**** BEGIN LOGGING AT Tue Feb 28 10:59:56 2006 Feb 28 11:00:19 pb__: was your train on time ? Feb 28 11:00:33 lrg: does Wolfson have any DACs with tone control built in? One of our customers is looking for a way to do some kind of basic EQ on their audio output. Feb 28 11:00:41 lrg: yup, all went very smoothly Feb 28 11:02:37 pb__: oh, good question. I'll need to ask our sales dept. I'll get back to you Feb 28 11:02:43 pb__: I pointed the nokia people to your integer pixbuf patch, so lets hope they update it and push it to mainline Feb 28 11:03:14 lrg: the pops my ipod produces aren't 'small' ;) Feb 28 11:03:30 lrg: thanks Feb 28 11:03:32 koen: rock Feb 28 11:03:41 lrg: although I suspect my amp+woofer might be sensitive in that region Feb 28 11:04:37 lrg: hi, Feb 28 11:04:51 lrg: success with fixing that ugly bug? Feb 28 11:05:10 koen: imho you amp is probably amplifying the pop to be quite loud Feb 28 11:05:20 greentux: hi, yes. patch sent to RP Feb 28 11:05:36 koen: is it loud with hp only ? Feb 28 11:06:41 lrg: no, with hp it's just very noticable Feb 28 11:07:55 lrg: fine, will ask rp... Feb 28 11:08:14 lrg: I also discovered that it depends on the dock which output you get... Feb 28 11:08:54 the nano dock uses something without volume control, the universal one uses an output with volume control Feb 28 11:09:06 koen: i'm afraid it's something atm thats damn near impossible to get rid off. new codecs are making this less noticable :) Feb 28 11:09:12 :) Feb 28 11:09:46 lrg: the only way to stop it is to fully power up the device, and thats not really good for partable devices Feb 28 11:10:19 true Feb 28 11:12:27 build a kernel + root fs for a freescale powerpc platform - runs nicely :) Feb 28 11:13:01 the kernel staging dir needs some additional header files though, submitted a patch for it. Feb 28 11:13:48 _law_: how goes keylaunch things? Feb 28 11:21:39 hi zecke Feb 28 11:23:28 hey Feb 28 11:23:40 mickey|zzZZzz: I turned in a crappy draft of this paper Feb 28 11:23:50 mickey|zzZZzz: at least I have \usepackage{pdfdraftcopy} in it Feb 28 11:24:01 hey zecke Feb 28 11:24:26 <[lala]> hi zecke Feb 28 11:25:01 <[lala]> zecke: what would i have to do to get any directory visible in the documents tab in opie? Feb 28 11:25:24 [lala]: change the code :) Feb 28 11:25:35 [lala]: the doctab would not scale well though Feb 28 11:26:25 <[lala]> zecke: well i would do with just *some* directory. say i would want to loop mount something and want that to appear in the doctab Feb 28 11:27:41 [lala]: okay this involves still changing code Feb 28 11:28:07 <[lala]> zecke: i noticed that after insertion of a cf card a dialog is popping up Feb 28 11:28:09 [lala]: the device needs to be recognized as 'removable' Feb 28 11:28:21 <[lala]> hmm Feb 28 11:28:26 <[lala]> how would i get that? Feb 28 11:28:32 [lala]: QPE has a class to identify FileSystems and removables... Feb 28 11:29:06 <[lala]> and which count as removeable in that scenario? Feb 28 11:29:26 [lala]: http://handhelds.org/cgi-bin/cvsweb.cgi/opie/library/storage.cpp?rev=1.25&content-type=text/x-cvsweb-markup Feb 28 11:30:26 [lala]: feel free to merge the code from Qtopia. So that one only needs to edit a configuration file in the future Feb 28 11:31:21 <[lala]> i see Feb 28 11:32:29 <[lala]> and what about files in my regular home directory on the zaurus (/home/root for instance)? those files are not meant to appear either? Feb 28 11:33:04 [lala]: they are not Feb 28 11:33:14 [lala]: Documents appear Feb 28 11:33:24 and as this is a Documents Tab this sounds about right Feb 28 11:33:52 [lala]: you might be interested in implementing the New Document Model found in the wiki Feb 28 11:33:59 it would kick ass to have that implemented Feb 28 11:34:11 <[lala]> :) Feb 28 11:36:27 greentux: I'll probably sort lrg's patch out this afternoon. I have a couple of things to deal with first Feb 28 11:37:10 [lala]: You could also watch for mount events through udev - the automount script could be set to notify something whenever anything is mounted... Feb 28 11:39:42 <[lala]> RP: that sounds as an interesting idea Feb 28 11:40:25 <[lala]> RP: are those sound profiles for headset, headphone and speaker supposed to be included in the zaurusd? Feb 28 11:49:38 03koen 07org.oe.dev * reda55732... 10/packages/linux/ (ep93xx-kernel/derevo6.diff ep93xx-kernel_2.6.15.bb): Feb 28 11:49:38 ep93xx kernel: update to derevo6 Feb 28 11:49:38 * usb host works(ish) Feb 28 11:49:38 * serial on the glomation board requires an extra patch Feb 28 11:51:36 yay only a single head in the db :) Feb 28 11:53:56 ~seen mjg59 Feb 28 11:54:00 mjg59 was last seen on IRC in channel #gpe, 319d 32m 52s ago, saying: 'Ah, the joypad works'. Feb 28 11:54:50 [lala]: zaurusd knows how to adjust the mixer in each of those cases Feb 28 11:57:14 RP: any plans expirement with HAL for that kind of stuff? Feb 28 11:57:25 pb_: have a look at http://wolfsonmicro.com/products/digital_audio/dacs/. The 8751 should fit your requirements, although others from the list may better fit your exact requirements, depending on power, rate , snr etc. You could also email apps@wolfsonmicro.com with your exact requirements and an Apps engineer will be able to better advise than me. Feb 28 11:57:53 pb_: hth Feb 28 12:00:18 koen: Not sure. I dislike the size and weight of hal. I suspect we could get the same functionality with some simpler udev -> dbus gateway Feb 28 12:01:01 koen: I have a plan at the back of my mind to implment such a thing in zaurusd. I'd make sure the events were compatible with hal's data format etc. Feb 28 12:01:32 RP: I agree, hal is probably too heavy (I remember the speaker saying < 64MB was not good) Feb 28 12:02:06 the problem with hal is that is has files for every hardware device available Feb 28 12:02:11 lrg: There are a lot of people who disagree with that though (my boss included) ;-) Feb 28 12:02:20 removing 80% off the .fdi files might help Feb 28 12:02:32 koen: Yes, and no structure to allow it to me slimmed down Feb 28 12:02:38 s/me/be/ Feb 28 12:02:50 RP: yeah, mallum convinced me to test hal + gnome-power-manager on the hx Feb 28 12:03:23 hal did pretty ok Feb 28 12:03:29 i.e. no OOMing ;) Feb 28 12:04:02 http://pastebin.com/576477 - why "mt merge" sucks badly Feb 28 12:05:03 I also had this conversation with Nokia. Since they've done a lot of running of desktop software on handhelds, they don't see why hal should be any different... Feb 28 12:05:03 heh Feb 28 12:05:51 hrw: Mad :) Feb 28 12:09:14 ~lart glibmm Feb 28 12:09:14 * ibot strangles glibmm with a 9-pole serial cable Feb 28 12:10:03 I also heard that nokia is going softvfp Feb 28 12:11:06 RP: mad like mad.. suxx Feb 28 12:11:48 koen: Interesting. Switching causes some issues for them so it'll be interesting to see if they do Feb 28 12:12:24 I did prove which was faster to them... Feb 28 12:12:43 RP: they have also tested EABI stuff Feb 28 12:12:51 so I wonder what they end up doing Feb 28 12:12:58 koen: How did they find it? Feb 28 12:13:37 all I heard was "it boots and X works" Feb 28 12:14:28 I had planned to discuss it with nchip on fosdem, but I had to leave early Feb 28 12:14:31 interesting. EABI was missing from my tests but that was a while ago... Feb 28 12:15:38 eabi vs softvfp shouldn't make to much difference Feb 28 12:15:42 (none actually) Feb 28 12:16:23 koen: but if you're going to switch, better just go straight for EABI... Feb 28 12:16:31 indeed Feb 28 12:16:49 but eabi requires either the infamous csl branch or glibc+gcc head Feb 28 12:17:06 they tested with the csl branch last I heard Feb 28 12:18:36 both options are kind of icky from a management POV Feb 28 12:18:54 yes. icky from lost of POVs :-/ Feb 28 12:19:02 s/lost/lots/ Feb 28 12:19:21 OTOH they do use scratchbox ;) Feb 28 12:19:25 which is also icky Feb 28 12:19:58 yes. very icky... Feb 28 12:20:08 qOEmu is much better ;) Feb 28 12:20:49 In case anyone hasn't noticed, I commited some nice changes to OE yesterday. We can now run arm OE images under qemu system emulation ;-) Feb 28 12:21:37 I want oz 3.5.4 shipped.... then one month free of oz work (maybe will run .dev on Z) Feb 28 12:22:10 hrw: I can understand that. How close is it to shipping now? Feb 28 12:22:54 RP: hm... keylaunch work by _law_ need to be finished, merged, pushed. Feb 28 12:23:11 RP: I piss on apm problem - some users say it works, some say it does not Feb 28 12:23:35 We should think about having a way to run the emuator. Have we any ideas on how to select which kernel and initrd to use? I guess we could stash a script somewhere but I'd like a neater solution... Feb 28 12:24:38 hrw: Quite an apt typo :) (at least I think you meant pass but you might not :) ) Feb 28 12:25:02 piss is what I meant Feb 28 12:25:04 RP: when it has a network or block driver we could finally run stuff like localegen offline Feb 28 12:25:16 hrw: fair enough :) Feb 28 12:25:29 koen: It has a network driver Feb 28 12:25:35 morning Feb 28 12:25:41 * hrw want to declare poodle and tosa as UNMAINTAINED machines Feb 28 12:26:14 hrw: Give up on the apm problem... Feb 28 12:26:23 koen: why cant localgen be run scratchbox stylee? Feb 28 12:26:36 XorA: hacky Feb 28 12:27:02 RP: not really, if OE makes qemu-native as a tool Feb 28 12:27:39 XorA: that's what I meant Feb 28 12:27:42 XorA: The binary emuation is hack as you have to create a rootfs for it. If you use system emulation, its good Feb 28 12:27:53 * RP would like to see the images just run through qemu system emualation to complete the postinsts Feb 28 12:28:05 RP: we have 90% of a rootfs in staging Feb 28 12:28:07 much simpler Feb 28 12:28:12 XorA: load up an image, run postinst stuff and pull out the needed stuff Feb 28 12:28:15 <[lala]> RP: