**** BEGIN LOGGING AT Fri Jun 12 02:59:56 2009 Jun 12 05:00:44 03Rolf Leggewie  07org.openembedded.dev * r973a007b74 10openembedded.git/recipes/xorg-lib/xtrans-native_1.2.1.bb: xtrans: add native 1.2.1 package Jun 12 05:00:55 03Rolf Leggewie  07org.openembedded.dev * rb23e6b9e4d 10openembedded.git/recipes/xorg-proto/xextproto-native_7.0.3.bb: xextproto: add native 7.0.3 package Jun 12 06:14:14 cbrake_away: locale problem on melo fixed Jun 12 06:16:43 I have fixed in issue in the origin/stable/2009 branch, can anyone use "git format-patch -s --subject-prefix="PATCH][STABLE" Jun 12 06:16:55 to send a patch to the mailing list? Jun 12 06:17:03 an issue Jun 12 06:18:04 or does that just give me a preformatted mail that I should send myself? All new to git **** ENDING LOGGING AT Fri Jun 12 06:20:50 2009 **** BEGIN LOGGING AT Fri Jun 12 06:21:13 2009 Jun 12 06:29:05 * * OE Bug 5155 has been created by erik.boto(AT)gmail.com Jun 12 06:29:07 * * Taglib doesn't stage all necessary files - patch fix included (origin/stable/2009 branch) Jun 12 06:29:09 * * http://bugs.openembedded.net/show_bug.cgi?id=5155 Jun 12 07:05:31 erbo: Is the fix already in .dev? Jun 12 07:09:53 Laibsch: no Jun 12 07:10:08 then push it to .dev first Jun 12 07:10:13 google our wiki for stable Jun 12 07:10:26 search our wiki for stable Jun 12 07:13:45 I did read it but maybe I missed something about pushing to .dev first (allthough it seem very logical).. too much work lately :) Jun 12 07:13:52 i'Ãll read it again Jun 12 07:17:16 erbo: you're right it doesn't say "push to .dev first", there Jun 12 07:17:31 but that is kind of the policy Jun 12 07:17:48 It's a wiki, somebody should update that page Jun 12 07:20:15 good morning Jun 12 07:26:05 morning Jun 12 07:39:25 morning Jun 12 07:53:32 morning Jun 12 07:55:14 Laibsch: added note Jun 12 07:55:48 hrw: thanks Jun 12 08:26:03 anyone working with an at91sam9263ek board? Jun 12 08:26:29 b0fh: I was Jun 12 08:26:46 but my board broke and do not want to power on Jun 12 08:27:45 hrw: that's too bad.. They are pretty expensive, can't fix it? Jun 12 08:30:01 b0fh: I got mine as gift from Atmel. Jun 12 08:30:23 now it is at my friend office and wait for his free time to check how to get it working Jun 12 08:30:50 Crofton: I'm out a a few hours right now, but please let me know when we can do the DNS stuff to move bugzilla to melo. Jun 12 08:31:06 I finally beat that Debian package into shape enough that it seems to be working ;-) Jun 12 08:31:39 hrw: lucky you.. (for getting it for free). I really like the board, OE is running great on it, was wondering if anyone was applying the patches by Nicolas Ferre et al to OE. (from linux4sam.org) Jun 12 08:32:15 which patches? Jun 12 08:32:42 at91 + at91exp are used Jun 12 08:34:04 hrw: those apply to the kernel don't they? I still miss functional backlight and suspending the processor. And do favor the most recent OE release over that old release linux4sam.org points at with their patches (http://www.linux4sam.org/twiki/bin/view/Linux4SAM/OpenEmbeddedAngstromBuild#AT91_OpenEmbedded_Angstrom_overl) Jun 12 08:38:07 b0fh: their patch is for very old OE Jun 12 08:38:44 hrw: I know... but those patches, combined with a recent kernel/patches produce a more functional kernel :-( Jun 12 08:40:07 b0fh: so prepare patch and send to OEML Jun 12 08:40:51 I will review/ack/merge Jun 12 08:44:49 i have build kernel modules outside kernel sources if i change the kernel using make menuconfig. should i need to rebuild the modules again? Jun 12 08:45:53 vsr|work: I suggest yes Jun 12 08:46:10 hi flameman - which arm device you play with now? Jun 12 08:46:45 hrw :-) I'll take an effort.. I was (perhaps silly) hoping the linux4sam folks would forward their patches upstream. Jun 12 08:48:26 b0fh: Ulf Samuelsson did a lot of merging in past Jun 12 08:48:48 hrw zaurus akita Jun 12 08:49:14 hrw unfortunately i get a bit of issue with the sound card Jun 12 08:49:55 someone should really drop 2.6.2x kernels for zaurus and take effort to get 2.6.30 on it Jun 12 08:51:26 hrw well .... 2.6.30 is broken, too ... anyway i have 2.6.23 and 2.6.30-rc6 here Jun 12 08:51:45 hrw: Well, to be honest I was surprised what OE produced by just adding at91sam9263ek as target. Jun 12 08:52:27 b0fh: ;) Jun 12 08:52:40 b0fh: in plus or in minus? Jun 12 08:53:12 hrw: alsa is working ok, ethernet, touchscreen all work ok. It lacks the suspend and backlight (didn't test i2c etc). PLUS! :-) Jun 12 08:54:15 it took me a while to findout why kernel oops on alsa playing Jun 12 08:54:33 b0fh akita has a good status of art in 2.6.23: lan, uart, irda, hda, kb, fb: working .... what is not working is sound, touch screen Jun 12 08:54:46 also the i2c bus is bad implemented Jun 12 08:54:48 hrw: tried to get a nxp lpc2478 a few days, to no avail, nothing. But my skills are more software than hardware. Jun 12 08:55:16 heya Jun 12 08:56:00 Any ideas why HOME is not set when in devshell? And do you think this may be related to shasum-native failing when I have ccache? Jun 12 08:56:30 flameman: I'll give i2c a go, this is a 2.6.28 kernel with a recent OE. Jun 12 08:58:08 vhallac: fedora? Jun 12 08:58:58 slackware Jun 12 08:59:17 A collegue installed it, and I inherited the build failure. :) Jun 12 09:00:10 Funny thing