**** BEGIN LOGGING AT Thu Feb 22 02:59:58 2007 Feb 22 04:12:22 hi * Feb 22 04:12:39 my old distro using inherit image_ipk in bb file Feb 22 04:14:58 anybody can tell me this package has been rename ? Feb 22 04:15:12 hmm bbclass file in classes, Feb 22 07:37:28 gm Feb 22 08:21:26 hi Feb 22 08:36:40 morning all Feb 22 08:36:54 saladino: hi Feb 22 08:37:04 hi do13_ Feb 22 08:37:14 hi Ifaistos Feb 22 08:37:26 hey Dirk, morning all Feb 22 08:37:28 morning Feb 22 08:37:33 hey Liam, Marcin Feb 22 08:38:41 hrw : Have you noticed any libtool problem for x86 and uclibc ? in my case gettext fails compaining about a missing m4 file Feb 22 08:39:03 Ifaistos: I do not remember when last time I did uclibc build Feb 22 08:39:09 and it was for mipsel Feb 22 08:42:21 morning Feb 22 08:44:13 hi XorA Feb 22 08:47:55 Ifaistos Feb 22 08:48:05 what was the distro you were compiling? Feb 22 08:48:13 hey XorA & do13_ Feb 22 08:48:32 yo koen Feb 22 08:48:43 saladino: my own ;) its called oplinux Feb 22 08:49:03 ahm Feb 22 08:49:06 but uclibc powered? Feb 22 08:49:40 saladino: both. there is a glibc (glibc 2.5) and a uclibc (uclibc 0.9.28-svn) version Feb 22 08:50:06 saladino: i am testing it on x86, powerpc and mips (next week) Feb 22 08:50:10 Ifaistos with what gcc version? did you see my mail to oe list? Feb 22 08:50:20 saladino: gcc 4.1.1 Feb 22 08:50:29 Ifaistos: .29 release after fosdem :) Feb 22 08:50:44 koen : cool :) Feb 22 08:50:54 Ifaistos i tried with that version and i get a "internal compiler error" O_O Feb 22 08:51:24 saladino: i haven't touch arm builds yet Feb 22 08:51:34 ahm Feb 22 08:52:05 im trying to solve it but im a bit stucked Feb 22 08:52:18 saladino: my main focus is on powerpc and x86 targets at the moment Feb 22 08:52:54 and is all compiling correctly? Feb 22 08:53:30 Ifaistos: gcc 4.1.1, glibc 2.5 you say... Feb 22 08:53:42 saladino: well it depends on how you define "correctly" Feb 22 08:53:47 Ifaistos: any plans to join Angstrom distro team? Feb 22 08:54:04 Ifaistos, no terrible errors like internal compiler error not? Feb 22 08:54:59 hrw : i think i partially did :) i got some work/packages on efika submitted to the feed :) Feb 22 08:55:50 saladino: no. they only problem i have and try to solve at the moment is that soft-float support is crappy (at least for non arm targets) Feb 22 08:56:49 mickey|defence: fingers crossed Feb 22 08:56:54 thanks!!! Feb 22 08:57:03 when i'm back it's over --- in any case :) Feb 22 08:57:07 c'ya Feb 22 08:57:23 hrw : i am now trying to figure out which packages work for glibc 2.5, so we can get a angstrom-packages.bb Feb 22 08:59:06 gl mickey|defence Feb 22 09:01:04 Ifaistos: angstrom-packages.bb is a bad idea Feb 22 09:01:14 Ifaistos: it's better to fix the stuff that doesn't work Feb 22 09:02:13 hrw koen : i was thinking of doing monthly "fresh' builds for all supported targets Feb 22 09:03:04 hrw koen : so we can get Q&A on a regular basis on what is working and what not Feb 22 09:04:39 hrw koen : what do you think ? Feb 22 09:14:12 hello! maybe someone can help to solve apparently simple problem with angstrom build error of glibc-intermediate-2.5-r4: task do_configure Feb 22 09:19:10 03koen 07org.oe.dev * r4195ab00... 