**** BEGIN LOGGING AT Mon Aug 20 02:59:56 2007 Aug 20 06:40:36 good morning all Aug 20 07:12:50 morning Aug 20 07:18:15 hrw : hi ! Aug 20 07:29:17 steliosk: I plan to go to greece for my honeymoon. probably crete Aug 20 07:31:53 hrw : cool :) When is the wedding ? Aug 20 07:32:03 steliosk: it was 14.08.07 ;D Aug 20 07:33:00 hrw : My best wishes ! Aug 20 07:33:16 thx ;) Aug 20 07:33:22 hrw : When you plan to go in Crete ? Made reservations ? Aug 20 07:33:41 steliosk: not yet - second part of september Aug 20 07:34:02 steliosk: first have to check which hotel to choose etc Aug 20 07:40:25 steliosk: one of friends suggested also some city near atheens but I forgot it name Aug 20 07:41:28 hrw : if you like Santorini, i have a few contacts there Aug 20 07:42:25 steliosk: maybe next time - this time I will rather take one of organized trips instead of organizing something itself Aug 20 07:44:57 hrw : yeah i know the feeling :) Crete is a big island you need to spend *months* to see it all Aug 20 07:45:25 hrw : and because its south, even in September weather is very nice Aug 20 07:47:18 thats why go to greece Aug 20 07:47:27 in Poland september is usually cold Aug 20 07:48:30 03hrw 07org.oe.dev * r2940479b... 10/ (1 site/ix86-common): site/ix86-common: added glib_cv_sizeof_system_thread definition to get glib 1.2 built Aug 20 07:48:48 morning all Aug 20 07:49:48 hi richard Aug 20 07:50:54 good morning Aug 20 07:51:24 hrw : This is where we stayed last year -> http://www.annio.com/ Aug 20 07:52:44 RP : Hi ! Aug 20 08:02:32 steliosk: creete? Aug 20 08:03:25 hrw : no Santorini. I was in Crete 2 years ago, but don't have the link to hotel handy. I'll check it if you want Aug 20 08:05:54 steliosk: if it will be no problem.. Aug 20 08:06:01 RP: If the build fails for any reason i see a lot of *.lock files in the package dirs. Is that normal ? Aug 20 08:06:15 steliosk: svn up bitbake Aug 20 08:08:31 RP: here bitbake doesn't set the shell return value anymore, not sure if this is due to my switching to python 2.5 Aug 20 08:08:53 "bitbake foo || echo whoops" used to work, but now it doesn't anymore Aug 20 08:09:25 pH5: That'll be issues in trunk Aug 20 08:13:44 hrw : this is the hotel in Maleme i stayed 2 years ago ->http://www.thehotel.gr/castro-beach-hotel Aug 20 08:19:43 thx steliosk Aug 20 08:24:17 hrw : Let me know if you need anything else. I will be in Athens till around the 20th of September so if you stop here to see Athens, it would be a good opportunity to meet Aug 20 08:25:13 RP: how can this be fixed? there are different codepaths involved, it seems. NoProvider fires of an unknown (to knotty) event, ParseError doesn't even fire an event. both exit with 0 Aug 20 08:27:26 morning all Aug 20 08:30:26 steliosk: I will gibe you info if I will be plan to be there Aug 20 08:41:38 pH5: Ideally they should both generate a "commandFailed" response and then knotty can exit with error status Aug 20 08:47:59 Can someone check #2764 and see if the mentioned .bb files are attached, because i can't locate them and probably the poster had forgot to attach them Aug 20 09:10:07 hi all Aug 20 09:10:26 got an error with rxvt-unicode-7.9-r0 Aug 20 09:10:50 `/mnt/hd1/oe/ipaq/tmp/staging/x86_64-linux/bin/tic: symbol lookup error: /mnt/hd1/oe/ipaq/tmp/staging/x86_64-linux/bin/tic: undefined symbol: _nc_check_termtype` during install phase Aug 20 09:11:50 re Aug 20 09:11:57 hej koen Aug 20 09:11:58 hi koen Aug 20 09:12:03 hrw: a belated congrats Aug 20 09:22:37 thx koen Aug 20 09:23:12 good morning Aug 20 09:25:58 * * OE Bug 2807 has been created by gremlin(AT)gremlin.it Aug 20 09:26:00 * * rxvt-unicode fail compile terminfo Aug 20 09:26:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2807 Aug 20 09:53:33 hey mickeyl Aug 20 09:53:53 hi koen Aug 20 09:53:54 morning Aug 20 09:54:14 it's evening for my biological clock :) Aug 20 09:54:34 hi Aug 20 09:54:41 just reformatting etc my C3200 Aug 20 09:55:34 hi mickeyl, koen Aug 20 09:55:45 hey RP Aug 20 09:56:01 RP: I'll give trunk a spin later this week Aug 20 09:56:06 koen: ok, thanks Aug 20 09:56:30 does .dev require 1.8.8 already? Aug 20 09:56:54 koen: mickeyl was suggesting a delay of two weeks before switching. We really need to send something to the list about it Aug 20 09:57:36 koen, how was the festival Aug 20 09:57:50 Crofton|home: great weather, people and music Aug 20 09:58:06 hi people Aug 20 09:58:06 excellent, no mud wrestling then Aug 20 09:58:19 Crofton|home: none that I have seen :) Aug 20 09:58:27 I just went through some bugs in bugzilla. What happened to x11vnc? Aug 20 09:59:24 seems the bb files for it dissappeared. Has it been replaced? Aug 20 09:59:31 philippe_work: use bonsai Aug 20 09:59:32 http://bugs.openembedded.org/show_bug.cgi?id=1126 Aug 20 09:59:54 zecke: I am just interested in closing some bugs :-) Aug 20 10:00:00 http://www.openembedded.org/bonsai/query/file/x11vnc/?scm_id=3 Aug 20 10:01:05 http://www.openembedded.org/bonsai/view/rev/17821/ Aug 20 10:02:02 zecke: ok. So I get that libvncserver has an own client which overrides x11vnc... Aug 20 10:02:11 Which means I can close the bug. Aug 20 10:06:05 * * OE Bug 1126 has been RESOLVED (WONTFIX) by philippedeswert(AT)scarlet.be Aug 20 10:06:05 * *  updated bb file for x11vnc 0.8.1 Aug 20 10:06:05 * * http://bugs.openembedded.org/show_bug.cgi?id=1126 Aug 20 10:21:58 * * OE Bug 2808 has been created by philippedeswert(AT)scarlet.be Aug 20 10:22:01 * * zten bb file seems to have wrong md5sum Aug 20 10:22:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2808 Aug 20 10:24:26 lrg: any word from robin? Aug 20 10:30:22 koen: ne Aug 20 10:30:23 not yet Aug 20 10:30:27 koen: i'll email him again. I'll buy another board as well. Aug 20 10:30:35 bbl Aug 20 10:41:23 