is, my Debian box doesn't have $HOME set either Jun 12 09:00:22 But ccache -V works in one, and not in the other Jun 12 09:00:39 vhallac: remove ccache? Jun 12 09:00:51 Yup. But I want to use it? Jun 12 09:01:17 hrw: BTW fedora 11 runs OE with no issues Jun 12 09:01:26 XorA: good to know Jun 12 09:01:27 To be clear, if I set CCACHE="", and rename ccache (or remove it), the build works fine Jun 12 09:01:30 XorA: binary locales too? Jun 12 09:01:35 hrw: yes Jun 12 09:01:40 XorA: finally Jun 12 09:01:41 hrw: fedora 11 here Jun 12 09:02:08 no more 'if host-os == fedora then compile-own-kernel' Jun 12 09:02:11 hrw: is there a way to avoid recompilation of kernel modules Jun 12 09:02:12 hrw: I had to make some small fixes due to really bad C code in some of our native utils, but thats about all Jun 12 09:02:44 b0fh have you got akita ? if so, have you ever had success with the sound card ? Jun 12 09:03:15 vsr|work: some kernel structures can be changed - it is better to recompile external modules afte kernel config change Jun 12 09:03:33 b0fh about pxa linux, the i2c bus has been reimplemented kernel >=2.6.28 Jun 12 09:03:52 * b0fh googles akita Jun 12 09:04:03 the problem is we dont have sources of kernel modules it has given against one specific kernel Jun 12 09:04:12 so we change modules and it wont insmod Jun 12 09:04:39 i remember some option in make menuconfig that is there. i dont remember it at the moment Jun 12 09:04:39 flameman: I've got the evalkit from atmel: http://tinyurl.com/lmbbkd (first version) Jun 12 09:06:15 vsr|work: missing symbols? Jun 12 09:08:59 hrw: i dont find out. not sure Jun 12 09:09:11 hrw: some option should be there isnt it Jun 12 09:09:42 some work around, some guys dont give sources you have to use, they build it against a specific kernel version only Jun 12 09:11:31 03Koen Kooi  07org.openembedded.dev * ra9bba85d50 10openembedded.git/recipes/nautilus/nautilus_2.26.1.bb: nautilus: add libunique to DEPENDS Jun 12 09:11:32 03Koen Kooi  07org.openembedded.dev * r32f0c50271 10openembedded.git/recipes/libdlo/kernel-module-udlfb/ (udlfb.c udlfb.h): kernel-module-uldfb: sync with mainline, no functional changes Jun 12 09:11:32 03Koen Kooi  07org.openembedded.dev * r9c3ce6cf46 10openembedded.git/recipes/fontconfig/fontconfig_2.6.99.bb: fontconfig: add fedora branch Jun 12 09:16:30 flameman: both kernel 2.6.27 and 2.6.28 list I2C_AT91 as BROKEN :-( Jun 12 09:16:50 flameman: description states: Using low I2C clock rates may partially work..... Jun 12 09:23:18 flameman: grmpf.. I removed the '&& BROKEN' from the Kconfig but the at91_i2c doesn't show up in make menuconfig *confused* Jun 12 09:25:46 flameman: Hmm, in 2.6.27 it does show up. Bummer.. I was hoping the 2.6.28 i2c would be available.. Jun 12 09:26:31 Got it. Bad build of ccache on system - kind of. Jun 12 09:39:00 florian: good morning Jun 12 09:39:22 good morning Jun 12 09:43:33 03Koen Kooi  07org.openembedded.dev * r00a803f6a9 10openembedded.git/recipes/imagemagick/imagemagick_6.4.4-1.bb: image-magick: fix do_configure Jun 12 09:44:50 fscking beagleboard and its crappy usb controller Jun 12 09:46:22 I really wonder why they don't manage to get it supported properly. Jun 12 09:48:17 sometimes reboot helps Jun 12 09:48:21 but only sometimes Jun 12 09:52:12 Moin moin Jun 12 09:53:18 that does sound a bit weird. how hard can usb be? Jun 12 09:53:25 presumably it's just ehci or something, right? Jun 12 09:54:35 ehci Jun 12 09:56:09 right, in common with 100 million PCs. seems like that driver should be pretty well debugged by now. Jun 12 09:57:11 sure Jun 12 09:57:26 but it is 'musb' (what ever it means) Jun 12 09:57:29 morons usb? Jun 12 09:57:36 heh, it seems Jun 12 09:58:19 ~seen mickeyl Jun 12 09:58:20 mickeyl was last seen on IRC in channel #oe, 2d 18h 30m 52s ago, saying: 'thanks'. Jun 12 09:58:41 florian: do we have some flyers for linuxtag? Jun 12 09:59:09 hrw: not yet Jun 12 10:02:45 who is Yuri ? anybody knows his name ? he should be involved in akita (zaurus) Jun 12 10:05:57 florian: l can order flyers/business cards etc in Poland for quite low prices Jun 12 10:06:07 florian: 2500 A5 flyers for 50 eur Jun 12 10:06:54 hrw: that's cheap indeed. in fact less would be ok since information is likely to change Jun 12 10:07:19 florian: most print companies starts from 2500 ;D Jun 12 10:07:29 but I saw some with 1000 for start Jun 12 10:07:31 heh Jun 12 10:09:25 mickeyl: good morning Jun 12 10:09:25 mickeyl: hi Jun 12 10:09:39 mickeyl: sources for OE poster/flyer - you have them? Jun 12 10:09:43 i can checkout the sources to the flyer and poster, but you need MacOSX to work on them Jun 12 10:09:46 morning folks Jun 12 10:09:52 s/checkout/checkin/ Jun 12 10:10:10 hi mickeyl Jun 12 10:10:15 fsck Jun 12 10:10:15 yo florian Jun 12 10:10:23 mickeyl: so we need new ones Jun 12 10:10:27 on the other hand, you can tell me which changes you want Jun 12 10:10:35 and if it's not too much, i can do them Jun 12 10:10:42 [time-wise] Jun 12 10:11:10 mickeyl: we need update - at least new logo Jun 12 10:11:34 as I can print flyers for LinuxTag but need materials for monday Jun 12 11:05:13 03Koen Kooi  07org.openembedded.dev * r649b44a943 10openembedded.git/recipes/dsplink/ (30 files in 3 dirs): dsplink: move dir to 'dvsdk' to make merging