**** BEGIN LOGGING AT Mon Jun 25 02:59:56 2007 Jun 25 04:13:55 any qemu experts here? Jun 25 05:52:12 morning Jun 25 05:52:31 Zero_Chaos: look at binconfig.bbclass for magic-config mangling Jun 25 05:53:13 Zero_Chaos: inherit autotools and do not put do_configure in recipe to get autoreconf started to regenerate configure script Jun 25 05:54:15 good morning Jun 25 05:56:03 hi Jun 25 05:57:13 * djay-il is anxious to get his device working today Jun 25 05:58:31 it would mean some $$$ Jun 25 06:07:53 ;) Jun 25 06:25:07 question: would the enablement of AEABI be enough to make kernel EABI? Jun 25 06:25:12 hmmmm, my task depends on util-macros, and it builds it when try to make my image, but at the end it errors out and says it can't resolve the dependency for util-macros Jun 25 06:28:07 djay-il: enough Jun 25 06:28:18 hm Jun 25 06:28:29 wvdschel: and you have util=-macros packaged? Jun 25 06:28:49 yes, I think so, you mean as an ipk? Jun 25 06:29:19 hmmm, removing the dependency and cleaning my image and task solves it Jun 25 06:29:23 I hope it runs Jun 25 06:29:59 I only seem to have util-macros-dev Jun 25 06:30:12 hrw: should 2.6.16 support it? Jun 25 06:31:56 \ Jun 25 06:32:03 djay-il: yes Jun 25 06:32:10 eabi was added to 2.6.16 Jun 25 06:32:33 hrw: it just seems that I have some sort of problem with it Jun 25 06:34:58 wvdschel: right - I got only dev and dbg packages of util-macros too Jun 25 06:35:28 ok so it doesn't find it because there's no plain runtime version Jun 25 06:35:40 it's a build time dependency for libxinerama Jun 25 06:35:49 so I guess it shouldn't hurt not having it Jun 25 06:39:17 wvdschel: libxinerama depends on util-macros-native rather then util-macros Jun 25 06:39:27 ah Jun 25 06:40:01 well it didn't build, and when I built util-macros it started building, but I suppose that's because util-macros depends on util-macros-native? Jun 25 06:43:03 let me look Jun 25 06:46:52 I have >100 revs to fetch.. but see the problem Jun 25 06:47:36 2hi kun Jun 25 06:50:22 I've now compiled kernel with EABI but now neither old-based rootfs nor new-based do not work...anyone can help please? Jun 25 06:52:07 djay-il: enabi eabi and oabi support in kernel Jun 25 06:52:26 djay-il: we have it in many our arm kernels (if not in all 2.6.16+ ones) Jun 25 06:52:34 hrw: I have. Jun 25 06:52:45 I've enabled both Jun 25 06:52:56 and it still doesn't work... Jun 25 06:54:25 still nfs boot? Jun 25 06:54:31 hrw: yes Jun 25 06:54:44 its seems that kernel and rootfs are not compatible Jun 25 06:55:59 just to make sure: after I build the kernel I find it under /staging ? Jun 25 06:56:26 no - in tmp/deploy/ Jun 25 06:56:50 djay-il: think of staging/ as of 'OE internal - do not touch' Jun 25 06:57:28 oh, ok Jun 25 06:57:58 so, the kernel would be *.bin ? Jun 25 06:58:00 it will simplify usage Jun 25 06:58:16 djay-il: kernels, images ends in deploy/*/images dir Jun 25 06:58:46 ok... Jun 25 06:58:49 djay-il: how kernel is named depends on its recipe and kernel.bbclass Jun 25 06:59:13 ok, got that.. Jun 25 06:59:22 * djay-il checks whether he uses the right kernel Jun 25 07:00:16 I do... Jun 25 07:00:50 what's the difference between .bin kernel and .img kernel? Jun 25 07:01:15 read your recipe? Jun 25 07:01:41 my recipe os a short one :-( Jun 25 07:02:43 is eabi faster like this http://wiki.emqbit.com/eabi-oabi-benchmark on all archs? ie on spitz? Jun 25 07:03:11 hrw: I think .img is an initrd image Jun 25 07:04:16 djay-il: use "file" command? Jun 25 07:04:38 JaMa: eabi is softfloat on most ARM cores Jun 25 07:04:47 JaMa: oabi can be hardfloat or softfloat Jun 25 07:05:07 hardfloat on arm without fpu use kernel fpu emulator == slow Jun 25 07:05:31 JaMa: EABU is not faster. it just uses sane float-point calling conventions. so, it's oabi which was slow, eabi is merely a baseline. Jun 25 07:05:33 softfloat use libc internal code to handle fpu stuff so no kernel exception is created == faster Jun 25 07:06:00 hrw: it just says "data" Jun 25 07:09:43 psokolovsky__: for those which used oabi only with debian or familiar eabi is faster due to softfloat Jun 25 07:11:53 hrw: I checked that eabi is disabled in my kernel.. (dunno if spitz has FPU, or how to choose hard/soft float), I'll google it.. thanks Jun 25 07:12:23 JaMa: spitz lack fpu Jun 25 07:12:42 JaMa: and if your kernel == OZ kernel then no eabi (iirc) Jun 25 07:35:19 good morning all Jun 25 07:35:58 good morning all Jun 25 07:36:16 koen : hi ! Jun 25 07:36:49 psokolovsky__: EABI saves a byteswap on every FP op as well Jun 25 07:36:55 hey steliosk Jun 25 07:40:39 morning' Jun 25 07:45:30 4.1M cpio minimal image Jun 25 07:45:33 big ;( Jun 25 07:45:51 * hrw -> breakfast Jun 25 07:45:57 without without kernel? Jun 25 07:46:03 yes Jun 25 07:46:33 * koen needs some breakfast as well Jun 25 07:46:41 libc package takes most Jun 25 07:46:57 ldconfig is >500K, some libs are not needed too Jun 25 07:47:06 hrw: Smacznego Jun 25 07:47:18 koen: same to you Jun 25 07:51:23 * ljp is eating dinner Jun 25 07:52:07 ~bon appetit Jun 25 07:52:11 bon appetit is, like, smacznego. Guten Appetit. Eet Smakelijk. God Appetitt. Buon Appetito. Buen apetito Bom Apetite. buen apetito Smaklig måltid!. Hyvää ruokahalua. Bo Proveito Jun 25 07:53:11 morning Jun 25 07:53:51 hey XorA Jun 25 08:01:42 heh Jun 25 08:01:56 