**** BEGIN LOGGING AT Tue Feb 01 02:59:57 2011 Feb 01 06:39:19 just installed power kernel - now unable to scan/see any access points that I used to connect to ok Feb 01 06:39:46 but mobile hotspot works ok! - how can I get back to normal wifi access? Feb 01 07:52:40 * ieatlint stares in astonishment at this single line of code: QPointer _layout = new QVBoxLayout; Feb 01 07:57:20 ieatlint: What about it? Feb 01 08:00:36 mooo Feb 01 08:01:30 RobbieThe1st: not saying it's not valid, but to me it's a wtf, why not just use QVBoxLayout*? Feb 01 08:02:05 Have you tried replacing it with that and see if it works the same? Feb 01 08:02:40 no, but i see the entire use of that var, and there's nothing else odd about it Feb 01 08:03:06 except that it's the first time the original author declared a layout (or a widget) on the heap instead of the stack :( Feb 01 08:03:45 odd Feb 01 08:34:26 Morning, all Feb 01 08:34:36 moo Jaffa Feb 01 08:36:21 ieatlint: sorry, that's way beyond what I understand about friggin C++ Feb 01 08:38:10 maybe it's about type of _layout var, to be a generic Qpointer rather that a particular QVBoxLayout ? Feb 01 08:38:51 err QVBoxLayout* Feb 01 08:40:02 or maybe the simple C style deref pointervar* semantics doesn't work for Qobjects and their Qpointers? Feb 01 08:42:14 I.E. QVBoxLayout* is a simple C pointer, not a QPointer ? Feb 01 08:42:39 meh, nevermind Feb 01 08:42:51 yeah, qvboxlayout* is just like the c equiv Feb 01 08:43:07 qpointer, in my semi-limited understanding, is equiv to gpointer Feb 01 08:43:53 "A guarded pointer, QPointer, behaves like a normal C++ pointer T *, except that it is automatically set to 0 when the referenced object is destroyed (unlike normal C++ pointers, which become "dangling pointers" in such cases). T must be a subclass of QObject" Feb 01 08:44:13 heh, ok, so there is potential for that form being useful Feb 01 08:44:15 so that's it :-D Feb 01 08:44:25 but it still makes no sense in this context :P Feb 01 08:46:10 to give an idea, the main() function for this builds 95% of the gui itself, with all widgets are on the heap (and qt follows a similar parent-child relationship as gtk, so that's a big no-no) Feb 01 09:00:48 ieatlint: I know what's heap and stack, but I miss to follow your hidden rationale about widgets living in either of those Feb 01 09:03:05 ah, well if you care, it a hierarchy.... eg, you have a window widget, so you place a box in it, and then a label in the box. then you destroy the window, so it destroys it's child (the box), which destroys its child, and so forth Feb 01 09:03:22 memory management Feb 01 09:03:51 fine, so what's the problem with heap here? Feb 01 09:05:12 nothing, heap is mandatory... if you put something from the stack on there, the parent will try to free it at some point Feb 01 09:05:18 and that's a segfault Feb 01 09:05:26 as long as each object's ~destructor is called correctly I don't see the problem Feb 01 09:05:43 bad programmers get away with it sometimes because simple guis don't destroy objects until you exit Feb 01 09:06:01 but if you've ever had a gui app (linux or windows) that crashes when you quit... Feb 01 09:06:07 oh yeah, obviously objects won't love to live on stack Feb 01 09:07:45 eh, it can be more dangerous than leaking a bit of memory Feb 01 09:10:20 sure, I guess ~destructor would try to give the (stack!) memory back to heap manager, and when this one would accept it then it occasionally gets reallocated or whatever - you mustn't do this Feb 01 09:11:20 that's why you create objects with new() (=heap), rather than instantiating a simple new var (=stack) Feb 01 09:11:20 doesn't ever