regarding zaurusd - so i have only turn up the volume and enjoy? :) Feb 28 12:28:35 [lala]: yes Feb 28 12:28:41 RP: a lot of postinst require a real machine Feb 28 12:28:48 RP: stuff like blueprobe Feb 28 12:28:57 lrg: right-left volumes are locked now in the mixer. thats fine! Feb 28 12:30:00 koen: I can see that being slightly problematic. I'd still much rather use the system emulation rather than starting to emulate individual binaraies though... Feb 28 12:30:12 greentux: yep, they are now in 1 control, but can still be independently changes if need be Feb 28 12:30:37 <[lala]> RP: so what would i have to do to use the opie-mixer to turn up the volume? ;-) Feb 28 12:30:51 RP: some extra magic var like NEEDS_REAL_MACHINE_FOR_POSTINST which only gets used in qemu system emu Feb 28 12:30:53 RP: when it comes down to it, you are more likely than me to write the bbclass, so Ill try and not annoy you by arguing :-) Feb 28 12:31:46 RP: alternatively we could just patch qemu to present a 'real' /proc/cpuinfo for each machine in OE Feb 28 12:31:58 [lala]: use alsa mixer atm, I'm not sure if opie understands alsa, so it may use oss emu Feb 28 12:32:00 that would solve 90% of the machine specific scripts Feb 28 12:32:36 <[lala]> lrg: that would be fine with me - but i cannot actually ask my end users to use alsamixer ;-) Feb 28 12:32:36 [lala]: i'll be writing a patch soon to alsa-lib that will make the mixer smaller Feb 28 12:33:00 <[lala]> lrg: they need to use a gui mixer embedded in opie Feb 28 12:33:40 [lala]: I don't know much about the gui mixer in opie :( Feb 28 12:34:06 heh Feb 28 12:34:27 (qt)opi(e)(a) don't know about alsa afaik Feb 28 12:35:43 [lala]: ok, i'll look at whats involved with oss mixer emulation, and try an map the hp vol correctly Feb 28 12:35:50 koen: so do gpe? Feb 28 12:35:51 <[lala]> koen: i was afraid you would say that ;-( Feb 28 12:36:07 <[lala]> lrg: that would be great Feb 28 12:36:46 hrw: afaik gpe doesn't know about alsa either Feb 28 12:38:55 RP: according to the nokia benchmarks gtk2.8/softfloat is 2% faster as gtk2.6/hardfloat :) Feb 28 12:39:44 bleh Feb 28 12:39:47 * koen is out of cheese Feb 28 12:40:02 so gtk2.8 suxx Feb 28 12:40:12 koen: Those postinst's just need /proc/cpuinfo?! That sounds like it could be done totally differently and avoid the issue entirely then... Feb 28 12:40:43 RP: most of the time only /proc/cpuinfo is used to set stuff Feb 28 12:40:50 XorA: Just trust me that emualting indiviual binaries would just turn us into scratchbox and all its hacks. We can do better. Feb 28 12:40:56 RP: but sometimes they probe hw (e.g. blueprobe) Feb 28 12:41:58 morning Feb 28 12:42:01 hi chouimat Feb 28 12:42:20 koen: We should probably see whether something like zaurusd/devmand could handle all the machine speicifc bits... Feb 28 12:42:51 koen: I made those benchmarks ;-) Feb 28 12:42:58 RP: time for rpd where we all get a mini RP inside our machines to automatically fix kernel for our hardware Feb 28 12:43:01 looks like we will end in one rootfs to rule them all... Feb 28 12:43:14 RP: see http://handhelds.org/cgi-bin/cvsweb.cgi/gpe/base/blueprobe/blueprobe.init?rev=1.12&content-type=text/x-cvsweb-markup Feb 28 12:43:50 (the h2200 case) Feb 28 12:43:52 koen: Im 99% sure hal could have all the desktop specific h/w stuff stripped out Feb 28 12:44:14 mallum: that's kind of what I said :) Feb 28 12:44:19 koen: :) Feb 28 12:44:41 koen: RP is allergic to it though ;) ... Feb 28 12:44:48 hmmm, I can't find the screenies of hal-devicemanager running on the hx Feb 28 12:45:12 http://handhelds.org/cgi-bin/cvsweb.cgi/gpe/base/blueprobe/blueprobe.init?rev=1.12&content-type=text/x-cvsweb-markup Feb 28 12:45:16 ops Feb 28 12:45:30 http://www.oesf.org/forums/index.php?showtopic=17632&view=findpost&p=116363 Feb 28 12:45:38 ~lart X11 for two clipboards Feb 28 12:45:39 * ibot changes X11's permissions to 0777 and tells the world for two clipboards Feb 28 12:46:02 looks like we got someone who want to maintain getting poodle for OE/OZ Feb 28 12:46:05 koen: That blueprobe hardware specific stuff is an ideal candidate for zaurusd/devmand. I'm sure we could teach it how to handle not breaking anything if run under qemu as well Feb 28 12:46:42 RP: now you know why I said "don't call it zaurusd" :) Feb 28 12:47:17 * RP wonders if mallum heard that... Feb 28 12:49:05 RP: any concrete plans to add some dbus glue to devmand? Feb 28 12:49:57 RP: LA LA LA LALA LALA LA Feb 28 12:50:01 koen: mallum will have be try and slim down hal instead I expect. Feb 28 12:51:21 Its getting to the point where I have to write the dbus glue just to prove how simple it is though ;-) Feb 28 12:54:03 RP: rename it to small-hall Feb 28 12:54:08 or shallow-hal Feb 28 12:54:13 heh Feb 28 12:55:00 * koen wanders to the kitchen to bake some eggs Feb 28 12:57:40 Actally, I can see a case for splitting hal in two. One part providing notifications, the other details of attached hardware. The notifications might lack certain information without the hardware decoding part but it would make it a lot more useable on handhelds Feb 28 13:00:11 and everybody is happy! Feb 28 13:00:37 but no ones said why HAL isn't usuable on handhelds and I though zaurusd and hal were two different things Feb 28 13:01:57 mallum: mostly the zillion .fdi files and hal parsing those is what bothered me Feb 28 13:02:11 HAL provides all kinds of info we don't really need on a handheld taking CPU, memory and disk space away from things which are more important Feb 28 13:02:33 and the lake of C tools to manipulate HAL Feb 28 13:02:37 ~lart python Feb 28 13:02:37 * ibot slaps python around with a large trout Feb 28 13:02:44 s/lake/lack/ Feb 28 13:03:53 also, HAL appears to be quite entwined with system policy and gconf. I did hear comments about there being issue with it if you weren't running X. Feb 28 13:04:15 A true HAL should work for something like opie... Feb 28 13:04:24 RP: gconf is per user Feb 28 13:04:39 so you have to be in a session with gconf running Feb 28 13:05:02 but that's mostly about stuff talking to the user and hal Feb 28 13:05:09 hal it self is just a daemon Feb 28 13:06:00 RP: mjg59 was saying that all the cool hal stuff (gnome-power-manager) needs gconf since it's a per user thing Feb 28 13:06:15 which isn't so clearcut with handhelds Feb 28 13:10:16 koen|food: but we agreed fdi files could be removed and lack of c tools isn't an argument - there are no c tools for zuarusd either Feb 28 13:10:42 RP: If opie supports dbus I cant see there being too much of a problem making it also work with hal Feb 28 13:10:48 someone want to maintain poodle support? Feb 28 13:10:59 RP: it has equivilants to X and gconf right ? Feb 28 13:11:37 mallum: opie doesn't have dbus but I agree dbus makes sense so opie would need to develop that, fair enough. Feb 28 13:14:09 mallum: showing of a poky image running uHAL + dbus + avahi + eds-dbus might attract some necessary attention for gtk+ on handhelds Feb 28 13:14:28 I think what bugs me is that HAL was designed for its device manager like functionality which is what most of the code is about. It also happens to export a handful of events. Those events are what we're really interested in (and why HAL is so popular) and yet you either get all or nothing Feb 28 13:16:34 I just want to avoid reinventing the wheel, keep in line with desktop tech and API's and make nice things like network-manager just work Feb 28 13:17:37 I know. Same aim, different approaches. Feb 28 13:17:49 hmmm Feb 28 13:17:59 I'm so not going to a PSPICE to OE Feb 28 13:18:08 or SPICE-ng Feb 28 13:18:29 s/a/add/ Feb 28 13:21:05 RP: having hal export information specific to handhelds would be very useful though - display dpi for example Feb 28 13:21:05 the new omap 3430 looks pretty sweet Feb 28 13:21:24 it even has a real FPU Feb 28 13:22:42 lrg: the new omap uses the cortex core I was getting at saturday evening Feb 28 13:24:54 koen: cool, i like the fact it has a real fpu, will be useful for cairo Feb 28 13:25:36 I still haven't figured out if the multimedia unit is a DSP or asic Feb 28 13:41:30 lrg: 'dynamic bass boost' + pop is also quite interesting Feb 28 13:43:23 koen: the pop will be right in the bass boost gain area. Feb 28 13:48:59 lrg: somewhere around 100-200Hz? Feb 28 13:49:33 koen: i think so. Feb 28 13:50:24 koen: On the borzoi at least, you can choose the freq.. 130Hz or 200Hz @48kHz Feb 28 13:52:35 lrg: about broken silicon: the ep93xx has 2 gpios for leds which are used to drive an RC network to reset the chip if the red one stays on too long Feb 28 13:53:02 to work around lockup while booting up the cpu Feb 28 13:54:39 koen: interesting watchdog Feb 28 13:54:57 koen: WTF? Feb 28 13:58:02 XorA: indeed Feb 28 13:58:37 koen: is that a fault with the cpu, or the demo software they are working around Feb 28 14:00:27 XorA: cpu Feb 28 14:00:39 XorA: it's during cpu init Feb 28 14:00:57 no bootloader or kernel running at that point Feb 28 14:03:12 bbl Feb 28 14:08:07 someone has spare Audi RS4? Feb 28 14:08:15 heh Feb 28 14:08:49 I don't even have a car of my own Feb 28 14:09:00 I'm updating audi.pl website now with audi rs4 infos Feb 28 14:09:13 420 horsepower == 4.8s to 100km/h Feb 28 14:18:59 <[lala]> RP: does the fb on the borzoi behave like a normal console or is there some buffering involved? i'm asking because i couldn't see all printk before crash Feb 28 14:19:36 [lala]: Its a normal console Feb 28 14:20:48 <[lala]> RP: hmm, any idea why not all printks made it to the console before crash? Feb 28 14:21:43 [lala]: Were the log levels correct? (KERN_DEBUG might not be printed if the console log level wasn't high enough) Feb 28 14:22:39 <[lala]> hmm Feb 28 14:23:25 RP: 8 is highest or 1 