a non-autotools package fails ot build if you don't inherit autotools Jun 25 08:02:07 koen: interesting Jun 25 08:02:31 the new version is autotools based anyway Jun 25 08:02:57 * XorA thinks someone doesnt quite get kernel modules in #2556 Jun 25 08:03:20 checking for random stuff to make you feel better... failed Jun 25 08:04:20 hi Rod Jun 25 08:04:42 rwhitby-n800: n800 with treo650 as bt phone? Jun 25 08:06:14 * XorA reads mailing list, cries and despairs at the state of education Jun 25 08:07:14 good morning Jun 25 08:08:29 * hrw decided tonot read mails on the move Jun 25 08:10:20 * koen decided to look at guinnea pigs instead Jun 25 08:10:27 hey florian Jun 25 08:13:02 hi Jun 25 08:14:13 guinea pigs have more intelligence than some people are showing on mailing list at moment Jun 25 08:16:27 do those people in the mailinglist keep jumping into the hay dispenser as well? Jun 25 08:16:43 is there a "generic" kernel in OE? Jun 25 08:16:58 that I can try... Jun 25 08:18:15 linux_2.6.21.bb is kind of 'generic' which is used by 4-6 machines now (each one with own settings added to it) Jun 25 08:18:16 koen: get used to it, its their favourite game Jun 25 08:18:30 koen: get a rat sling and hang it from top of cage, ours loved that Jun 25 08:18:34 djay-il: you still did not get compulab270 booting? Jun 25 08:18:36 hmmm Jun 25 08:19:14 time to get a new dispenser, this one has a too high gallows factor Jun 25 08:19:29 * koen gets some pliers Jun 25 08:39:32 * florian is confused Jun 25 08:39:47 glibc-intermediate's do_configure fails Jun 25 08:41:48 i wonder why it tries to build this at the end of the whole build process... Jun 25 08:44:31 hrw: yea Jun 25 08:48:09 hi pH5 Jun 25 08:48:20 hi florian Jun 25 08:49:43 hrw: its like there's a problem that doesn't want to be solved Jun 25 08:50:39 djay-il: Cliff Brake has working compulab boards so getting system working on them is not a problem Jun 25 08:50:43 it looks Jun 25 08:51:40 hrw: yea, but I think he got them working with oabi Jun 25 08:52:09 because I have his configs and EABI is not enabled there Jun 25 08:53:41 hm...wouldn't you say that a gcc of 4.1 needed for EABI support? Jun 25 08:54:03 hrw: do you know why pimlico icons are broken with mb-desktop-1? Jun 25 08:54:16 koen: mb-desktop-1? someone still use it? Jun 25 08:54:30 hrw: everyone except o-hand basically Jun 25 08:54:48 since mb-desktop-2 is also quite broken on OE x11 images Jun 25 08:54:52 koen: can you make a shot? Jun 25 08:55:04 one tab full of panel applets is not what I call useable Jun 25 08:55:17 hrw: all pimlico apps have the "no icon" icon Jun 25 08:55:19 koen: thats going to change soon Jun 25 08:55:53 yay Jun 25 08:56:34 hrw: I think mb-desktop-1's parsing of .desktop files is broken Jun 25 08:56:54 since mb-desktop-2 on the same machine does show the icons correctly Jun 25 08:57:11 koen: official answer: mb-d1 is dead project but patches welcome. Jun 25 08:57:20 that's what I feared Jun 25 08:57:36 and mallum says "mb2 is experimental and shouldn't be used" Jun 25 09:19:27 gah I have a whole lot of catching up todo on Lego Space Jun 25 09:25:39 XorA: Hehe.. Anything interesting? ;) Jun 25 09:30:32 CM: everything since Classic Space Jun 25 10:21:38 03tmbinc 07org.oe.dreambox * r0c9a14d8... 10/ (16 files in 4 dirs): python-2.5.1: add python 2.5.1 Jun 25 10:22:23 03tmbinc 07org.oe.dreambox * r95b11400... 10/ (3 files in 2 dirs): twisted: update to 2.5.0 Jun 25 10:22:30 03tmbinc 07org.oe.dreambox * r78c1b66b... 10/ (1 packages/twisted/twisted-web_0.7.0.bb): twisted-web: update to 0.7.0 Jun 25 10:22:35 03tmbinc 07org.oe.dreambox * r51062514... 10/ (3 files in 2 dirs): zope-interface: update to 3.3 Jun 25 10:22:40 03tmbinc 07org.oe.dreambox * rcfb7bfc0... 10/ (1 site/mipsel-linux): site/mipsel-linux: add ac_cv_sizeof_{double,float,fpos_t},ac_cv_file__dev_{ptmx,ptc} Jun 25 10:22:45 03tmbinc 07org.oe.dreambox * r3f549ca9... 10/ (1 packages/twisted/twisted-web2_0.2.0svn.bb): twisted-web2: update to 0.2.0svn Jun 25 10:22:54 03tmbinc 07org.oe.dreambox * rbb30f1ac... 10/ (1 classes/distutils-base.bbclass): classes/distutils-base.bbclass: fix for python 2.5 Jun 25 10:22:57 03tmbinc 07org.oe.dreambox * re2d80c08... 10/ (1 conf/distro/opendreambox-1.5.conf): conf/distro/opendreambox.conf: add opendreambox-1.5 distro Jun 25 10:23:02 03koen 07org.oe.dev * r00a85a4f... 10/ (3 files in 3 dirs): libpcap: fix shared.patch and add a comment about it inside the recipe Jun 25 10:23:10 03polyonymous 07org.oe.dev * r862020d5... 10/ (1 packages/sysvinit/sysvinit_2.86.bb): sysvinit: fix build for those passing DISTRO via environment Jun 25 10:23:14 03koen 07org.oe.dev * r5b34085f... 10/ (3 files in 2 dirs): gpsbabel: add usb support and update to 1.3.3 Jun 25 10:23:18 03rwhitby 07org.oe.dev * rcdcbe26e... 10/ (1 packages/linux/ixp4xx-kernel_2.6.21.1.bb): ixp4xx-kernel: Updated 2.6.21.1 to include dsmg600 auto-power-on hack support Jun 25 10:25:42 ~oebug 2296 Jun 25 10:26:23 !oebug 2296 Jun 25 10:26:24 * * Bug 2296, Status: NEW, Created: 2007-05-10 06:17 Jun 25 10:26:25 * * andrew.paulsen(AT)gmail.com: Java updates Jun 25 10:26:26 * * http://bugs.openembedded.org/show_bug.cgi?id=2296 Jun 25 10:38:23 re Jun 25 10:43:08 ~blame koen Jun 25 10:43:08 * ibot blames koen (and Canada) for all the evil in the world Jun 25 10:43:18 * lrg_ agrees Jun 25 10:44:38 yeah Jun 25 10:44:49 in two weeks you can blame me in person :) Jun 25 10:44:55 heh Jun 25 10:45:14 turn him over to Edin mafia Jun 25 10:45:38 "OE gains mysterious investors" Jun 25 10:45:58 and "protection" Jun 25 10:46:18 What causes this: "NOTE: preferred version 2.5 of glibc-intermediate not available" Jun 25 10:46:19 XorA: did you have any time to play around with the nk.bin stuff? :> Jun 25 10:46:21 ? Jun 25 10:46:25 Jin^eLD: no Jun 25 10:46:30 me neither :P Jun 25 10:46:51 hm Jun 25 10:47:22 "using kernel version less than 2.6.17 for EABI is not recommended" Jun 25 10:47:25 I wonder Jun 25 10:48:52 djay-il: we were using eabi on 2.6.15 even Jun 25 10:48:58 Jin^eLD: I cant remeber which browser I left the link in Jun 25 10:58:34 XorA: I can look it up again+ Jun 25 11:00:21 XorA: http://winterm.gaast.net/downloads/make-nk-0.5.tar.gz Jun 25 11:00:41 and thats the corresponding page: http://winterm.gaast.net/main.php/news.html Jun 25 11:01:14 Jin^eLD: cheers Jun 25 11:02:07 np Jun 25 11:03:56 hrw: so why can't I? :-( Jun 25 11:07:17 djay-il: no idea Jun 25 11:09:57 now, 2.6.20 doesn't load at all.... Jun 25 11:18:45 Hi! Jun 25 11:18:55 well, I know have blog too %): http://pfalcon-oe.blogspot.com/ Jun 25 11:26:27 psokolovsky_: contact koen/florian on #linuxtogo to get it added to planet ltg Jun 25 11:26:49 ok, thanks for hint Jun 25 11:27:21 * florian just found a link to the feed Jun 25 11:27:29 psokolovsky_: i'll add you Jun 25 11:27:38 florian: thanks! ;-) Jun 25 11:27:40 * hrw -> coffee Jun 25 11:32:36 psokolovsky_: http://planet.linuxtogo.org/ Jun 25 11:33:41 looks nice, thanks! Jun 25 11:35:46 psokolovsky_: yw Jun 25 11:39:10 psokolovsky_: re "power management for gpios" - should uart gpios be set to output/clear in the pxa-serial driver, too? currently I'm doing this in the phone driver via the hh.org specific pxa-serial hooks. Jun 25 11:41:03 pH5: I'd think yes, but I'm not hw guy, and I'm not sure how universally this rule applies (does it has something to do w/opoen drain, etc.). So, I'd be cautions about changing it w/o exploring area a bit more ;-) Jun 25 11:48:05 hi everybody Jun 25 11:48:33 hey thesing Jun 25 11:57:01 * * OE Bug 2559 has been created by thommy(AT)tabao.de Jun 25 11:57:03 * * PATH doesn't contain /sbin and /usr/sbin for root Jun 25 11:57:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2559 Jun 25 11:59:53 how the hell does chanserv work here, cant invite him to a channel and he doesnt answer help requests Jun 25 12:00:01 * * OE Bug 2559 has been RESOLVED (WORKSFORME) by Jun 25 12:00:03 * *  PATH doesn't contain /sbin and /usr/sbin for root Jun 25 12:00:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2559 Jun 25 12:08:13 koen: su != su - ? Jun 25 12:08:51 one uses the env from your sessions, the other inherit it from the user you are switching to Jun 25 12:13:43 I mean that this bug was about it Jun 25 12:14:40 it must be Jun 25 12:14:58 since I saw no ldconfig errors after flashing an image last week Jun 25 12:15:34 koen: do you know some kind of ldconfig replacement? Jun 25 12:15:39 something not 0.5M big Jun 25 12:15:49 no, sorry Jun 25 12:16:02 but is ldconfig needed on a running system? Jun 25 12:16:16 can't you pregerenate it at image time? Jun 25 12:19:06 need to test it Jun 25 12:25:21 * Jin^eLD just got a DaVinci board Jun 25 12:25:26 hopefully I get some time to try it out :> Jun 25 12:25:36 Jin^eLD: the evm? Jun 25 12:25:43 "evm" is what? Jun 25 12:25:50 a davinci board :) Jun 25 12:26:00 ah you mean what kind? :) I did not look closely yet Jun 25 12:26:28 DM6446, EVM yes Jun 25 12:26:29 :) Jun 25 12:26:50 03koen 07org.oe.dev * rd9256db6... 10/ (1 packages/librsvg/librsvg_2.16.1.bb): librsvg 2.16.1: fix packaging Jun 25 12:27:16 I was a little surprised about the security camera, kind of cool though heh Jun 25 12:28:48 03koen 07org.oe.dev * r730a1eb9... 10/ (1 packages/gtk-webcore/midori_0.0.2.bb): midori: add version 0.0.2 of yet another gtk-webcore based web browser Jun 25 12:29:12 is anyone actively using it with OE? I know some of you guys have a board lying around, but last time I asked noone really did anything serious? Jun 25 12:29:37 crofton and I are waiting for TI to ship us some boards :) Jun 25 12:29:52 ah :) Jun 25 12:30:08 likewise has a (some?) board(s) at work Jun 25 12:30:30 I think he did not do much OE'ing with it yet Jun 25 12:30:55 no, he was looking for people to do that for him :) Jun 25 12:31:01 hehe Jun 25 12:31:31 when do you expect to get yours? Jun 25 12:31:40 in a few months Jun 25 12:31:49 the board we are getting isn't in production yet Jun 25 12:31:55 I see.. Jun 25 12:32:20 I guess I'm on my own then heh :) not sure when and if I will get a go to start hacking it though Jun 25 12:32:36 it's basically "just another ARM board" Jun 25 12:32:46 well, its the DSP that I am concerned about Jun 25 12:32:53 need to talk to it somehow Jun 25 12:32:57 yeah, ,that part is currenty a mess Jun 25 12:33:06 and you guys said it was using MV Linux which did sound scary Jun 25 12:33:11 TI is trying to push a new driver/interface architecture Jun 25 12:33:26 koen: 3rd one? Jun 25 12:33:46 dspgateway and dspfs all sucks according to rishi Jun 25 12:34:03 you can't do e.g. fast-forward with it Jun 25 12:34:12 mhm Jun 