accept it.. will always segfault Feb 01 09:12:20 yeah, and when you want to confuse new programmers, you use alloca() Feb 01 09:12:21 but pointers to objects can live on stack, except if they are objects themselves ;-P Feb 01 09:14:11 ooh, I ***love*** C++ Feb 01 09:14:14 i only recently learnt of alloca(), the way to make variably-sized arrays on the stack without breaking c89 rules Feb 01 09:14:24 Feb 01 09:15:13 heh, yeah, and i suck at c++ :( Feb 01 09:15:29 i'm on the stage defined as "oh, it's like c, but with classes!" Feb 01 09:15:47 same here Feb 01 09:17:25 that's why I found it rather funny when MohammadAG asked how to embed c into c++ - though it's maybe a valid question if only I understood enough of c++ Feb 01 09:17:45 it is, i do it Feb 01 09:18:33 for me a c++ code is 60..90% plain c, with a few special extensions Feb 01 09:18:54 so I think you don't need any special embedding Feb 01 09:18:57 to use c in c++, you do: extern "C" { see? ignorant here Feb 01 09:19:42 and it's common to then put that in a preprocessor of "#ifdef __cplusplus" Feb 01 09:19:55 * DocScrutinizer waves and heads off for breakfast out Feb 01 09:19:58 so that you can use the same code for regular c then without ever running into issue Feb 01 09:20:13 enjoy :) Feb 01 09:22:25 can someone tell me where does maemo store wifi passowrd ? Feb 01 09:22:53 gconf Feb 01 09:23:27 how to get it ? Feb 01 09:24:29 search on the forums/wiki for that enumerating a list of possible gconf settings Feb 01 09:24:38 mhm Feb 01 09:24:39 look around in /home/user/.gconf perhaps Feb 01 09:25:24 or maybe that's .config, hehe Feb 01 09:28:01 that reminds me, any of you guys noticed that under gconf, it has said that the warranty is labelled as invalid? Feb 01 09:44:57 how does one install meego with a non-standard kernel on the n900? Feb 01 09:45:47 Toadpole, you start by asking in #meego or #meego-arm, but in general the meego wiki pages explain differentiation and building images with kernel adaptions included Feb 01 09:46:46 the official team are currently looking at (afaik) the .37 kernel and putting the nice pvr/pm modules and stuff in Feb 01 09:47:17 great, thanks lcuk Feb 01 09:50:06 where can I go for help/information on a wifi problem with the power users kernel for the n900 ?? Feb 01 09:50:59 whats the problem? Feb 01 09:51:55 no managed mode wifi, but works ok with mobile-hotspot. flash old kernel back and can connect to AP again Feb 01 09:52:18 cant see what problem is Feb 01 10:03:25 Well. Moo. Feb 01 10:21:22 Ah, technology! http://qdb.us/305451 Feb 01 10:48:06 kerio: ping Feb 01 10:51:06 pong? Feb 01 10:52:06 kerio: you are good with Python, right? Feb 01 10:52:44 kinda Feb 01 10:52:45 why? Feb 01 10:52:57 I need some help regarding a Python thing Feb 01 10:54:18 go right ahead Feb 01 10:58:32 kerio: pm'd you Feb 01 11:16:29 soltys: http://talk.maemo.org/showthread.php?t=68063 [on popular demand] recover key/passphrase of WLAN Feb 01 11:37:13 DocScrutinizer: thanks Feb 01 11:37:33 click thanks on wiki ;-D Feb 01 11:42:40 hrm, anyone noticed the media player swapping left and right audio channels for no apparent reason? Feb 01 11:45:18 hard to tell. You need to check the whole chain, esp your headset etc Feb 01 11:46:11 attaching a weak 1.5V battery to tip and base of headphone jack will make the real left earpiece click Feb 01 11:47:20 then you need a media file (.wav, .mp3) that's definitely correct with test signals' assignment to channels Feb 01 11:49:06 haltdef: I've seen ~50% of headphones having wrong channel assignment Feb 01 11:49:46 that's the thing, it's not a permanent thing, it just decides to