http://worsethanfailure.com/Articles/The-Cool-Cam.aspx is briliant! Aug 20 10:42:43 question I have about python Aug 20 10:43:13 is it worth to switch to python 2.5 with 1.8 bitbake or keep 2.4 still? Aug 20 10:44:43 Have we ever taled to anyone at gumstix about getting some boards? Aug 20 10:45:25 hrw: python is one of the things where every consecutive version gets faster Aug 20 10:45:30 albeit hardly noticable, but still Aug 20 10:45:33 hi lrg Aug 20 10:45:39 Crofton|home: no idea Aug 20 10:46:00 someone know how goes OE foundation thing? Aug 20 10:46:08 I fixed the build, but apparently it does not work well Aug 20 10:49:24 dialy wtf roxx Aug 20 11:23:41 mickeyl: what do you think about splitting task-openmoko - or should everybody build the native-sdk? Aug 20 11:23:50 pH5: agreed. native sdk should be a seperate task Aug 20 11:23:50 pH5: could you do that ? Aug 20 11:23:54 i also want to add something like task-openmoko-feed Aug 20 11:23:56 where we add all the packages from world which we want Aug 20 11:24:31 mickeyl: I'll do that. ok to split out both -native-sdk and -debug? Aug 20 11:24:33 pH5: right. thans Aug 20 11:24:34 thanks, even Aug 20 11:26:11 mickeyl: task-native-sdk exists already Aug 20 11:31:19 hrw: good point - task-openmoko-native-sdk be based on task-native-sdk? Aug 20 11:32:07 the additional things that it installs currently are cvs, ldd, libtool, pkgconfig and xoo from a quick look **** BEGIN LOGGING AT Mon Aug 20 11:33:59 2007 Aug 20 11:39:57 i'm trying openembedded, but i've problem with timezone Aug 20 11:42:03 is enough setting the TZ variable, or i've to do something else to set the timezone? Aug 20 11:49:51 * mickeyl starts adding a 'functions' file for our shell scripts Aug 20 11:49:58 enough with duplicating helpful stuff Aug 20 11:55:29 hi Aug 20 11:56:04 how can I made a package of a download file Aug 20 11:56:24 there is no need to unpack and compile it Aug 20 11:56:30 its only a jar file Aug 20 11:56:49 woglinde: Just use cpio to create the ipk? (iirc) Aug 20 11:58:17 NAbyss the problem is to get the jar-file under /work/bla/package/ Aug 20 11:58:45 or I dont know the sane way to do it Aug 20 11:59:10 woglinde: Look at do_package for .ipks.. see how it creates the .ipk - do it outside of bitbake Aug 20 12:00:57 hellp Aug 20 12:00:58 nabyss thats not an option Aug 20 12:01:00 *hello Aug 20 12:01:08 hi olegfink Aug 20 12:01:48 woglinde: Hmm, then I'd do a custom .bb, override do_staging with a simple cp Aug 20 12:02:22 solved that mount problem (in russia we say "you're your own Cruel Pinoccio for youself"), but now I can't deal with busybox init Aug 20 12:02:40 context: initrd, aim: start up the system Aug 20 12:03:07 03Laibsch 07org.oe.dev * r46647ef8... 10/ (1 packages/zten/zten_1.6.2.bb): Aug 20 12:03:07 zten: correct md5sum. Closes 2808. Aug 20 12:03:07 Thanks, Philippe for spotting it. Aug 20 12:03:12 03pH5 07org.oe.dev * r366ae3a9... 10/ (1 classes/task.bbclass): Aug 20 12:03:12 task.bbclass: add a task class that sets common task recipe Aug 20 12:03:12 variables and automates adding -dev and -dbg packages Aug 20 12:03:16 03pH5 07org.oe.dev * rbdb564d0... 10/ (4 files in 2 dirs): task-openmoko: split out -debug and -native-sdk tasks into own recipes Aug 20 12:03:20 busybox documentation for init says it's designed to be run only by root -- what does it mean for me? Aug 20 12:03:20 03pH5 07org.oe.dev * r80856f57... 10/ (4 files in 2 dirs): Aug 20 12:03:20 task-openmoko: inherit task.bbclass to enable generation of -dbg and Aug 20 12:03:20 -dev versions of all task packages Aug 20 12:03:26 03pH5 07org.oe.dev * r64019590... 10/ (1 packages/tasks/task-openmoko.bb): Aug 20 12:03:26 task-openmoko: remove -debug and -native-sdk task packages, they are Aug 20 12:03:26 split out into their own recipes Aug 20 12:03:45 Nabyss hm yes Aug 20 12:04:03 seems I need to put a wget in the package Aug 20 12:04:52 olegfink what's the real problem? Aug 20 12:05:12 in an initrd/initramfs you usally only have on user Aug 20 12:05:13 root Aug 20 12:05:34 woglinde: iirc, the default do_fetch should handle the wgetting for you Aug 20 12:06:02 woglinde: at the moment it's my wondering whose duty is to create /dev/initctl Aug 20 12:06:08 nabyss yes but what me angers is the copy from archive-dir Aug 20 12:06:25 woglinde: init complains about its missing Aug 20 12:06:27 olegfink yours? Aug 20 12:06:49 can i get oe kernel compiles to just show me the warnings ? Aug 20 12:06:51 woglinde: really? in initrd? Aug 20 12:07:25 woglinde: can I look at any OE-based userland _with_ initrd and see where does it create initctl and how? Aug 20 12:08:41 olegfink yes you can Aug 20 12:09:03 aeh you have to look at the oe-repos for creating initramfs Aug 20 12:09:25 woglinde: can you point me to any? Aug 20 12:09:59 woglinde: manual page for sysvinit suggests me looking at its sources to read about initctl. Aug 20 12:10:21 http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/initrdscripts Aug 20 12:10:27 does anybody know how to set the timezone in OE? Aug 20 12:10:50 diego in oe? Aug 20 12:11:09 diego oe is only the repository Aug 20 12:11:09 using open embedded Aug 20 12:11:10 diego : install tzdata and friends Aug 20 12:11:16 for build the packages Aug 20 12:11:35 diego : and then set the TZ variable Aug 20 12:11:41 hi steliosk Aug 20 12:11:46 i've installed timezones-* Aug 20 12:11:59 woglinde : hi Aug 20 12:12:37 export TZ=CEST Aug 20 12:12:51 an date output utc time Aug 20 12:13:22 diego : yes, also check the initscripts (can't remember which) you can set TZ there permenent Aug 20 12:14:38 woglinde: sorry, can't find anything about initctl here. Aug 20 12:14:56 olegfink hm but for creating /dev Aug 20 12:15:36 woglinde: /dev _devices_, while initctl is a named pipe Aug 20 12:15:45 (atleast on my desktop) Aug 20 12:20:45 oleg hm okay than it is in classes Aug 20 12:20:48 let me see Aug 20 12:22:57 seems that at least sysvinit does this automatically Aug 20 12:23:23 hm yes but in initrd you dont have sysvinit Aug 20 12:23:31 sure Aug 20 12:23:47 (by the way I'm starting not initrd, but userland init) Aug 20 12:24:07 olegfink aehm why than you ask about initrd? Aug 20 12:24:23 woglinde: because I'm doing it in initrd Aug 20 12:24:43 Thought that initrd way of running init is somehow different Aug 20 12:27:12 woglinde: okay, my more general question is how to run busybox init from initrd Aug 20 12:28:18 hi Aug 20 12:29:25 ~pastebin Aug 20 12:29:29 olegfink normaly the init is only a shellscript Aug 20 12:30:21 i've a problem with ipkg. Installing tzdata he says that conflict with timezone pkg, but when i remove it, it says that timezone is not installe Aug 20 12:31:26 woglinde: good god! Aug 20 12:31:45 woglinde: in my OE userland I actually have _sysv_init Aug 20 12:31:55 woglinde: do_install() { install -d ${D}/somewhere install -m 0644 ${WORKDIR}/file.jar ${D}/somewhere} FILES_${PN} += "/somewhere" Aug 20 12:34:10 ~lart python-evas for not building Aug 20 12:35:35 hrw okay Aug 20 12:35:50 That was what I am think of Aug 20 12:35:57 but wanted to be sure Aug 20 12:37:06 well Aug 20 12:37:28 shouldn't I use switch_root instead of pivot_root? Aug 20 12:39:02 olegfink depends of if you use initrd or initramfs Aug 20 12:39:22 switch_root is more fore initramfs and nfsboot Aug 20 12:39:29 olegfink: busybox init run all from /etc/init.d/ in alphabetical order Aug 20 12:42:12 woglinde: I use initrd, that means I _shouldn't_ use switch_root? Aug 20 12:42:52 hrw: figured out my userland uses sysvinit (ask OE about jlime-image if you want to know WHY) Aug 20 12:43:24 what does it means this ipkg message error: Aug 20 12:43:27 Multiple packages (tzdata and tzdata) providing same name marked HOLD or PREFER. Using latest. Aug 20 12:44:27 olegfink what is the goal with your initrd/initramfs? Aug 20 12:44:30 *sigh* Aug 20 12:45:01 woglinde: long story, in short I need it to set up loopback root Aug 20 12:46:00 switch_root says it's not rootfs Aug 20 12:46:14 however I _know_ it is (from output of mount) Aug 20 12:47:18 olegfink, a couple of year ago i solved a similar problem, using pivot_root Aug 20 12:47:39 diego yes for kernel 2.4 and initrd Aug 20 12:47:57 with kernel 2.6 its better to use initramfs and switch_root Aug 20 12:48:00 diego: if you could find your linuxrc (or complete initrd) it would be very helpful for me Aug 20 12:49:03 woglinde: what does initramfs have what I will be missing in initrd? Aug 20 12:49:29 woglinde: and do I need to change anything in kernel config to use it? Aug 20 12:50:57 hm olegfink I googled a lot to find all the infos about pivot_root and switch_root and how it works Aug 20 12:51:48 I've glanced through their source code in busybox Aug 20 12:51:58 but haven't understood too much Aug 20 12:52:52 maybe this helps you Aug 20 12:52:57 http://www.busybox.net/lists/busybox/2007-April/027067.html Aug 20 12:54:30 or read http://bugs.archlinux.org/task/3402?getfile=438 Aug 20 12:54:36 okay I have to go now Aug 20 12:54:38 bye Aug 20 12:56:17 has anyone here successfully created a package with a wxwidget based gui? Aug 20 12:56:19 archlinux? hm Aug 20 12:56:23 for what i remember pivot_root just switch between the old root and the new root Aug 20 12:56:52 diego: I'm trying to do what is described in kernel Documentation/initrd.txt Aug 20 13:06:24 03mickeyl 07org.oe.dev * re44796d1... 10/ (6 files in 4 dirs): python-pygtk-1.2: catch up with renaming Aug 20 13:06:35 03mickeyl 07org.oe.dev * rbdbfa6d4... 10/ (3 files in 2 dirs): python-pytester: fix SRCNAME to make it fetch Aug 20 13:06:43 03mickeyl 07org.oe.dev * r7dbcd6b1... 10/ (3 files in 2 dirs): Aug 20 13:06:43 task-openmoko: include gpe-scap Aug 20 13:06:43 task-python-everthing: catch up w/ python-pytester renaming Aug 20 13:06:53 03mickeyl 07org.oe.dev * r384c03fc... 10/ (3 files in 2 dirs): disapproval of revision '7dbcd6b1eac989239df942a9072b3fe7044e977c' Aug 20 13:07:11 03mickeyl 07org.oe.dev * ra11f318a... 10/ (3 files in 3 dirs): Aug 20 13:07:11 initscripts: add a functions script that is supposed to be sourced from individual init scripts providing some useful functions. Aug 20 13:07:11 added cpuinfo_id() and killproc() for a start Aug 20 13:07:29 03mickeyl 07org.oe.dev * r32a741e2... 10/ (1 packages/tasks/task-python-everything.bb): task-python-everything: catch up with python-pytester name change Aug 20 13:07:36 03mickeyl 07org.oe.dev * ra380dceb... 10/ (1 packages/tasks/task-openmoko.bb): task-openmoko: add gpe-scap Aug 20 13:07:41 03mickeyl 07org.oe.dev * rfd012d75... 10/ (3 files in 2 dirs): libsdl-net: update to 1.2.7 Aug 20 13:08:04 olegfink, i've found my old script Aug 20 13:10:01 03hrw 07org.oe.dev * r4287770e... 10/ (1 packages/libtiff/tiff_3.7.2.bb): tiff: more granulated packaging Aug 20 13:10:12 diego: cool Aug 20 13:11:56 http://pastebin.ca/664367 (it should be the relevant part) Aug 20 13:13:42 diego: nice, what I really want to know is what this $* is Aug 20 13:14:01 because running without args results in init: usage: 12345s.... Aug 20 13:16:02 i.e. which argument should I pass to init on chroot Aug 20 13:16:57 ok, my solution was quite tricky, couse i use to pass to kernel the argument: init= Aug 20 13:17:09 hm Aug 20 13:17:16 I can show my linuxrc Aug 20 13:17:16 so the argument of my script are the argument passed by kernel Aug 20 13:18:17 diego: please take a look: http://rafb.net/p/MGKPn888.html Aug 20 13:20:15 you know that everything after 'exec ... ' will never been executed? Aug 20 13:20:58 btw, just to find out if it works, u can replace $@ with 2 Aug 20 13:21:22 I've set 5 and it doesn't Aug 20 13:21:31 complains about missing /dev/initctl Aug 20 13:21:55 crate it ^^ Aug 20 13:22:08 created in userland, still the same Aug 20 13:23:00 permission denied? Aug 20 13:23:16 hmm Aug 20 13:23:25 okay, I have to run away now Aug 20 13:23:32 thanks for the help Aug 20 13:23:35 will try later Aug 20 13:56:30 03mickeyl 07org.oe.dev * r38c630cb... 10/ (7 files in 7 dirs): ode / python-pyode: move to nonworking, needs a lot of gcc4 fixes to compile again Aug 20 13:56:35 03mickeyl 07org.oe.dev * rfd51b44f... 10/ (1 conf/distro/include/sane-srcdates.inc): sane-srcdates: add entries for python EFL bindings Aug 20 13:56:40 03mickeyl 07org.oe.dev * r9b67d266... 