**** BEGIN LOGGING AT Tue Aug 15 02:59:58 2006 Aug 15 06:36:52 'morning Aug 15 07:43:22 good morning Aug 15 07:51:34 florian_kc: I got upto gdb-cross when building task-sdk Aug 15 07:51:47 umm, meta-sdk that is Aug 15 07:55:45 http://www.linuxdevices.com/news/NS8030785497.html Aug 15 07:55:46 lol Aug 15 08:02:07 morning Aug 15 08:02:44 morning all Aug 15 08:03:34 RP : Morning. Found some free time and i am truing to sync owmnr to the latest OE Aug 15 08:04:06 RP: If nothing major comes up i will post the patches today and tommorow Aug 15 08:05:08 Ifaistos: Sounds good! :) Aug 15 08:07:39 guys i am thinking of seting up a "cluster" of 10-15 machines to increase compile speed at the office. I am trying to think if there is a way to make it availiable to OE developers also. Any ideas/thoughts ? Aug 15 08:08:24 zecke: ehh... the most obvious thing is that they must have spent a huge amount of money in this architecture graphic. that one is in *every* qt/e related article... Aug 15 08:17:08 Ifaistos: even without making it available to the OE developers, such a cluster can be of great help to ensure a better OE quality, I mean spotting build regression - i.e. performing tests and doing clobber and 'from scratch' build of OE packages and distributions and to report the builds and tests status to tinderbox Aug 15 08:18:40 unless someone think of a better use of such power ... Aug 15 08:19:32 i was thinking of setting up a Xen machine with accounts for the developers and from there they could have access to the build system Aug 15 08:19:46 florian_kc: yes Aug 15 08:19:59 florian_kc: but is it Qtopia(Core) not QtE Aug 15 08:20:21 i will be upgrading the dsl line to dsl2+ (24Mbits/3Mbits) so i think the bw will be there also Aug 15 08:20:45 zecke: well... Aug 15 08:26:32 Ifaistos: ok. so I guess a simple ssh access + rights to send report to tinderbox is enough :) Aug 15 08:29:31 cyrilRomain: Yes something in this context. I am thinking of using 2.5Ghz Celerons with 1Gig Ram on each box. Any idea what distro would be better suited for such a task ? Aug 15 08:29:37 03florian 07org.oe.oz354x * r27678a75... 10/ (3 files in 2 dirs): gpe-conf: add 0.2.2 Aug 15 08:34:47 Ifaistos: any distro can be used as soon as you can easily install the OE required software. I would use Debian or Gentoo though. Aug 15 08:35:32 * cyrilRomain feel sick when thinking about rpm-based distro Aug 15 08:42:32 Ifaistos: or if you feel it, you could create a Knoppix-based OE liveCD, so that you just have to burn 1 cd per machine (no install required) :) Aug 15 08:43:19 i.e. a Knoppix liveCD with OE required software Aug 15 08:44:24 morning all Aug 15 08:44:46 i was thinking for making the machines netboot but it seems that it will effect the throuput if no local hd is present, and if you end up with a HD on every machine then no need to have a cd also Aug 15 08:44:51 morning Aug 15 08:45:22 hi lrg Aug 15 08:49:34 hey lrg Aug 15 08:53:57 hi Liam Aug 15 08:54:10 lrg: I'ev not got around to doing the ASoC testing yet, sorry Aug 15 08:54:36 RP: np Aug 15 08:55:14 RP: I think we may have to set pop_time = 0 for corgi due to the board noise :( Aug 15 09:00:12 lrg: I think that's looking like the likely result :-/ Aug 15 09:00:46 RP: at least they fixed it in spitz :) Aug 15 09:00:55 lrg: yes :) Aug 15 09:01:29 RP: Dell still cant fix board noise. Even my new PC is noisy when I move the mouse :( Aug 15 09:05:05 * lrg has finally upgraded to mt 0.28 and is impressed by it's speed Aug 15 09:05:48 mickey|thesis: ping Aug 15 09:05:54 mickey|thesis: The greenphone is funny Aug 15 09:06:04 hail zecke Aug 15 09:06:19 mickey|thesis: TT advertises Qtopia with its Openness Aug 15 09:06:30 lrg: Built in sound cards never seem to work well for me :-/ Aug 15 09:12:10 ljp: Ship me a greenphone and I do not mean the PSD's :) Aug 15 09:25:43 is there any web repository for oe? Aug 15 09:26:58 katossi: See links on openembedded.org Aug 15 09:32:21 zecke: Do you have any example of using the icecream.class ? Aug 15 09:33:44 echo 'INHERIT += "icecc"' >> conf/local.conf Aug 15 09:33:58 but I think it will fail to pack the toolchain together Aug 15 09:35:08 Scary thought for the day - if we have a set of identical machines, the multithreading code could run the tasks on different machines. There's some scary possibilities there Aug 15 09:36:21 RP: Think of Xen Aug 15 09:36:27 RP: then we all would have the same machine! Aug 15 09:37:16 zecke: With a rootfs built by OE to build within? :) Aug 15 09:37:41 and the next generation of man-kind will ask Aug 15 09:38:01 "How did they solve the Chicken-Egg problem. One needs bitbake to create bitbake?" Aug 15 09:38:34 zecke: and can it not go faster? :) Aug 15 09:38:41 hehe Aug 15 09:39:00 RP: bastard ;) Aug 15 09:56:47 zecke: thanks for the tip. I just came back from vacations and I'm blind Aug 15 09:57:00 03pb 07org.oe.dev * r42a6f628... 