change them around on its own Feb 01 11:50:10 seems esp cheap chinese brands do not care at all. Often there's no labels or marks whatsoever for left and right, leaving it up to your guess what's the correct orientation Feb 01 11:50:52 decent pair of sony earphones, happens on every other earphones I've tried too Feb 01 11:51:03 haltdef: errm, what? You say mediaplayer is swapping channels for the very same .mp3, from one playback to next? Feb 01 11:51:42 different mp3s, different times Feb 01 11:52:20 I'd accuse the .mp3 to be inverse stereo Feb 01 11:53:07 even if it sounds normal most of the time but swaps over what seems to be randomly? only on the n900? Feb 01 11:53:29 why do you assume .mp3 to be correct and mediaplayer to do nonsense? Feb 01 11:53:39 because the mp3 is fine everywhere else Feb 01 11:53:57 I change track and channels are still swapped Feb 01 11:54:45 mediaplayer doesn't do anything to play music Feb 01 11:54:49 mafw's playing everything Feb 01 11:55:04 err, as long as you don't have a single known good test mp3 file to select for playback every time you suspect something fishy, I'd not follow up on this report Feb 01 11:56:11 I hear you, will grab a left/right test mp3 from somewhere Feb 01 11:56:41 please ping me with your results and findings Feb 01 11:57:35 I'd not deny possibility of a bug in PA or even ALSA, maybe regarding hw codec/mixer setup or sth Feb 01 11:58:00 noticed the n900 making my music sound different a while ago, only hit me last week the stereo channels were getting swapped :P Feb 01 11:58:04 there are several places where you could swap channels Feb 01 11:58:11 mm Feb 01 11:58:29 hi there Feb 01 11:58:45 is it possible to port cmake 2.8 to maemo ? Feb 01 12:00:54 ~wtf mm Feb 01 12:00:55 Gee... I don't know what mm means... Feb 01 12:01:00 ~mm Feb 01 12:01:01 Shared Memory Library. URL: http://www.engelschall.com/sw/mm/ Feb 01 12:02:29 got a wav on n900 with a voice saying "left right" over and over, normal now, just have to wait until weirdness happens and try then :P Feb 01 12:09:55 X-Fade: time has come to kick mscp into gear Feb 01 12:10:00 achipa: Ok here then ;) Feb 01 12:10:34 achipa: So do I need to push some packages through the repo chain? Feb 01 12:11:14 yes Feb 01 12:11:32 so, in order: block promotions of stuff that depends on libqt4-bearer-hotfix, qtquickcompat Feb 01 12:11:50 and then push those two to extras Feb 01 12:12:02 then we test mscp, and if it's okay, we push that too Feb 01 12:12:13 at least that's the plan :) Feb 01 12:12:49 achipa: What is the reasoning behind block them first? Feb 01 12:13:40 just to avoid a race condition, but if you think you can do the promotion reasonably soonish, then the order is irrelevant Feb 01 12:14:02 the point being that we should not have apps directly dependent on them in testing or extras Feb 01 12:14:04 I can script it. So it could be fast. Feb 01 12:15:16 achipa: Can you provide a list of exact package names and versions that need to get pushed? Feb 01 12:16:34 X-Fade: qtquickcompat 0.1.0 and libqt4-bearer-hotfix 4.7.0~git20101112-0maemo1 Feb 01 12:17:16 of course, these might get updated in which case I'll poke you again :) Feb 01 12:17:43 achipa: Sure. Feb 01 12:18:21 (the important part of these updates being promoted in a synced manner, not by automatic dependency pulling - hence the need for a block) Feb 01 12:18:36 does anyone know why with maemo and using cmake, sometimes no .so files are generated? Feb 01 12:19:07 achipa: yeah, but if I push them then they will be there instantly. Feb 01 12:19:15 exactly Feb 01 12:19:44 also, bergie said something about poking you if I need