is highest? (log levels) Feb 28 14:23:49 8 is KERN_DEBUG Feb 28 14:23:55 thx Feb 28 14:25:31 <[lala]> RP: might it be that printk gets confused if the kernel stack is corrupt? Feb 28 14:25:46 <[lala]> RP: actually we had a stack overflow bug Feb 28 14:26:01 [lala]: That's very possible yes, especially if the oops itself is being corrupted Feb 28 14:26:26 or it could be the error took out parts of the console of framebuffer code... Feb 28 14:28:47 <[lala]> ok Feb 28 14:29:35 has everybody complained to linuxdevices that OE is missing in their 2006 survey? Feb 28 14:29:35 http://linuxdevices.com/contactus.html Feb 28 14:30:32 koen: they do not want us to tell users to vote for us Feb 28 14:30:46 'this year OE will not fake votes' Feb 28 14:32:37 hrw: nobody can vote for us Feb 28 14:32:44 we aren't in the damn survey Feb 28 14:32:55 that's the point Feb 28 14:32:57 I know Feb 28 14:32:59 Is there an "other" category? Feb 28 14:33:22 RP: yes, and that tells you to contact them Feb 28 14:37:06 * RP contacts them Feb 28 14:38:05 koen: question 7&8? Feb 28 14:38:39 <[lala]> RP: if i want some of the mixer controls to have a higher default value -> change the default asound state in /etc/asound.state? Feb 28 14:38:52 [lala]: correct Feb 28 14:39:08 [lala]: zaurusd only populates that file if it doesn't exist Feb 28 14:39:24 and should preserve user settings in it Feb 28 14:39:30 <[lala]> RP: and how do i automatically switch between the differen settings for the mixer? headphone etc.? Feb 28 14:41:19 [lala]: It can't automatically detect what was inserted so it assumes an insertion is headphones. It tries to call /usr/bin/mixer-handler first which could ask the user though Feb 28 14:41:53 <[lala]> but that mixer-handling is not included yet i assume Feb 28 14:42:19 [lala]: That would have to be an opie applet or similar so its beyond the scope of zaurusd Feb 28 14:42:29 <[lala]> RP: i see Feb 28 14:43:06 RP: let's hope the next zaurus has servos so you can make it swivel using dbus commands :) Feb 28 14:43:13 hey tnb Feb 28 14:43:22 hi koen Feb 28 14:43:32 [lala]: The applet would call something like "zaurus-mixer-callback Headset Off" once the user had indicted what was connected Feb 28 14:43:47 koen: That would be interesting :) Feb 28 14:44:36 if you can turn a laptop into a robot you can turn a zaurus into a robot :P Feb 28 14:44:46 koen: you could attach web cams to it a then use them to survey the area remotely ;) Feb 28 14:44:52 user pressed "power on" so let open hinge Feb 28 14:48:16 reminds me ... Feb 28 14:48:44 does anyone know if teh gnutls wpa_supplicant is making it into familia 0.8.3? Feb 28 14:49:25 since shadows got it done ... i havent tested it to see if it behaves on 0.8.3 yet ... Feb 28 14:50:22 emte: you'd have to ask france_, he'll pass the question to the underground group working on familiar in complete secrecy Feb 28 14:50:39 ~lart PID controllers Feb 28 14:50:39 * ibot stamps PID controllers on the forehead with the official Troll marker Feb 28 14:50:55 s/Troll/Trolltech/ Feb 28 14:51:39 koen, lol Feb 28 14:51:53 emte: that wasn't a joke Feb 28 14:52:18 yeah i been wondering if i should make my own "copy" for the h36 and start mangling it Feb 28 14:54:06 anyway bbl Feb 28 14:54:40 koen: you wont have time for Fam with Angstrom up and running anyway Feb 28 14:55:31 http://linuxdevices.com/contactus.html any one seen one of these? Feb 28 14:56:14 XorA: it would have been nice if *I* decided that I didn't have enough time Feb 28 14:56:33 XorA: but I guess people 6k km away know my schedule better than me Feb 28 14:57:08 koen: I dont know the details, but it does sound kinda shitty Feb 28 14:58:09 XorA: angstrom will obsolete everything else when it comes out ;) Feb 28 14:58:28 koen: and that will be cool Feb 28 14:59:25 koen: every day I have used OE, my Z has become more like a desktop Feb 28 14:59:29 koen: and that is good Feb 28 14:59:38 It is *the* opportunity to have a zillion ABI breaks Feb 28 14:59:58 XorA: I miss pda functionality on my Z :( Feb 28 15:00:17 gcc3 -> gcc4, *fpa -> softvfp, cruft -> EABI, qt -> qt-mt, etc Feb 28 15:00:26 hrw: I have PDA, I just use almost the same software on both now Feb 28 15:00:50 * XorA needs to buy a second Z Feb 28 15:00:51 koen: 101 random hardware hacks -> devmand? Feb 28 15:01:08 or a PSP Feb 28 15:01:12 RP: maybe more than 101 Feb 28 15:01:19 XorA: gp2x Feb 28 15:01:56 koen: yeah forgot that one, it'll run mplayer Feb 28 15:01:59 XorA: for me c7x0 is hard to use as pda Feb 28 15:02:15 need once device for movies, and one for devel Feb 28 15:10:17 hrw: you need to use an Ipaq running WinCE, once youve had to edit the registry using a virtual keyboard one too many times you will find the c7x0 a dream Feb 28 15:12:43 emte: there is no secret. All I said is last week that after the kernel project was back on track, that we would be working on the distributions. I would like to see familiar move back from a developer focus to a more user focus. It has taken 3 weeks to get the kernel folks on the same page. Until the basic work is done to get everyone un-scattered, I do not even know what the changes will be. A while back koen wanted to merge Feb 28 15:12:43 familiar in to his new project angstrom (deprecate familiar), I do not believe this is a good idea. Familiar has a good brand name. I wish him well on angstrom. I would be happy to host the angstrom project. Competing projects is usually good. For example gpe vs opie. Two different concepts tested in the real world. Feb 28 15:13:15 yay Feb 28 15:13:31 hh.org people are twisting my words *again* Feb 28 15:13:56 XorA: too much time to just open Z, navigate etc.. Feb 28 15:13:58 * koen will shut up now so people have nothing to twist Feb 28 15:14:41 03koen 07org.oe.dev * r0386ace1... 10/conf/distro/angstrom.conf: angstrom.conf: use -Os by default, see the comments in the file for more info Feb 28 15:17:17 RP: http://www.kdedevelopers.org/node/1829 Feb 28 15:19:04 koen: interesting. So opie support for dbus might happen eventually... Feb 28 15:19:41 <[lala]> koen: is there some website explaining the new stuff about angstrom? Feb 28 15:20:31 RP: I do not think that this will happen in opie 1.2.x tree Feb 28 15:20:45 RP: if opie will move to qt4/x11 then yes. but not before Feb 28 15:21:26 hrw: hence "eventually" :-/ Feb 28 15:24:34 hrw: ultimately it is darwinism, Opie keeps up, finds itself a niche, or dies Feb 28 15:25:09 personally I hope for first option Feb 28 15:26:08 [lala]: the bottom part of http://openzaurus.org/wordpress/newsletters/ooo-newsletter-200508/ is your best bet Feb 28 15:26:29 where 'keeps up' means 'ignore TT australia FUD' Feb 28 15:26:47 XorA: did you hear about the konq/e update? Feb 28 15:26:57 koen: konq/e? Feb 28 15:28:27 XorA: the qt/e version of konqueror Feb 28 15:28:45 koen: I dont remeber anything about an update Feb 28 15:28:57 koen: but that doesnt mean I didnt say something Feb 28 15:29:12 svn trunk can now run on qt/e 2.3.x Feb 28 15:29:15 * RP ponders why hal would end up depending on gcc-cross-sdk :-/ Feb 28 15:29:20 koen: ah, sorry, I totally misread your original sentence Feb 28 15:29:25 * XorA needs a bigger font Feb 28 15:29:33 XorA: terminus/48? Feb 28 15:29:39 hrw: Im on *doze Feb 28 15:29:44 so opie can now have 1 decent browser Feb 28 15:29:50 XorA: ttf-terminus/48 Feb 28 15:29:54 decent = recent + open source Feb 28 15:29:54 koen: that is good news Feb 28 15:30:06 koen: and compilable unlike minimo Feb 28 15:31:12 XorA: and hopefully gcc4 safe Feb 28 15:31:30 <[lala]> has anyone tried it out yet? Feb 28 15:33:41 hmm, samba is a little fscked, I get an infinite tree of folders instead of my expected files Feb 28 15:35:01 does OE have capability to have multiple gcc in staging? Feb 28 15:35:15 if I compile my oe stuff in a multiprocessor machine, is there any way pass a -j3 or something like that to the compiler? Feb 28 15:35:46 PARALLEL_MAKE="-j3" Feb 28 15:35:49 katossi_uni: PARALLEL_MAKE = "-j3" in your local.conf Feb 28 15:36:45 <[lala]> koen: have u tried the new konq/e? Feb 28 15:36:46 hi Feb 28 15:42:03 thanks for the -j3 tip! i was wondering about this last night... Feb 28 15:46:08 hi, folks. is it possible for me to compiling with my own toolchain ? Feb 28 15:54:06 leoncamel: in theory yes Feb 28 15:54:32 hrw, but, how ? Feb 28 15:54:57 is there anybody had done this before ? Feb 28 15:55:05 leoncamel: never tried that way Feb 28 15:55:21 what's rp's email, someone? please privmsg it to me, i would like to send some debug output he asked for Feb 28 15:55:27 you can use ASSUME_PROVIDED and add gcc-cross, glibc etc to it Feb 28 15:55:45 jnc: pastebin it and privmsg him url to it? Feb 28 15:55:49 RP: i have the output from the serial console for ya Feb 28 15:56:05 jnc: hi, pastebin is good Feb 28 15:56:19 RP: okay. will do. Feb 28 15:56:20 or I can give an email address, I don't mind Feb 28 15:58:59 hrw, the ASSUME_PROVIDED only skip the toolchain downloading. the question is, do the bitbake always using "arm-linux-gcc" to compile package ? Feb 28 15:59:30 hrw, but I want use my own toolchain which named "arm-iwmmxt-linux-gnu-gcc" ... Feb 28 15:59:37 hrw, is it possible ? Feb 28 16:02:33 leoncamel: ln -s arm-iwmmxt-linux-gnu-gcc arm-linux-gcc Feb 28 16:03:40 re Feb 28 16:05:52 XorA, hmm, sounds good. Feb 28 16:06:26 XorA, but I think it is not a best idea.. Feb 28 16:06:48 why not use tune-iwmmxt.conf ? Feb 28 16:07:46 koen, ok, I will check it . Feb 28 16:09:24 koen, it is not what I want. I want use different toolchain, instead of append option for current toolchain. Feb 28 16:10:09 03koen 07org.oe.dev * r2ac7aa00... 10/packages/libgpepimc/libgpepimc_0.6.bb: libgpepimc: update to 0.6 Feb 28 16:11:51 03koen 07org.oe.dev * r22a2e0e4... 