25 12:34:32 I'll be happy if can reproduce the functionality of their SDK in OE Jun 25 12:34:57 they are planning to opensource a big part in ~july Jun 25 12:35:08 that would be nice! Jun 25 12:35:11 indeed Jun 25 12:38:03 koen: btw, how far are you with bfin? Jun 25 12:46:05 Jin^eLD: not far, it seems everything I stumple upon is broken, and the fix is in their svn, but not as patch (forked gcc, forked uclibc, forked kernel, etc) Jun 25 12:46:43 doh.. sounds messy Jun 25 12:46:53 it is Jun 25 12:47:11 I didn't expect anything else from uclinux Jun 25 12:47:38 I'm planning to see if their buildroot works and try to extract patches from that Jun 25 12:48:01 the uclinux community must be really small? their webpage looked somewhat abandoned and this whole thing did not make a solid impression Jun 25 12:50:45 I suspect arm9 lured people into MMU land Jun 25 12:51:31 uclibc people used arm7 usually? maybe also some coldfire.. Jun 25 12:51:34 well, MMU land seems to be a much more comfortable place :> Jun 25 12:51:45 yeah Jun 25 12:52:03 the bfin screwage revolves around shared-flat and fdpic for shared libs Jun 25 12:52:37 s/and/or/ Jun 25 12:56:29 thesing: I just registered an #alsa-soc channel Jun 25 12:59:39 XorA: thanks. Jun 25 13:04:14 * hrw goes away Jun 25 13:04:21 bye Jun 25 13:07:01 * * OE Bug 2326 has been RESOLVED (FIXED) by hma(AT)syd.odn.ne.jp Jun 25 13:07:03 * *  minimo not aware of Fn key Jun 25 13:07:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2326 Jun 25 13:10:24 does anyone knows whether moko devices are available? Jun 25 13:10:30 I mean Neo Jun 25 13:13:27 djay-il: try #openmoko Jun 25 13:13:55 djay-il: devs (e.g. me) have a neo Jun 25 13:15:25 zecke: thanks Jun 25 13:28:56 morning Jun 25 13:29:00 * * OE Bug 2326 has been REOPENED by hma(AT)syd.odn.ne.jp Jun 25 13:29:02 * *  minimo not aware of Fn key Jun 25 13:29:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2326 Jun 25 13:37:08 djay-il: did you check the .config in your work directory to make sure the kernel built has the EABI enabled? Jun 25 13:38:35 djay-il: tmp/work/compulab-pxa270-angstrom-linux-uclibcgnueabi/linux-2.6.21-r3 Jun 25 13:38:37 /linux-2.6.21/.config Jun 25 13:42:21 * djay-il checks Jun 25 13:43:34 cbrake: yes Jun 25 13:44:39 djay-il: I don't know then. The jtag debugger and the EABI have been the only two things I have run into that will cause init not to run. Jun 25 13:44:49 I now have the following: 2.6.20 kernel with EABI, stops again on "Freeing init...". System seems alive. If I use the same kernel but compulab's rootfs, it loads fine. chrooting then into my rootfs works fine Jun 25 13:45:17 djay-il: you using jffs2 for filesys? Jun 25 13:45:28 cbrake: nfs mount Jun 25 13:46:04 djay-il: ohh. I've not tried that in a long time with the cm-x270, so I don't have much experience there. Jun 25 13:46:22 and no, I use ext2 Jun 25 13:46:26 djay-il: nod Jun 25 13:47:09 djay-il: well, I guess I would keep trying things until you get it narrowed down. Maybe putting the image in flash and see if you see the same thing, etc. Jun 25 13:47:48 cbrake: well, it shouldn't matter I guess. kernel does mount NFS root, but init doesn't start. How can I debug it? Jun 25 13:49:09 djay-il: good point. Debugging init problems is fairly difficult as there is a lot going on at that poing. Jun 25 13:49:13 *point Jun 25 13:50:05 that is the question at the moment. I'm starting to suspect something weird, as chrooting "proves" that rootfs is fine Jun 25 13:51:50 djay-il: if its nfs that wireshark is good to see what files are being read from Jun 25 13:51:53 djay-il: I would typically try different senarios and eventually something will pop out. This is the difficult part with embedded Linux -- things don't work, but it is way too complex to debug by tracing line-by-line :-\ Jun 25 13:52:35 cbrake: what would you suggest to start with? Jun 25 13:55:06 djay-il: XorA has a good suggestion. dunno, I would probably load a small JFFS2 filesys in flash or initramfs. I often don't get ideas until I start "doing", so it is difficult to advise. Jun 25 13:58:54 ok, I've got this: adding init=/bin/ash gave me 1. /bin/ash can't access tty; job control turned off; 2. I've got shell :-) Jun 25 14:00:04 maybe I have access problem with rootfs? Jun 25 14:00:27 no_root_squash? Jun 25 14:01:58 Good afternoon Jun 25 14:02:04 hey sirfred Jun 25 14:02:05 yo sirfred Jun 25 14:02:21 Hi friends Jun 25 14:02:36 steliosk: do you also work 50% now, or is that reserved for the bureaucrats? Jun 25 14:04:23 A question about font rendering on angstrom. A little confused about what seem to be different pango backends, there is pango-cairo, pango-xft, pango-ft2,... Jun 25 14:04:33 Are all them actually used? Jun 25 14:05:34 I'm asking that because I'm trying to find where is xrender really used in the whole picture. Jun 25 14:05:40 pango-cairo for the gtk+ we use I think Jun 25 14:07:02 XorA: Perhaps that's eventually calling the xrender stuff. Jun 25 14:07:26 anyone tell me how to force autoreconf before configure? Jun 25 14:07:42 Zero_Chaos: by not overriding do_configure Jun 25 14:07:43 Zero_Chaos: hrw already answered your question Jun 25 14:07:58 Hmm, a new webkit based browser: midori Jun 25 14:08:19 sirfred: already in OE :) Jun 25 14:08:27 koen: and I read that, but it