some tweaks for the REST interface part Feb 01 12:19:55 you up for a few mods ? :) Feb 01 12:20:24 nothing big, just want to avoid multiple queries in kisstester that make it slow Feb 01 12:22:16 * alterego wonders why every one and their mother has to create a new thread every time there is a new phone announced (that's not even Nokia ffs) Feb 01 12:22:28 achipa: Sure, can you file a bug with those so I don't forget? Feb 01 12:22:57 X-Fade: the qthello wart still occurs Feb 01 12:23:07 X-Fade: "size mismatch" for v1.1.1 Feb 01 12:23:12 I think Nokia knew that t.m.o was going to become a crap hole and decided, in MeeGo, to make it much more clear in regards to developer and user seperation. Feb 01 12:23:26 X-Fade: aye Feb 01 12:23:40 lunch brb Feb 01 12:23:44 You are saying tmo is a craphole. I am saying it is a living community. Feb 01 12:24:31 I may be being a bit harsh, I'm not in the best of moods. Feb 01 12:24:35 And if Nokia does not like communities consisting of real, chatting, conjecturing, bullshitting, etc people, it is their problem Feb 01 12:24:45 Heh Feb 01 12:24:58 Yes, I never really thought of it like that ;) Feb 01 12:25:24 And yes I understand that from the management point of view it is way better to have a community made of corporate employees told to go and create a community Feb 01 12:25:29 It would certainly be easier to follow though if it was just the 10 active maemo app developers :D Feb 01 12:25:44 /j #kinect Feb 01 12:27:32 RST38h: There's a culture on TMO though of "ooh, shiny thing" without being sufficiently worried about what it'll do to their primary phone. Feb 01 12:27:50 RST38h: And then said user moans about newbie complexity when somehting else breaks. Feb 01 12:27:57 RST38h: And, like typical users, they don't read the docs. Feb 01 12:28:04 * Jaffa is also perhaps not in the best of moods ;-) Feb 01 12:28:22 * lcuk slides bacon buttie over Feb 01 12:28:40 Ta. Much better. Feb 01 12:28:40 Ooh, I'm actually eating a fish finger sandwich Feb 01 12:28:46 Aww, unfair. Feb 01 12:29:03 * Jaffa had a Swedish sausage, topped with cheese and stuff, with gravy and mash potato. Feb 01 12:29:09 But that was two hours ago :-( Feb 01 12:29:21 haltdef: speaker-test http://paste.debian.net/106207/ alas probably not available and also maybe not even applicable for maemo armel & PA Feb 01 12:30:09 Heh Feb 01 12:30:37 Jaffa: Yes,I am well aware of the notion that 90% of tmo posters aremindless vermin Feb 01 12:30:44 I am bit too hungover for life today Feb 01 12:30:49 Jaffa: Also aware of the Sturgeon's law Feb 01 12:31:38 Jaffa: Still, you want a community made of live people? Here is your community. Want to segregate it by merit a little bit? No problem, look at devianart for how it can be done, for example. Feb 01 12:31:44 * DocScrutinizer passes alterego a glass of vodka with lemon Feb 01 12:32:01 Urgh, no thanks Feb 01 12:32:08 Too much vodka last night Feb 01 12:32:46 vodka + guinness + sambucca Feb 01 12:32:58 gj Feb 01 12:33:31 RST38h: I don't suggest that 90% of tmo posters are mindless vermin, FWIW. Feb 01 12:33:32 Jaffa: And if you just want all of them to shut up, you have got meego.com :) Feb 01 12:33:35 RST38h: Maybe 90% of posts ;-) Feb 01 12:33:54 Heh Feb 01 12:34:05 Jaffa: Nice, quiet, positive, business like, and you can fire people for talkingtoo much! Feb 01 12:36:26 alterego: trust me - I know how to deal with that :-D Feb 01 12:36:59 * DocScrutinizer trows an Aspirin into that vodka lemon Feb 01 12:37:05 Hahah Feb 01 12:37:11 Nice :P Feb 01 12:37:54 #meego-handset Feb 01 12:38:03 Erm ... whoops Feb 01 12:42:02 MohammadAG: heya, poked people so