10/ (3 files in 3 dirs): linux-ezx: remove localversion, we already have it in PR Feb 22 09:23:52 gena_: which arch? Feb 22 09:24:47 x86.conf + c3-tune.conf Feb 22 09:24:57 | checking for kernel header at least 2.6.17... too old! Feb 22 09:24:59 | configure: error: *** The available kernel headers are older than the requeste Feb 22 09:24:59 d Feb 22 09:25:24 where does it look for kernel headers actually? angstrom downloads 2.6.18 actualy Feb 22 09:30:18 do13_: oops, sorry, it was generic who'd downloaded kernel 2.6.18, when building angstrom - tu the time when glibc-intermediate is built there is no unpacked kernel in the work dir, which kernel did it take then? Feb 22 09:30:44 gena_: I have no build for x86 here. look for linux-libc-headers Feb 22 09:31:17 gena_: my progear/angstrom build works. Feb 22 09:31:24 progear is x86 based Feb 22 09:32:18 do13_: hmm, strange, in the work dir there's linux-libc-headers-2.6.15.99-r4. does 2.6.15 stand for the kernel version here? Feb 22 09:40:37 did anyone have "kernel-too-old" problem when building glibc-intermediate with angstrom? Feb 22 09:54:28 hrw: hi, how are you? Feb 22 09:54:41 Hi! why could "cannot find -lgcc_eh" happen when building glibc-intermediate? Feb 22 09:54:54 sorry for bugging everyone, but I'll sum up the problem once more: Feb 22 09:54:55 hrw: I'm currently sitting at the firmware download stuff, now that I have time again for the libertas CF card Feb 22 09:54:55 x86,angstrom,task-base Feb 22 09:54:55 uses linux-libc-headers-2.6.15.99-r4 Feb 22 09:54:55 while do_configure of glibc-intermediate-2.5-r4 Feb 22 09:54:55 I receive configure error: Feb 22 09:54:55 checking for kernel header at least 2.6.17... too old! Feb 22 09:54:57 configure: error: *** The available kernel headers are older than the requested Feb 22 09:54:59 any clue/fix? Feb 22 09:55:35 arch is arm, distro is angstrom Feb 22 09:55:48 mickey|defence: hi! Not yet Dr. ? Feb 22 09:57:21 gena_: try building newer linux-libc-headers Feb 22 10:01:52 schurig: hi Feb 22 10:02:07 bb in 5 minutes Feb 22 10:05:03 do13_: thx, will try... Feb 22 10:14:18 re Feb 22 10:14:28 schurig: he has defence today Feb 22 10:15:16 do13_: looks like I found what's the problem: Feb 22 10:15:27 schurig: I just subscribed to libertas-dev recently. I spend most of time on working for OH now Feb 22 10:15:28 do13_: PREFERRED_VERSION_linux-libc-headers_i686 ?= "2.6.18" Feb 22 10:15:29 PREFERRED_VERSION_linux-libc-headers_powerpc ?= "2.6.18" Feb 22 10:15:29 PREFERRED_VERSION_linux-libc-headers ?= "2.6.15.99" Feb 22 10:15:50 hrw: OH ? Feb 22 10:16:14 this is in angstrom.conf, my arch is i586 and 2.6.18 is only installed when building for i686, do I see it right? Feb 22 10:16:40 gena_: yep Feb 22 10:16:41 openedhand Feb 22 10:16:52 do13_: thank you very much indeed! Feb 22 10:17:40 gena_: does it work now? Feb 22 10:18:34 schurig: OpenedHand Feb 22 10:19:19 do13_: will try in a minute Feb 22 10:21:08 do13_: need an RTFM about the difference between xxx = yyy and xxx =? yyy in *.conf files, could womeone give a link? Feb 22 10:24:12 gena_: ?= is 'set if not set' Feb 22 10:24:31 thx! Feb 22 10:34:34 03koen 07org.oe.dev * r2be33b1e... 10/ (1 conf/distro/angstrom-2007.1.conf): angstrom: add linux-libc-headers version for i586 and i486 Feb 22 10:47:26 do13_: 2.6.17 is last working tosa kernel? Feb 22 10:49:24 hrw: last with usb host Feb 22 10:49:51 ok Feb 22 10:50:50 do13__: I will pin 2.6.17 for tosa then Feb 22 10:51:47 hrw: ok Feb 22 10:52:19 * koen waits for that cset before building a tosa image Feb 22 10:56:21 koen: can you change OE snapshot to be 0.30 generated? Feb 22 10:57:05 OE.mtn you mean? Feb 22 10:57:15 yes Feb 22 10:57:28 I got another debian user which cant use snapshot Feb 22 10:57:32 You'd need to come up with a