10/ (1 packages/gcc/gcc-cross-initial_3.3.4.bb): gcc-cross-initial 3.3.4: clobber PACKAGES to avoid shlibs imbroglio Aug 15 10:10:25 RP: I am trying to build bash-3.0 and it failed, while trying to figure out what is wrong i noticed that the bash-3.0-fixes patches where not applied because the bash-3-0/lib/intl source files are unpacked as read only... Aug 15 10:14:37 RP : -r--r--r-- 1 stelios stelios 1856 Dec 9 2003 dcgettext.c Aug 15 10:15:08 is there any other site except pastebin.com ? its TOO slow the last few days Aug 15 10:19:05 RP : http://pastebin.com/768873 Aug 15 10:25:28 Ifaistos: you can use http://rafb.net/paste/ Aug 15 10:29:27 RP: I unpacked the tar file manualy and it seems that the problem is there. the lib/intl dir files are all set to readonly Aug 15 10:36:39 Ifaistos: You could add a task between unpack and patch which makes them all writeable? Aug 15 10:39:51 RP: Not sure how i should do it :( post_unpack () ? Aug 15 10:41:55 Ifaistos: Add a do_make_writeable() { chmod a+w somepath/* }, then "addtask mark_writeable before do_patch after do_unpack" Aug 15 10:42:02 s/mark/make/ Aug 15 10:45:35 mornng Aug 15 10:50:57 hey Graeme Aug 15 10:51:10 hey lrg Aug 15 10:57:57 hi XorA Aug 15 11:13:56 RP: I think the problem with the bash-3.0 is the way the patches were grouped together after my initial patch. Aug 15 11:14:36 RP: The initial "fixes" for bash-3.0 had a pnum=2 Aug 15 11:14:59 RP: While the other patch that causes the build failure has no pnum... Aug 15 11:15:16 RP: and these were grouped together Aug 15 11:16:39 Ifaistos: pnum should have no effect on whether the files are writeable? Aug 15 11:17:37 RP: I know but the funny thing is that the dir was not writable also before... but the patch worked... Aug 15 11:18:03 Ifaistos: You weren't running as root were you? Aug 15 11:18:36 RP: Nop i am using Ubuntu... so no root here ;) Aug 15 11:19:20 Ifaistos: Presumably quilt is clever enough to make the file writeable to apply the patch then? Aug 15 11:20:11 RP: Not sure i will try it again, using the original patches and see if it works or not Aug 15 11:20:40 RP: pretty sure it is clever enough indeed. sets it writable, then puts it back after applying the change Aug 15 11:20:54 * kergoth kicks his build in an attempt to encourage it to go faster Aug 15 11:21:10 hi kergoth Aug 15 11:21:15 hey Aug 15 11:21:16 * chouimat wants this http://www.linuxdevices.com/news/NS8030785497.html Aug 15 11:23:27 kergoth: I've seen quilt shoot itself in the foot before with file modes. Can't remember exactly what and I think a later version fixed it... Aug 15 11:23:51 heh, wouldnt surprise me, that area of it never seemed entirely polished Aug 15 11:30:49 Hello. Aug 15 11:41:25 hi sirfred Aug 15 11:41:36 RP: Hello. Aug 15 11:42:04 sirfred: Did you have any further success with the alpha blending? Aug 15 11:42:04 mr_nice_slacker1: with the bostmodified 2.4.x bootldr i could check my ram right? Aug 15 11:42:17 RP: Yes, I have it working. Aug 15 11:42:43 RP: At least with an 4BPP alpha plane. Aug 15 11:42:57 RP: And also with a global alpha value. Aug 15 11:43:35 RP: I'm now trying to do a StretchBlt of a surface in host memory, in YUV format to an overlay showing card surface. Aug 15 11:43:38 sirfred: excellent :) Aug 15 11:43:56 RP: Still some problems, because StretchBlt only support RGB surfaces. Aug 15 11:44:18 RP: But it seems that it can be done using two StretchBlt operations, one for the Y channel and other one for the UV channels. Aug 15 11:44:28 RP: Each one using 8BPP. Aug 15 11:44:55 sirfred: Interesting. That could start to give a good speed improvement for video playback :) Aug 15 11:45:00 RP: I want to setup an overlay + scale for YUV using the new library. Aug 15 11:45:09 which library? Aug 15 11:45:20 x29a: I new one I'm developing. Aug 15 11:45:28 im looking for a fast image processing library, not for oe tho Aug 15 11:45:28 x29a: A new one I'm developing. Aug 15 11:45:44 sirfred: any publications yet? Aug 15 11:45:52 sirfred: You have the overlay working already? or you're using stretchbt to implement it? Aug 15 11:45:57 x29a: Still not, I'm thinking about it. Aug 15 11:46:04 RP: Overlay working. With RGB and YUV Aug 15 11:46:21 RP: Well, I've tested it with YUV422 and works fine. Aug 15 11:46:32 RP: Also with RGB565 Aug 15 11:46:38 so we can accelerate video playback :) Aug 15 11:46:39 RP: But there're some strange things. Aug 15 11:47:04 RP: When using RGB surfaces, overlay honours the video_portrait bit in VIDEO_CTRL Aug 15 11:47:53 RP: But I was not able to put it to work using YUV. So, we cannot rotate the surface using hardware. Aug 15 11:47:53 sirfred: You really seem to have an understanding of how the chip works now :) Aug 15 11:48:09 RP: Well, I've learned a little. :) Aug 15 11:48:28 RP: I'm thinking about giving the current library public access. Aug 15 11:48:44 RP: It's not using any microcode, neither any part directly copied from the reverse engineered work. Aug 15 11:49:07 sirfred: Sounds like it might be possible to release it then Aug 15 11:49:08 RP: Still a lot of work to do, of course. But perhaps some other people gets involved. Aug 15 11:49:17 Other interest would be good Aug 15 11:49:28 w100? Aug 15 11:49:42 XorA: yes Aug 15 11:49:44 XorA: Yes, it's a library for the ATI IMAGEON 100 Aug 15 11:49:55 Id be willing to look at mplayer Aug 15 11:50:16 sirfred: XorA watches video on his 760 ;-) Aug 15 11:50:17 RP: About the speed of the thing, I have some doubts. Aug 15 11:50:38 RP: I also would like to do so. Aug 15 11:50:58 