the TI OE trees easier Jun 12 11:29:32 hrw, what about the poster likewise did for FOSDEM? Jun 12 11:30:28 I do not know where it is Jun 12 11:34:37 Do you want me to ping likewise when he appears? Jun 12 11:35:26 would be great Jun 12 11:35:34 will do Jun 12 11:37:50 thx Jun 12 11:39:03 http://www.flickr.com/photos/32615155@N00/3260848665/in/set-72157613444405963/ Jun 12 12:23:35 printing 1000 or 2,500 often hardly makes a difference in price, not much difference at least Jun 12 12:44:35 hrw: ping Jun 12 12:45:00 hrw: it seems Pavel Macheck would need a serial cable for Z Jun 12 12:50:37 I do not have spares Jun 12 12:50:45 ant_work: Laibsch is our serial cable man Jun 12 12:50:55 oki thx Jun 12 12:55:39 Anyone knows the recipe to create a x11 demo image? I used to bitbake angstrom-x11-image-demo Jun 12 12:57:45 cd recipes; find | grep x11 | grep bb Jun 12 12:59:18 | nbench0.c:(.text+0x154): undefined reference to `__adddf3' Jun 12 12:59:20 ;( Jun 12 12:59:50 missing a fp lib ? Jun 12 13:00:20 x86 non-fpu Jun 12 13:01:01 yeah, then fp instructions are simulated in sw and you are not linking in the lib with those functions Jun 12 13:01:29 part of soft-fp Jun 12 13:01:47 google says: http://www.koders.com/c/fidF2E93C31E288F79D62420221FE06F36FE73C4CCA.aspx Jun 12 13:03:24 effem: ok, fair enough. Where should SPLASH be defined? bitbake x11-image now bails as ${SPLASH} isn't defined. Do I add this to the local.conf as psplash-angstrom ? Jun 12 13:22:47 hrw: about distcc... http://www.oesf.org/forum/index.php?showtopic=26676&pid=182527&st=0&#entry182527 Jun 12 13:22:49 :) Jun 12 13:24:29 ant_work: Koen made beaglecube for distcc Jun 12 13:24:44 yea, months ago Jun 12 13:24:51 but without Z :) Jun 12 13:29:05 (Z speedup seems to be of 3secs over 4mins 52'') lol Jun 12 13:52:13 Hi all, I have a problem with compiling a code using the OE's toolchain, the error is : http://pastebin.com/m12c38df7 Jun 12 13:54:48 but I didn't assign the /lib directory as library directory in eclipse Jun 12 14:08:13 is there anyone to help me? Jun 12 14:10:15 "Unable to start filter "pdftoraster" - Success." Jun 12 14:10:28 why this is success is beyond me ;( Jun 12 14:18:44 03Stanislav Brabec  07org.openembedded.dev * r50bb21e804 10openembedded.git/ (3 files in 3 dirs): gammu: Update to version 1.24.0. Jun 12 14:22:13 03ghost  07org.openembedded.dreambox * re2ac418e6a 10openembedded.git/packages/gstreamer/gstreamer_0.10.23.1.bb: gstreamer_0.10.23.1.bb: use patch Jun 12 14:27:20 03Phil Blundell  07org.openembedded.dev * r99aba1b132 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@new.openembedded.org:openembedded into org.openembedded.dev Jun 12 14:27:21 03Phil Blundell  07org.openembedded.dev * r5510cb2417 10openembedded.git/recipes/binutils/ (binutils-2.19.51/ld-stub-crash.patch binutils_2.19.51.bb): binutils: fix crash when linking uclibc in thumb mode Jun 12 14:33:33 does anyone know what the relationship between DISTRO and HCI_SECMGR is? Jun 12 14:34:19 none, as far as I know Jun 12 14:34:35 DISTRO is an oe variable. HCI_SECMGR is just a constant defined in a header file, isn't it? Jun 12 14:36:00 that's what i thought, too. when i change DISTRO to our soon-to-be-custom-fork name, bluez breaks its build with an HCI_SECMGR error. Jun 12 14:36:38 sounds like your distro must be selecting a set of bluez packages that are incompatible with each other or have bits missing. Jun 12 14:36:52 I think you'll just have to debug this build failure from first principles and then figure out what distro bits to change. Jun 12 14:37:53 yup :-) Jun 12 14:38:48 i made a local overlay, copied all the angstrom distro files, and started with small modifications first. it was when i changed the DISTRO variable that everything started breaking. Jun 12 14:40:29 kg4ysn, why mess with the distro files anyway? Jun 12 14:41:23 kg4ysn: it looks like there are some specific angstrom overrides in the bluez.bb files. I guess that's probably what's causing your upset. Jun 12 14:41:26 crofton: the company i work for is creating their own embedded system ... not a phone ... and needs a lot of stuff stripped out of the angstrom distro. Jun 12 14:41:41 I'm curious why? Jun 12 14:41:46 pb: thanks for the help! I'll take another look at the bluez recipes. Jun 12 14:41:52 don't you really need to make your own image Jun 12 14:42:33 in general, messing with the distro files is not needed Jun 12 14:43:03 crofton: it's possible that i'm going about this in an inefficient manner. :-) we're not making a handheld, but rather an embedded power management system. things like opie aren't required and we don't have space for it. Jun 12 14:43:30 that has nothing to do with Angstrom Jun 12 14:43:33 kg4ysn: you might want to look at the "micro" distro, this is intended for deeply embedded systems. Jun 12 14:43:44 I really think you are going about things the wrong way Jun 12 14:43:50 pb_, rubbish Jun 12 14:44:11 build micro-image with angstrom distro and build up to what you need Jun 12 14:44:58 Crofton|work: this will pull in some things like avahi you don't want Jun 12 14:45:07 Crofton|work: eh? what do you think "micro" is for if not embedded? Jun 12 14:45:11 using micro-image? Jun 12 14:45:29 Crofton|work: yes.. it pull this in every image Jun 12 14:45:36 hmm Jun 12 