10/ (4 files in 2 dirs): openmoko-terminal2: lower font size Aug 20 13:56:48 03mickeyl 07org.oe.dev * rf8b2adf1... 10/ (3 files in 3 dirs): mrxvt: change default fonts and allow more smaller fonts Aug 20 13:56:55 03mickeyl 07org.oe.dev * rff470c07... 10/ (1 conf/distro/include/sane-srcdates.inc): sane-srcdates: include common variable for EFL Aug 20 13:57:01 03mickeyl 07org.oe.dev * rb4249d62... 10/ (1 conf/distro/include/sane-srcdates.inc): sane-srcdates: fix typo in value for eds-dbus Aug 20 14:01:37 hmm avahi breaks to find rpl__realloc on mipsel .. even adds in site/mipsel-linux doesnt fix that Aug 20 14:07:09 morning Aug 20 15:01:08 The guy working with the gumstix and angstrom asked the gumstix people to send me one :) Aug 20 15:01:28 Aparently, they are aware of OE and have considered using it Aug 20 15:05:29 Crofton|home: Did they give a reason why they're not using it? Aug 20 15:06:25 RP: they use buildroot for few years Aug 20 15:07:48 I suspect inertia Aug 20 15:09:08 and NIH? Aug 20 15:09:32 change is always difficult Aug 20 15:09:47 especially when change is big Aug 20 15:09:58 this is why we need to encourage them to supply hw so we can do some testing Aug 20 15:10:03 and avoid bit rot Aug 20 15:14:00 ~lart cmake Aug 20 15:14:01 * ibot beats cmake over the head with a microkernel Aug 20 15:21:09 It seems like we need to document what we support Aug 20 15:21:28 for example, we have boards donated to the project Aug 20 15:21:35 so we should support those well Aug 20 15:21:56 boards core devs have access to via work, so we should be OK there Aug 20 15:21:57 I plan to run angstrom on alix in this week Aug 20 15:22:32 then we have boards owned by core dev indivually, so we do not lose access if they change job Aug 20 15:22:58 and then we have stuff theere are configs for, but we are dependent on tohers to test Aug 20 15:23:30 Crofton|home: some of boards which core devs have access at work are not in OE Aug 20 15:23:54 well, they should be beaten then :) Aug 20 15:24:09 hrw, I am only thinking of what we have in OE Aug 20 15:24:11 no - NDA limits Aug 20 15:24:20 since that is what we have to maintain Aug 20 15:24:29 hrw I undertand Aug 20 15:24:30 on guadec OH shown Poky on ST Nomadik and Intel^WMarvell Zylonite Aug 20 15:24:35 but that is outside teh scope Aug 20 15:25:05 sure Aug 20 15:25:32 I am thinking how we can show vendors it is in their interest to feed us Aug 20 15:26:31 and give users an idea of how effectively we support the hardware they use Aug 20 15:26:38 anyway, time to walk to work Aug 20 15:26:39 first we need to have list of supported devices on page Aug 20 15:26:54 then clean rules how to get board supported Aug 20 15:27:08 ls conf/machine/*.conf :) Aug 20 15:28:34 no Aug 20 15:28:40 not a list of devices in OE Aug 20 15:28:47 but list of really supported devices Aug 20 15:28:56 devices which someone take care of Aug 20 15:29:06 and a way to give credit to vendors that support us Aug 20 15:29:28 so omap5912osk is supported, but progear is not (because it is packed and on shelf due to lack of space on my desk) Aug 20 15:29:57 but OSK is only supported because I have access via work Aug 20 15:30:08 if I change jobs, support quality drops Aug 20 15:30:56 sure Aug 20 15:31:04 Crofton|home : You did not notice that when you signed on as OE dev, there was a clause stating that you are not allowed to change jobs :) Aug 20 15:31:20 unless the other dev's agree Aug 20 15:31:26 Crofton|home: but if vendor or someone other wants to keep osk supported then he can hire you or provide board Aug 20 15:31:32 exactly Aug 20 15:31:32 steliosk: really? Aug 20 15:31:43 steliosk: my job change gave better support for OE :D Aug 20 15:32:01 hrw : That;s why nobody objected :) Aug 20 15:32:12 ok, I need to go Aug 20 15:32:16 Basicaly I have a vision Aug 20 15:32:31 of a table of devices, with vendors who are good to us at the top Aug 20 15:32:33 :) Aug 20 15:32:54 aha table full of devices :) Aug 20 15:33:53 and stuff like Zauruses under the desk near trashcan? Aug 20 15:34:00 it would be much better if there are specific devs for each board etc Aug 20 15:35:00 steliosk: like you or me for alix? Aug 20 15:35:01 steliosk: btw - did you tried alix build? Aug 20 15:35:33 hrw : yes something like that. Aug 20 15:35:36 03mickeyl 07org.oe.dev * ra7156792... 10/ (1 packages/efl1/evas.inc): evas.inc: configure patch no longer applies, seems to be no longer necessary as well Aug 20 15:35:40 03mickeyl 07org.oe.dev * r741a12f5... 10/ (1 packages/openmoko2/openmoko-terminal2_1.0.0.bb): openmoko-terminal2: enable antialiasing again, reduce menu bar font Aug 20 15:35:45 03mickeyl 07org.oe.dev * r1aa9d5d2... 10/ (3 files in 2 dirs): libpng[-native]: update to 1.2.19 Aug 20 15:35:52 03hrw 07org.oe.dev * r580d2297... 