RP: My idea is to setup an vidix driver for mplayer using this library. Aug 15 11:51:05 sirfred: Its only going to be a speed increase if we can use the main processor for other things whilst the w100 handles its tasks Aug 15 11:51:25 RP: At least YUV->RGB is going to be made in the GPU Aug 15 11:52:00 sirfred: Which is a good bit of load to lose from the main processor... Aug 15 11:52:01 RP: Well, there is a problem with overlays in external memory. Aug 15 11:52:02 sirfred: should be able to modify the exsting w100 driver away from libaticore Aug 15 11:52:21 XorA: That's the idea, to drop libaticore. Aug 15 11:52:43 XorA: What's the state of the aticore vidix driver? Aug 15 11:52:52 sirfred: its not vidix anymore Aug 15 11:52:58 XorA: What's now? Aug 15 11:53:25 RP: I don't know if you was aware of this problem, But when the surface is in external memory, overlay is not working fine. Aug 15 11:53:29 sirfred: it just accesses it without X11/console Aug 15 11:53:54 XorA: Direct framebuffer access? Not using libaticore accelerated functions? Aug 15 11:54:18 sirfred: I wasn't. We can only fit QVGA in internal memory if I remember rightly... Aug 15 11:54:31 RP: That's right. Aug 15 11:54:45 sirfred: no, it uses libaticore, but without any form of locking with the console or X11 Aug 15 11:55:08 sirfred: does your library allow overlay bigger than 320x240? Aug 15 11:55:25 XorA: No. It's not working properly, as I'm telling to RP Aug 15 11:55:37 sirfred: If anything I would have thougth the opposite - external would work and internal wouldn;t :-/ Aug 15 11:55:42 sirfred: ah right, I understand now Aug 15 11:56:16 RP: Visually, the surface seems cutted in some horizontal lines, and some parts get repeated. Aug 15 11:56:29 RP: I remember this problem also from the tests with the libaticore. Aug 15 11:56:32 * XorA has seen that effect before Aug 15 11:56:46 It seems some kind of timing problem, or something so. Aug 15 11:57:28 sirfred: I've also seen this when I experimented with half the framebuffer in internal memory and half external... Aug 15 11:57:49 I wonder if we could tweak the memory timings in anyway... Aug 15 11:57:53 RP: Perhaps we need some DISP_DEBUG bits to be set. Aug 15 11:58:21 RP: At least when using RGB surfaces, the original aticore code makes a lot of black magic with that register. Aug 15 11:58:33 Depending on the BPP, but it's not touched for YUV surfaces. Aug 15 11:58:44 RP: I need to experiment a lot more with this. Aug 15 11:59:05 RP: Also, I'm using a different approach for host->card copy. Aug 15 11:59:13 sirfred: I'm too out of touch to help with it :-( Aug 15 12:00:07 RP: I seen it in some RADEON driver. Instead of using the mmap zone, I'm using HOST_DATA0-HOST_DATA7 registers. It looks like some kind of bus master, because when I make some mistake, it's able to hung the zaurus. :) Aug 15 12:00:40 RP: But it seems no more fast than using the mmaped zone, at least in my first tests and with little surfaces. Aug 15 12:01:05 RP: Well, your help and knowledge will be always welcomed Aug 15 12:01:20 RP: I have to leave now. Glad to talk with you again. Aug 15 12:01:30 sirfred: I'd expect it to come down to the memory timings so its probably the same speed Aug 15 12:01:48 sirfred: Its nice to see its still moving forwards! :) Aug 15 12:01:53 RP: Yes, it looks the speed is the same. Aug 15 12:02:06 Anyone know what those screenshots fuchs posted up with screencap were from? Aug 15 12:02:14 RP: I think we could have something usable in little time. Aug 15 12:02:29 sirfred: That is nice to hear :) Aug 15 12:02:44 ( http://handhelds.org/scap/land.16428.png ) (there were better ones than this too) Aug 15 12:10:21 * florian_kc smiles innocently Aug 15 12:10:38 I can't use glibc 2.4 Aug 15 12:10:53 bitbake says that it don't find this version Aug 15 12:11:02 but the .bb file exists Aug 15 12:11:26 Cobelius: its a bug Aug 15 12:11:31 Cobelius: for intermediate right? Aug 15 12:11:50 intermediate? Aug 15 12:12:19 Cwiiis: Its a small software study here... will be public soon i hope. Aug 15 12:12:41 florian_kc: Cool, I hope so :) Aug 15 12:13:06 I want that glibc 2.4 is used in my image Aug 15 12:13:20 but is uses 2.3.5+cvs Aug 15 12:13:21 Cwiiis: its not rocket science... just plain gtk with some nice pixmaps :-) Aug 15 12:13:42 okay, the panel might be a nice toy Aug 15 12:14:10 florian_kc: Would be nice to have a mobile-phone like API for that sort of interface too, if that doesn't provide one Aug 15 12:15:18 Cwiiis: yes indeed... we have quite some good stuff to do this but its quite some work to create a complete framework. Aug 15 12:15:29 Cobelius: try and adding PREFERRED_PROVIDER_virtual/yourarch-libc-for-gcc = "glibc-intermediate" Aug 15 12:15:52 in my case thats PREFERRED_PROVIDER_virtual/sh3-linux-libc-for-gcc = "glibc-intermediate" Aug 15 12:16:22 florian_kc: Good to know someone's working on it though, we need something to compete with http://www.linuxdevices.com/news/NS8030785497.html :) Aug 15 12:16:56 Cwiiis: we have the Gimp already. It competes pretty good with photoshop Aug 15 12:16:57 Cwiiis: well that still needs to turn into reality Aug 15 12:17:09 okay I'll try Aug 15 12:17:30 Cwiiis: We get is sponsored by