very good reason Feb 22 10:57:54 since the 0.32 static build works on debian Feb 22 10:58:04 koen: on powerpc? Feb 22 10:58:56 eh, no Feb 22 10:59:05 but people with 0.30 can download the 0.30 db Feb 22 10:59:10 that's why it's there, right? Feb 22 10:59:18 03hrw 07org.oe.dev * r6c1ed5cf... 10/ (1 conf/machine/tosa.conf): tosa: prefer 2.6.17 kernel as this is last with working USB Host support Feb 22 10:59:55 what I can do is update the 0.30 snapshot to the current head Feb 22 11:00:15 koen: sure. but how many docs already point to only one snapshot? how many people does not even know which version of monotone they have? Feb 22 11:00:46 how many lazy idiots are there in the world? Feb 22 11:00:54 we can't make everyone happy Feb 22 11:01:18 but we can decrease amount of noice from them with one 'mv' Feb 22 11:01:29 The canonical doc (GettingStarted) points to snapshots Feb 22 11:02:20 does any of the images generate a "minimum" image for X ? (i.e just loggin) Feb 22 11:02:21 hrw: an mv won't help Feb 22 11:02:35 every morning it takes a snapshot of the db the server uses Feb 22 11:02:43 which is running 0.32 on debian Feb 22 11:05:03 ok Feb 22 11:06:00 fsck. progear bios does not support keyboard conencted via hub;( Feb 22 11:15:13 hrw: http://www.angstrom-distribution.org/unstable/images/20070222/tosa/ Feb 22 11:16:04 thx Feb 22 11:16:19 koen: can you also build gpe-test? Feb 22 11:16:28 sure Feb 22 11:16:51 koen: where is the opie image? :) Feb 22 11:17:06 opie isn't supported Feb 22 11:17:34 it seems no-one wants to support it properly Feb 22 11:20:09 why, opie slowly progresses. I'm going to add config for 1.2.3-pre soon Feb 22 11:40:50 03koen 07org.oe.dev * rc3a4255c... 10/ (1 packages/gnome/libsoup_2.2.98.bb): libsoup: fix staging Feb 22 11:47:06 ~change 18 gbp to pln Feb 22 11:47:11 18.00 Pound (GBP) makes 103.795 Zloty (PLN) (from http://www.xe.com/) Feb 22 12:02:35 2002 ? Ss 0:00 /usr/bin/dbus-daemon --fork --print-pid 4 --print-address 6 --session Feb 22 12:02:36 root@progear:~$ dates Feb 22 12:02:37 ideas? Feb 22 12:02:40 (dates:2005): GConf-WARNING **: Client failed to connect to the D-BUS daemon: Feb 22 12:03:00 looks like dbus on my progear does not want to work at all Feb 22 12:04:17 hrw: you don't have the env var pointing to the session bus Feb 22 12:04:46 idea why? I just did /etc/init.d/dbus start Feb 22 12:04:55 that's the system bus Feb 22 12:05:21 what runs session bus? Feb 22 12:05:36 I cant get ANY login/session manager from OE working Feb 22 12:05:55 eval `/usr/bin/dbus-launch --auto-syntax --exit-with-session` Feb 22 12:06:20 now better Feb 22 12:06:50 now other part of system needs work Feb 22 12:07:44 hrw: if you don't mind being root you can use the o-hand 'login' scripts Feb 22 12:07:52 they are in OE as well Feb 22 12:08:05 03coredump2 07org.oe.dev * r792a595e... 10/ (3 files in 3 dirs): nfs-utils: Fix compile for kernels > 2.6.18 Feb 22 12:13:30 ...... Feb 22 12:13:48 I so know why I avoided .dev in the past Feb 22 12:14:08 because you suck? Feb 22 12:14:39 because .oz actually build and doesn't get fucked up every second day Feb 22 12:14:59 anybody remembers how to fix /home/slapin/openedhand/tmp/cross/lib/gcc/arm-angstrom-linux-gnueabi/4.1.1/../../../../arm-angstrom-linux-gnueabi/bin/ld: cannot find -lgcc_eh Feb 22 12:15:37 probably I miss some package in my installation? it is during build of glibc-intermediate-2.5 Feb 22 12:15:50 distro is angstrom Feb 22 12:17:15 * CoreDump|afk dunnos and goes off to work