10/ (4 files in 2 dirs): gammu: updated to 1.13.0 Aug 20 15:35:57 hrw : not yet. need to finish some other stuff first Aug 20 15:36:00 03hrw 07org.oe.dev * re36bd25b... 10/ (1 packages/python/python-gammu_0.21.bb): python-gammu: unbreak it - we have recent gammu now Aug 20 15:36:37 re Aug 20 15:40:40 woglinde: wb Aug 20 15:41:51 hi florian Aug 20 15:50:56 Be nice to the gumstix guy on the angstrom list, he is doing some good PR work for us. Aug 20 15:51:59 hi crofton Aug 20 15:55:07 uf.. task-python-everything fetchalled. now time to build Aug 20 16:01:23 gm woglinde Aug 20 16:01:36 03hrw 07org.oe.dev * r3fd52ff8... 10/ (1 packages/tasks/task-python-everything.bb): task-python-everything: remove python-pyode which is in nonworking Aug 20 16:01:40 03hrw 07org.oe.dev * r6951a4bf... 10/ (1 conf/distro/include/angstrom-glibc.inc): angstrom-glibc: prefer glibc-initial Aug 20 16:02:44 koen, ping Aug 20 16:03:07 bye all Aug 20 16:03:10 gn Aug 20 16:03:48 hmm Aug 20 16:04:11 how does pkgconfig now where to get its data from? Aug 20 16:04:18 i.e. non-native versus -native? Aug 20 16:04:58 and did that change recently? Aug 20 16:09:42 diego: here? Aug 20 16:15:54 hrw / crofton: what other diecent arm devices are they aside from the c3000/c3100/c32000 zarus devices? Aug 20 16:16:31 what do you want to do with one? Aug 20 16:16:59 built in keyboard touchscreen wifi bluetooth alsa sound at least 640x480 Aug 20 16:17:38 cf is fine for wifi bt if it has to be, gps would be nice too Aug 20 16:18:53 I do not do much with devices that have displays :) Aug 20 16:19:17 the only comparable devices i've seen are not arm, the oqo2 and sony vaio ux, but they're way out of my range, twice what the c3000 yes Aug 20 16:19:58 yes=was Aug 20 16:24:56 okay folks Aug 20 16:25:13 I need some ideas on how do I write a linuxrc for initrd Aug 20 16:25:29 I need to free it after it's over Aug 20 16:25:57 is kernel Documentation/initrd.txt way up-to-date and usable? Aug 20 16:27:29 koen : Could you check what the gcc-cross for efika displays for --target-help ? Aug 20 16:27:43 koen : i removed tmp ;( Aug 20 16:27:57 any docs, howtos or other channel suggestions would of great help for me Aug 20 16:29:52 *would be Aug 20 17:14:47 hello all Aug 20 17:15:01 hello likewise Aug 20 17:15:19 likewise: hi Aug 20 17:15:43 olegfink: could you find a solution yet? Aug 20 17:15:54 for THAT problem Aug 20 17:16:01 the initrd stuff? Aug 20 17:16:04 it was _exacltly_ proc missing Aug 20 17:16:12 but now I'm facing new one Aug 20 17:16:20 (one line below in linuxrc :) Aug 20 17:16:30 my exec chroot ... line Aug 20 17:16:39 what should I start in it? Aug 20 17:16:43 /sbin/init? Aug 20 17:17:28 olegfink: probably yes Aug 20 17:17:47 it says that I should pass an argument to it Aug 20 17:17:52 but that's not all Aug 20 17:18:03 when I write init 5 (just to check) Aug 20 17:18:19 I'm not really the person with the most experience on initramfs's and pivot_root's. Never used them. Aug 20 17:18:34 it first claims about missing /dev/initctl, and when I create one, it says it's timeout o0 Aug 20 17:18:44 likewise: my problem is with init now =) Aug 20 17:18:54 olegfink: I would really Google for these errors, you are not the first one to hit this. Aug 20 17:21:45 okay Aug 20 17:40:52 likewise: YAY Aug 20 17:41:03 olegfink: found it? Aug 20 17:41:08 likewise: you suggestion, though obvious, was really helpful Aug 20 17:41:10 ;) Aug 20 17:41:23 found on LQ Aug 20 17:42:19 likewise: however the solution is strange Aug 20 17:42:36 kernel commandline I used to have (relevant part): Aug 20 17:42:41 root=/dev/loop0 Aug 20 17:42:55 kernel commandline I have now (relevant part): Aug 20 17:43:03 root=/dev/ram0 init=/linuxrc Aug 20 17:43:20 _why_ does it trigger init having PID 1? Aug 20 17:44:35 olegfink: dunno... Aug 20 17:44:51 init=/linuxrc looks really idiotic for me Aug 20 17:45:11 sure it _is_ and _was_ /linuxrc, how else my initrd worked before? Aug 20 17:51:47 likewise, omap5912osk builds on F7 with libtool 1.5.10 and 1.5.24 ... Aug 20 17:52:53 Crofton: ok, can you tell we your OE revision? cd org.openembedded.dev && mtn automate get_base_revision_id Aug 20 17:52:59 s/we/me Aug 20 17:53:21 i think we have 2 heads in the repo Aug 20 17:53:32 Crofton: I will then test again with that revision on Ubuntu 7.04. (I still suspect host-dependent stuff creeping into the build) Aug 20 17:53:39 mtn: updating along branch 'org.openembedded.dev' Aug 20 17:53:39 mtn: multiple update candidates: Aug 20 17:53:39 mtn: 93974111864d318a6e96f9646847c5a9647aa819 pH5@openembedded.org 2007-08-20T17:45:29 Aug 20 17:53:39 mtn: f7fb409ce8abfb0a3e4ef36fe058a6fbb429a7b6 oe@openembedded.org 2007-08-20T17:47:51 Aug 20 17:55:01 03mickeyl 07org.oe.dev * ra76f71ae... 10/ (5 files in 4 dirs): evas/ecore: reinstantiate configure patches - without them EFL picks up wrong CFLAGS/LIBS Aug 20 17:59:01 Check bug tracker, nothing there, libxml2 if failing on a "bitbake bootstrap-image" any ideas ? worked a couple of days ago. Aug 20 17:59:20 03mickeyl 07org.oe.dev * ra76f71ae... 10/ (5 files in 4 dirs): evas/ecore: reinstantiate configure patches - without them EFL picks up wrong CFLAGS/LIBS Aug 20 18:00:58 likewise, 7003f77db62eaca322374d2eed9a7a32815a3d65 Aug 20 18:01:12 likewise, agreed Aug 20 18:01:27 it is always worthwhile reducing stuff like that Aug 20 18:01:55 my native libtool (from rpm) is 1.5.22 Aug 20 18:04:01 Crofton: "The program 'libtool' is currently not installed." Aug 20 18:04:10 hmmm Aug 20 18:04:14 Crofton: I don't have any native installed (I could later) Aug 20 18:04:52 olegfink, yes the same i have done: init=/script Aug 20 18:05:09 tell you what why don't i do a mtn update and mtn pull (Someone slap me) Aug 20 18:05:50 diego_: works now, but umount still doesn't want to work out /initrd Aug 20 18:06:24 it means that someone is still using it Aug 20 18:06:54 do u have fuser o lsof? Aug 20 18:07:17 no, but I may try getting it Aug 20 18:07:24 does busybox have either? Aug 20 18:07:49 it has fuser, ok Aug 20 18:09:50 otherwise, u