didn't seem to work. when I completely remove do_configure it seems to use the existing configure no regen Jun 25 14:08:33 s/webkit/webcore/ Jun 25 14:08:38 koen: Yes, I've just updated and see the adition Jun 25 14:08:47 XorA: Are they different? Jun 25 14:09:10 one is new and swanky and maintained the other isnt Jun 25 14:09:20 webkit is 'upstream' and webcore is nokia's gtk version Jun 25 14:09:35 Thanks. I thought they were the same thing. Jun 25 14:10:06 webcore is basically a C api for webkit nowadays Jun 25 14:10:25 but Zecke is working on gtk support fro webkit Jun 25 14:11:45 The tcpmp guy anwered me about their ati hardware idct implementation. Jun 25 14:11:52 Just reverse engineering. :-( Jun 25 14:12:06 better than a kick in the stones Jun 25 14:12:35 XorA: Slightly Jun 25 14:13:35 On the other side, I was also considering xrender support. But since the imageon only support 4bpp alpha channels, I don't know if it worths the effort Jun 25 14:13:54 I see corecodecs still havent managed to get coremedia player released a year and a half after announcing it Jun 25 14:14:17 XorA: I've also asked him about that. Jun 25 14:14:35 tcpmp works nice on my Omap phone Jun 25 14:14:41 XorA: He say they have a still private linux port. The open source release coming soon. Jun 25 14:15:12 * koen is getting a bit allergic to 'soon' Jun 25 14:15:33 koen: Like canola going opensource? Jun 25 14:15:36 * XorA wishes linwizard project was alive Jun 25 14:15:45 sirfred: yeah Jun 25 14:15:48 ~lart nokia legal Jun 25 14:15:48 * ibot frags nokia legal with his BFG9000 Jun 25 14:15:52 koen: :) Jun 25 14:17:15 koen it won't work ... you need to use silver bullets or a stake in the heart ;) Jun 25 14:17:41 03pfalcon 07org.oe.dev * rb02088ae... 10/ (1 conf/machine/htcwallaby.conf): htcwallaby.conf: Bare config for HTC Walalby (very first XDA) machine. Jun 25 14:17:52 03pfalcon 07org.oe.dev * re16d41b7... 10/ (11 files in 10 dirs): Jun 25 14:17:52 merge of '730a1eb946899965bd1703a6dcb9e962751d914a' Jun 25 14:17:52 and 'b02088ae9e5847432d378e1127122006d2d82fa9' Jun 25 14:18:56 sirfred: didn't you try to look at wince version in the meantime? Jun 25 14:19:18 psokolovsky_: Yes, I've got a toshiba e740 dll Jun 25 14:19:40 psokolovsky_: The good thing is that it has a iDCT api Jun 25 14:20:04 psokolovsky_: The bad thing is that almost everything relays in the card microcode Jun 25 14:21:03 hm, weird Jun 25 14:21:18 Using the microcode should be better than just the cmdfifo, as xserver-kdrive-imageon is currently doing, but, we need to load that thing into the card. Jun 25 14:21:27 sirfred: and don' they have just normal avi, etc. decoding hand-optimized for pxa for example? Jun 25 14:22:07 psokolovsky_: Not sure, but the main problem for the imageon is the bandwidth, imho Jun 25 14:22:37 psokolovsky_: If we have to decode the idct and send to the card the whole decompressed frame, we are dead. Jun 25 14:22:38 sirfred: ah well, you're working on X optimization, not video? Jun 25 14:23:00 * * OE Bug 2561 has been created by utx(AT)penguin.cz Jun 25 14:23:01 psokolovsky_: Yes, the xv port. Jun 25 14:23:02 * * qemu links with zlib, but zlib is compiled after qemu Jun 25 14:23:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2561 Jun 25 14:23:36 ok, jsut trying to a be a little bit in loop ;-) as usual, I'm glad you're working on that stuff! Jun 25 14:23:52 psokolovsky_: But as far as it only works on plain imageon 100, I'm afraid there's no much interest. Jun 25 14:44:41 djay-il: looking back through the logs -- did you figure out the diff between bin and img batteries? The img files have a 4 byte header that the cm-x270 bootloader uses. Jun 25 14:45:31 djay-il: I'm open to other naming ideas if there is something more clear -- maybe kernel.cmx270 ? Jun 25 14:47:20 cbrake: yea, "kernel" would be better. What can .img be used for? Jun 25 14:47:52 djay-il: that is the format of the file in flash. Jun 25 14:48:05 djay-il: if you download the file using tftp in the cm-x270, it attaches the header for you. Jun 25 14:48:26 djay-il: if you program the file using other means (cat > /dev/mtd2, etc), use the *.img files Jun 25 14:48:58 does anybody know how to turn an arbitrary hardware address range (in this case the pxa27x SRAM) into a mmappable region? Jun 25 14:49:19 I have just used ioremap to put the pxafb framebuffer in SRAM, but now /dev/fb0 can't be mmapped anymore. Jun 25 14:50:54 cbrake: I just found that my rootfs doesn't have /dev/initctl. Is that OK? Jun 25 14:54:07 pH5: I thought SRAM had a driver in latest 2.6 kernels Jun 25 14:54:26 pH5: you used that to allocate SRAM memory Jun 25 14:56:51 really? I'm not aware of that. I just ioremapped the 256k at 0x5c000000(SRAM_PHYS) to get a virtual address and and set the dma source address to it SRAM_PHYS. Jun 25 14:58:54 03tmbinc 07org.oe.dreambox * r19508614... 