10/conf/machine/ipaq-pxa270.conf: ipaq-pxa270.conf: add an option to use iwmmxt optimizations Feb 28 16:13:32 ping pb__ Feb 28 16:15:08 koen: I've been asked by a journalist for a copy of my presentation. Have you been asked. I'm trying to suss him out Feb 28 16:16:05 lrg: I haven't been asked yet Feb 28 16:16:23 koen: is your presentaion on the web ? Feb 28 16:16:40 03koen 07org.oe.dev * rf144cf89... 10/packages/libgpevtype/libgpevtype_0.16.bb: libgpevtype: update to 0.16 Feb 28 16:16:43 it is since this morning Feb 28 16:27:12 ~lart imake Feb 28 16:27:12 * ibot gives imake a "free" copy of Windows and then charges double for "Upgrades" Feb 28 16:29:43 koen: where exactly ? Feb 28 16:34:00 cu Feb 28 16:44:08 * Philippe is back (gone 16:40:13) Feb 28 16:45:35 obergix[work]: http://dominion.kabel.utwente.nl/koen/talks/ Feb 28 16:45:40 hey Philippe Feb 28 16:45:51 hallo koen Feb 28 16:46:14 hey Philippe Feb 28 16:46:33 hello liamv :-) Feb 28 16:46:35 koen: thx, sorry, I seem to have asked twice ;) Feb 28 16:46:39 Philippe: I won the OSCON tickets in the raffle :) Feb 28 16:47:15 lrg, that is definitely good news! Enjoy the travel! Seems Fosdem was particulary good for you :-) Feb 28 16:47:22 indee :) Feb 28 16:47:25 indeed Feb 28 16:51:00 koen: Apparently openembedded.org is now listed on the linuxdevices.com survey as part of handhelds.org Feb 28 16:51:15 heh Feb 28 16:51:21 ~lart linuxdevices Feb 28 16:51:21 * ibot flings poo at linuxdevices Feb 28 16:51:28 you and me both, brother Feb 28 16:53:57 wow Feb 28 16:54:04 CosmicPenguin: AMD has been pretty busy Feb 28 16:54:26 koen: just being lazy, I guess Feb 28 16:54:30 its not often that we push usptream Feb 28 17:00:17 ? Feb 28 17:04:58 JustinP: I just got a huge patch from AMD Feb 28 17:05:45 lrg: yo Feb 28 17:06:02 sorry, was at a meeting in Hemel Hempstead for a while there Feb 28 17:06:34 192kB of update goodness Feb 28 17:08:26 pb_: I've requested some product brochures for you. They should help you with codec selection. Feb 28 17:09:35 excellent, thanks Feb 28 17:10:29 pb_: np, in the meantime, the wm8751 may be what your are looking for, but I'm not sure of your exact requirements Feb 28 17:12:03 koen: updates to OE? Feb 28 17:12:28 JustinP: yes Feb 28 17:13:11 http://pastebin.com/576905 is the changelog Feb 28 17:13:28 I wasn't aware that AMD does OE stuff Feb 28 17:14:09 Its our only build system Feb 28 17:14:28 We even use it to build Gentoo and Redhat packages for legacy testing Feb 28 17:14:46 heh Feb 28 17:14:48 awesome Feb 28 17:14:56 looks like some good updates Feb 28 17:15:44 hi folks Feb 28 17:15:48 hey mickeyl Feb 28 17:15:56 hey mickeyl Feb 28 17:16:06 abend mickeyl Feb 28 17:16:39 :) so you all had a safe trip back home Feb 28 17:17:34 CosmicPenguin: who should I credit, AMD or Ray? Feb 28 17:18:13 morning mickeyl Feb 28 17:18:18 hi JustinP! Feb 28 17:18:32 JustinP: there are some pretty cool things appearing in the E cvs Feb 28 17:18:44 e.g. epdf, exhibit and enterminus Feb 28 17:19:17 mickeyl: what do the last 2 do? Feb 28 17:19:40 exhibit is an etk-based picture viewer (better UI than the other thing) Feb 28 17:19:51 enterminus is a show-off / benchmarking for Evas Feb 28 17:20:01 (IIRC) Feb 28 17:20:14 i like etk a lot Feb 28 17:20:25 it feels much more like a GUI toolkit than ewl ever did Feb 28 17:20:28 hail mickeyl Feb 28 17:20:33 hi phil! Feb 28 17:21:01 pb_: i managed to buy some leffe brun at the airport, but I didn't dare to open one of the six bottles yet :) Feb 28 17:21:11 hey incinerator Feb 28 17:21:11 mickeyl: I forgot to look at your edje + extras demo Feb 28 17:21:39 incinerator: going to edlug tomorrow ? Feb 28 17:21:59 koen: ah right. well, i will take care about adding it into a cvs as promised Feb 28 17:22:04 hi mickeyl Feb 28 17:22:15 hey RP Feb 28 17:22:20 thanks for commiting qemu Feb 28 17:22:22 can' Feb 28 17:22:26 t wait to try Feb 28 17:24:28 mickeyl: I haven't had a chance to look into the colours bug but will aim to do so soon Feb 28 17:25:01 cool! Feb 28 17:25:13 * mickeyl does the Linuxdevices poll Feb 28 17:25:33 I should try opie on it :) Feb 28 17:25:45 RP: yeah, that's a good plan :) Feb 28 17:26:01 koen: credit ray Feb 28 17:26:02 lrg: yup, the 8751 looks pretty good. Feb 28 17:26:28 CosmicPenguin: ok Feb 28 17:28:28 mickeyl: hehe Feb 28 17:32:14 hi reenoo Feb 28 17:32:33 evening Feb 28 17:32:35 hi REdOG Feb 28 17:32:37 oops Feb 28 17:32:38 hey pb__ Feb 28 17:32:42 ~lart tab completion Feb 28 17:32:42 * ibot eats tab completion and falls over dead Feb 28 17:32:44 hi reenoo# Feb 28 17:33:09 moin mickeyl Feb 28 17:33:45 koen: monotone: misuse: you have no private key to make signatures with Feb 28 17:33:55 tried to add some files Feb 28 17:34:17 sounds like you didn't add your privkey Feb 28 17:34:47 ls ~/.monotone/keys/ Feb 28 17:35:51 koen: there is no .monotone should i create one Feb 28 17:37:28 koen: monotone: error: Extraneous data in key store. Feb 28 17:40:11 hmm first none, then too many. odd. which version are you using again? Feb 28 17:41:41 ~seen hrw|gone Feb 28 17:41:44 hrw|gone is currently on #oe #openzaurus #opie, last said: 'CoreDump|install, JustinP: feel free to share tst5 URL location'. Feb 28 17:44:02 mickeyl: cool.I'll see if I can get those in Feb 28 17:44:18 JustinP: that would rock Feb 28 17:44:33 epdf needs some ultra bleeding edge version of poppler though Feb 28 17:44:37 hi Feb 28 17:44:40 hail hrw! Feb 28 17:44:45 hey hrw Feb 28 17:44:55 mickeyl: has poppler fixed it's loading algo? Feb 28 17:45:04 mickeyl: it used to put the complete pdf into meme Feb 28 17:45:12 03koen 07org.oe.dev * r006bbeb2... 10/site/mipsel-linux: Feb 28 17:45:12 site/mipsel-linux: updates courtesy Advanced Micro Devices, Inc. Feb 28 17:45:12 * samba, vim, sudo, and intercom build now Feb 28 17:45:16 03koen 07org.oe.dev * r3905ca21... 10/classes/base.bbclass: base.bbclass: Handle crlf with DOS ZIP file, courtesy Raymond Danks Feb 28 17:45:21 03koen 07org.oe.dev * r408486dc... 10/classes/kernel.bbclass: kernel.bbclass: Add kernel-modules meta package for PARALLEL_INSTALL_MODULES, courtesy Raymond Danks Feb 28 17:45:22 koen: not sure about that. i need to do some memory benchmarks Feb 28 17:45:23 w8 a moment I look into backlog Feb 28 17:45:37 which sucks If I have to consult my feyneman book (~700M pdf) Feb 28 17:45:45 true, that's a no go Feb 28 17:46:13 mickeyl: that's why I have both gpdf and evince installed :) Feb 28 17:46:14 i can't believe someone writes anything new from scratch that loads files completely into memory for anything else than a prototype Feb 28 17:46:20 koen: hehe Feb 28 17:46:52 700Mb is a hell of a book though Feb 28 17:47:06 do they embed AVIs nowadays? Feb 28 17:47:07 :) Feb 28 17:47:35 mickeyl: http://www.amazon.com/gp/product/0201021153/002-3530199-0217648?v=glance&n=283155 Feb 28 17:47:46 whats the point of a book with embedded avis? :) Feb 28 17:47:55 it a compilation of his lecture notes for the 1960's Feb 28 17:47:59 Somewhere Gutenberg is rolling in his grave Feb 28 17:48:15 I've seen a pretty cool bookreader on fosdem... but i can't remember the name Feb 28 17:48:17 anyone do? Feb 28 17:48:40 a thin hardware gadget roughly a4 size Feb 28 17:49:08 libre? Feb 28 17:50:04 hmm possible. is that the name or the vendor? Feb 28 17:50:20 sony Feb 28 17:50:41 http://en.wikipedia.org/wiki/Sony_Librie_EBR-1000EP Feb 28 17:50:43 no that wasn't it Feb 28 17:50:48 it was more dark gray Feb 28 17:50:52 mickeyl: have you looked at apm threads on oz-devel? Feb 28 17:52:07 hrw: i have. i'm pretty confused though - do you have an idea of which device works now and which one doesn't? Feb 28 17:52:43 mickeyl: poodle, collie, tosa works, other does not even noticed Feb 28 17:56:03 so what testing do we need before 3.5.4? Feb 28 17:56:14 (i wouldn't include CXK in the 3.5.4 release at all) Feb 28 17:56:18 (but it's your call) Feb 28 17:56:27 mickeyl: I want keylaunch changes by law_ Feb 28 17:57:11 ok Feb 28 17:57:14 _law_: hurry up ;) Feb 28 17:57:39 I will probably get few of them, try to get them in packages and ignore rest Feb 28 18:02:30 mickeyl: you want to block cxk until 2.6 is done? Feb 28 18:02:38 JustinP: it _is_ done. Feb 28 18:02:49 mickeyl: well, yeah..... Feb 28 18:02:56 JustinP: it just needs some time for RP to port it back to branch Feb 28 18:03:03 JustinP: or am i missing something? Feb 28 18:03:05 except one thing, but I guess that's still my fault.... Feb 28 18:03:18 it works pretty well in my tests, yeah... Feb 28 18:03:22 better than 2.4 Feb 28 18:03:24 however Feb 28 18:03:34 actually, a few things are still broken IIRC Feb 28 18:03:51 RNDIS when connected to Windows Feb 28 18:03:52 i don't think it makes sense to release 3.5.4 with crappy CXK support and then lose again our reputation before some weeks later we could have 3.5.4.1 with proper CXK support Feb 28 18:03:56 g_file_storage Feb 28 18:04:04 true Feb 28 18:04:10 makes sense Feb 28 18:04:19 I'll go with that Feb 28 18:04:25 ok. me too Feb 28 18:04:44 cool. i think the CXK users will understand that as well Feb 28 18:04:46 the major stopping point for me with 2.6 is that I can't do USB-eth with Windows Feb 28 18:05:06 and it was incredibly slow when connected to a mac Feb 28 18:05:15 I have clues about getting that working in my inbox. Just need time Feb 28 18:05:18 * RP has to go Feb 28 18:05:24 ah yes...the fabled inbox Feb 28 18:05:25 RP: cool. cu Feb 28 18:05:27 RP: ;-) Feb 28 18:05:37 (RNDIS thats is, not the Mac - no idea whats wrong there) Feb 28 18:05:42 * JustinP waits patiently Feb 28 18:05:52 I can kexec, so I'm still ok ^_^ Feb 28 18:05:56 hehe right Feb 28 18:06:02 do13 showed me kexec on his tosa Feb 28 18:06:04 I'm impressed! Feb 