can try umount -n Aug 20 18:11:41 -n as in don't write to /etc/mtab? Aug 20 18:11:45 how would it help? Aug 20 18:20:37 olegfink, sometimes mount fail just becouse root is read only ^^ Aug 20 18:21:14 not my case Aug 20 18:21:23 I'm looser Aug 20 18:21:45 ran find /initrd -exec fuser {} \; and forgot to umount /initrd/proc Aug 20 18:22:46 don't worry, i had to make a lot of mistake, before i can make it work ^^ Aug 20 18:24:27 diego_: Aug 20 18:24:49 umount: couldn't umount /initrd: No such file or directory Aug 20 18:25:42 and the directory is still there? ls /initrd? Aug 20 18:25:53 sure Aug 20 18:26:07 the whole initrd contents are under it Aug 20 18:26:09 also Aug 20 18:26:17 $ fuser /dev/ram0 Aug 20 18:26:28 before i forget, i see there was an error in your script Aug 20 18:26:29 $ freeramdisk /dev/ram0 Aug 20 18:26:43 freeramdisk: Device or resource busy... Aug 20 18:26:55 diego_: yeah? however I changed it a lot Aug 20 18:27:29 diego_: http://rafb.net/p/GnQ8gi82.html Aug 20 18:28:54 looks like I will see some gumstix Aug 20 18:29:25 uhm, but the new root, is new_root or initrd? Aug 20 18:29:42 new-root Aug 20 18:31:48 ok, so line 7 should be mount .... /initrd Aug 20 18:31:52 khem, ping Aug 20 18:32:05 olegfink, sorry Aug 20 18:32:16 it should be new-root Aug 20 18:32:24 line 10 should be pivot_root . /new-root Aug 20 18:32:27 initrd exists inside new-root Aug 20 18:32:34 Crofton: gm Aug 20 18:32:57 I'm going to delete the patch and push Aug 20 18:32:59 diego_: no, pivot_root is Aug 20 18:33:08 I am pretyt sure you are the only guy using the file :) Aug 20 18:33:24 i have to go... cya Aug 20 18:33:30 so new root is . (pwd=/new-root) and old root goes to be /initrd Aug 20 18:34:45 Crofton|work_: OK Aug 20 18:34:58 Crofton|work_: Looks like that Aug 20 18:35:33 * khem is doing clean build with bb 1.8.8 Aug 20 18:36:39 Khem: I didn't follow koen and your efforts to get eglibc in oe. is there something to play with? Aug 20 18:37:24 yeah likewise its working for me. Aug 20 18:38:08 likewise: if you want to try it then add ANGSTROM_MODE = "eglibc" Aug 20 18:38:14 in your local.conf Aug 20 18:38:36 Khem: thanks. I will familiarize with eglibc first a bit. Aug 20 18:38:55 a bit, first, rather Aug 20 18:38:56 nothing much to familiarize if you know glibc Aug 20 18:39:08 but slowly its gonna change Aug 20 18:39:57 * khem thinks of cross locale generation Aug 20 18:40:34 ah, so it's not radically different, except better configurable? Aug 20 18:40:40 * likewise reading the website Aug 20 18:42:20 likewise: yeah and supports embedded arches like mips and arm respectfully :) Aug 20 18:42:29 Khem: ppc as well? Aug 20 18:42:44 Khem: or must we push Freescale? :-) Aug 20 18:43:03 yes Aug 20 18:43:15 actually PPC FPUless Aug 20 18:43:37 normal ppc is well supported in upstream glibc Aug 20 18:48:03 khem : the question here is which "ppc" :) Aug 20 18:48:22 steliosk: mpc for sure Aug 20 18:48:28 ppc as well :) Aug 20 18:49:25 Khem: why the difference between ppc and ppc fpu-less, I do not quite follow? Aug 20 18:49:37 likewise : the fpu ;) Aug 20 18:49:47 likewise: ppc systems that does not have FPU Aug 20 18:50:22 * khem needs softfloat Aug 20 18:50:33 glibc has some issues with soft-floats on ppc Aug 20 18:50:53 yes, I understand what fpu-less is, but does eglibc support both fpu-full and fpu-less powerpc cores? Aug 20 18:51:04 likewise yes Aug 20 18:51:13 it supports both Aug 20 18:51:14 likewise : yes it does Aug 20 18:51:19 Khem: ok, then I misunderstood your answer. thanks Aug 20 18:52:02 likewise : hopefully,someday, the soft-float/fpu code will find its way into glibc so we don't have to hunt patches Aug 20 18:52:13 who started to work for openmoko :) Aug 20 18:52:26 steliosk: You will start using eglibc it means :) Aug 20 18:52:49 khem, the mtn diff shows a change in the arm1136 tune file, but the patch is empty Aug 20 18:52:53 you may want to revert Aug 20 18:53:09 hmmm ok Aug 20 18:53:39 khem : did some minimal testing using buidlroot/eglibc as i was looking for the float point performance Aug 20 18:53:41 I am going push the dropped patch Aug 20 18:54:11 arg, my priv key is at home .... Aug 20 18:54:14 steliosk: ah cool so you could get buildroot working Aug 20 18:54:25 I never got one working so I trashed it Aug 20 18:55:36 khem : it took several days but at the end we ended up using oe/glibc Aug 20 18:56:05 oe/eglibc is on the (short) todo list Aug 20 18:56:11 steliosk: yeah OE is cool man atleast it compiles and boots Aug 20 18:56:56 steliosk: I am intending to do cross-locales which will be a huge win for eglibc but depends how long my daughter sleeps during the night Aug 20 18:57:32 Khem: she doesn't sleep => you can develop, or she doesn't sleep => less time for development? :-) Aug 20 18:57:40 yay my greenphone is here Aug 20 18:58:02 khem : for embedded guys talking about locales is a joke (as i found out when i tried to explain why OE takes that long to build glibc) Aug 20 18:58:07 likewise: if she is awake I have to cater to her :) Aug 20 18:58:44 Khem: then stick with glibc which takes all night to build :-) Aug 20 18:58:59 khem : I am preparing for similar situation. My wife is 7months pregnant now Aug 20 18:59:04 this is why I disable Binary Local :) Aug 20 18:59:10 likewise: :) well eglibc is no less but Aug 20 18:59:52 steliosk: hmm yeah you will see the child's controlling powers Aug 20 18:59:53 khem : for which board did you build/try oe/eglibc Aug 20 19:00:35 steliosk: osk5912 omap2430sdp with oe/eglibc Aug 20 19:01:32 I think we havethe complete mpc/ppc developers group of oe complete, am I right? :-) Aug 20 19:01:34 khem : I am all ready feeling it. Everything allready revolves around the child (and mother) :) Aug 20 19:02:01 steliosk: I have sucessfully