10/ (5 files in 3 dirs): liboil: fix for embedded powerpc without altivec and without fpu Jun 25 14:59:25 cbrake: I have to go, could I talk to you tomorrow? Jun 25 14:59:41 djay-il: sure, I'll be here :-) Jun 25 15:00:09 cbrake: cool, thanks Jun 25 15:00:13 laters all Jun 25 15:03:07 pH5: maybe it was never merged in Jun 25 15:06:15 pH5: ah, seems to be an RP patch http://www.rpsys.net/openzaurus/patches/tosort/0206_linux-2.6.16-pxa27x-sram.patch Jun 25 15:06:28 ah, many thanks! Jun 25 15:14:23 ~ibot Jun 25 15:14:24 i guess ibot is a blootbot written in perl run by TimRiker on his server. logs on http://ibot.rikers.org// , ibot, jbot, apt are all the same process. It uses sqlite, but mysql or other SQL storage is also supported. Jun 25 15:23:25 koen : why work 50% less ? Jun 25 15:24:33 steliosk: the news said that goverment workers in athens were working halfdays due to the heat Jun 25 15:25:28 koen : 50% less means they don't work at all :) Jun 25 15:27:03 :-D Jun 25 15:27:32 Government workers are the same no matter what continent they may be found upon, it seems. Jun 25 15:33:28 heh i think even on another galaxy it will be the same :) Jun 25 15:38:56 03koen 07org.oe.dev * r04a37ec5... 10/ (3 files in 2 dirs): osb-nrcore: package icons as well Jun 25 15:45:38 XorA: that patch doesn't solve my problems, but I've found the "Re-mapping SRAM to user space" discussion on linux-arm-kernel thanks to it. Jun 25 18:29:59 koen, are you here? Jun 25 18:32:34 step2back: hi! here's issue with palmld patches: marex also submitted some changes. you guys need to figure out which one (or combined) to apply Jun 25 18:34:55 hm.... psokolovsky_, i'm so sorry for our ... conflict. But I think without farcaller it's not resolve Jun 25 18:35:35 I wait for him and a work will be continue Jun 25 18:35:54 step2back: I'm not sure there's conflict, but some duplication of effort for sure. Jun 25 18:39:05 I went here to get reason for not appling patches for machine config of LD. What do I do wrong? Jun 25 18:41:02 * * OE Bug 2562 has been created by tomtechguy(AT)gmail.com Jun 25 18:41:04 * * X server crash on poodle Jun 25 18:41:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2562 Jun 25 18:41:26 I built the "alpha release"... well used it and find bugs in palmld.conf, :) Jun 25 18:42:33 step2back: if there're 2 competing patches, tehir authors first need to resolve issues among themselves, and provide 1 patch. Jun 25 18:43:23 what is second patch? Jun 25 18:43:57 I miss something in our subject.... Jun 25 18:44:33 psokolovsky_, do you use opie? Jun 25 18:45:22 well, both patches aren't ready for inclusion, so there's no problem yet Jun 25 18:46:32 koen, hello. Jun 25 18:46:53 I don't understand where is second patch?.. Jun 25 18:47:13 it the bug tracker Jun 25 18:47:18 in* Jun 25 18:47:44 http://bugs.openembedded.org/buglist.cgi?quicksearch=palmld Jun 25 18:48:22 ok... I find it. Jun 25 18:49:02 * * OE Bug 2556 has been marked as DUPLICATE of bug 2533 by Jun 25 18:49:04 * *  PalmLD: new machine file ( based on hx4700 which should be correct) Jun 25 18:49:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2556 Jun 25 18:49:58 step2back: yep, kinda use Jun 25 18:50:31 * koen tries some more clearer language Jun 25 18:50:55 , please, explain "even ship them?" Jun 25 18:51:31 marek doesn't include them in task base, so they won't end up in the image Jun 25 18:51:44 so autoloading them doesn't make sense Jun 25 18:52:21 ok Jun 25 18:52:26 people listen up: Jun 25 18:52:33 what? Jun 25 18:52:37 * mwester listens up Jun 25 18:52:37 re Jun 25 18:52:47 "remove" means that you delete that what is referenced Jun 25 18:52:58 it doesn't mean "put it in your updated patch" Jun 25 18:53:32 remove what I should? Jun 25 18:54:32 I will be happy If somebody make modules "build-ined" Jun 25 18:54:55 I will be happy if people stop putting distro crap in machine.confs Jun 25 18:55:59 I remove BOOTSTRAP_EXTRA_RDEPENDS Jun 25 18:56:13 http://bugs.openembedded.org/attachment.cgi?id=1669 Jun 25 18:56:32 step2back: regarding modules, I offered Marex global solution to such issues: just add palmld support to defconfigman, and then it will be as good (or as bad) as bunch of other machines. then, all improvments will benefit all affacted machines, not just one. Jun 25 18:57:27 psokolovsky_: to quote hrw loosely 'some people do not want to cooperate' Jun 25 18:58:35 koen: please don't frustrate them. I spend lotsa time just motivating them to submit patches to OE, instead of digging own hole. let them take their time to learn all that. Jun 25 18:59:05 Guys, My english and my developer status|knowledge's do not allow understand you... :(... I want cooperate Jun 25 18:59:07 psokolovsky_: right, but maybe they want their own hole Jun 25 18:59:34 I tried to get the palm people to use OE, but they refused 'because it is to wince centric' Jun 25 18:59:36 koen: no, they don't ;-) and we too Jun 25 18:59:58 then it was 'OE sucks, we want g3nt00' Jun 25 19:00:20 OE is gentoo without the benefit of use flags Jun 25 19:00:22 then I put them on my 'morons, ignore' list and was happily ever after Jun 25 19:00:29 koen: that's not the case, they move in the right direction - after all, they were playing too long with all that and started to get tired of that ;-) Jun 25 19:00:32 Zero_Chaos: it isn't Jun 25 19:00:40 sure looks like it Jun 25 19:00:48 koen: times change Jun 25 19:00:55 Zero_Chaos: does your zaurus spend eons recompiling to 'update'? Jun 25 19:01:11 koen you can crosscompile gentoo Jun 25 19:01:19 koen: no, my desktop does. And I said OE not Angstrom. Jun 25 19:01:24 , later H&D supports own familiar repository... I don't want that happen with Angstrom Jun 25 19:01:29 chouimat: yes, but that's not what they meant Jun 25 19:02:10 koen I see no problem if they use the new 5GHz arm based chip ;) Jun 25 19:02:42 chouimat: palm devices :) Jun 25 19:02:59 some have 16 megs of ram Jun 25 19:03:02 koen 3 minutes of battery life ;) Jun 25 19:03:15 step2back: I hope we'll handle palmld issue. but you really should talk with marex and compare your patches. Jun 25 19:03:41 * chouimat won't change his palm m125 for any of the new ones ... I have 40 days on 2 AAA batteries :) Jun 25 19:04:19 chouimat: palm V is even better. 