tt, paint it blue install GPE ;) Aug 15 12:18:02 brb, food Aug 15 12:18:06 "However, the company has promised to make the phone available at a reasonable cost to all Qtopia licensees" Aug 15 12:18:12 what is the differece ob glibc and glibc-intermediate? Aug 15 12:21:00 florian_kc: I don't think the BOM is too expensive on this phone Aug 15 12:32:22 zecke: the Greenphone will be more open than any other phone Aug 15 12:33:44 ljp: hey Aug 15 12:33:47 ljp: how are you? Aug 15 12:34:32 ljp: more open than the universal device? Aug 15 12:34:43 ljp: including hardware semantics and datasheats? Aug 15 12:36:29 i would not call the universal open Aug 15 12:40:39 ljp: hehe, you can flash it Aug 15 12:41:01 perhaps, but the manufacturer does not support that Aug 15 12:41:15 03kristoffer 07org.oe.dev * rf31bfd26... 10/ (1 packages/xorg-app/xterm_207.bb): Aug 15 12:41:15 packages/xorg-app/xterm_207.bb: Addition of DEPEND LibXaw Aug 15 12:41:15 * Xterm requires LibXaw as depend. Added. Aug 15 12:41:53 simply because someone hacked linux onto it does not mean its open Aug 15 12:46:58 ljp: well :) Aug 15 12:47:34 hi ljp Aug 15 12:51:07 morning kergoth Aug 15 12:55:13 hey chouimat Aug 15 12:55:50 ljp: I wish you good luck with pre production and delivering the devices Aug 15 12:57:26 thanks Aug 15 12:58:13 hi ljp Aug 15 12:58:32 ljp: this also means you will release Qtopia Phone as GPL (compatible) in September? Aug 15 13:01:00 http://news.yahoo.com/s/nm/20060815/tc_nm/telecoms_mobile_reprogrammable_dc_1 Aug 15 13:01:00 ? Aug 15 13:35:44 /cr/cr Aug 15 13:35:49 morning Aug 15 13:36:00 hey Marcin Aug 15 13:38:10 03Giorgio 07org.oe.dev * r7b0e01a7... 10/ (1 packages/keymaps/keymaps_1.0.bb): keymaps: added h2200 support - close #1304 Aug 15 13:38:35 it was --author commit Aug 15 13:39:58 RP, do13, lrg, mickey|thesis: what do you think about 'metabug: Linux 2.6 kernel for Zaurus' in OE bugzilla? Or component for it? or other solution to track 2.6/Z bugs Aug 15 13:42:41 a question: what pda/smartphone/mobile device would you choose for a project where java, location awareness (e.g. gps) and camera is needed? Aug 15 13:42:43 hey hrw Aug 15 13:43:29 katossi: I'm thinking about getting Sagem My-700X as my next phone Aug 15 13:43:52 hrw: It would mean I could find them easier. It might be an idea to have an OZ 2.6 kernel component Aug 15 13:44:33 RP: metabug give as also dependency tree Aug 15 13:44:38 hrw: but small screen and no gps Aug 15 13:45:15 any experiences with the motorolas? Aug 15 13:45:29 katossi: had L6, V3 in hands - both suxx Aug 15 13:45:35 lrg: is Markku back in the office now? I haven't heard anything from him lately. Aug 15 13:45:45 hrw: no I mean the a180 Aug 15 13:46:47 pb_: yep he should be. Although, I'm working from home today and he probably has a full inbox after his hols. It's probably worth giving him a ping again. Aug 15 13:46:58 righto, thanks Aug 15 13:54:14 katossi: You mean A780 Aug 15 13:54:34 stefan_schmidt: yes, I forgot the precise model name Aug 15 13:55:29 katossi: I've one. But never played with the java stuff on it. Aug 15 13:55:49 stefan_schmidt: is the camera running? Aug 15 13:56:20 katossi: With java? I think so. Aug 15 13:57:10 katossi: Not sure about gps and java, but you can get the gps information through the linux system. :) Aug 15 13:57:38 but does it have a built in gps? Aug 15 13:58:32 katossi: yes Aug 15 13:58:50 and is it recognized? Aug 15 13:59:46 if I had something to read from under /dev/ i'd be great Aug 15 14:00:32 I see some nice changes in .dev to add into .oz354x Aug 15 14:00:40 anyone aware of the "udev needs hotplug, not started" message? Aug 15 14:00:46 thought that udev would supply hotplug Aug 15 14:02:33 how old udev you are using? Aug 15 14:02:50 stefan_schmidt: I see no gps in the a780 specs Aug 15 14:03:09 hrw: 0.92 Aug 15 14:03:13 latest Aug 15 14:03:15 katossi: iirc there was 2 versions of a780 - with or without gps Aug 15 14:03:21 Kristoffer: should work without Aug 15 14:03:27 katossi: Of course it is recognized from the official linux system. I've a code snippet around to dump the gps info. moment Aug 15 14:03:33 aha, anyone could tell me the price of a a780 in germany? Aug 15 14:03:39 hrw: hmm, thats what I thought. Just weird why it doesnt.. Aug 15 14:03:42 katossi: Only the european models have gps. Aug 15 14:03:59 nice Aug 15 14:03:59 katossi: it is not manufactured anymore;( Aug 15 14:04:13 katossi: http://wiki.openezx.org/A780#GPS Aug 15 14:04:14 too bad Aug 15 14:04:28 any replacement for that? new model or something? Aug 15 14:04:31 koen|away: Thinking about buying a zaurus, whats good to buy? Want one with keyboard Aug 15 14:04:36 katossi: A780 is out of sale. Aug 15 14:05:00 Kristoffer: sl-c1000 or sl-c3200 Aug 15 14:05:11 Kristoffer: 3200 has 6G microdrive Aug 15 14:05:28 hrw: ah, goodie. do they got wifi btw? or do I need CF card for it? Aug 15 14:05:31 what did they sell the a780 for such short time? Aug 15 14:05:45 Kristoffer: both lack any wireless other then irda Aug 15 14:06:02 hrw: oki Aug 15 14:06:26 katossi: They sold it some time before in asia Aug 15 14:06:49 umm, nokia 770 + bluetooth gps module ? Aug 15 14:06:54 stefan_schmidt: damn, that that everything I wanted