Feb 22 12:18:48 Rob Kramer just found a nice bug in the PACKAGE_INSTALL handling. Feb 22 12:18:50 psokolovsky_, koen: You should read the reply I just sent to the list... Feb 22 12:22:18 hrw: gpe-image uploaded Feb 22 12:22:53 * koen heads to the lab Feb 22 12:24:44 RP: I knew it's not that simple... ;-) I just wished that OE namespaces were clearly described in docs (they weren't when I read tehm in the beginning ;-) ) Feb 22 12:24:54 koen|away: thx Feb 22 12:26:32 RP: and other issue I see with image building is that RRECOMMENDS are always pulled in. I understand that's ipkg mostly, wonder if it's worth to add switch to tell ipkg to don't do that (if it has such) Feb 22 12:27:18 psokolovsky_: It pulls them in if they're there. You're describing RSUGGESTS behaviour... Feb 22 12:28:07 RP: yes, but if you build recipes, they are *always* there. what if I want to build minimal image (initrd, etc)? Feb 22 12:28:42 psokolovsky_: I know what you mean, it is an ipkg issue really though Feb 22 12:28:49 yep Feb 22 12:29:06 psokolovsky_: The lack of docs is mainly due to the namespaces being properly defined only more recently Feb 22 12:29:11 ah, hi paul :) just responded to your mail Feb 22 12:29:27 psokolovsky_: You probably don't remember the bad old days before some of my bitbake changes... Feb 22 12:30:23 RP: only hazy ;-) Feb 22 12:40:11 ~change 1200 pln to eur Feb 22 12:40:16 1,200.00 Zloty (PLN) makes 310.100 Euro (EUR) (from http://www.xe.com/) Feb 22 13:00:51 hi Cliff Feb 22 13:01:00 RP: libraries in PACKAGE_INSTALL haven't been a problem since sane people don't ship unused libs.... Feb 22 13:10:06 hrw: hello Feb 22 13:11:10 fsck.. I have a tool to calibrate touchscreen in progear but it use ncurses. and vesafb does not give any output ;( Feb 22 13:14:29 koen|uni: It should just work even if the user wants to do something insane though... Feb 22 13:15:21 hrw: okay, firmware downloading now works Feb 22 13:15:47 it's very different on CS as opposed on USB, so I could not re-use much of OLPC libertas code Feb 22 13:15:58 schurig: you extracted firmware from windows driver? Feb 22 13:16:17 ~seen florian Feb 22 13:16:20 no, from the wifistix binary only linux module Feb 22 13:16:46 florian was last seen on IRC in channel #openezx, 13h 30m 13s ago, saying: 'good night'. Feb 22 13:16:52 but I guess extracting it from windows CE driver would be possible as well Feb 22 13:17:00 schurig: ah. good - they used 5.0.16/17 driver from marvell so it is quite fresh Feb 22 13:17:40 schurig: prplaggue had problems with firmwares from other cards iirc Feb 22 13:17:44 but I do not remember details Feb 22 13:18:34 hrw: did you point prplaggue to the linuxtogo project? I've just found out yesterday via google search that he is doing something like me Feb 22 13:19:48 hrw: the OLPC libertas code is written for Firmware v5.1, and by accident someone from Marvell sent in a PDF with the firmware spec Feb 22 13:20:07 schurig: he is in project Feb 22 13:20:34 schurig: David Anders == prplaggue Feb 22 13:21:16 hrw: once I got it all running I need to pepper the OLPC code with "is firmware = V50 then this else that". The OLPC people agreed to that approach. Feb 22 13:22:10 hihi all. I am kindly sorry but can someone please explain me how to make MTN understand that I am behind proxy ? Feb 22 13:23:52 kaafree: google for "monotone proxy" :-) Feb 22 13:24:52 schurig, did it but couldn't find anything for brain-paralized like me. Can you give a couple of links please ? Feb 22 13:25:20 hrw: having the FW spec is quite nice. However, I'd would be even nice if I would have a complete doc about the registers and the bits :-) Feb 22 13:25:40 schurig: sure Feb 22 13:26:12 ~lart frontpath for using siliconmotion gfx chip Feb 22 13:26:12 * ibot stuffs frontpath into a shiny new tin can and vacuum seals it for using siliconmotion gfx chip Feb 22 13:27:05 hehe Feb 22 13:30:32 how to tell X that there is no core pointer? Feb 22 13:31:14 hrw: use the void input driver Feb 22 13:32:12 to have nice touchscreen under X I need to start X Feb 22 13:32:26 Section "InputDevice" \n Driver "void" \n Option "CorePointer" \n EndSection Feb 22 13:33:19 doing it now (had to build driver first) Feb 22 13:33:33 hrw: I have a system where X doesn't know about touchscreen and mouse (because the evdev sucks in my case). So I load X this way and handle the touchscreen from userspace via XTest extension Feb 22 13:38:55 fsck. touchcal does not want to talk with touchscreen Feb 22 13:43:34 koen|uni: http://www.fsf.org/blogs/community/iphone <- quite funny Feb 22 13:53:58 hrw : have you tried to build lirc-modules for 2.6.20 ? Feb 22 13:54:07 no Feb 22 13:54:47 i think we have another 'victim' of the 2.6 20 api change Feb 22 14:00:41 building.. Feb 22 14:01:46 | make: Entering directory `/home/hrw/devel/build/progear/tmp/work/progear-angstrom-linux/lirc-modules-0.8.1-r0/lirc-0.8.1/drivers' Feb 22 14:01:49 | make: *** No targets. Stop. Feb 22 14:06:37 hrw: i've done some updates for openembedded-essential Feb 22 14:07:09 where the heck is "Brussels Midi Railway station" Feb 22 14:08:34 hrw: http://rafb.net/p/yqVGTn88.html Feb 22 14:08:42 interested? Feb 22 14:09:11 hrw : same here. it outputs these messages during config -> http://rafb.net/p/NWhUbL74.html Feb 22 14:10:51 robtaylor: I do not like postinst part Feb 22 14:10:59 hrw: why not? Feb 22 14:12:09 robtaylor: it has bad and good side. good one is that it remove problems with OE. bad one is that it change behaviuor of few apps in system Feb 22 14:12:56 hrw: well,only if you've got GNU Interactive Tools installed Feb 22 14:13:07 robtaylor: want to takeover that package? I will be glad to give it to someone else Feb 22 14:13:21 hrw: it could be done with a Conflicts,i guess Feb 22 14:13:32 hrw: sure, don't mind at all Feb 22 14:13:34 and bash vs dash Feb 22 14:13:49 fwiw, o-hand has removed a lot of bashisms from their tree Feb 22 14:13:57 might be worth merging into OE Feb 22 14:13:58 yeah Feb 22 14:14:03 koen|uni: contra zsh, tcsh, pdksh probably too Feb 22 14:14:08 its good to remove bashism Feb 22 14:14:32 i'd prefer if evything could work with dash,tbh Feb 22 14:14:45 see http://www.openembedded.org/bonsai/view/scm/6/ Feb 22 14:14:50 though it isn't exactly bug-free Feb 22 14:16:00 why are ohand doing their own version of oe, ooi? Feb 22 14:17:04 robtaylor: No company in their right mind would offer to support .dev ;-) Feb 22 14:17:24 robtaylor: Poky contain very limited subset of packages Feb 22 14:17:33 ah, i see :) Feb 22 14:17:57 robtaylor: it also support only one distro (poky) and only few machines. and works out-of-box Feb 22 14:19:09 03ifaistos 07org.oe.dev * rbee9b207... 10/ (3 files in 2 dirs): Feb 22 14:19:09 packages/memtester: Fix download URL for memtester_4.0.5.bb Feb 22 14:19:09 Add memtester_4.0.6.bb Feb 22 14:19:13 03ifaistos 07org.oe.dev * rcf825d08... 