the marge request should be in Feb 01 12:44:02 marge request: http://cache2.allpostersimages.com/p/LRG/8/830/F52Y000Z/poster/the-simpsons-marge.jpg Feb 01 12:44:14 ;-D Feb 01 13:07:29 hmmpf, bugs in alsaped.conf - wonder whether and where to report Feb 01 13:08:38 grep tvut /usr/share/policy/etc/rx51/alsaped.conf Feb 01 13:21:40 ok one of the fun things to type into youtube is "Feynman" Feb 01 13:22:05 grep -n 'ihfandheadset = @outband_cancellation@' /usr/share/policy/etc/rx51/alsaped.conf Feb 01 13:22:31 hey Feynman, my alter ego Feb 01 13:23:30 :) Feb 01 13:23:38 Adventures of a Curious Character Feb 01 13:35:23 A friend of mine donnated me the book "you must be joking, Mr Feynman". He told me the first few chapters he thought the book is about me :-) Feb 01 13:40:39 DocScrutinizer: he must be joking, Mr DocScrutinizer. Feb 01 13:44:43 achipa, cool :D Feb 01 13:46:00 achipa: packages pushed. Feb 01 13:47:47 Heh, I'm not even planning to promote Columbus but it's already gor 9/10 votes :D Feb 01 13:48:30 MohammadAG: hi , the package of hildon-desktop with the desaturize instead of blur get keybinding broke Feb 01 13:48:38 or maybe i miss something Feb 01 13:49:54 i mean , now i can't use ctrl+back for taskmanager call , or just beging type for contact search Feb 01 13:53:33 E0x, gconf-schemas --register hildon-desktop-keys.schemas Feb 01 13:53:46 run that as root and reboot/killall hildon-desktop Feb 01 13:56:05 MohammadAG: cool , thx work just fine.. Feb 01 13:56:22 gconf-schemas --register - WAAAAAH a new g* spell Feb 01 13:56:39 ~nuke g* Feb 01 13:56:40 * infobot prepares 100 missle silos, and targets them at g* ... B☢☢M! Feb 01 14:18:23 lol DocScrutinizer Feb 01 14:43:57 how bright are compilers? Feb 01 14:44:10 lardman, hmm Feb 01 14:44:25 lardman: 42 Feb 01 14:44:50 e.g. is it better to declare my inner for loop iterator inside the for loops or declare once at the start of the fn Feb 01 14:44:56 and then add 42 of course Feb 01 14:46:21 usually a good compiler should optimize out such things, unless explicitly instructed not to do Feb 01 14:46:43 ok, well we'll see how good MATLAB's lcc compiler is then Feb 01 14:46:48 gcc -O11 -ffast-math -funroll-all-loops -fblazing-balls-of-holy-awesomeness Feb 01 14:47:00 Windows I'm afraid Feb 01 14:47:06 though I could install gcc I guess Feb 01 14:47:54 Look at the generated code. Feb 01 14:48:04 always a good idea Feb 01 14:48:14 yep Feb 01 14:48:26 or simply write sensible code Feb 01 14:48:38 well that's why I was asking Feb 01 14:49:11 basicaly there's for sure no *advantage* to do things that need to be done once, inside a for loop Feb 01 14:49:36 except if it makes it easier for the compiler to optimise it away Feb 01 14:49:43 that was my only thought Feb 01 14:50:33 so how could that open opportunities to optimize - except moving that crap outside of the loop where it belongs nayway? Feb 01 14:50:53 two loops, one inside the other Feb 01 14:51:59 define "declare inner for loop iterator" Feb 01 14:52:47 for(x=x_start;x so whether to declare y within the first loop, in the hope of making the compiler's life easier Feb 01 14:53:16 as it will remain a register variable anyway Feb 01 14:53:47 Is there any reason you're doing ++y? Feb 01 14:53:58 like in for(int x=x_start;x no, typo Feb 01 14:55:26 you'll probably need to compare assembler output of both versions Feb 01 14:56:18 ok, as I feared Feb 01 14:56:27 but not to worry, cycles are cheap Feb 01 14:56:46 Or bench Feb 01 14:57:00 bench = PITA Feb 01 14:57:15 asm listing FTW Feb 01 14:57:30 The difference in time will be miniscule, I was just wondering as I move this code from MATLAB