14:45:45 actually, avahi is desired for ipv4 autoconfig. Jun 12 14:45:45 florian, at least you have good responses Jun 12 14:45:55 heh Jun 12 14:46:32 kg4ysn, I think you are having the standard confusion between images and distros Jun 12 14:46:52 one distro can have many images, each targeting diverse hardware needs Jun 12 14:47:16 I nornally build console-image, which has lots of stuff, but no local screen Jun 12 14:47:41 but, you do not need a new distro to build without opie Jun 12 14:47:48 hopefully, this makes sense Jun 12 14:49:44 crofton: do you recommend any references for this? I've already been through the OE user manual, but apparently i missed something. Jun 12 14:50:05 heh Jun 12 14:50:07 sadly no Jun 12 14:50:30 this is a frequent source of confusion and I know of no good reference to clear it up :( Jun 12 14:50:47 lovely :-) seems like a deserving FAQ entry on someone's wiki. Jun 12 14:50:50 heh Jun 12 14:51:12 have you tried console-image on your hw? Jun 12 14:51:27 kind of hard to see what you need in my crystal ball :) Jun 12 14:51:53 LOL, yes and no. we have an eval board while our actual hw is being built up. Jun 12 14:51:58 i've been using that for the moment. Jun 12 14:52:02 heh Jun 12 14:52:07 i got the AT91SAM9263 image running on it. Jun 12 14:52:18 cool Jun 12 14:52:28 but there is too much in that? Jun 12 14:52:42 I am not familiar with that image Jun 12 14:52:51 well, there were certain things that aren't needed, and there are other things that we do need. Jun 12 14:52:58 hmm Jun 12 14:53:00 what image? Jun 12 14:53:14 or does this come from an overaly of some form? Jun 12 14:53:29 let me check Jun 12 14:53:59 Would anybody know how to produce a .fb with a ppm for splashscreen? Jun 12 14:54:15 crofton: angstrom-x11-image-demo Jun 12 14:54:25 ah Jun 12 14:54:27 ok Jun 12 14:54:36 do you need the display stuff? Jun 12 14:54:52 yes, we have an 18-bit LCD Jun 12 14:55:13 the best thing to do is look at the bb file that creates the image Jun 12 14:55:42 I still can't find a bb file with that name :) Jun 12 14:56:01 ainulindale: is this for psplash? Jun 12 14:56:14 yep supposedly Jun 12 14:57:15 ainulindale: might want to look at bmp2ppm ... search on google. no guarantees. Jun 12 14:57:27 Well I already have a ppm, I need *.fb files Jun 12 14:57:48 ainulindale: ppm2fb? :-) Jun 12 14:58:16 Well looking for that yes :-) Jun 12 14:58:45 crofton: so i need to create a recipe called "my-cool-new-distro-image.bb", put it in my local overlay, and add all my INHERIT flags as desired? Jun 12 14:58:53 morning Jun 12 14:58:58 yo kergoth Jun 12 14:59:03 hi kergoth Jun 12 14:59:11 kg4ysn, yes Jun 12 14:59:17 basically :) Jun 12 14:59:35 kg4ysn: "crofton: do you recommend any references for this?" If I am not mistaken, you need help on custom image creation, maybe this might help : http://bec-systems.com/web/content/view/79/9/ Jun 12 14:59:37 ainulindale, http://wiki.openmoko.org/wiki/Configuring_the_boot_splash_screens Jun 12 14:59:49 crofton, cool thanks! Jun 12 15:00:01 see rkirti's link also Jun 12 15:00:08 cbrake's articles are very good Jun 12 15:00:11 rkirti, IMAGE_INSTALL is what i'm looking for? Jun 12 15:00:19 heh Jun 12 15:00:38 look in the images directory and find something close to what you want Jun 12 15:00:50 then modify it, we should be able to answer questions Jun 12 15:01:09 ya'll rock :-) thanks so much for the help! Jun 12 15:01:19 kg4ysn: just found out how :-) Jun 12 15:01:21 IMAGE_INSTALL is more for hacking Jun 12 15:03:15 kg4ysn, unless you have extremely specialized requirements, you should just need to edit the image file Jun 12 15:04:03 hrw: was hrw-pcmcia-ids2-r2.patch your old patch to add wifi-id's ? Jun 12 15:04:05 crofton, beyond adding/removing some files, it's not a major change. i'll start parsing through the image recipes and find one that most closely matches. Jun 12 15:07:14 kg4ysn, let us know if you have any questions Jun 12 15:07:20 03Julien 'Ainulindale' Cassignol  07shr/import * r428fbb72c6 10openembedded.git/recipes/shr/shr-splash-theme-dociswatchingyou_git.bb: Jun 12 15:07:20 shr-splash-theme-dociswatchingyou: New recipe. Jun 12 15:07:20 Signed-off-by: Julien 'Ainulindale' Cassignol Jun 12 15:07:20 03Julien 'Ainulindale' Cassignol  07shr/import * rceb92864cb 10openembedded.git/: Merge branch 'shr/import' of ssh://git@git.openembedded.net/openembedded into shr/import Jun 12 15:07:22 03Julien 'Ainulindale' Cassignol  07shr/import * r7f0177a8ad 10openembedded.git/recipes/tasks/task-shr-feed.bb: Jun 12 15:07:25 task-shr-feed: Added shr-splash-theme-dociswatchingyou to the feed. Jun 12 15:07:27 Signed-off-by: Julien 'Ainulindale' Cassignol Jun 12 15:07:27 the task stuff and machine features can be confusing at first Jun 12 15:07:37 Crofton|work: definitely will :-) this irc channel has proved to be one of the best OE resources i've come across. Jun 12 15:07:41 you can create images that do not use them Jun 12 15:08:18 gl Jun 12 15:08:34 Crofton|work: thanks, i'll need it (and a lot of caffeine!) Jun 12 15:09:29 ant_work: was Jun 12 15:09:40 flameman: ^^^ Jun 12 15:09:41 ant_work: such patches were merged each time Jun 12 15:09:44 hrw: thx Jun 12 15:10:10 hrw: flameman is playing with 2.6.16 Jun 12 15:10:18 o fsck Jun 12 15:10:22 :/ Jun 12 15:12:06 ~hail uvesafb Jun 12 15:12:10 ~curse intel badly Jun 12 15:12:46 I'm attempting to build an initramfs image for my device, and all seems to go well, but when swith_root is callled, I get "not rootfs" and the kernel stops Jun 12 15:13:07 I've built the image as ext2.gz..do I need to do it with a different image? Jun 12 15:13:25 I'm using the initramfs-bootmenu-image Jun 12 15:16:03 hey guys Jun 12 15:16:28 does anyone know if there is a meta package that only installs a web browser? Jun 12 15:16:37 no other goodness Jun 12 15:18:18 03Julien 'Ainulindale' Cassignol  07shr/import * r3f197d55eb 10openembedded.git/conf/distro/include/shr-autorev.inc: Jun 12 15:18:18 shr-autorev: Set rev for shr-splash-theme-dociswatchingyou. Jun 12 15:18:18 Signed-off-by: Julien 'Ainulindale' Cassignol Jun 12 15:21:04 Crofton: can you switch DNS for bugs.openembedded.org? Jun 12 15:21:13 to where? Jun 12 15:21:22 to melo Jun 12 15:21:27 I'll let you know when Jun 12 15:21:35 Could be very soon Jun 12 15:21:51 let me know Jun 12 15:22:03 ok Jun 12 15:22:04 cool Jun 12 15:40:56 When I compiling an app with the meta-toolchain, I get error like this arm-angstrom-linux-gnueabi/bin/ld: cannot find /lib/libc.so.6 , does anyone know the solution? Jun 12 15:44:31 sx: more details Jun 12 15:45:31 hrw, http://pastebin.com/m2dd82df3 this is my console make log Jun 12 15:45:36 which package, show error log etc Jun 12 15:46:13 sx: "arm-angstrom-linux-gnueabi-gcc -L/tftpboot/192.168.10.6/usr/lib" - where from that -L is?? Jun 12 15:47:24 hrw, tftpboot/192.168.10.6/ is my NFS filesystem, -L is library search path Jun 12 15:47:41 hrw, my .so files are there for directfb and gstreamer.. Jun 12 15:47:45 sx: how that goes into CFLAGS? Jun 12 15:48:09 hrw, in the CGFLAGS ? What should I do Jun 12 15:48:12 re mickeyl Jun 12 15:48:18 hi hrw Jun 12 15:48:34 sx: OE did not adds random dirs to CFLAGS - you had to edit local.conf or recipe for that Jun 12 15:49:16 hrw, thanks, I will try it Jun 12 15:52:24 ls Jun 12 16:03:35 03Koen Kooi  07org.openembedded.dev * r6cf6511c75 10openembedded.git/classes/image.bbclass: Jun 12 16:03:35 image bbclass: add install_linguas method Jun 12 16:03:35 * this method will install all linguas in $IMAGE_LINGUAS for all packages installed into the rootfs Jun 12 16:03:35 * currently opkg only, guarded with a check Jun 12 16:03:36 03Koen Kooi  07org.openembedded.dev * r81b636b1af 10openembedded.git/conf/distro/include/angstrom.inc: Angstrom: set BUILDNAME Jun 12 16:03:41 03Koen Kooi  07org.openembedded.dev * re44d97c025 10openembedded.git/recipes/images/beagleboard-demo-image.bb: beagleboard demo image: use install_lunguas Jun 12 16:03:46 03Koen Kooi  07org.openembedded.dev * raa96a85c64 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.org:openembedded into org.openembedded.dev Jun 12 16:08:15 I ♥ uvesafb Jun 12 16:08:40 wow, someone actually setting BUILDNAME. Jun 12 16:08:46 finally 1024x768 framebuffer working on intel hardware Jun 12 16:11:44 03Michael 'Mickey' Lauer  07fso/milestone5.5 * ra7a9a7d414 10openembedded.git/recipes/freesmartphone/ (cornucopia.inc libfsoframework_git.bb): libfsoframework: add dependency on misc-vapi Jun 12 16:13:55 if hw.vendor == 'intel' and hw.age > 2y and hw.name != 'i945' then printk 'user: f... you!\nloving intel' Jun 12 16:21:06 heh, crazy angstrom, they should just fix the translation support in opkg. Jun 12 16:26:36 pb__: Koen has better skills in shell then in opkg hacking Jun 12 16:26:51 does someone uses irda in laptop? Jun 12 16:29:31 Crofton: can you please make the switch for bugs.oe.org and bugzilla.oe.org to the melo IP? Jun 12 16:30:00 mickeyl: can you please make set bugs.oe.org and bugzilla.oe.org as CNAME to their .org counterparts? Jun 12 16:30:21 mickeyl: can you please set bugs.oe.net and bugzilla.oe.net as CNAME to their .org counterparts? Jun 12 16:30:30 ok Jun 12 16:34:20 Laibsch, what is the ip of melo? Jun 12 16:34:48 140.211.169.165 Jun 12 16:35:03 hmm, the website is still on amethyst, only got moved Jun 12 16:36:09 done Jun 12 16:36:20 it will take a while to propagate Jun 12 16:39:30 the new virtual host is already set up Jun 12 16:39:44 so melo should take over once the DNS stuff has propagated Jun 12 16:39:56 I can set up amethyst to redirect as necessary, too Jun 12 16:40:08 I'm having some more fundamental issues at the moment, though Jun 12 16:40:54 bye Jun 12 16:41:13 gn Jun 12 17:23:14 CoreDump|cf-18: we moved the bugzilla server, what needs to be done for cdbot2? Jun 12 17:23:37 moved again? Jun 12 17:24:04 !oebug 4711 Jun 12 17:24:19 * * Bug 4711, Status: HTTP transfer failed :( Jun 12 17:39:19 CoreDump|cf-18: why again? Jun 12 17:39:38 maybe you think about some other service Jun 12 17:39:48 We are migrating them one by one to melo from amethysst Jun 12 17:40:16 the reason !oebug does not work currently is probably because the .org DNS update has not propagated yet Jun 12 17:40:27 probably thinking of git :) Jun 12 17:40:41 That kind of information pull should likely continue to work without any changes Jun 12 17:40:57 I'm wondering about the notification when a bug is opened or closed Jun 12 17:41:25 CoreDump|cf-18: does cdbot2 get it's own mail for that or does the bot grab the info off the mailing list? Jun 12 17:41:53 03Phil Blundell  