but then just see a blank screen with the messages about g_ether at the top... Aug 05 21:44:14 A GTA02 with qi, another with uboot Aug 05 21:44:17 and a GTA01 with uboot Aug 05 21:44:24 all are able to properly boot into SHR Aug 05 21:44:33 So I'd be interested in logs if you have any Aug 05 21:44:41 as for g_ether Aug 05 21:44:53 Did you build your image yourself? Aug 05 21:44:58 Or did you take it from the buildhost? Aug 05 21:45:19 no - I grabbed shr-shr-image-glibc-ipk--20090703-om-gta02.rootfs.jffs2 from the downloads site. Aug 05 21:45:37 the g_ether warning appears always with Qi and is harmless Aug 05 21:45:48 yeah Aug 05 21:45:54 BUt this "no screen" issue sounds as no evas Aug 05 21:46:01 It should boot into illume Aug 05 21:46:04 After a while Aug 05 21:46:07 yep Aug 05 21:46:09 I think it means that Qi has successfully found a kernel and tried to boot it, right? Aug 05 21:46:13 If it doesn't after 5 minutes there's a problem that obvious Aug 05 21:46:16 +'s Aug 05 21:46:22 wjbaird: yep Aug 05 21:46:27 try to remove µSD Aug 05 21:46:30 I don't see any of the flashing buttons or whatever that Qi uses to flag a problem... Aug 05 21:46:30 and boot with qi Aug 05 21:46:32 just to check Aug 05 21:46:43 ok - i can try that... Aug 05 21:53:26 strange - it is certainly behaving differently without the uSD card... I'm seeing a lot more messages and it seems to be booting properly... But like I said, I was able to boot android off the flash fine with the uSD card in... Aug 05 21:57:54 multiple partitions on the flash? Aug 05 21:58:25 yeah, a small fat partition and a big ext2 (maybe ext3?) partition. Aug 05 21:58:50 well that's because you tried android and android is bad :-) Aug 05 21:58:50 not flashed the current kernel to the kernel partition of the internal flash? (i think you need to do that, right?) Aug 05 21:59:25 but more seriously I'm tired Aug 05 21:59:30 I'll go and sleep for a while Aug 05 21:59:37 G'night Aug 05 21:59:38 same here , good night Aug 05 21:59:43 Hmm. However, it *still* hasn't booted. been about 5 minutes. last message was "JFFS2 warning: (445) jfs2_do_read_inode_internal: Truncating #6257 to 1415960 bytes failed..." Aug 05 21:59:49 I have a 10 hours flight tomorrow Aug 05 21:59:57 I flashed both the kernel and root images... Aug 05 22:00:08 matched from the same date on the download server. Aug 05 22:00:16 spaetz: Ouch... good luck on the flights... Aug 05 22:00:38 anyways, I have to head home from work also... Aug 05 22:00:46 no clue then, good luck... and good night Aug 05 22:00:49 I'll stick with u-boot for now to play with SHR Aug 05 22:00:56 and sort it out later... Aug 06 01:15:43 hi Aug 06 01:16:57 good morning Aug 06 01:17:05 in some remote part of the world, probably Aug 06 01:17:30 :) **** ENDING LOGGING AT Thu Aug 06 02:59:57 2009