28 18:06:13 yeah, works like a charm Feb 28 18:06:27 I'm using it when I test 2.4 stuff now Feb 28 18:06:35 although I hopefully shouldn't need it soon :-) Feb 28 18:07:12 and my final "problem" with 2.6 is lack of inline audio remote support.... Feb 28 18:07:26 but since I seem to be the only Cxk owner with one who uses OE/OZ.... Feb 28 18:07:29 JustinP: we have someone working on that ;) Feb 28 18:07:39 koen: that's me Feb 28 18:07:49 koen: or is someone else working on it that I don't know of? Feb 28 18:07:51 I know, hence the smiley Feb 28 18:07:55 heh Feb 28 18:08:05 I have a driver that compiles and gets interrupts Feb 28 18:08:10 mickeyl: keylauch already works for c7x0 Feb 28 18:08:22 but the values returned by the max1111 aren't unique for the keys.... Feb 28 18:08:31 law_: commit it.... Feb 28 18:08:34 * JustinP needs some hours to try to trace into it Feb 28 18:09:01 hrw: shouldnt we wait for gpe-conf before ? Feb 28 18:09:19 law_: then try to resolve it with Florian Feb 28 18:09:42 law_: I can make ugly patch to gpe-conf just to get c7x0 working with keylaunch Feb 28 18:10:07 law_: and revert it after tagging Feb 28 18:10:19 heh Feb 28 18:10:22 hrw: my altgr patch ? Feb 28 18:10:37 law_: no - rather key-layout things Feb 28 18:10:47 law_: altgr need work also from ipaq people Feb 28 18:10:54 ok Feb 28 18:11:20 law_: oz354 is too delayed Feb 28 18:15:58 * mickeyl cheers at zecke Feb 28 18:16:01 yay! dual core mac mini Feb 28 18:16:03 hey zecke Feb 28 18:16:24 koen: x86 one? Feb 28 18:16:29 re zecke Feb 28 18:16:29 hrw: yes Feb 28 18:16:36 koen: where when? Feb 28 18:16:55 hail zecke Feb 28 18:16:55 zecke: 4 minutes ago, Apple Media Event Feb 28 18:17:17 zecke: time to practice your dutch: http://www.macosx.nl/?p=showarticle&art_id=2238 Feb 28 18:17:45 'new feature: shared libraries.' Feb 28 18:17:45 pb__: yes my lord Feb 28 18:17:46 heh Feb 28 18:18:29 koen: can you put your OE talk slides somewhere online? Feb 28 18:18:58 * mickeyl ponders whether to apply for a presentation @ froscon 2006 Feb 28 18:19:05 28 17:48 < koen> obergix[work]: http://dominion.kabel.utwente.nl/koen/talks/ Feb 28 18:19:11 excellent Feb 28 18:19:42 we had wiki page for it iirc Feb 28 18:20:39 http://oe.handhelds.org/cgi-bin/moin.cgi/PapersAndPresentations Feb 28 18:20:47 * mickeyl adds koens talk Feb 28 18:21:21 mickeyl: could you add both of them? Feb 28 18:21:37 ok Feb 28 18:21:58 "The OpenMotif Project Team has released Beta 2 of OpenMotif 2.3, the most significant version of OpenMotif since it was released to the open source community in May 2000. OpenMotif 2.3 includes major feature enhancements and over 25 bug fixes requested by developers of enterprise applications." Feb 28 18:22:03 lol Feb 28 18:22:32 25 bugfixes.. Feb 28 18:22:36 sounds like pretty depressing bugs if it took them that long Feb 28 18:24:32 heh Feb 28 18:26:26 I'm sure Jörg Schilling loves Motif Feb 28 18:28:48 maybe they mean, "25 bug fixes requested by [people who matter]. Untold multitudes of bugfixes requested by [people who don't matter]" Feb 28 18:29:23 'since we got it to compile again a week ago we fixed 25 bugs' Feb 28 18:29:35 heh Feb 28 18:30:30 * zecke wonders if Qtopia5 will be based on OpenMotif Beta3 Feb 28 18:30:52 mickeyl: do you use multimachine config? Feb 28 18:31:02 mickeyl: I'm finishing work on 3.5.4.conf now Feb 28 18:33:05 hrw: i had nothing but problems with multimachine yet Feb 28 18:33:12 probably i'm doing something wrong Feb 28 18:34:12 mickeyl: http://ewi546.ewi.utwente.nl/tmp/hrw/conf/ - my configs from 3.5.4-test phase (multimachine) Feb 28 18:34:25 mickeyl: but ofcourse you can do normal ones Feb 28 18:35:15 openzaurus-3.5.4.conf | 29 +++++++++++++++++++---------- Feb 28 18:35:15 1 file changed, 19 insertions(+), 10 deletions(-) Feb 28 18:44:59 8 revs goes into .oz354fam083 branch Feb 28 18:54:57 and few more Feb 28 18:55:28 yay Feb 28 18:57:36 and openzaurus-3.5.4.conf Feb 28 18:57:41 CoreDump|install: ping Feb 28 18:58:25 CoreDump said that he will update altboot in branch;) Feb 28 18:59:00 http://blog.hentges.net/?p=49 Feb 28 19:00:12 that'd be cool Feb 28 19:00:22 yep Feb 28 19:00:53 mickeyl: Daniel Stone told that he will not work on rboot anymore - he saw altboot in 3.5.4-test1 Feb 28 19:02:02 and now I will go to pub where I meet people from work and drink some beer(s) - maybe it is not same as fosdem beers but they will be beers for sure Feb 28 19:03:13 :) Feb 28 19:03:28 heh.. c7x0/2.6, cxx00/2.6, tosa/2.6, poodle/2.6... but why collie/2.4 ;( Feb 28 19:03:41 cu Feb 28 19:03:56 let CIA-12 floods you with commit messages Feb 28 19:04:11 hrw|gone: collie will possibly get 2.6 eventually Feb 28 19:04:44 ... including the infamous ess dee Feb 28 19:09:10 <_stalker_> hi everyone, Feb 28 19:09:31 ~kill ess dee Feb 28 19:09:33 * ibot shoots a charged pseudopositrino gun at ess dee Feb 28 19:09:42 <_stalker_> could anyone help with package hostap-4.4 that refuses to build for ipaq-pxa270 Feb 28 19:09:44 <_stalker_> ? Feb 28 19:10:34 try compiling 0.3.9 Feb 28 19:10:39 0.4.4 never worked for me Feb 28 19:10:41 <_stalker_> right Feb 28 19:10:53 03hrw 07org.oe.oz354fam083 * r4481f024... 10/packages/meta/meta-opie-all.bb: meta-opie-all: removed Gentium fonts Feb 28 19:10:57 03hrw 07org.oe.oz354fam083 * re5596fed... 10/packages/meta/meta-opie-all.bb: meta-opie-all: removed snes9x-qpe due to build problems Feb 28 19:11:01 03hrw 07org.oe.oz354fam083 * rb70cb460... 10/packages/apmd/apmd-3.2.2/zaurus24.patch: Feb 28 19:11:01 apmd/zaurus24.patch updated to remove /var/run/apmd.pid when it goes to kill apmd Feb 28 19:11:01 - we will not use that patch starting from OZ 3.5.4 but maybe someone find it needed Feb 28 19:11:05 03hrw 07org.oe.oz354fam083 * r323f253f... 10/packages/apmd/apmd_3.2.2.bb: Feb 28 19:11:05 apmd: remove zaurus24.patch so Zaurus/2.4 machines will not kill apmd on suspend Feb 28 19:11:05 - this workaround is not needed with our recent 2.4.18-crapix kernels Feb 28 19:11:06 and we will not release OZ 3.5.4 for cxx00 machines (2.4.20 is too broken) Feb 28 19:11:09 03hrw 07org.oe.oz354fam083 * r6f8f44dd... 10/packages/network-suspend-scripts/ (files/usbnet network-suspend-scripts.bb): Feb 28 19:11:13 network-suspend-scripts: set PACKAGE_ARCH to 'all' as they are simple shell scripts Feb 28 19:11:14 - Zaurus 2.4 usbnet modules are unloaded on suspend, loaded on resume now Feb 28 19:11:19 03hrw 07org.oe.oz354fam083 * r822741ab... 10/packages/ipkg/ipkg_0.99.159.bb: ipkg: added 0.99.159 - we used this version in OZ 3.5.4-test5 and noone complain about it Feb 28 19:11:22 03hrw 07org.oe.oz354fam083 * r81e9ba85... 10/packages/ttf-fonts/ (ttf-dejavu_2.2.bb ttf-dejavu_2.3.bb): ttf-dejavu: updated to 2.3 Feb 28 19:11:24 03hrw 07org.oe.oz354fam083 * rbd97405c... 10/packages/ (9 files in 3 dirs): Feb 28 19:11:27 changed SRC_URI in my packages to ewi instead of hrw.one.pl Feb 28 19:11:30 - I'm moving with hosting and have transfer limits Feb 28 19:11:33 <_stalker_> koen: PREFERRED_PROVIDERS = " virtual/hostap:hostap-0.3.9"? Feb 28 19:12:17 how about PREFERRED_VERSION_hostap = "0.3.9" Feb 28 19:12:26 or even better: hostap-modules Feb 28 19:12:35 <_stalker_> sweet Feb 28 19:12:41 <_stalker_> thx mickeyl Feb 28 19:12:46 np Feb 28 19:13:12 hrw|gone: you sure that ipkg is ok? I had someone trying to install an altboot ipk and it was refusing to... or could that be because I built with a different version than he was installing with? Feb 28 19:15:04 JustinP: it's not ok, but it's the version with the least amount of bugs atm Feb 28 19:20:37 mickeyl: http://pastebin.com/577112 Feb 28 19:21:41 any bets on when the first xscale ipod will hit the street? Feb 28 19:22:03 d'oh Feb 28 19:22:48 OT: can latex beamer have rows as well? Not just columns? Feb 28 19:22:49 which # is that from? Feb 28 19:23:18 mickeyl: #ep93xx Feb 28 19:23:35 zecke: well, technically every line is a seperate row - so what would you need a an additional row concept for? Feb 28 19:23:52 mickeyl: I need fancyboxes around these rows? Feb 28 19:28:09 03hrw 07org.oe.oz354fam083 * rdf0aab40... 10/packages/hostap/ (files/no-sandisk-connect-twice.patch hostap-modules_0.3.9.bb): hostap-modules 0.3.9: let Sandisk ConnectPlus be only once in config - with disable ide-cs part Feb 28 19:28:13 03hrw 07org.oe.oz354fam083 * re8e4acf9... 10/conf/distro/openzaurus-3.5.4.conf: Feb 28 19:28:13 openzaurus 3.5.4 conf: preparing to final release Feb 28 19:28:13 - fixed all not fixed versions Feb 28 19:28:13 - used FEED_BASE_URI to show where all feeds belong Feb 28 19:28:14 - set some GPE/X11 providers to preferred once Feb 28 19:28:15 - configured images dir to be machine related Feb 28 19:49:59 koen|tv: there's a guy called ChanKoLee regularly sending me a monotone key. Do you know anything about that? Feb 28 19:56:11 mickey|tv: never heard of him Feb 28 19:56:14 mickey|tv: regularly? that sounds fishy Feb 28 19:57:54 sounds like he accidentally created a cronjob or something Feb 28 20:02:14 yeah...which sounds like someone who's trying to "do something" Feb 28 20:04:44 Ciao all Feb 28 20:04:58 hi pigi Feb 28 20:05:17 hi pb_ quite a lot.... Feb 28 20:07:49 "--> 000 STATUS lists.random.goatsex (UNSEEN MESSAGES)" lol Feb 28 20:12:40 zecke: ! Feb 28 20:14:47 pb_: dwmw2 bitching about evolution Feb 28 20:15:08 ah, heh Feb 28 20:16:17 pb_ I have received a complain ( and a patch ) about ipkg not checking against md5 for the whole package. Feb 28 20:16:59 I have intention to add this check, but I'm scared that a lot of packages could have a worng md5 in control, and this could break the work of ipkg ( and oe ) Feb 28 20:17:09 what do you think about this ? Feb 28 20:17:16 s/worng/wrong Feb 28 20:17:21 I guess you should add it and see what happens. Feb 28 20:17:48 heheh. trial_and_fix method ? nice..... Feb 28 20:17:57 I don't expect many (maybe any) packages will actually have the wrong md5sum. If it does turn out to be a problem, I guess you could add a --force-corrupt-packages option or something. Feb 28 20:19:02 that's a good advice, but I wouldn't force every oe build to force a "corrupt" something while building... I'll keep that idea in mind BTW. Feb 28 20:19:32 I've been far for some time, for work. Whats the ipkg state in oe now ? Feb 28 20:19:33 if OE is generating wrong md5sums, that's definitely a bug in oe and should just be fixed. Feb 28 20:19:42 I mean which version is in ? Feb 28 20:19:50 no idea, sorry Feb 28 20:19:55 I have been mostly out of touch as well recently Feb 28 20:20:09 I guess mickeyl or zecke would know Feb 28 20:20:21 np. I'll try to sync my db. ( Last time I haven't been able btw heh ) Feb 28 20:22:01 Pigi: sorry don't know about the PREFERRED ipkg version Feb 28 20:22:18 np zecke ... thx Feb 28 20:28:05 03koen 07org.oe.dev * r83f19fb9... 