07org.openembedded.dev * rf29aea68c3 10openembedded.git/recipes/base-files/base-files_3.0.14.bb: base-files: handle DISTRO=micro-uclibc correctly Jun 12 17:42:55 03Phil Blundell  07org.openembedded.dev * red20360f42 10openembedded.git/recipes/uclibc/uclibc-initial_0.9.30.1.bb: uclibc-initial: respect ${CC} Jun 12 17:44:08 * kergoth makes a little python function to print an exception, including a traceback only if bitbake debugging is enabled Jun 12 17:45:10 03Phil Blundell  07org.openembedded.dev * r1e02d5cf1d 10openembedded.git/conf/distro/micro-uclibc.conf: micro-uclibc: make sure "micro" is in OVERRIDES; use eabi for armeb Jun 12 17:48:30 Crofton: does bugzilla.openembedded.org have a DNS entry? Jun 12 17:51:03 it should now Jun 12 17:52:44 cool, thanks Jun 12 17:53:10 Laibsch: cdbot is subscribed to the ML Jun 12 17:53:21 OK Jun 12 17:53:38 It looks like right now the mailing list is not yet receiving the mails Jun 12 17:53:49 mickey|DepecheMo: Can you please take a look? Jun 12 17:54:51 If I am not mistaken, the mailing list was locked down pretty hard and it may not accept the mails from the new bugzilla Jun 12 18:11:40 03Philip Balister  07org.openembedded.dev * r82efacb2b2 10openembedded.git/recipes/gnuradio/gnuradio_3.2.0.bb: gnuradio-3.2 : Use -program-prefix to remove prefix from programs. Jun 12 18:16:15 wow, those nameresolve.com servers are really slow to update. They themselves still respond with the old IP Jun 12 18:35:07 03Phil Blundell  07org.openembedded.dev * r1f72f3ed3a 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@new.openembedded.org:openembedded into org.openembedded.dev Jun 12 18:35:08 03Phil Blundell  07org.openembedded.dev * r6c47b3e373 10openembedded.git/recipes/xcb/ (libxcb.inc libxcb/configure.patch): libxcb: look for xproto.xml in staging, not in the host environment Jun 12 18:55:56 hm, this is weird, x11-trim is crashing with sigfpe in makekeys Jun 12 18:56:17 I'm sure I've built that code hundreds of times in the past and never had a problem with it. Jun 12 18:56:25 Laibsch, hi Jun 12 18:57:53 ah, hm, no, libx11 and diet-x11 both disable makekeys, it's just -trim that runs it. I wonder why that is. Jun 12 19:06:34 kergoth: ${BUILD_OPTIMIZATION} is just for things that are built natively, right? Jun 12 19:37:26 pb__: i *think* so.. it's been too long for me to remember :) Jun 12 19:37:36 hi kergoth Jun 12 19:39:22 kergoth: yeah, that's what I thought too. iirc, BUILD_OPTIMIZATION is for the build host, and SELECTED_OPTIMIZATION (which goes into TARGET_CFLAGS) is for the target. seems weird, then, that there are so many places where BUILD_OPTIMIZATION is set to crazy things. Jun 12 19:39:28 hi pb Jun 12 19:39:49 that is weird indeed if that's the case, guess it needs to be thoroughly reviewed Jun 12 19:40:02 fr'instance, angstrom-uclibc.inc sets BUILD_OPTIMIZATION to "-Os". seems hard to imagine that even uclibc users care that much about the size of their host binaries. Jun 12 19:41:00 and tune-ep9312.inc sets it to "-O" (default is -O2). might possibly be some plan to make the cirrus target seem faster by hobbling the optimisation of host binaries, but that's a bit of a stretch. Jun 12 19:41:13 pb hm? Jun 12 19:41:23 hi woglinde Jun 12 19:41:27 whats the problem with host binaries? Jun 12 19:41:52 well, the actual problem is that overriding BUILD_OPTIMIZATION from local.conf doesn't do what you would expect Jun 12 19:42:08 hm okay Jun 12 19:42:18 since, although bitbake.conf sets it (correctly) with ?=, these various other include files then clobber it with their own wacky values Jun 12 19:42:19 but why we set BUILD_OPT in local.conf? Jun 12 19:42:56 the particular one that's screwing me up in this instance is the (apparently inaptly-named) sane-toolchain-glibc.inc, which sets it to -O2. Jun 12 19:43:26 but, on inspection, this is a widespread problem: just try grepping for BUILD_OPT in conf/ Jun 12 19:45:43 woglinde: same reason you'd set anything in local.conf, I guess. in this case I wanted to build with more debugging and less optimisation. Jun 12 19:46:14 actually, I'd have thought that "-O2 -g" would be a saner default, seems no point building undebuggable binaries if you don't have to. Jun 12 19:47:34 yes, -g would be good Jun 12 19:48:38 hm Jun 12 19:50:55 I'll send a note to the list about that. Jun 12 19:53:06 nite rkirti Jun 12 20:00:28 re Jun 12 20:00:56 florian: wb Jun 12 20:01:13 he florian Jun 12 20:01:32 florian: do you happen to know why libx11-trim runs makekeys, when diet-x11 and regular x11 don't? Jun 12 20:01:52 still pursuing my goal of a buildable gpe-image :-} Jun 12 20:03:37 ah, dinner is served Jun 12 20:03:38 bbiab Jun 12 20:04:04 pb__: no, sorry Jun 12 20:05:29 till later pb Jun 12 20:06:14 pb__: an improved gpe-image is some kind of a dream. we should check how we can make use of the modern matchbox components Jun 12 20:06:30 pb__: enjoy Jun 12 20:06:55 pb__: i built gpe-image some time ago... Jun 12 20:23:26 florian, better dafe than sorry :) Jun 12 20:23:50 :-) Jun 12 20:28:37 hms Jun 12 20:29:09 I cant consider if I should reserve me a ticket for linuxtag Jun 12 20:29:22 florian are the tickets bounded to persons? Jun 12 20:30:22 woglinde: usually yes... but i don't think they check this. Jun 12 20:30:54 hm I will ask tomorrow if