5-6 weeks on internal battery Jun 25 19:04:23 psokolovsky_: if step2back remove all {DISTRO,COMBINED}_FEATURES it's ready to go in Jun 25 19:04:40 , ok. I understand the target Jun 25 19:04:41 Philippe: bah Jun 25 19:05:11 koen: I'm ready to do such cleanups, and look over palm configs, while there're no hackndev member in OE team Jun 25 19:05:29 chouimat: no need to meddle with annoying batteries Jun 25 19:06:02 Philippe: as I said I won't change it anyway ... Jun 25 19:06:16 , ${@base_contains("COMBINED_FEATURES", "irda", "opie-irdaapplet", "",d)} \ not work.... :( Jun 25 19:06:24 psokolovsky_: commit #2533 without {DISTRO,COMBINED}_FEATURES Jun 25 19:06:30 step2back: it does Jun 25 19:06:41 chouimat: :-) Jun 25 19:06:46 koen: ok Jun 25 19:07:02 step2back: at least 2 OE core developers disagree with you, how many more does it take for you to get the hint? Jun 25 19:07:12 If it's be removed I don't miss something. Jun 25 19:07:17 koen 328 Jun 25 19:08:01 * * OE Bug 2507 has been RESOLVED (WORKSFORME) by papercrane(AT)reversefold.com Jun 25 19:08:03 * *  building for angstrom-2007.1 tries to build glibc 2.3.2 Jun 25 19:08:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2507 Jun 25 19:08:45 ... I try to get a facts... until I stop talking about it. Jun 25 19:09:20 , wait! Jun 25 19:09:42 2 string must be there ... Jun 25 19:10:13 step2back: please upload final version to tracker Jun 25 19:10:29 one minutes please... Jun 25 19:20:42 , ready. commit please only palmld.conf: defconfig change with new kernel version Jun 25 19:21:22 step2back: what new version? what about current version? Jun 25 19:21:54 not matter. Jun 25 19:22:02 * * OE Bug 2563 has been created by discdonker(AT)gmail.com Jun 25 19:22:04 * * GPE pointer does not accurately track stylus on Zaurus SL-6000 Jun 25 19:22:06 * * http://bugs.openembedded.org/show_bug.cgi?id=2563 Jun 25 19:26:01 * * OE Bug 2534 has been REOPENED by Jun 25 19:26:03 * *  Hackndev project move to 2.6.21 Jun 25 19:26:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2534 Jun 25 19:32:04 step2back: then close ticket with it yourself Jun 25 19:36:40 which ticket? Jun 25 19:38:19 step2back: your ticket with defconfig attached Jun 25 19:40:36 it's a actual ticket... but I don't make new defconfig, because new version of kernel release few times ago Jun 25 19:41:11 Marex put in svn broken defconfigs.... :( Jun 25 19:41:35 prolific usb to serial adaptor. Any module available for that in OE? Jun 25 19:42:30 hi Jun 25 19:42:50 dcordes: pl2303 kernel module Jun 25 19:43:25 dcordes: lots of machines compile it by default: http://www.angstrom-distribution.org/repo/?action=details&pnm=kernel-module-pl2303 Jun 25 19:43:29 step2back: broken defconfigs in svn should be discussed in #hackndev and commits to OE should be with agreement of developers, shouldn't it? Jun 25 19:45:06 , hmmm I'm banned on h&d first and I tried discussed broken defconfig later in form. You know what happen after that Jun 25 19:45:38 koen, oh ok so I guess my akita already has it Jun 25 19:46:02 How can I simply test my gps device's function? Jun 25 19:46:15 step2back, you are banned because you were insulting people on #hackndev, you put naughty words on hackndev website and you dont know how to behave towards other people Jun 25 19:46:32 psokolovsky_: could you enable all usb and webcam modules in defconfig man for machine with usbhost (e.g. builtin or sl811-cs based) ? Jun 25 19:47:15 step2back, you never put anything to svn and if that defconfig is broken just report it and I will fix it, or do you want to humiliare someone again? Jun 25 19:47:25 Marex: calm down, please :) Jun 25 19:48:49 koen: ok, will look into that Jun 25 19:48:50 Sleep_Walker, I wont let someone who didnt write single line of useful kernel code humiliate me who took the port where it is now :E Jun 25 19:49:21 Marex and Sleep_Walker if you want continue I can join to #hnd.... Jun 25 19:49:32 step2back: 1] I believe that this situation can be solved peacefully 2] you crossed the line of tiff and published in trac very impolite content Jun 25 19:49:38 step2back: yes, please Jun 25 19:50:21 step2back, you dont know how to behave towards people and you're spreading lies in public trying to throw bad light on others :E Jun 25 19:50:51 Marex: this discussion should take place at #hackndev .... Jun 25 19:51:25 Sleep_Walker, no, this discussion should take place here so anyone can see MY point of view about this stuff Jun 25 19:52:01 Marex: if you think that this wont throw bad light on others... :b Jun 25 19:52:18 Marex: yes, please move to project channel, discuss issue of *palmld*, and let oe maintainers know Jun 25 19:52:54 ;-E Jun 25 19:55:24 since over here I get the opposite impression Jun 25 19:55:45 step2back behaving and marex