Aug 15 14:07:24 katossi: http://people.openezx.org/stefan/tmp/gpsdump/ Aug 15 14:07:25 nchip: that's an option, but a built in bt woul be nicer Aug 15 14:07:45 katossi: it has built in bt Aug 15 14:07:59 katossi: the imcosys phone has built in gps, too. Aug 15 14:08:01 nchip: sorry I means built in gps Aug 15 14:08:15 gps and camera are a must Aug 15 14:08:24 stefan_schmidt: imcosys? Aug 15 14:09:00 katossi: http://www.imcosys.com/ Another linux smartphone. Aug 15 14:09:03 moto a780 still seems available Aug 15 14:09:32 nchip: Yes, sometimes still in stock Aug 15 14:10:07 hey loks good Aug 15 14:10:24 any success with oe and the imcosys? Aug 15 14:10:49 katossi: I don't know if they already started to ship it. Aug 15 14:10:53 wow, it has even wifi! Aug 15 14:11:33 katossi: They don't released any code or doc yet. Aug 15 14:11:52 I think it will be a case for harald Aug 15 14:12:32 stefan_schmidt: but did anyone contacted them? Aug 15 14:12:51 woglinde: Yes, perhaps. Last time i talked about this phone to him, he planned to buy one. Aug 15 14:13:02 katossi: Yes. Aug 15 14:13:16 there are no prices on the website Aug 15 14:13:46 katossi: They don't sell on their one. Aug 15 14:13:52 ok I found it for 250 in germany Aug 15 14:13:57 350 Aug 15 14:13:59 not bad Aug 15 14:15:04 stefan_schmidt: thanks for the info and help Aug 15 14:15:15 katossi: np Aug 15 14:15:19 I've got to go home Aug 15 14:15:25 see you all Aug 15 14:15:28 bye katossi Aug 15 14:15:30 katossi: http://www.schellens.de/smartphone.php Read the fourth FAQ entry. Aug 15 14:15:34 hm only 200 mhz Aug 15 14:15:40 bye katossi Aug 15 14:16:06 Es handelt sich um ein geschlossenes (keine GPL) Embedded Linux Aug 15 14:16:07 hahah Aug 15 14:17:02 idiots Aug 15 14:17:16 poor astrid Aug 15 14:17:17 stefan_schmidt: " (keine GPL) Embedded Linux" ? why? Aug 15 14:17:26 anyway see you and thx again Aug 15 14:17:47 I bet she isnt really reasonable for this Aug 15 14:17:51 they told her this crap Aug 15 14:18:26 03hrw 07org.oe.oz354x * rca93e493... 10/ (1 packages/aspell/aspell-native_0.50.5.bb): Aug 15 14:18:26 aspell-native_0.50.5.bb : add native that is required by aspell language packs Aug 15 14:18:26 applied changes from 70d230573c9b9cecb6616046b4c5b24e2c5ad07f Aug 15 14:18:26 through 9145fc8b71029fb9ef74ec089e40d4b29737efe8 Aug 15 14:18:31 03hrw 07org.oe.oz354x * re72ece76... 10/ (1 packages/aspell/aspell5-pl_6.0-20060811.bb): aspell: added Polish dictionary Aug 15 14:18:36 03hrw 07org.oe.oz354x * rcde358dd... 10/ (3 files in 2 dirs): aspell: added English dictionary from .dev Aug 15 14:18:46 woglinde: ask them where they got the "keine GPL" Linux Aug 15 14:19:06 presumably from some cheap-ass company Aug 15 14:19:29 A friend of mine was at the cebit and talked to imcosys about code. They told him, they changed anything in the linux kernel and don't have to release any code. We will see. Aug 15 14:19:40 lol Aug 15 14:20:45 I'm not sure they understand the difference between the used GPL software and their own written software. Aug 15 14:28:37 03cbrake 07org.oe.oz354x * r84ec85d5... 10/ (1 packages/python/python-pysqlite2_2.0.2.bb): Aug 15 14:28:37 python-pysqlite2 2.2.2: add python-datetime to RDEPENDS Aug 15 14:28:37 applied changes from 1b17f8afca24771aa900e415899c6189576c2977 Aug 15 14:28:37 through 113954a10c1eecc2f688fdb881e40dcd951cf598 Aug 15 14:28:42 03hrw 07org.oe.dev * rf6751285... 10/ (1 packages/mysql/mysql_4.1.18.bb): mysql: fix SRC_URI - close #1319 Aug 15 14:31:54 anyone trying out konqueror? using the 20060404 version, but when I start it it just gives me a misformed tiny window Aug 15 14:32:07 hrw: what version did you say worked best? Aug 15 14:32:33 du -h Aug 15 14:33:04 doh Aug 15 14:33:31 RP: I'm doing the releases now Aug 15 14:34:34 there was another linux phone released to market recently in asia i think - an orange one Aug 15 14:34:50 Kristoffer: I use 084 in oz 354 Aug 15 14:34:51 anybody seen that? i might be able to dig up a link Aug 15 14:35:00 if noone else has it handy Aug 15 14:36:20 hrw: and that didnt need hotplug other than that udev supplied? Aug 15 14:36:22 hi reenoo Aug 15 14:36:39 Kristoffer: have to check it Aug 15 14:37:37 poushag: I don't heard of it yet. Aug 15 14:38:02 hey pb_ Aug 15 14:38:10 afternoon all Aug 15 14:38:57 Kristoffer: hm. my rootfs has linnux-hotplug and udev 084 so cant help Aug 15 14:39:09 hrw: oki, thx anyways Aug 15 14:39:51 03zecke123 07bitbake-1.4 * r565 10/ (ChangeLog bin/bitbake lib/bb/__init__.py): bitbake-1.4/bin/bitbake: Assign release number 1.4.4 Aug 15 14:40:31 zecke: new bitbake release coming up? Aug 15 14:40:46 Kristoffer: well, no Aug 15 14:40:54 Kristoffer: it is just to release this obsolete stuff Aug 15 14:41:16 okidoki Aug 15 14:41:26 Kristoffer: RP has a radical rework of the task handling Aug 15 14:41:34 Kristoffer: and we want to ship enough stable versions Aug 15 14:41:55 zecke: ah, understand. Btw, im using 1.5.0 any reason to update in the near future? Aug 15 14:42:09 Kristoffer: You should not be using 1.5.0 Aug 15 14:42:15 you might want to upgrade to 1.6.0 then Aug 15 14:42:22 zecke: so everyone keeps telling me, but it works well Aug 15 14:42:40 Kristoffer: it might not work tomorrow, or the day