10/packages/ipkg/ipkg_0.99.159.bb: ipkg: add .159 (cross) version Feb 28 20:28:16 hi koen ! Feb 28 20:28:27 hey Pigi Feb 28 20:28:49 .159 is still broken, but less broken as .155 Feb 28 20:29:07 what's broken in 159, koen|tv ? Feb 28 20:29:09 .159 removed itself when on my h5550 Feb 28 20:29:29 I had a r0 and r1 Feb 28 20:29:39 then some other package had an error Feb 28 20:30:09 A user also had problems installing ipk that I built...it would clear the want_flag and such but not install... Feb 28 20:30:20 not sure abt ipkg versions, though Feb 28 20:31:08 hey mallum Feb 28 20:31:09 that's rather strange koen|tv as the remove part is just few lines before the install ones, so if a pkg get removed, it should go installed also if others fail Feb 28 20:31:28 koen|tv: hey. I just committed some layout fixes to mb-kb Feb 28 20:31:44 koen|tv: if they appear better to you I win spin a 0.1 release this week Feb 28 20:31:52 yay! Feb 28 20:32:03 and a libmb + mb-wm release as well? Feb 28 20:32:07 and mb-panel Feb 28 20:32:37 koen|tv, what about "monotone: warning: discarding revision data packet with unmet dependencies" ? Feb 28 20:32:52 http://www.chaos.org.uk/~wookey/ <- luckily wookey can look worse than me as well Feb 28 20:33:03 sounds like your db has rogue packets Feb 28 20:33:36 (not so ) nice to know, I suppose :) Feb 28 20:34:37 koen|tv: yeah trying for mb + libmb + panel release this week also Feb 28 20:34:43 mallum: cool Feb 28 20:34:50 koen|tv: there is just some wierdness in wm I need to sort Feb 28 20:35:00 mallum: can mb-kb use a recent cairo? Feb 28 20:35:07 e.g. the one gtk 2.8 uses? Feb 28 20:39:21 mallum: does mb have any bugs wrt Composite, or might the bugs be in our kdrive? Feb 28 20:39:34 hey mreimer Feb 28 20:39:41 hey koen|tv Feb 28 20:41:03 koen|tv: yeah I think so. I've disabled cairo support unless its explicitly asked for for now Feb 28 20:41:43 mreimer: havn't touched mb composite code in a lonnnng time Id expect it to have bugs Feb 28 20:42:27 mreimer: I think the old "invisible window" bug is still there. Feb 28 20:42:50 it's hard to say for sure whether matchbox or kdrive is to blame, but my suspicion is that matchbox is probably at fault there. Feb 28 20:42:52 hmmmm - quilt2git Feb 28 20:43:29 but, since the matchbox compositor doesn't implement argb windows anyway, there isn't really much benefit in enabling composite in the first place. best to just keep it switched off. Feb 28 20:45:38 ok, thanks for the answers Feb 28 20:47:08 pb_: it does implement argb windows last time I checked Feb 28 20:47:19 oh, does it? that's cool. Feb 28 20:47:32 I guess I haven't tried it for ages. Feb 28 20:47:36 me niether Feb 28 20:47:45 dialogs can be argb iirc Feb 28 20:48:08 mallum: if someone were to pick away at composite bugs, what part of the code would you point him to? Feb 28 20:48:25 cute. maybe I should play with it again. Feb 28 20:49:01 mreimer: the voodoo is in composite-engine.c Feb 28 20:49:05 thanks Feb 28 20:49:33 mallum: quick question for you: is it intentional that mb-keyboard looks in ~/matchbox instead of ~/.matchbox? Feb 28 20:49:55 mreimer: no, that sounds like a bug Feb 28 20:50:11 bleh. Feb 28 20:50:27 hey hey kergoth Feb 28 20:50:44 morning kergoth Feb 28 20:50:50 hey Feb 28 20:51:48 been messing with pcap and libnet a fair bit lately, and found http://netdude.sourceforge.net/index.html .. pretty slick. libraries for navigation and manipulation and creation of pcap format dump files Feb 28 20:51:52 heh Feb 28 20:52:10 cute Feb 28 20:52:45 quite. useful on occasion to be able to take your packet construction and dump it directly into a file for analysis rather than putting it on the wire Feb 28 20:52:54 kergoth: did you see that vhci over ip project? Feb 28 20:53:09 nope, /me googles Feb 28 20:53:13 http://usbip.naist.jp/ Feb 28 20:53:30 ooh, neat Feb 28 20:57:39 n8 all Feb 28 21:04:56 what's the current best open source soultion for text to speech? is festival still the best bet? Feb 28 21:06:17 kergoth: yeah, festival is the only real option Feb 28 21:06:53 I'd really like a halfway decent speech recognition thing for some home control stuff now that I'm buying a house Feb 28 21:07:09 festival doesn't really look maintained :\ Feb 28 21:07:44 kergoth: maybe it's complete ;) Feb 28 21:07:49 * kergoth rolls eyes Feb 28 21:08:59 yeah, 18 months since last beta release Feb 28 21:12:00 mreimer: fixed the ~/.matchbox thing. should probably fix it further to use XDG_DATA_DIR or somesuch Feb 28 21:12:15 thanks mallum Feb 28 21:12:58 mallum: I'm not able to type '<' with mb-kb using the default layout. Does it work for you? Feb 28 21:13:07 I get '>' instead. the .xml looks right Feb 28 21:13:27 mreimer: one sec Feb 28 21:14:24 mreimer: hmm works for me Feb 28 21:15:04 gah Feb 28 21:15:08 * CosmicPenguin throws things at OE Feb 28 21:15:15 mallum: ok, I'll try building a newer version. I was also having problems with a couple of other characters but tweaking the .xml fixed it. Maybe it's my locale Feb 28 21:17:49 * CosmicPenguin can't figure out why glibc-intermediate is building Feb 28 21:19:16 CosmicPenguin: do you have another glibc that will provide libc-for-gcc? Feb 28 21:19:21 mreimer: okey, well its pretty new so probably a few buglets. please send any patches for tweaks you make Feb 28 21:19:29 pb_: maybe not Feb 28 21:19:37 mallum: will do. I really like mb-kb Feb 28 21:19:40 I just couldn't figure out where the connection was being made Feb 28 21:20:54 mreimer: I plan to rewrite the compositor in m-w-m-II btw. the plan with the code so far is to definetly structure it so it makes things like compositors easier to plug in Feb 28 21:21:32 mallum: that's good to hear. Sure makes things pretty. And it's nice to show up the PocketPC folks :-) Feb 28 21:21:55 mallum: is that on your to-do list for the near-, middle-, far-term? Feb 28 21:22:35 * CosmicPenguin figured it out Feb 28 21:22:58 mreimer: middle/far I guess. The problem is I dont know of any handheld hardware thats really up to running composite well Feb 28 21:23:46 mallum: does the n770 have blending hardware, or just yuv conversion? Feb 28 21:24:41 hey gang. what's the "right" way to get files included in a package? I've listed the directories that appear in the image/ hierarchy in FILES_${PN}, but this doens't seem to work unless i'm quite specific about the actual file, rather than the path to a dir i want picked up. Feb 28 21:25:04 mreimer: just yuv conversion *I think* dont quote me Feb 28 21:25:47 mallum: the other day I stumbled across Intel's IPP library of go-fast code. I wonder if that would help make composite useable on existing hardware? Feb 28 21:26:39 hmm, anyone have a current 'dontdiff' file? (for diff -X, lists the generated files in a kernel tree to exclude from the diff) Feb 28 21:26:42 ~dontdiff Feb 28 21:26:44 extra, extra, read all about it, dontdiff is http://movementarian.org/dontdiff Feb 28 21:26:47 heh Feb 28 21:26:49 ibot: thanks Feb 28 21:26:49 kergoth: my pleasure Feb 28 21:27:05 aw, 404 Feb 28 21:27:26 mreimer: I've heard a rumour that nico said he could beat IPP Feb 28 21:27:53 mreimer: its not just speed but memory too. current render 'wastes' quite a bit of memory on a 16bpp display Feb 28 21:28:08 koen|tv: competition is good :-) Is he planning to make it available? Feb 28 21:28:18 mallum: yeah I'd bet Feb 28 21:28:29 koen|tv: he should put it into liboil :) Feb 28 21:28:34 yes Feb 28 21:28:45 there's no ARM code in liboil yet, is there? Feb 28 21:28:47 kergoth: http://www.moses.uklinux.net/patches/dontdiff Feb 28 21:28:58 mreimer: not yet Feb 28 21:29:34 mreimer: part of his stuff should be in gcc now for softfloat Feb 28 21:29:40 hmm, according to googling, dontdiff is in DOcumentation/ in current kernel trees Feb 28 21:29:42 * kergoth looks Feb 28 21:29:56 koen|tv: excellent Feb 28 21:30:19 Does anyone know how I can create a 'clean' copy of a checked out git repository so that all of the git files don't exist? I'm sure you can do it, like I'm sure you could with CVS, I just can't work out how to do it. Feb 28 21:30:43 koen|tv: the IPP libraries also include code to do fast JPEG, MP3, and video decoding, which I doubt Nico's working on Feb 28 21:30:56 * RP just remembers the arm website links to dontdiff... Feb 28 21:32:12 http://www.kernel.org/git/?p=linux/kernel/git/torvalds/linux-2.6.git;a=blob_plain;h=24adfe9af3ca92a169b9f7a7fc368e9098dac487;f=Documentation/dontdiff Feb 28 21:32:16 heh Feb 28 21:32:50 Offhand, does anyone know how to disable the touchscreen under opie and make it use a mouse? Feb 28 21:33:15 I can confirm the qemu framebuffer colour corruption is in qemu... Feb 28 21:33:28 kergoth: Ok, cool thanks Feb 