we have enought tickets at tarent Jun 12 20:30:57 but I guess yes Jun 12 20:31:00 woglinde: usually we get enough anyway... currently 12 are left Jun 12 20:31:07 oh okay Jun 12 20:33:36 woglinde, you work for tarent? the company making the free java for embedded devices? Jun 12 20:33:47 oops hi Jun 12 20:33:56 gnutoo yes Jun 12 20:34:02 wow Jun 12 20:34:06 *g* Jun 12 20:34:56 Tarrent is awesome Jun 12 20:35:00 what does tarrent apart of porting the normal java to arm for the bug device? Jun 12 20:35:05 tarent Jun 12 20:35:08 only one r Jun 12 20:35:10 oops Jun 12 20:35:11 s/tarrent/tarent do Jun 12 20:35:13 sorry Jun 12 20:35:16 hm Jun 12 20:35:28 well, they bought us all dinner once, that goes a long way :) Jun 12 20:35:37 we have seval projects Jun 12 20:35:43 most have to do with java Jun 12 20:36:20 crofton yes Jun 12 20:36:25 I ask robert for that Jun 12 20:36:28 asked Jun 12 20:36:46 and 1 1/2 year later I was at taren too Jun 12 20:36:47 *g* Jun 12 20:38:01 I'm curious...what are theses projects? (if you are allowed to talk about them) Jun 12 20:38:13 hm Jun 12 20:38:17 I cant say much Jun 12 20:38:20 ok Jun 12 20:38:28 some have to do with embedded stuff Jun 12 20:38:35 some have to do with nx Jun 12 20:38:46 nx? Jun 12 20:38:56 nx from no-machine Jun 12 20:39:00 ah ok Jun 12 20:39:17 my first project was to port the qtnx client to the nokias nxx Jun 12 20:39:22 ok Jun 12 20:39:23 oh, is nomachine still going? I haven't heard anything about nx for ages. Jun 12 20:39:30 pb hehe Jun 12 20:39:35 the run really nice Jun 12 20:39:56 are there any open clients for nx nowadays? Jun 12 20:40:12 features for the next versions are browser plugins and iphone support Jun 12 20:40:16 pb 2 Jun 12 20:40:20 for now Jun 12 20:40:35 theire is x2go Jun 12 20:40:43 but they have their own stuff Jun 12 20:40:53 wit server and client written in qt Jun 12 20:40:58 freenx is stalled Jun 12 20:41:03 as server Jun 12 20:41:07 qtnx as client too Jun 12 20:41:10 but qtnx works Jun 12 20:41:31 the nx maintainer from ubuntu is rewriting freenx in python Jun 12 20:41:34 and a client too Jun 12 20:41:46 very good Jun 12 20:42:06 maybe I should look at it again, I could use a better compression scheme for remote x11 when I want to access my office desktop from home Jun 12 20:42:14 but the guis only settings the params for nxagent Jun 12 20:42:22 running firefox over regular ssh forwarding is fairly painful Jun 12 20:42:24 aeh nxproxy Jun 12 20:42:30 yes Jun 12 20:42:31 jepp Jun 12 20:42:31 it is Jun 12 20:42:34 nx rocks Jun 12 20:42:37 in this area Jun 12 20:42:40 I should try it... Jun 12 20:42:45 (nx) Jun 12 20:42:53 pb__: what kind of trouble do you face with gpe-image? Jun 12 20:42:54 in lan even flash videos works Jun 12 20:43:41 pb the only problem is that the nxserver X called nxagent is based on old X server 6.9 Jun 12 20:43:59 I hope they will update it at some point Jun 12 20:44:00 florian: well, previously I had the unfetchable libidn thing. it turned out that it was actually webkit-gtk dragging that in. Jun 12 20:44:14 I wrote a new recipe for a current version of libidn (1.15 or some such) and that seemed to work Jun 12 20:44:41 now, I have a problem where bitbake seems to want to build this weird "libx11-trim" thing for no reason that I can see, which fails with a divide-by-zero trying to run makekeys. Jun 12 20:45:20 I'm just trying to work out why it's picking libx11-trim over regular libx11 (which builds fine) or diet-x11 (likewise) Jun 12 20:45:29 ah right, we need to introduce a check that prints out a big warning if some source is present but not fetchable any more Jun 12 20:45:56 yeah, that'd be handy Jun 12 20:48:35 hm, seems that it was just using x11-trim to satisfy virtual/libx11 Jun 12 20:48:47 I set PREFERRED_PROVIDER_virtual/libx11 and that has apparently cleared it up Jun 12 20:49:05 phew Jun 12 20:49:16 which is weird, because I would have expected it to prefer the already-staged libx11 automatically. I guess bitbake must have lost that ability. Jun 12 20:49:19 * florian needs to decide for a kernel or userland project now Jun 12 20:49:31 kernel Jun 12 20:49:48 userland! Jun 12 20:49:59 *g* Jun 12 20:50:33 you are a great help ;) Jun 12 20:51:00 any time Jun 12 20:51:33 I could spend ages in fixing linux support for hardware companies claim to have linux support. Jun 12 20:51:44 if only ibot had the ability to toss a coin it'd help with breaking these kinds of ties Jun 12 20:52:22 heh, that would be a good job for the bot :) Jun 12 20:52:23 florian: yeah, this is a common problem Jun 12 20:53:04 writing a complete linux bsp is, sadly, just too big a job for a lot of vendors. Jun 12 20:53:05 The company that ported linux to the Topas910 is famous for Amiga stuff... Jun 12 20:53:26 I know we still haven't really sorted out the alsa sound driver situation on our own hardware, so I speak from bitter experience of this :-} Jun 12 20:55:32 pb__: I face the situation where one person with limited experience seems to be reponsible for porting and BSP including documentation... for three boards at least. Jun 12 20:56:39 doh Jun 12 20:56:40 ouch Jun 12 20:58:47 I do wish the kernel people would try to stop changing things at random all the time. As a hardware vendor it does make life much more annoying if