being an annoying git who insults people Jun 25 19:56:16 * chouimat is waiting for blood Jun 25 19:56:25 time to go get some pop corn Jun 25 19:58:34 Why am I not laugh? Jun 25 19:58:40 koen, shall I look up a proof for you? Jun 25 19:59:02 Marex: I only care about #oe Jun 25 19:59:43 koen, I care about " Marex put in svn broken defconfigs.... :(" and I care about who did the kernel porting job Jun 25 19:59:59 http://hackndev.com/node/754 and please see this ... Jun 25 20:00:31 Marex: the stuff work well in qemu? Jun 25 20:00:54 chouimat, what stuff? Jun 25 20:01:06 Marex: the hackndev Jun 25 20:01:27 chouimat, you mean hackndev/qemu-pxa/palmXY? Jun 25 20:01:37 yes Jun 25 20:01:56 chouimat, palmtc can even boot palmos original roms to some effort Jun 25 20:03:21 Marex: I will have to look at this then .. Jun 25 20:04:00 chouimat, me too since I killed my palmtc while writing mtd device driver ... no comments please ;) Jun 25 20:04:21 Marex: done worse ... Jun 25 20:05:09 chouimat, I guess I found jtag pads or I have at least idea where they might be so if I didnt trashed to flash it might be repairable Jun 25 20:06:57 Marex: last years I think I killed aroun 45 development boards ... thanks it wasn't all my fault and thanks the board never got into production ... and the nutcase (hardware designer/manager) never wanted to implement the jtag in the boards Jun 25 20:07:58 chouimat, the worst is that device inst actually even mine :-/ Jun 25 20:08:11 Marex: ouch Jun 25 20:08:27 I borrowed it from a friend to finish palmtc/wifi and I really had to try mtd ;) Jun 25 20:24:18 03koen 07org.oe.dev * r425d2b72... 10/ (4 files in 2 dirs): bluez: add 3.12 Jun 25 20:58:02 03koen 07org.oe.dev * ra7b55000... 10/ (1 packages/bluez/bluez-gnome_0.9.bb): bluez-gnome: add 0.9 Jun 25 20:58:09 03koen 07org.oe.dev * r6e0bf65a... 10/ (1 packages/bluez/bluez-hcidump_1.37.bb): bluez-hcidump: add 1.37, that actually builds against a recent bluez-libs Jun 25 21:10:15 morning Jun 25 21:12:12 'mornin' Jun 25 21:13:10 hey hrw Jun 25 21:14:27 mwester: 23:14 here Jun 25 21:15:10 07:15 here Jun 25 21:16:07 * chouimat|tv goes watch some Star trek Voyager and CSI (Las Vegas) and eat something Jun 25 21:16:36 hrw: Ok, you're just 46 min short of morning,then :) Jun 25 21:16:54 hrw: morning sir Jun 25 21:18:21 * hrw came to see what is going on Jun 25 21:53:39 Good night. Jun 25 22:03:48 hi Jun 25 22:22:39 bye Jun 25 22:45:00 * * OE Bug 2533 has been RESOLVED (FIXED) by Jun 25 22:45:02 * *  PalmLD new machine config Jun 25 22:45:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2533 Jun 25 22:59:20 03Stepan 07org.oe.dev * r7733cf94... 10/ (1 conf/machine/palmld.conf): Jun 25 22:59:20 palmld.conf: Elaborate and update per current state. Jun 25 22:59:20 * Closes #2533. Jun 25 22:59:25 03kristoffer 07org.oe.dev * r01ba9ef8... 10/ (1 packages/keymaps/files/jornada6xx/keymap-uk.map): packages/keymaps/files/jornada6xx/keymap-uk.map: Minor changes to uk keymap Jun 25 22:59:34 03pfalcon 07org.oe.dev * rb8b628bf... 10/ (19 files in 12 dirs): Jun 25 22:59:34 merge of '01ba9ef84a43848013480079effda2abb4d3b581' Jun 25 22:59:34 and '7733cf946eb165c244b4967fd9a2e3bf3b801ceb' Jun 25 23:07:01 * * OE Bug 2535 has been RESOLVED (WONTFIX) by Jun 25 23:07:03 * *  PalmLD defconfig for 2.6.21 Jun 25 23:07:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2535 Jun 25 23:13:01 * * OE Bug 2534 has been RESOLVED (FIXED) by Jun 25 23:13:03 * *  Hackndev project move to 2.6.21 Jun 25 23:13:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2534 Jun 25 23:23:00 * * OE Bug 2564 has been created by kjkillick(AT)hotmail.com Jun 25 23:23:02 * * bitbake opie-image fails to build Jun 25 23:23:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2564 Jun 26 00:17:01 * * OE Bug 2565 has been created by longislandsmell(AT)gmail.com Jun 26 00:17:03 * * segmentation fault on hx4700_acx load Jun 26 00:17:04 * * http://bugs.openembedded.org/show_bug.cgi?id=2565 Jun 26 00:53:39 will qmake every be available for the Zaurus? Jun 26 01:14:41 03coredump2 07org.oe.dev * rbeb9be9b... 10/ (8 files in 5 dirs): Remove hentges meta-data as syncing svn with mtn is too much work (and the overlay works perfectly anyway) Jun 26 01:14:49 03coredump2 07org.oe.dev * r990d1b4c... 10/ (25 files in 15 dirs): Jun 26 01:14:49 merge of '862020d539b671d14575834bfda296adc8ef20ec' Jun 26 01:14:49 and 'beb9be9b22208b02578f8cc4ad29efb48be52977' Jun 26 01:14:55 03coredump2 07org.oe.dev * ra7a2a49a... 10/ (3 files in 3 dirs): fbreader: Enable large icons for SL-Cxx00 devices & add first version of zaurus-VGA.patch, fixing font sizes Jun 26 01:15:01 03coredump2 07org.oe.dev * rd2b977e2... 10/ (3 files in 2 dirs): altboot: Add SVNR66, fixing mounts of partitions w/o UUID in 05-boot-HDD. Thanks goes to summatusmentis for nailing down this error Jun 26 01:15:10 03coredump2 07org.oe.dev * rc21a3380... 10/ (27 files in 14 dirs): Jun 26 01:15:10 merge of 'b8b628bfe4824f594298abb40c7545cefd3aa41a' Jun 26 01:15:10 and 'd2b977e2b47520e4d1f916319fbbf8f5f7202b7c' Jun 26 02:26:01 * * OE Bug 2566 has been created by  Jun 26 02:26:03 * * Build of ecore-native 0.9.9. 038 fails with missing Evas_Engine_Software_X11.h Jun 26 02:26:05 * * http://bugs.openembedded.org/show_bug.cgi?id=2566 **** ENDING LOGGING AT Tue Jun 26 02:59:56 2007