28 21:34:22 ashak: i wasnt asnwering your question, that was regarding my search for a current version of dontdiff Feb 28 21:34:31 ashak: you're looking for git-tar-tree, I believe Feb 28 21:34:33 RP: its the QWS_MOUSE env var, but i cant recall the format of it Feb 28 21:34:43 i'd check the qt/e kernel code Feb 28 21:34:44 kergoth: Oh :) Feb 28 21:35:13 CosmicPenguin: Thanks :) Feb 28 21:35:15 in-kernel qt/e? eleet Feb 28 21:35:28 or very very wrong Feb 28 21:36:21 heh, qt/e's kernel Feb 28 21:36:27 qt/src/kernel/ Feb 28 21:36:49 Is it possible to control Opie without a touchscreen? Maybe with just keypad presses? Feb 28 21:37:15 koen|tv: wow I finally made a new release of one of my softwares ;) Feb 28 21:37:24 mallum: yay! Feb 28 21:37:48 kergoth: heh Feb 28 21:38:06 ashak: maybe, at least QtE has a keypad mode Feb 28 21:38:25 ashak: it should be possible with a mouse. Feb 28 21:38:31 * RP pokes around qte... Feb 28 21:38:39 RP: I don't have a mouse, I only have a keypad :) Feb 28 21:39:20 ashak: The award wining Qtopia Phone runs on QtE. With the QtE keymode you can navigate through widgets etc... Feb 28 21:39:27 zecke: If it does that will be cool, as I can get Opie to run, but I can't get past the touch the screen to continue screen Feb 28 21:39:33 zecke: Ah cool Feb 28 21:39:52 ashak: I have never tried it on Opie though, but obviously patches are welcome Feb 28 21:40:15 zecke: Heh... patches it where it may become a problem ;) Feb 28 21:40:52 zecke: Any idea where I can start looking for information on the subject? Feb 28 21:41:05 ashak: you could fix tslib or your driver as well? (to make the touchscreen work) Feb 28 21:41:11 ashak: ./configure --help of QtE Feb 28 21:41:19 zecke: The device doesn't have a touchscreen :( Feb 28 21:42:12 zecke: Which is why I enquired about controlling it with a keypad Feb 28 21:43:38 make sure you do not have QPE_NEED_CALIBRATION defined Feb 28 21:43:53 ljp: hey Feb 28 21:44:02 hi Feb 28 21:44:20 ljp: summer is ending in aussie land? Feb 28 21:45:09 ljp: how does QtE know which keys to use for what kind of actions (switching focus between widgets) Feb 28 21:45:16 ljp: are these documented? Feb 28 21:46:44 $QPEDIR/etc/defaultbuttons.conf Feb 28 21:47:07 hmm. not it really Feb 28 21:47:13 grep -rni getenv("QPEDIR") * | wc -l in QtE results in 0 Feb 28 21:48:20 zecke: probably need to look at the hardcoded keymap for the keyboard in question in the QWS keyboard handler. check the source Feb 28 21:48:24 heh Feb 28 21:48:29 see what qt keycodes they map to? Feb 28 21:48:30 keypad mode is really only specific to phone Feb 28 21:48:38 er no, you want higher level than that Feb 28 21:48:39 * kergoth shrugs Feb 28 21:49:08 just for completeness... my device is a phone Feb 28 21:49:41 kergoth: right, I wanted to point ashak to some kind of document :) Feb 28 21:49:57 look in src/kernel/qkeyboard_qws.cpp Feb 28 21:49:57 ashak: you might consider licensing the award winning Qtopia Phone then? Feb 28 21:50:06 hehe Feb 28 21:50:31 zecke: Ah, it's not free? Feb 28 21:50:42 not at this time Feb 28 21:50:43 ashak: sadly not Feb 28 21:50:55 ashak: QtE keyboard mode is free though! Feb 28 21:50:55 bummer :( Feb 28 21:51:05 keypad Feb 28 21:51:24 ashak: right keypad mode (not keyboard well that is free as well) Feb 28 21:51:54 ok... I think i'm slightly more confused right now... but at least i'm making progress ;) Feb 28 21:51:57 ashak: a lot of Phone implementations in Qtopia are free as well (due #ifdef QTOPIA_PHONE #else QTOPIA_PDA #endif constructs) Feb 28 21:52:23 you wont get the phone libraries Feb 28 21:52:26 ashak: try compiling QtE with keypad mode, compile a simple application make that work Feb 28 21:52:52 ljp: well the ContextManager looks like it could be implemented quite easily :) Feb 28 21:53:05 but that is getting Off Topic here Feb 28 21:53:33 * ljp looks around for the irc police Feb 28 21:53:45 zecke: compile it manually? Feb 28 21:54:17 ashak: if you use OE you could adjust the bb file Feb 28 21:54:28 zecke: Cool, that's what I though. Feb 28 21:54:28 ashak: I figure you have crated a 'machine.conf' for your machine? Feb 28 21:54:37 zecke: Sort of... Feb 28 21:55:11 zecke: It may not be complete, but I have something. Feb 28 21:55:48 ashak: you could do something like this EXTRA_OECONF_CONFIG_ashaksthing = "-qconfig qpe -other-important-options" in the qte_2.3.10.bb file Feb 28 21:56:06 ljp: how is compositing performing in QtE4? Or was this removed again? Feb 28 21:58:51 composting? ;) Feb 28 21:59:26 not real sure, but 'qtopia core' needs more optimizing Feb 28 22:00:19 ljp: I thought I had viewed code that double buffered the complete window in framebuffer memory Feb 28 22:00:35 ljp: and all these double buffered windows was composited into the viewable area Feb 28 22:00:43 I might have been on crack - again Feb 28 22:03:50 i think they fixed the triple buffering, but am not too sure Feb 28 22:05:18 so.. without having qte to play with manually, is there an obvious place to look for the important options are available? Feb 28 22:05:19 oh well I just turned in my first drafts of the paper and presentation... I need to investigate more time on fixing them up Feb 28 22:05:36 ashak: vim ./configure Feb 28 22:05:46 ashak: search for keypad in it, or just do ./configure --help Feb 28 22:06:08 ljp: we have missed you at FOSDEM, where have you been? Feb 28 22:06:23 zecke: So really I need to extract the qte source somewhere? Feb 28 22:07:27 ashak: or switch into the qte directory in tmp/work/* Feb 28 22:08:52 here in brisbane. Feb 28 22:10:27 there were other trolls there Feb 28 22:10:37 I have 'met' simon Feb 28 22:10:53 the good ole sisihaus Feb 28 22:11:41 * ashak waits for qte to download Feb 28 22:12:50 did someone take notes at the embedded room? Feb 28 22:13:32 ljp: I think we didn't Feb 28 22:13:51 zecke: michaelo should have videos of all the presentations Feb 28 22:13:56 argh Feb 28 22:13:58 pright Feb 28 22:14:00 right Feb 28 22:14:04 stupid glibc-intermediate Feb 28 22:14:35 CosmicPenguin: isn't building toolchains fun? Feb 28 22:14:59 I can understand why it thinks 4.0.2 is a bad version, but I don't understand why this hasn't been breaking before Feb 28 22:16:08 gcc 4.0.2 that is Feb 28 22:16:09 Ah... -keypad-mode Feb 28 22:18:10 * ashak tries it Feb 28 22:19:10 hmm, there's keypad-input too Feb 28 22:19:18 RP: is your LED patch going to make it into mainline? Feb 28 22:22:25 mreimer: I hope so Feb 28 22:22:57 RP: me too. I haven't heard a peep about the kexec patch on l-a-k. Feb 28 22:23:02 * Philippe is away: visual contact - melancholic dreams Feb 28 22:23:36 good nite guys! Feb 28 22:24:00 zecke: night, thanks for the help Feb 28 22:24:20 you are welcome Feb 28 22:38:43 mreimer: I noticed that. Very strange Feb 28 23:05:17 Thanks for the help guys Feb 28 23:22:19 So, is it bad to claim gcc-cross-initial as 3.4.4 to make glibc-intermediate happy, even if I'm compiling a 4.0.2? Feb 28 23:22:38 nite all Feb 28 23:24:07 CosmicPenguin: if you're going to do gcc4, do it all the way through Feb 28 23:25:35 except that the current version of glibc-intermediate doesn't know anything about 4.X Feb 28 23:26:06 * CosmicPenguin should pay more attention in class Feb 28 23:40:21 does anyone have a url for 3.5.4 test images...? Feb 28 23:54:31 hrw might Feb 28 23:55:33 by the way, 760 should be better supported in 3.5.4 than most platforms :) Feb 28 23:55:54 3.5.4.1 is going to refine the port for cxk machines Feb 28 23:56:06 and older ones are gradually going to die off and be less supported Feb 28 23:56:26 C700/C750/C760/C860 are pretty similar in overall capability Feb 28 23:58:21 btw, if you want to be an official tester, send mail to openembedded@hrw.one.pl Feb 28 23:58:51 but the images are in: http://ewi546.ewi.utwente.nl/tmp/hrw/oz3.5.4-test-images/test5/ Feb 28 23:59:32 i sent an email to test354 or something.... it was an addy hrw put in a post on the 3.5.4 thread... i guess ill try the other one too Feb 28 23:59:32 not sure if its up to 2.6 kernel yet (maybe!) Mar 01 00:00:11 3.5.4.1 will probably have 2.6 updates; but that wont be out for awhile Mar 01 00:02:40 thank you Mar 01 00:04:59 raduga do you know if the oz/opie port of gaim has made it into the test feed? Mar 01 00:05:54 ahhh yes here it is Mar 01 00:15:50 ok wish me luck on flashing... Mar 01 01:05:34 Is the kernel in .dev known to be broken? Mar 01 01:21:05 NA|Zzz: which device and kernel? there are many kernels? Mar 01 01:22:21 JustinP: Yeah.. should've specified that.. it's too early :) borzoi (3100), kernel 2.6.15 Mar 01 01:23:24 ummm...not that I know of...IIRC it should build Mar 01 01:24:08 It builds.. I'm just getting oopses on the console relating to APM Mar 01 01:24:15 Let me boot it again.. Mar 01 01:26:00 wow ;D NA|Zzz can type while sleeping !! Mar 01 01:26:29 decca_work: Yeah, I call it work :) Mar 01 01:26:34 Weird... it's booting this time Mar 01 01:26:35 lol Mar 01 01:40:05 koen: ping Mar 01 01:54:19 njs: I feel I can point people to that e-mail :-) Mar 01 01:55:20 JustinP: oh, cool :-) Mar 01 02:20:03 * emte is curious how familar is going to become more "user focused" without more developer focus to fix all the problems ... Mar 01 02:39:33 * france is away: Away Mar 01 02:39:53 * france is back (gone 00:00:05) Mar 01 03:10:36 hi, folks. is the any soft float toolchain for OE ? Mar 01 03:14:21 I believe soft floats are used on targets requiring it.. Mar 01 03:23:40 NAbyss, what do you mean ? Mar 01 03:29:13 not all cpu have floating point registers Mar 01 03:29:17 so they need soft float Mar 01 03:31:40 ok. and what is VFP ? Mar 01 03:32:29 emte, and how does the linux kernel float emulator working ? Mar 01 03:46:35 * emte wonders which