you have to constantly run on this treadmill to stay in the same place. Jun 12 20:59:36 one of our boards is still on kernel 2.6.21, for example, because upgrading to a newer one breaks all our sound drivers and we haven't had time to fix them up. Jun 12 21:04:10 hm, why is sane-toolchain-glibc doing setting TARGET_OS? Jun 12 21:04:23 the more I look at this file the less it seems to deserve the "sane" epithet. Jun 12 21:05:35 a question for the gnome experts here Jun 12 21:05:56 I noticed a number of packages have DEPENDS on gconf-dbus Jun 12 21:06:29 do they really depend on the dbus version, or should they have a more general DEPENDS on gconf? Jun 12 21:11:11 dbus is evil Jun 12 21:13:33 this logicpd omap3 dev kit board is annoying Jun 12 21:13:38 03Phil Blundell  07org.openembedded.dev * r543f79f029 10openembedded.git/conf/ (4 files in 2 dirs): Jun 12 21:13:38 bitbake.conf: add -g to BUILD_OPTIMIZATION Jun 12 21:13:38 sane-toolchain-*.inc: remove assignments to BUILD_OPTIMIZATION Jun 12 21:13:38 Acked-by: Richard Purdie Jun 12 21:14:26 sakoman__: hmm... wich ones? Jun 12 21:14:38 prpplague: it can be pretty useful Jun 12 21:14:52 Crofton|work: which one? Jun 12 21:15:47 florian: i'm just not a fan of the complexity, dependencies, and overall size Jun 12 21:17:17 florian: gnome-bluetooth for example Jun 12 21:17:37 the logicpd zoom development kit Jun 12 21:17:45 I need to probe the gpmc bus Jun 12 21:17:54 I am hoping this is an easy way to do that ,,,, Jun 12 21:17:58 you should use an overo ;-) Jun 12 21:18:08 hehe Jun 12 21:18:11 sakoman__: good one Jun 12 21:18:16 * Crofton|work kicks sakoman__ Jun 12 21:18:21 ouch! Jun 12 21:18:31 bring gpmc out to where I can easily probe Jun 12 21:18:46 board layout resources are in short supply .... Jun 12 21:19:03 get a tobi and probe the lan9221 Jun 12 21:19:28 I should make a breakout board Jun 12 21:19:39 heh Jun 12 21:19:45 I am going to drink beer Jun 12 21:19:52 bet I could sell, hmm, well at least one! Jun 12 21:19:58 Crofton|work: please take me with you! Jun 12 21:20:00 I had to post to a "bulletin board", the horror Jun 12 21:21:05 sakoman__: i'm going to start doing some small run utility boards in the fall, if there is something you are looking for let me know Jun 12 21:21:52 prpplague: beagle utility boards? Jun 12 21:22:28 sakoman__: probably some, but more generic stuff Jun 12 21:22:36 sakoman__: breakout boards and such Jun 12 21:23:06 sakoman__: stuff like what sparkfun does, but more specific to SoC/Module dev Jun 12 21:23:12 sakoman__: that's a little bit strange indeed... i thought we had multiple providers for dbus Jun 12 21:24:26 florian: gconf-dbus is one of the providers of gconf, so I think those packages probably want gconf as the DEPENDS Jun 12 21:24:42 yeah, that makes sense Jun 12 21:24:58 I can't imagine that there are many/any packages that genuinely need the dbus version rather than any other Jun 12 21:25:31 eah, and if you really don't want gconf-dbus it becomes an issue :-) Jun 12 21:26:00 i can imagine that some maemo stuff might need it but i don't think so Jun 12 21:26:21 in that case they can set the preferred provider Jun 12 21:26:34 I think angstrom sets it to gconf-dbus by default Jun 12 21:27:38 sakoman__: something specific you are looking for to work with the beagle? Jun 12 21:28:04 prpplague: no, actually I was thinking of an Overo breakout board Jun 12 21:28:13 sakoman__: ahh Jun 12 21:28:32 sakoman__: thats doable to Jun 12 21:28:48 sakoman__: i'll be getting geared up soon , i'll see what i can do Jun 12 21:29:56 what ever became of the Beagle add-on board? Jun 12 21:30:30 sakoman__: its in limbo, the owner of TinCanTools has been busy on other projects Jun 12 21:30:54 sakoman__: the board is done, the drivers need some testing, but its ready to go into production Jun 12 21:31:05 sakoman__: i think he's lost too much time on it Jun 12 21:31:32 sakoman__: if he'd gone into production back in october/november of 2008 it would be good Jun 12 21:32:01 sakoman__: i want to do some small simple boards that i can get to market quickly and easily Jun 12 21:32:20 sakoman__: i.e. less than 20 days Jun 12 21:32:50 * cbrake checks out the TinCanTools site ... Jun 12 21:33:00 prpplague: makes sense Jun 12 21:52:43 * prpplague heads home for the day Jun 12 21:53:34 * pb__ gives enchant_1.3.0.bb a funny look Jun 12 21:54:05 FILES_${PN} = "/usr/bin/* /usr/lib/enchant/*.so /usr/share/enchant /usr/lib/libenchant*.so.*" Jun 12 21:54:12 I wonder what that's in aid of **** ENDING LOGGING AT Fri Jun 12 22:02:30 2009 **** BEGIN LOGGING AT Fri Jun 12 22:03:25 2009 Jun 12 22:13:15 pb__: There is lots of insanity around sadly Jun 12 22:13:31 pb__: Even poky's recipes are making me cry atm :/ Jun 12 22:23:34 hi,is there an option in one of the binutils tools to know the compiler with which it was compiled? Jun 12 23:00:09 anyone recall which recipe sets up the initial users? Jun 12 23:13:47 Ok, I know I asked this before, but I really mean it now. :-) Is there a (nice) way to get Java up on Angstrom? Jun 13 02:01:44 03Rolf Leggewie  07org.openembedded.dev * r624edfdc6c 10openembedded.git/ (conf/checksums.ini recipes/gpephone/libidn_0.5.19.bb): libidn: old SRC_URI has become unavailable **** ENDING LOGGING AT Sat Jun 13 02:59:57 2009