booted several ppc and mips targets too x86 is seftunderstood with eglibc but not with oe Aug 20 19:02:08 steliosk: If you feel neglected, just login on #oe :-) Aug 20 19:02:16 * khem havent mucked arounf with ppc much Aug 20 19:02:24 likewise : I am :) Aug 20 19:02:36 * koen wakes up Aug 20 19:03:02 likewise : as for neglected, can't say that.... There so many things they want me to do :) Aug 20 19:03:04 * likewise hits koen with a bat - go sleep! ;-) Aug 20 19:03:18 koen, the gumstix guy from angstrom-users sounds like he has begged gumstix to send me some gumstix Aug 20 19:03:27 Crofton: cool Aug 20 19:03:30 yeah Aug 20 19:03:42 koen: enjoyed your ll weekend? Aug 20 19:03:44 hopefully he gets one of the fast ones ... Aug 20 19:03:48 Crofotn : What flavour ? :) Aug 20 19:03:52 dunno Aug 20 19:03:54 likewise: very much! Aug 20 19:03:59 beggars can't be chosers Aug 20 19:04:11 lots of scantily clad hippy chicks? Aug 20 19:05:06 khem : We need to get eglibc build/tested for some of the ppc boards in OE Aug 20 19:05:19 khem, I pushed the linux-omap2 change Aug 20 19:05:35 khem : i will do that, but need to finish some Xilinx stuff first Aug 20 19:05:55 Crofton: cool Aug 20 19:07:10 khem Crofton : Got any numbers on how big eglibc based system is ? compared to glibc.uclibc Aug 20 19:07:23 03likewise 07org.oe.dev * rb396aa55... 10/ (1 packages/makedevs/makedevs-1.0.0/makedevs.c): makedevs: Add support for creating directories. Aug 20 19:07:25 Crofton: why, you want to visit lowlands next year? Or hold a OE conference there? Aug 20 19:07:27 03crofton 07org.oe.dev * ra4bed52e... 10/ (1 packages/linux/linux-omap2_git.bb): linux-omap2 : Drop lcd patch from kernel bb file. Patch is in git now. Aug 20 19:07:32 03mickeyl 07org.oe.dev * ra76f71ae... 10/ (5 files in 4 dirs): evas/ecore: reinstantiate configure patches - without them EFL picks up wrong CFLAGS/LIBS Aug 20 19:07:34 steliosk: I bet it wont be any less than glibc as of now Aug 20 19:07:36 03pH5 07org.oe.dev * rb7ecd3fb... 10/ (1 packages/linux/linux-handhelds-2.6/magician/defconfig): linux-handhelds-2.6: update magician defconfig for pasic3 changes Aug 20 19:07:40 03pH5 07org.oe.dev * r93974111... 10/ (3 files in 3 dirs): Aug 20 19:07:40 libgsmd: band-aid. stage common/linux_list.h and fix up the mlbuf Aug 20 19:07:40 changes until this is repaired upstream Aug 20 19:08:10 steliosk: I can test on a MPC8313RDB board soonish, and EFIKA. Aug 20 19:08:45 I am always looking for excuses to travel Aug 20 19:08:51 likewise : the mpc is the board you have told me ? Aug 20 19:09:16 steliosk: yes, http://www.micetek.com/englishweb/product/new/PQIIproDevelopmentProducts/8313E-RDB.htm Aug 20 19:09:38 steliosk: I couldn't get free boards, only minimal discount. I could ask Freescale... Aug 20 19:09:57 but I do not have close contacts w/ Freescale. Aug 20 19:10:17 steliosk: MPC8313 is FPU-full Aug 20 19:10:47 Crofton : BTW They moved the astricon in the the desert :/ https://reservations.ihotelier.com/crs/g_reservation.cfm?groupID=58541&hotelID=13181 Aug 20 19:10:51 steliosk: http://www.micetek.com/englishweb/product/new/PQIIproDevelopmentProducts/MPC8323E-RDB.htm is FPU-less Aug 20 19:11:28 * Crofton is afraid to check the price Aug 20 19:11:46 likewise : yep. That's the boards. Aug 20 19:11:54 Crofton : 120$ Aug 20 19:12:16 not as bad as I expected Aug 20 19:13:28 Crofton : the Power.org dev is 149$ single/double room Aug 20 19:13:55 you should be ok with the temperature also :), I was in Phoenix in late June and it was 110F Aug 20 19:14:30 i am used to that temperatures Aug 20 19:14:49 just heard at the news, they expect a new heatwave the next few days Aug 20 19:15:05 talking allready about 41-42+ Celcius Aug 20 19:18:41 likewise: the freescale boards have SPEs I think not the standard PPC FPUs Aug 20 19:19:32 Khem: looking Aug 20 19:43:57 * * OE Bug 2809 has been created by gfellerpatrik(AT)gmx.net Aug 20 19:43:59 * * zudoku .bb did not install & package all necessary files Aug 20 19:44:01 * * http://bugs.openembedded.org/show_bug.cgi?id=2809 Aug 20 19:52:45 hmm Aug 20 19:53:58 * * OE Bug 2810 has been created by carbonfreeze(AT)gmail.com Aug 20 19:54:00 * * at daemon: /var/spool/at is not created on installation Aug 20 19:54:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2810 Aug 20 20:01:45 pgfeller, and if you have problems/questions with this zudoku issue, feel free to ask here too ;-) Aug 20 20:02:16 03mickeyl 07org.oe.dev * r75a70f98... 10/ (1 classes/distutils.bbclass): classes/distutils.bbclass: add a staging task (to be called explicitly) Aug 20 20:02:25 03mickeyl 07org.oe.dev * rab1eefbf... 10/ (1 packages/python/python-evas_cvs.bb): python-evas: make it build reliably (again) Aug 20 20:02:30 03mickeyl 07org.oe.dev * rf1d39ac0... 10/ (1 packages/python/python-efl.inc): python-efl.inc: bump PV Aug 20 20:02:57 polyonymous: thanks a lot; I'll try to define the var in the distro conf. & report back to bugtracker if it works Aug 20 20:03:29 polyonymous: unfortunatela those are not yet in the oe userdocu (@ least I did not find them) Aug 20 20:03:46 pgfeller, great. It's just that I can't build and test sharprom-compat stuff here... Aug 20 20:04:01 pgfeller, yeah, that's relatively new stuff, I don't think it's anywhere ;-) Aug 20 20:04:24 pgfeller, and maybe I should read oe docu and perhaps contribute to it too - I haven't ever read it yet ;-) Aug 20 20:04:47 polyonymous: np ... I try to figure as much I can from there Aug 20 20:05:03 Good, thanks. Aug 20 20:05:31 polyonymous: as I still use some comercial apps on my collie I'm very happy that sharp-compat distro exists :-) ... gives me access to great SW. Aug 20 20:05:40 polyonymous: thanks again & cu Aug 20 20:05:48 pgfeller, good luck ;-) Aug 20 20:05:58 * * OE Bug 