after tomorrow Aug 15 14:42:58 Kristoffer: as we will do a check in that breaks everything and we will be busy until the next month to fix it up Aug 15 14:43:17 zecke: aha :) eek. Oki, will update to 1.6.0 when its out then Aug 15 14:43:24 http://linuxdevices.com/news/NS2220873893.html - this isnthte one i mentioned earlier but looks interesting Aug 15 14:45:27 zecke: btw, you said that you are working on the parse section, would that mean any increases in speed? Aug 15 14:46:22 mickey|thesis: Do you have any thoughts on improving dependency tracking in the python packages? Aug 15 14:47:05 mickey|thesis: I was trying to get idle to work last night, and I would install a package, try and then figure out what else was missing from the error messages. Aug 15 14:48:18 Kristoffer: well this means SIGSEGV and new errors :) Aug 15 14:49:44 mickey|thesis: I was thinking that maybe OE could parse the python files and dynamicly updatae the RDEPENDS or something along those lises. That might also help with releasing python apps later. Aug 15 14:49:58 zecke: you dont think you are going abit microsoftish there? :D Aug 15 14:50:47 Kristoffer: it currently segfaults Aug 15 14:53:10 03zecke123 * r566 10bitbake/ (ChangeLog bin/bitbake lib/bb/__init__.py): bitbake: Assign release number 1.6.0 to the trunk and update the ChangeLog Aug 15 14:54:16 which is the recommended version of python for oe? Aug 15 15:03:29 03zecke123 * r567 10/tags/ (16 files in 7 dirs): Bitbake: Add version 1.4.4 and 1.6.0 Aug 15 15:04:40 03zecke123 * r568 10/branches/ (12 files in 6 dirs): BitBake: Bump and add Versions 1.4.5 and 1.6.1 Aug 15 15:05:28 03zecke123 * r569 10bitbake/ (bin/bitbake lib/bb/__init__.py): bitbake: Trunk is now at version 1.7.0 Aug 15 15:08:50 furlongm_: hey Aug 15 15:09:01 furlongm_: I'm searchinf a picture Aug 15 15:09:05 searching even Aug 15 15:09:34 furlongm_: http://linuxtogo.org/~zecke/Web-Site/Me.html I think none of them qualify Aug 15 15:09:47 furlongm_: so I have to charge the baetteries of my digi cam Aug 15 15:15:44 ~lart cairo Aug 15 15:15:49 ~slart cairo Aug 15 15:19:10 zecke: 1.4.4 released as archive? Aug 15 15:20:05 hrw: not yet Aug 15 15:20:13 hrw: I try to remember where to upload the tarballs Aug 15 15:20:34 ok Aug 15 15:20:47 zecke: I will bump minimal version for oz354x to 1.4.4 then Aug 15 15:24:12 hrw: hmm released Aug 15 15:24:20 I did something wrong as the file size is big... Aug 15 15:25:45 okay both are released Aug 15 15:25:48 ok Aug 15 15:25:49 will blog+release mail Aug 15 15:26:54 okx2 Aug 15 15:31:32 blogged.... Aug 15 15:32:42 afk Aug 15 15:37:35 03hrw 07org.oe.oz354x * r081b5732... 10/ (7 files in 3 dirs): (log message trimmed) Aug 15 15:37:35 xmms: sync with changes in .dev: Aug 15 15:37:35 - added xmms-mad Aug 15 15:37:35 - added xmms-tremor Aug 15 15:37:35 - add an icon and .desktop file for xmms Aug 15 15:37:35 Call via a wrapper script so we can insert XLIB_SKIP_ARGB_VISUALS=1 before Aug 15 15:37:37 xmms so that double size mode works correctly (a tip gleaned from gentoo FAQ) Aug 15 15:37:45 03koen 07org.oe.oz354x * r8c6bf6d5... 10/ (1 packages/mikmod/mikmod_3.2.2-beta1.bb): Aug 15 15:37:45 mikmod: depend on ncurses Aug 15 15:37:45 applied changes from d528cddcf71a3f46101bd8d52a87f28485ceb5e1 Aug 15 15:37:47 through 7af26a86041e7825fc69028e9e1b1fa04c5a8ccb Aug 15 15:37:56 03hrw 07org.oe.oz354x * rc210287d... 10/ (4 files in 2 dirs): openzaurus*.conf: moved common stuff from 3.5.4.{1,2} to openzaurus.conf, bumped minimal BitBake version to 1.4.4 Aug 15 15:48:41 later Aug 15 15:58:13 I'm trying to build php5 but I get an error during the config stage: "libiconv not found" I built libiconv with bitbake and did a clean on php5, rebuilt it and still get the same error message. Aug 15 16:30:43 how do you disable the disable or skip certain tests when compiling xorg? i'm cross compiling on an x86 to my Z and it's not going so well Aug 15 16:37:56 i've already found the line that leads to failure: if test "${ac_cv_sys_linker_h+set}" = set; then Aug 15 16:38:04 how do i make that eval to true, short of hacking the configure script? Aug 15 17:55:38 re all Aug 15 17:56:09 having a problem with update-rc.d, i cant seem to run it from a udev script Aug 15 17:56:35 and that appears to be messing up ipkg install Aug 15 17:58:38 that is unfortunate Aug 15 17:58:41 is there an error message? Aug 15 18:03:08 ahh, udev does not include /usr/sbin or sbin in the path Aug 15 18:03:32 i failed to fully qualify the pathname of update-rc.d Aug 15 18:03:43 so it did not run, no error either Aug 15 18:04:06 maybe stderr is >/dev/null somewhere in the udev chain Aug 15 18:14:08 ok, i have dropped the output to a log, and i get this: Aug 15 18:14:16 Installing fusb (3.0.0-r9) to root... Aug 15 18:14:16 Installing fusb (3.0.0-r9) to root... Aug 15 18:14:16 Installing fusb (3.0.0-r9) to root.. Aug 15 18:14:26 perhaps 6 times Aug 15 18:14:45 the final one actually seems to go... Aug 15 18:15:22 this does not happen if i run the udev script by hand at the prompt Aug 15 18:30:59 zecke: hmm... Aug 15 18:31:24 zecke: this one looks ok -> http://linuxtogo.org/~zecke/Web-Site/Me_files/Foto%2045.jpg can i use this one? Aug 15 18:32:27 ah, he's gone. doh. Aug 