to C Feb 01 14:58:10 lardman: You do not want to free yourself from MATLAB runtime? =) Feb 01 14:58:34 MATLAB is good for visualisation, so this is in a mex file Feb 01 15:00:54 Do several implementations. Pick which one wins at first run. Feb 01 15:01:13 This way you have all sorts of fun hardware dependant bugs, depending on random stuff. Feb 01 15:01:23 as the code takes ~8min to run and this is just setup code, I think it probably won't matter too much Feb 01 15:05:28 * lardman curses the fact MATLAB uses () for array elements Feb 01 15:05:55 <`Ika_Musume> uhm guys yesterday i was asking howto format my sd rite? but how come my phone name changes to this http://img828.imageshack.us/img828/2404/screenshot2011020116043.png Feb 01 15:06:04 <`Ika_Musume> how can i change that Feb 01 15:06:47 hmm Feb 01 15:06:48 Funky Feb 01 15:07:27 well you'll never confuse it with a humble flash drive when you've got it attached to your desktop Feb 01 15:07:50 <`Ika_Musume> huh? Feb 01 15:09:43 <`Ika_Musume> anyaways can anyone help me change it? Feb 01 15:10:30 Mount it on a desktop PC and change the name there? Feb 01 15:10:45 dosfslabel on the n900 should do it Feb 01 15:10:51 <`Ika_Musume> well there its called n900 Feb 01 15:11:41 dosfslabel /dev/mmcblk0p1 fish Feb 01 15:11:54 Will rename the thing fish Feb 01 15:13:05 fishy Feb 01 15:13:24 <`Ika_Musume> lol Feb 01 15:13:52 <`Ika_Musume> it stay the same Feb 01 15:14:07 you wil need to unplug and replug it Feb 01 15:14:32 maybe even need to reboot it Feb 01 15:14:38 <`Ika_Musume> how its the main memory of n900 Feb 01 15:14:47 though I doubt Feb 01 15:14:55 <`Ika_Musume> reboot i can do Feb 01 15:15:20 The "main memory" is the partition you just tried to rename, i.e. ~/MyDocs Feb 01 15:16:05 <`Ika_Musume> well its the 32 internal memory Feb 01 15:16:10 <`Ika_Musume> not an sd Feb 01 15:16:17 umm, no. That's in HFM Feb 01 15:16:41 heh, maemo 5 sdk installer isn't working very well. it opens in text editor if opened by default. i set it as executable and running it does nothing Feb 01 15:16:44 <`Ika_Musume> if you say so Feb 01 15:17:00 HFM transcodes whatever is in partition label, it's using an own definition of that Feb 01 15:17:18 dunno what it's based on, maybe /etc/hostname Feb 01 15:17:24 "run in terminal" asks "Do you want to install missing package "python-qt4"? (Type "y" to install) "..and selecting y closes the installer Feb 01 15:18:01 cos^: running as root? Feb 01 15:18:05 there are other config files as well, for that Feb 01 15:18:08 lardman: of course not Feb 01 15:18:21 cos^: well iirc you have to run it with sudo.... Feb 01 15:18:55 lardman: ok.. now it asks to run as root after i installed python-qt4 manually Feb 01 15:19:07 not very great user experience Feb 01 15:19:12 they should have put their check earlier Feb 01 15:19:38 naturally it doesn't offer to escalate the privileges, so it has to be run from console anyway Feb 01 15:20:05 why bother write a install wizard, if it needs to be run from console Feb 01 15:25:15 `Ika_Musume: I suggest a ``grep -r BCM2048 /etc /home/user'' Feb 01 15:26:03 does the partition need to be umounted before that dos-namey thing will work? Feb 01 15:26:06 <`Ika_Musume> uhh Feb 01 15:26:16 <`Ika_Musume> ill try Feb 01 15:26:54 * lardman wonders how well that copy&replace MATLAB->C conversion just worked.... Feb 01 15:30:58 * DocScrutinizer wonders how `Ika_Musume managed to get such a weird name in HFM Feb 01 15:33:38 <`Ika_Musume> idk Feb 01 15:35:38 <`Ika_Musume> btw nothing happends Feb 01 15:37:07 hmm, home/user probably is useless anyway Feb 01 15:37:20 to grep there, that is Feb 01 15:38:22 <`Ika_Musume> ohwell Feb 01 15:40:55 ^C Feb 01 15:40:59 ;-) Feb 01 15:41:25 <`Ika_Musume> its only in the phone , mass memory is called n900 Feb 01 15:41:28 <`Ika_Musume> on pc Feb 01 15:42:06 if only I could recall where I've seen this friggin naming definitions for HFM Feb 01 15:42:20 Somewhere under settings iirc one can change the device name Feb 01 15:42:31 e.g. Bluetooth Feb 01 15:42:47 yeah, but only bluetooth Feb 01 15:42:56 perhaps it's used for more than that? Feb 01 15:45:01 <`Ika_Musume> lol it is Feb 01 15:45:09 lardman: nah Feb 01 15:45:12 <`Ika_Musume> so bluetooth name eh Feb 01 15:45:28 lardman: the actual name is also changeable via some file in /etcthough Feb 01 15:45:35 yep Feb 01 15:45:51 <`Ika_Musume> i changed bt and it worked Feb 01 15:46:07 meh, so why didn't the grep work then? Feb 01 15:46:15 <`Ika_Musume> dunno Feb 01 15:46:38 and your device will be fishy when you next use mass storage :) Feb 01 15:46:39 <`Ika_Musume> maby not the right plugins? Feb 01 15:47:04 <`Ika_Musume> well im happy that its back to its old name Feb 01 16:09:32 Heh...Apple barred Sony's eReader app from its app store. Feb 01 16:18:20 hi there Feb 01 16:18:26 can you tell me how to install strace? Feb 01 16:18:48 I used apt-get install strace but I get: tpackage strace is not available, but is referred to by another package. Feb 01 16:19:48 tools repository Feb 01 16:20:11 http://wiki.maemo.org/Documentation/devtools/maemo5 Feb 01 16:20:17 uhm Feb 01 16:21:12 package seems strace.. Feb 01 16:22:23 strace is there, once you enable the repo Feb 01 16:22:40 I enabled Feb 01 16:23:36 my sources.list contains: # fremantle tools Feb 01 16:23:36 deb http://repository.maemo.org fremantle/tools free non-free Feb 01 16:23:36 # fremantle tools sources Feb 01 16:23:36 deb-src http://repository.maemo.org fremantle/tools free non-free Feb 01 16:29:59 Indian students duped by Tri-Valley University in California have been fitted with GPS devices by US immigration authorities. Feb 01 16:36:57 t900:~# apt-cache policy strace Feb 01 16:36:59 strace: Feb 01 16:37:00 Installed: 4.5.18-1.1+0m5 Feb 01 16:37:02 *** 4.5.18-1.1+0m5 0 Feb 01 16:37:04 500 http://repository.maemo.org fremantle/tools/free Packages Feb 01 17:40:31 RST38h, silly american policy Feb 01 18:21:08 so, anyone gona port trem to the n900? Feb 01 18:21:18 trem? Feb 01 18:21:27 tremulous! Feb 01 18:23:22 has anyone else noticed the massive bug with kernel-config script yet? Feb 01 18:23:25 vi_, is the current trem ioquake3 based or did they move to xreal? Feb 01 18:23:37 it is ioquake Feb 01 18:23:53 but debian armel does not use opengles Feb 01 18:24:08 so there is no hardw. accel. Feb 01 18:24:23 making 0.25 fps Feb 01 18:24:57 you follow? Feb 01 18:25:02 yep Feb 01 18:25:11 so how do i make it use opengles? Feb 01 18:25:21 a magic symlink or two? Feb 01 18:25:27 rebuild from sratch I fear Feb 01 18:25:38 or does the code need to be altered? Feb 01 18:25:46 lame Feb 01 18:26:08 afaik there are asm parts in the x86 code wich you have to convert Feb 01 18:26:36 also you'd have to get around opt / rootfs limitations Feb 01 18:26:36 there is already an arm port of trem thatruns Feb 01 18:26:57 my rootfs is 10GB XD Feb 01 18:27:08 well common user rootfs is smallish :) Feb 01 18:27:38 but I think with the right tweaks in the fs code of the engine you could have all the data files in MyDocs Feb 01 18:28:02 there is an arm port of trem that runs but without HW accel. Feb 01 18:28:15 all trem data is stored in home directory Feb 01 18:28:40 only 400k binary is in rootfs space Feb 01 18:29:07 (excuse my tersness, i am on the n900) Feb 01 18:29:40 is the