2811 has been created by carbonfreeze(AT)gmail.com Aug 20 20:06:00 * * gpe-clock alarm resume disables keyboard when gpe-lock is used Aug 20 20:06:02 * * http://bugs.openembedded.org/show_bug.cgi?id=2811 Aug 20 20:20:21 I would like to add Intel Core 2 support. I propose i686-core2.conf and include/tune-i686-core2.conf but if someone thinks different, pls discuss. Aug 20 20:20:31 http://gentoo-wiki.com/Safe_Cflags#Intel_Core_2_Duo.2FQuad_.2F_Xeon_51xx.2F53xx Aug 20 20:22:00 Hmm, instead x86_32-core2 and x86_64-core2 seem more approriate. Aug 20 21:16:42 likewise: yes, please use x86--core2.conf Aug 20 21:17:36 mickey|tv: I think x86-prescott and x86_64-nocona are better, see my email to oe ml. Aug 20 21:17:57 mickey|tv, it sounds like gumstix will send me some boards Aug 20 21:18:04 ah Aug 20 21:18:10 mickey|tv: they specify the x86 architecture better and cover a wider range of processors. Aug 20 21:18:26 the gentoo wiki and gcc sources are very helpful. Aug 20 21:18:28 likewise: ok, however i'd prefer x86-32-prescott then Aug 20 21:18:38 the guy that was on the angstrom users list talking to me over the weekend seems "well connected" Aug 20 21:18:46 mickey|tv: ok, that makes sense Aug 20 21:18:49 we'll see if the stuff really arrives though :) Aug 20 21:19:03 Crofton: cool, *crossing fingers* Aug 20 21:19:09 yeah Aug 20 21:19:24 mickey|tv: however, I am confused about what my TARGET_ARCH must be, as linux-headers seems to use it Aug 20 21:19:29 We'll see what shows up. If I get more than I can keep up with, we can slit them up Aug 20 21:19:32 likewise: btw, what's a gain using this flags? Aug 20 21:19:59 i just wonder if one can notice such compiler settings Aug 20 21:20:00 ynezz: compile for a specific instructions set, and optimize for a processor pipeline. Aug 20 21:20:49 ynezz: I need it for media processing (coding) and there it matters (SSE3 and SSSE3). No, it does not differ very much, but I would like to find out. Aug 20 21:21:12 cool, i would like to know results also Aug 20 21:21:27 mickey|tv, we'll need a better booth at FOSDEM :) Aug 20 21:21:44 Crofton: tell me something new :D (see /topic) Aug 20 21:21:55 hi zecke Aug 20 21:22:01 http://www.nabble.com/gumstix---OpenEmbedded-t4264443.html Aug 20 21:22:07 :) Aug 20 21:22:52 I guess it is worth pushing some changes just to help a guy out :) Aug 20 21:28:31 ho Aug 20 21:29:02 hoho Aug 20 21:31:45 Crofton: stick 'm up :-) Aug 20 21:32:52 heh Aug 20 21:37:00 mickey|tv: If I use i686 as TARGET_ARCH, I wrongly tell these executable run on any i686 architecture. I'm clueless how to solve this. Aug 20 21:37:34 likewise: i need to think about that as well Aug 20 21:42:27 mickey|tv: that's both good and bad news :-) Aug 20 21:48:18 hrw|gone: NOTE: package task-python-everything-1.0: completed Aug 20 21:48:23 hrw|gone: on x86_64 buildmachine Aug 20 22:00:36 03mickeyl 07org.oe.dev * r37817863... 10/ (4 files in 3 dirs): python-lxml: add patch to make it use pkgconfig instead of binconfig to detect its dependencies Aug 20 22:07:15 in which package are the fonts used by GTK located? My fonts in gtk apps are gone. Aug 20 22:07:19 only seeing boxes Aug 20 22:11:08 good nite all Aug 20 22:11:48 dcordes: ttf-bitstream-vera e.g. Aug 20 22:12:06 or some dejavu package Aug 20 22:17:42 flo_lap: is that the gpe standard font? Aug 20 22:19:57 dcordes: it depends on the image, ttf-bitstream-vera is quite small while the devavu fonts are an improved version of the vera fonts that support way more charsets. Aug 20 22:22:27 flo_lap: is it possible to use wildcards in ipkg? Aug 20 22:23:17 I installed some of the fonts and still only get boxes in gpe. Maybe that's a bug Aug 20 22:24:12 dcordes: maybe a pango issue or a pango configuration problem Aug 20 22:24:30 dcordes: which image on what kind of device do you use? Aug 20 22:26:44 flo_lap: an angstrom image on zaurus akita Aug 20 22:28:52 dcordes: Try to install ttf-dejavu-sans Aug 20 22:29:21 maybe some configuration in angstrom relies on these Aug 20 22:32:49 flo_lap: was already installed Aug 20 22:33:58 dcordes: ok, when did that start? just after installing the image or later? Aug 20 22:35:50 flo_lap: later. I compilled the image about a month ago. Last week I did some upgrading with ipkg and during the last days that boxes-text appeared Aug 20 22:36:25 I didn't remove anything related to fonts, gtk or so and n'either did I install new gui stuff Aug 20 22:37:34 dcordes: do you remember what was upgraded? Aug 20 22:40:55 flo_lap: A lot- I have no idea Aug 20 22:41:12 flo_lap: ipkg does not keep any logs by default, does it? Aug 20 22:41:53 dcordes: hmm no... but the psotinst scripts Aug 20 22:43:13 flo_lap: could reinstalling gtk libs help? Aug 20 22:43:53 dcordes: maybe... or try this: Aug 20 22:43:55 pango-querymodules > /etc/pango/pango.modules Aug 20 22:44:19 in the case the pango modules postinst did not run correctly Aug 20 22:47:45 I've got bootstrap-image working with qemu, what steps would I need to take to get a apache installed on the image Aug 20 22:49:27 budgester: try "ipkg update" and "ipkg install apache" Aug 20 22:49:50 * flo_lap has no ides of angstrom ships apache Aug 20 22:54:42 flo_lap, is that on the dev machine ? or the booted image ? sorry a bit new to oe Aug 20 22:56:32 budgester: on the booted image - it updates your pango module configuration Aug 20 22:56:41 there is a .bb file for apache2 so i'll use that Aug 20 23:02:51 good night Aug 20 23:09:59 flo_lap: rewriting the pango modules conf file did the trick. thanks Aug 20 23:24:29 03mickeyl 07org.oe.dev * rea002ef2... 10/ (1 packages/python/python-openmoko_0.1.bb): add python-openmoko, Python Bindings to the OpenMoko Phone Platform Libraries Aug 20 23:30:12 bye **** ENDING LOGGING AT Tue Aug 21 02:59:56 2007