15 18:32:30 furlongm zecke isnt here right now Aug 15 18:32:43 yep, sorry for noise Aug 15 18:33:01 hm for what you need the picture? Aug 15 18:38:37 * chouimat is away: bbl Duck Dogers on tv :) Aug 15 18:38:45 woglinde: zecke's giving a talk at akademy and we have the abstracts and mugshots on the webpage -> http://conference2006.kde.org Aug 15 18:39:04 furlongm ah I see Aug 15 18:40:03 hi woglinde Aug 15 18:40:07 he pb Aug 15 18:41:47 pb the U-media auidomate wanst from your firm right? Aug 15 18:41:53 audio Aug 15 18:42:08 no, it wasn't Aug 15 18:42:17 hm okay Aug 15 18:45:21 how to fix this error building gpe-image on dev: "ERROR: Cannot satisfy the following dependencies for locale-base-de-de: glibc-binary-localedata-de_DE" Aug 15 18:45:28 hi, does gettext currently build for you guys, I get a fail in RUNSTRIP Aug 15 18:45:36 with a few errors of the type Aug 15 18:45:37 armeb-linux-objcopy: Unable to recognise the format of the input file /home/frans/OpenSlugHead/openslug/tmp/work/gettext-0.14.1-r3/image/usr/lib/gettext/gnu.gettext.GetURL Aug 15 18:45:38 is this a case-sensitivity problem? Aug 15 18:46:25 mreimer: hm, it sounds like a bug in glibc-package.inc Aug 15 18:46:42 by the looks of it, it isn't properly crunching the RDEPENDS string for locale-base. Aug 15 18:46:43 hey pb__. long time no see Aug 15 18:47:04 * mreimer looks at glibc-package.inc Aug 15 18:47:13 er, glibc-package.bbclass actually Aug 15 18:47:40 bb.data.setVar('RDEPENDS_%s' % pkgname, 'glibc-binary-localedata-%s' % glibc_name, d) Aug 15 18:47:58 I think that line is probably missing a call to legitimize_package_name(). Aug 15 18:48:08 i'm trying to cross compile xorg but the self tests run during teh configure stage report the error: checking for sys/linker.h... configure: error: cannot run test program while cross compiling Aug 15 18:48:15 how do i deal with this? Aug 15 18:48:33 either fix the test so that it works when cross-compiling, or put the appropriate definition in your site file Aug 15 18:48:36 using .dev for zaurus c7x0 compilation btw Aug 15 18:48:46 yeah i figure that much but how? Aug 15 18:49:03 that rather depends on the test Aug 15 18:49:17 we need someone to bump us to autoconf 2.60... Aug 15 18:49:19 you'd need to establish why it's trying to run this program, and then find another way to achieve the same result Aug 15 18:49:19 try run autoreconf Aug 15 18:49:21 pb__: so add legitimize_package_name(glibc_name) to that line? Aug 15 18:49:26 mreimer: yeah Aug 15 18:49:30 ok, trying Aug 15 18:50:32 pb__: the line of code that tests false, reporting the error, is: if test "${ac_cv_sys_linker_h+set}" = set; then how do i make that test true short of hackign the configure script? Aug 15 18:50:44 woglinde: how do i run autoreconf inside of OE? Aug 15 18:51:20 Interrobang: I think you need to take a slightly higher level view of the problem rather than focussing on that one line Aug 15 18:51:22 iterrobang inherit autotools in the .bb file Aug 15 18:52:44 pb__: yeah i figure that much too, but configure --help doesn't say anything about self tests or disabling them. so i'm trying to work my way up to the high level configuration, but the configure script isn't that intuitive Aug 15 18:53:37 woglinde: what line do i put in there? Aug 15 18:53:42 Interrobang: you need to find the part of the configure.in script that corresponds to the problematic test, probably something like AC_CHECK_HEADER(sys/linker.h) Aug 15 18:53:55 then find the corresponding part of the generated configure script, and work out what it's trying to achieve Aug 15 18:54:09 pb xorg uses configure.in rather than configure.ac? Aug 15 18:54:19 or configure.ac, whichever one you have Aug 15 18:54:25 they're just two different names for the same thing Aug 15 18:55:07 hm whats in linker.h defined I will take a short look Aug 15 18:55:21 hmm, gettext compiles some code with gjc and that gives a i836 binary Aug 15 18:55:56 what the hell linker.h cames from? Aug 15 18:55:59 836? Leet! Aug 15 18:56:09 oops 386 Aug 15 18:56:48 hm in debian unstable no package provides linker.h Aug 15 18:57:07 * eFfeM hugged too many CosmicPenguins :-) Aug 15 18:57:23 woglinde: sys/linker.h Aug 15 18:57:37 woglinde: mm, it might only exist on some crazy bsd system or the like Aug 15 18:57:40 maybe zecke would know Aug 15 18:58:24 the configure.ac does have comments pertaining to freebsd around it Aug 15 18:58:47 oh Aug 15 18:58:49 hm bsd Aug 15 18:58:59 is xorg being cmopiled for bsd arm by mistake? Aug 15 18:59:05 *compiled Aug 15 18:59:23 no Aug 15 18:59:31 maybe they compile it native Aug 15 18:59:41 there are some fast arm-machines Aug 15 19:02:35 i'm running linux, trying to compile for OZ, why is bsd material involved? Aug 15 19:03:13 yeah asked the fucking xorg guys which can read the autotools manuals Aug 15 19:03:40 can not Aug 15 19:03:41 sorry Aug 15 19:04:57 has any1 ever managed to compile xorg (not kdrive) for any Zaurus? Aug 15 19:05:28 interrobang koen has compiled xorg for some arm devices, so it builds Aug 15 19:06:09 woglinde: for arm i don't doubt, but specifically Z? also, when did he compile it? what version? Aug 15 19:07:44 * chouimat is back. Aug 15 19:08:30 interrobang sorry I dont compiled xorg yet Aug 15 19:08:39 so you have to wait for someone which did Aug 15 19:20:28 03rpurdie 07org.oe.dev * r7c82370d... 10/ (5 files in 2 dirs): udev: Fix packaging after -dbg changes (udev-utils wasn't populated) Aug 15 19:20:33 03rpurdie 07org.oe.dev * rc0da2127... 