data in /home/user or in /home/user/MyDocs Feb 01 18:29:54 because /home/user would still be limited to 2 gb Feb 01 18:30:02 anyway i have put a 50$ bounty on maemo talk for the first dev who can get me tremming. Feb 01 18:30:09 hrhr Feb 01 18:30:15 I think I have something to hack on :) Feb 01 18:30:22 the data will be in the users home directory Feb 01 18:30:40 it is only 100mb anyway Feb 01 18:30:41 talk link? Feb 01 18:31:23 uh, i am on irssi on n900. cant paste links so easy. just search maemo talk for the work bounty Feb 01 18:31:41 ^word Feb 01 18:32:12 xchat > irssi ;) Feb 01 18:33:02 irssi > * Feb 01 18:33:09 vi_, why you cant paste links easy? Feb 01 18:33:13 Just paste o.o Feb 01 18:33:41 click on the "X Terminal" and then paste o.o Feb 01 18:33:58 Not so hard Feb 01 18:34:02 fellu my main concern wit irssi on the n900 is, the time I tried it I couldn't get it to switch channel tab :D Feb 01 18:34:21 :D Feb 01 18:34:29 there is esc for that Feb 01 18:34:31 on the touch screen Feb 01 18:34:36 esc + 1 Feb 01 18:34:39 do same the alt + 1 Feb 01 18:34:48 esc is like alt in irssi Feb 01 18:35:39 hard to join to all channels again and again with different client I use at home .P Feb 01 18:40:53 whadya think bro? Feb 01 18:43:07 vi_: that your spellchecker is broken Feb 01 18:43:42 bleh Feb 01 18:44:01 well the offer is legit anyways Feb 01 18:44:26 well I'll give it a try anyway Feb 01 18:44:39 time I finally read that opengles book I own :) Feb 01 18:45:10 :) Feb 01 18:48:17 a Feb 01 18:54:29 i wish there was an email solution that didnt totally suck balls Feb 01 18:56:05 i wish there was an email solution that didnt totally suck balls Feb 01 18:57:38 anyone know of anything that can take assembly language (i.e. objdump output) and output pseudo-C ? Feb 01 18:58:02 lol Feb 01 18:58:47 what you are talking about is the mad dream of a feverish reverse engineer(ist) Feb 01 18:59:10 it's quite doable, just wondering if someone's actually bothered to do it Feb 01 18:59:29 actually, it isnt Feb 01 18:59:40 what architecture? Feb 01 18:59:41 rally? why not? Feb 01 18:59:42 int d1,d2,d3,d4; Feb 01 18:59:44 really even Feb 01 18:59:49 int *a1,a2,a3,a4; Feb 01 18:59:50 ;) Feb 01 18:59:59 * lcuk should add more *s Feb 01 19:00:06 lcuk: exactly, was just how the opo revtran output used to look :) Feb 01 19:00:18 * lcuk shudders Feb 01 19:00:41 mind you, i have written similar in visual basic Feb 01 19:00:43 * lcuk hides Feb 01 19:00:51 well the best you can get is ida pro (if you can convince them to sell it to you) Feb 01 19:01:08 yes, and you need the add on thingy afaik Feb 01 19:01:12 then lateraly ollie dbg, but for x86 only Feb 01 19:01:18 aqualung used to work, but not now Feb 01 19:02:00 well if ya already know why are you asking? Feb 01 19:02:02 is /sys/class/sec/ Android only, or do we see that under normal Linux? Feb 01 19:02:14 vi_: I'd like something that works better Feb 01 19:02:34 and without ida Feb 01 19:02:49 compilation isnt like a one-way-hash its more of a rubiks cube Feb 01 19:03:05 there are many ways to arrive at the final construct Feb 01 19:03:28 I know that, I don't expect to run it, tweak one line and be able to re-compile Feb 01 19:03:50 but it is tiring looking at assembly all the time Feb 01 19:03:52 well if there is a better prog then i dont know it Feb 01 19:03:57 yup Feb 01 19:04:13 get used to it bro Feb 01 19:04:28 I wish I'd not left my laptop at work, sounds like a decent little project while my wife's watching silent witness Feb 01 19:04:43 uk huh! Feb 01 19:04:49 yep Feb 01 19:04:53 woot Feb 01 19:05:04 what are you trying to reverse? Feb 01 19:05:15