question that was supposed to be ... Mar 01 04:01:26 leoncamel: i am guessing VFP means virtual floating point? Mar 01 04:48:20 * france is away: Away Mar 01 04:58:49 NAbyss: http://oe.pastebin.com/577917 Mar 01 04:59:12 my notes on the cxx00 keyboard configuration possibilities Mar 01 05:00:17 NAbyss: it's quite beautiful ;) Mar 01 05:00:29 Bloody hell.. Mar 01 05:02:45 Looks rather complete Mar 01 05:09:31 NAbyss: it's genius if you're "a westerner" (american) Mar 01 05:09:59 i preserve places for all sorts of english keys, foregoing the non-english ones =) Mar 01 05:12:07 like i even manage to squeeze in numlock and crap Mar 01 05:22:22 Yep :) Mar 01 05:25:49 scroll lock was needed because of SysRq Mar 01 05:28:18 who is Jenna Mar 01 05:32:16 jnc: ? Mar 01 05:32:29 keep popping in/out of channel Mar 01 05:32:34 i am curious about that Mar 01 05:32:47 hundreds of people go in and out of channels all day Mar 01 05:32:53 you expect us to know them all? Mar 01 05:32:54 who cares? Mar 01 05:48:55 * emte thinks kergoth secretly cares and has a hitlist Mar 01 05:49:01 :P Mar 01 05:49:29 *laugh* Mar 01 05:49:30 * johnX always thought that everyone knew everyone else on IRC Mar 01 05:49:44 johnX: it used to be that everyone did, in like 1992 Mar 01 05:49:53 i do Mar 01 05:50:35 every one is the pervert fat guy named bubba sitting in jail trying to pick up little kids, selling viagra, and is ebay's #1 seller Mar 01 05:51:14 ebay's *only* real seller Mar 01 05:51:32 lol Mar 01 05:51:42 ...is selling haunted pieces of french toast in the likeness of jesus Mar 01 05:52:01 lol Mar 01 05:52:17 the whole thing just consists of him laundering money through thousands of fake ebay accounts, and every so often ripping people off who believe in the reseller ratings Mar 01 05:52:20 hey are any of you capable of building openzaurus SharpROM kernels? Mar 01 05:52:48 anyone who has OE installed can Mar 01 05:52:58 i mean, one that will work with SharpROM Mar 01 05:52:59 just follow the instructions Mar 01 05:53:09 i'm trying to verify a bug Mar 01 05:53:20 my C3000 does not resume/suspend properly Mar 01 05:53:24 anyone who has OE installed can just follow the instructions the Sharp instructions are there Mar 01 05:53:28 RP's C3000 does resume/suspend properly Mar 01 05:53:36 so we're like, derr.. what's going on Mar 01 05:53:42 oh Mar 01 05:53:57 emte: i have amd64 build host, going to need ia32 or someone else to do it for me :( Mar 01 05:54:21 why not just ask RP hof his image? Mar 01 05:54:27 for* Mar 01 05:54:33 his SharpROM image? Mar 01 05:54:41 yea, he's kind of short on time Mar 01 05:54:45 it would verify where the problem is Mar 01 05:54:49 oh Mar 01 05:55:02 eitehr its image based or hardware Mar 01 05:55:05 i should mention that SharpROM appears to suspend/resume fine on my hardware Mar 01 05:55:14 so like i'm trying to figure out what the heck is going on Mar 01 05:55:27 i would hate to think that the reason it works is the closed-source SD module Mar 01 05:55:29 holy crap Mar 01 05:55:31 the SD module! Mar 01 05:55:45 that would be a difference hmm... Mar 01 05:56:51 anyhow RP raises a good point, which is that unless you actually build the kernel yourself, it's not going to be clear whether the binaries Sharp released will be equal to the source code supposedly used to make those binaries Mar 01 05:59:24 did sharp release their toolchain as well? Mar 01 06:00:16 hmm Mar 01 06:01:42 ooOoh http://kopsisengineering.com/kopsis/SharpZaurusSdkDsl Mar 01 06:03:50 he links to OE :P Mar 01 06:08:59 that is rather handy Mar 01 07:11:21 hi all! Mar 01 07:47:34 03mwester 07org.oe.dev * r28f5897f... 10/packages/linux/ (3 files in 2 dirs): unslung-kernel: added kernel module netconsole Mar 01 07:47:42 03mwester 07org.oe.dev * r6f4af617... 10/ (6 files in 4 dirs): unslung-rootfs: unsling fix, empty pcmcia dir fix - unsling script fixed for no-automatic mount, updated README and NOTES, change to remove empty pcmcia dir Mar 01 07:47:43 03mwester 07org.oe.dev * ra0581b19... 10/packages/linux/unslung-kernel_2.4.22.l2.3r63.bb: unslung-kernel: add missing netconsole patch file Mar 01 08:37:29 <_law_> hmm how can i get my private key from monotone db? Mar 01 08:38:21 <_law_> cant commit anything to OE Mar 01 08:45:10 morning Mar 01 08:45:23 _law_: monotone --help Mar 01 08:45:51 _law_: monotone privkey KEYID Mar 01 08:49:41 ~hail offlineimap Mar 01 08:49:45 * ibot bows down to offlineimap and chants, "I'M NOT WORTHY!!" Mar 01 08:50:57 <_law_> hrw|work, and how can i checkout what KEYID i have got? Mar 01 08:51:13 <_law_> monotone privkey law@openembedded.org doesnt work Mar 01 08:52:05 _law_: monotone keys Mar 01 09:00:58 <_law_> hrw|work, i dont have got an privkey :-( Mar 01 09:02:59 _law_: so if you lost your privkey then you will need to generate new one and discuss with koen+mickeyl Mar 01 09:04:56 morning Mar 01 09:05:45 hi XorA Mar 01 09:06:05 hrw|work: offlineimap rules my Z Mar 01 09:06:19 XorA: What client? Mar 01 09:06:43 hrw|work: btw if you import python-2.4.2-manifest.inc and offlineimap_4.0.11.bb into the branch you will get fully functional for oz3.5.4 Mar 01 09:06:51 NAbyss: muttng Mar 01 09:08:05 <_law_> koen, i noticed that i lost my private key (had an hd-crash 2 month ago), could you remove my existing key from monotone? Mar 01 09:08:43 morning all Mar 01 09:08:46 bad _law_ you disobeyed the rule about locking your key in a fire proof safe and copying it to 1000 machines for safe keeping :-) Mar 01 09:08:52 RP: morning Mar 01 09:09:20 good morning all Mar 01 09:09:36 RP: tf rocks Mar 01 09:09:44 hi RP koen Mar 01 09:10:24 XorA: http://pastebin.com/578068 Mar 01 09:10:27 hey hrw|work Mar 01 09:10:57 _law_: we can't actually remove keys from the db, so you have generate a key with a different name Mar 01 09:11:14 <_law_> koen, ok Mar 01 09:11:22 koen: sweet, abiword likes us Mar 01 09:11:56 koen: I was rather amused when you started complaining about abiword in the presentation as I knew he might fix things :) Mar 01 09:12:16 <_law_> XorA, so you have to wait a little bit longer for tinymail update,or should i post .bb + gtk-no-dox stuff on my webserver ? Mar 01 09:12:26 morning @all Mar 01 09:12:31 _law_: I fixed my self Mar 01 09:15:01 <_law_> XorA, ok perhabs you could post it to OE ? Mar 01 09:15:15 _law_: if you want, I could do that later Mar 01 09:18:01 RP: the linux-openzaurus.bb had a nice effect on the audience ;) Mar 01 09:18:52 koen: Yes. It is getting rather scary atm. I was told I was working too hard :) Mar 01 09:19:59 koen: I thought it all went very well and presented OE in a good light in a manner that people found find useful. I know tf found it a useful intro (as someone who Mar 01 09:20:04 's no used OE yet) Mar 01 09:22:28 RP: at Gulev tigrux said that complex subjects like this are better handled with interactivity Mar 01 09:23:18 koen: Was there anything special you needed to do to get hald running btw? hal-device-manager appears to want pygnome - is that in OE? Mar 01 09:24:19 it isn't in OE, but it's not hard to get it to make an ipkg Mar 01 09:24:33 it is hard to add it to OE in a clean way Mar 01 09:25:07 koen: I thought both the style and content worked well. It complemented mickeyl's presentation which aims to show different things (assuming I got the feel from his slides as I didn't see it in person) Mar 01 09:25:07 at least for a complete python novice as me Mar 01 09:26:38 I guess I can't describe myself as a complete python novice now... :) Mar 01 09:27:54 /cr Mar 01 09:29:30 <[lala]> hi guys Mar 01 09:29:45 <[lala]> RP: any luck with that asoc patch yet? Mar 01 09:46:20 [lala]: not had a chance yet. I'll aim to do that shortly Mar 01 09:50:06 <[lala]> RP: that would be great :-) thx in advance! Mar 01 10:00:29 koen: can you see if vanille is running please Mar 01 10:03:02 ade|desk: it is, I'm pushing to it right now :) Mar 01 10:05:28 03koen 07org.oe.dev * ra277efc6... 10/packages/vnc/vnc_3.3.7.bb: Add realvnc 3.3.7, courtesy Raymond Danks Mar 01 10:05:34 hi Mar 01 10:09:10 hi obergix[work] Mar 01 10:13:28 hi lrg Mar 01 10:14:28 hey greentux_alt2 Mar 01 10:15:03 lrg: i am missing a usb adapter in the borzoi case. before i start searching the office... Mar 01 10:15:15 lrg: do you forgot it? Mar 01 10:16:04 * lrg looks Mar 01 10:17:51 greentux_alt2: is it about 10cm long ? Mar 01 10:17:57 lrg: yes... Mar 01 10:18:14 greentux_alt2: found it my laptop bag Mar 01 10:18:27 lrg: is it possible for you to send it asap to a adress i mail you? Mar 01 10:18:38 ok Mar 01 10:18:40 lrg: simple in a letter case Mar 01 10:18:45 lrg: tnx Mar 01 10:20:17 lrg: mail sent Mar 01 10:41:06 ok cheer koen Mar 01 10:41:15 cheers even Mar 01 10:48:57 hi, folks. I just bitbake bootstrap-image. And, the /lib directory is empty. So. where is wrong ? Mar 01 10:49:49 And the "arm-linux-ldd busybox" indicate that it is really dynamicly linked. Mar 01 10:49:57 So. What shall I do ? Mar 01 10:51:34 leoncamel: you used own toolchain? Mar 01 10:51:49 hrw|work, yes. Mar 01 10:52:03 leoncamel: then copy all needed libs by hand Mar 01 10:52:50 leoncamel: bitbake did not have to build libc, libgcc1 etc so it does not package them and your packages does not depend on them Mar 01 10:53:45 is there a *good* reason to use your own toolchain with OE? Mar 01 10:54:37 hrw|work, So. which libs should I copying ? Mar 01 10:56:13 koen: heya Mar 01 10:56:26 hey njs Mar 01 10:56:52 hrw|work, I mean, how many libs should I copy to /lib in the new rootfs ? Mar 01 10:57:07 leoncamel: as many as ldd says are needed Mar 01 10:57:43 XorA, ok. Mar 01 10:58:22 koen: just wanted to get back to you about the nasty merge Mar 01 10:58:35 njs: I suspected as much **** ENDING LOGGING AT Wed Mar 01 10:59:56 2006