10/ (24 files in 7 dirs): udev: Prune old obsolete broken versions (058, 063, 065, 070, 071). Aug 15 19:20:38 03rpurdie 07org.oe.dev * r8c12c313... 10/ (1 packages/librsvg/librsvg_2.11.1.bb): librsvg: Disable mozilla plugins as otherwise their presence on the build system can confuse configure Aug 15 19:29:34 while compiling xserver-xorg 7.1 i get the error: /bin/sh: ./mkg3states: cannot execute binary file i assume that mkg3states is compiled for arm while cross compiling from x86. how do i address this? Aug 15 19:39:14 Interrobang: qemu-arm, see the emacs bb file Aug 15 19:41:36 JustinP: i heard a mention of the command "inherit autotools" is that what i'm looking for? or should i "merge" the whole emacs file with the xorg one? Aug 15 20:43:55 Interrobang: see the qemu-arm stuff in the emacs bb file....it would make no sense to merge emacs and xorg.... Aug 15 20:44:03 Interrobang: inherit autotools is one thing you can use, yes Aug 15 20:45:19 hi justin Aug 15 20:47:41 JustinP: but "\"merge\"" i meant swap out the emacs stuff for xorg code Aug 15 20:49:33 cu Aug 15 21:08:53 woglinde_: hey Aug 15 21:09:14 Interrobang: again...merge is just the wrong term. I mean for you to use the qemu stuff as an example...not to merge anything Aug 15 21:09:39 JustinP: yeah i know, i like ot make up words sometimes Aug 15 21:10:00 JustinP: ...don't reply to that... Aug 15 21:40:25 hey rp - bitbake stable seems to have gone to ver 1.6 - what happened to 1.5 ? Aug 15 21:40:42 why was it skipped over ? Aug 15 21:41:46 i read zeckes blog but it didnt say why - and i havent gotten mtn log to work for me yet Aug 15 22:00:09 pb__: ping Aug 15 22:00:30 poushag: 1.5 unstable became 1.6 stable Aug 15 22:01:17 poushag: The main reason is so we can hack on 1.7 and break a few things :) Aug 15 22:01:44 cool Aug 15 22:02:30 is 1.7 the multithreaded one? Aug 15 22:09:55 poushag: It will be Aug 15 22:10:06 nice Aug 15 22:10:50 do you have any idea how long the 1.7 development process will last, rp? Aug 15 22:11:22 poushag: as long as it takes ;-) Aug 15 22:11:57 yes - i guess i was hoping weeks rather than months is all Aug 15 22:12:53 poushag: bitbake hacking is done in my limited free time which has several competing demands. months might therefore be realistic. We'll see... Aug 15 22:13:02 I've been planning to do it for years :-/ Aug 15 22:13:16 do you blog about it? or otherwise announce progress other than in here? Aug 15 22:13:45 oe@ usually hears about it as does/should the bitbake mailing list Aug 15 22:14:00 * RP should really use his blog more Aug 15 22:14:03 rp i grok you - with the 3 little ones in my house i have the same constraints Aug 15 22:14:31 alright - thx for the contributions - bbl (dinner time) Aug 15 22:46:08 mreimer: hi! which lab did you want me to include? Aug 15 22:57:55 hi reenoo. I haven't had a chance to get LAB building in OE yet, so until I do, use the latest version 20060421 in handhelds.org/~mreimer Aug 15 23:05:28 must resist urge to build an image for my treo Aug 15 23:05:47 mreimer: source code? build instructions? linux is gpl... Aug 15 23:06:26 reenoo: the config file is there. it's just a kernel. the .bb worked at one point, and may still Aug 15 23:06:39 reenoo: http://handhelds.org/~mreimer/zImage-LAB-20060421.config Aug 15 23:08:29 mreimer: ah, ok. could you upload a tarball of the sources you used anyway? Aug 15 23:10:20 reenoo: I don't have the exact sources that I used to build that binary, because I've cvs up'd a few times since then. Aug 15 23:10:33 reenoo: the bb files are in .dev under packages/linux/LAB* Aug 15 23:10:37 I'll try building Aug 15 23:12:23 mreimer: ok, thanks Aug 16 01:37:34 JustinP: i have examined the emacs .bb and tried some test configurations modified for xorg with no success. there's a patch to the emacs build process for use qith qemu that is beyond my concern to try and convert Aug 16 01:47:43 Interrobang: you just have to make the build use qemu-arm to run the cross stuff Aug 16 01:48:02 Interrobang: or set up the config cache to have the correct answers without having to run a program (no, I don't know how to do this) Aug 16 01:49:05 JustinP: by "run the cross stuff", do you mean run the entire compilation or just the tests? if it's the tests you mean then that's beyond all worthwhile effort for me Aug 16 01:49:35 JustinP: i'd need some rather intimate knowledge of the xorg build process to pick out the tests' execution and send it to qemu Aug 16 02:08:09 I mean the stuff that needs to be run as arm Aug 16 02:08:10 of course Aug 16 02:08:38 and all it requires is runnung the damn thing over an dover and fixing those places that error out Aug 16 02:08:46 but whatever, I'm done with this... Aug 16 02:57:22 I need help with quilt for a patch Aug 16 02:57:37 What I tried is here: http://pastebin.ca/133808 Aug 16 02:57:52 I am trying to patch a file that is not in the {S} dir. **** ENDING LOGGING AT Wed Aug 16 02:59:57 2006