10/ (3 files in 2 dirs): packages/modphp/modphp_5.1.6.bb: modphp requires apxs from apache2 to build. Point to the correct dir where apxs is Feb 22 14:19:44 RP: unless that company employs 10 OE core devs ;) Feb 22 14:21:29 RP: do you guys push up to .dev pretty regularly? Feb 22 14:21:44 koen|uni: Maybe when we get to 10... :). Seriously, even then to have something stable/supported, you can only have a certain number of trusted committers. I trust the OE.dev people but you can't base something commercial off it... Feb 22 14:22:11 robtaylor: The intent is there, it doesn't happen as often as I'd like but it does happen Feb 22 14:22:32 RP: herh, i assumed the intent was there.. ;) Feb 22 14:26:08 ~lart svn Feb 22 14:26:09 * ibot shoots svn in the head Feb 22 14:26:31 ~lart monotone Feb 22 14:26:32 * ibot slaps monotone upside the head with a wet fish Feb 22 14:26:39 ~lart lart Feb 22 14:26:39 * ibot throws a AN/M-8 smoke grenade at lart Feb 22 14:33:34 is this a known issue ? http://pastebin.ca/367710 Feb 22 14:33:58 morning Feb 22 14:34:10 vlo: heh, i've just been talking about that with koen|uni Feb 22 14:34:46 koen|uni: did building with export PKG_CONFIG_PATH = "${STAGING_LIBDIR}/pkgconfig" work? Feb 22 14:35:16 robtaylor: and any idea of what's wrong here ? Feb 22 14:35:59 vlo: 12:35 < koen> robtaylor: the problem is insanity checks all .pc files, which in the the sanitized ones in share/pkgconfig and the broken ones in lib/pkgconfig Feb 22 14:37:04 vlo: 12:18 < koen> robtaylor: you can get around it by removing "insane" from INHERITS in conf/distro/includes/angstrom.inc Feb 22 14:38:37 hrw: madwifi-ng-1871 also fails for x86 and 2.6.20. there is a new version in their cvs Feb 22 14:40:25 ~me wonders if switching to 2.6.20 was a good thing... Feb 22 14:40:27 moi? Feb 22 14:40:27 robtaylor: thank you for the info Feb 22 14:41:44 koen|uni: sure, we can disable the check Feb 22 14:42:42 * zecke prints stickers with "Mark Gollahon can't read" and places them all over Berlin Feb 22 14:45:12 Jin^eLD: Morning :) Feb 22 14:45:22 hey hvontres|poodle :) Feb 22 14:45:57 Jin^eLD: Well, I never got a chance to look at that glibc stuff last night. Maybe sometime next week :( Feb 22 14:46:05 me neither, maybe this weekend Feb 22 14:46:57 which glibc stuff? Feb 22 14:47:34 zecke: on FC6 binary locale generation fails, qemu segfaults Feb 22 14:47:49 but hvontres|poodle figured its somehow related to incorrect paths/params being passed to quemu Feb 22 14:47:51 something like that Feb 22 14:49:38 is there anyy RFC RTFM on how to set up feeds for OE based distro's ? Feb 22 14:49:40 incorrect path can be, every os.path.join in glibc-package.? needs thinking Feb 22 14:50:07 Ifaistos: do it like you want Feb 22 14:50:14 hmm, pkgconfig has changed its tarball name to pkg-config Feb 22 14:51:15 zecke: it looks like somehow datadir gets set to /usr/share/, which cloobers treedir in path.join Feb 22 14:53:10 should i make the new bbs pkg-config, or can i set something to tell it the right unpack directory? Feb 22 14:59:15 hvontres|poodle: data = bb.getVar('${datadir}' and it starts with '/' and os.path.join is doing the rest Feb 22 14:59:34 see doku's example of os.path.join in the bug tracker Feb 22 15:01:03 CIA-3: speak Feb 22 15:06:28 03hrw 07org.oe.dev * r3fbc54a6... 10/ (1 classes/package.bbclass classes/rm_work.bbclass): classes: package, rm_work - fix bashisms (from Poky) Feb 22 15:06:33 03hrw 07org.oe.dev * re00e94a7... 10/ (4 files in 4 dirs): dropbear, sysvinit, linux-nokia770: fix bashisms (from Poky) Feb 22 15:06:37 03hrw 07org.oe.dev * r0e0265da... 10/ (5 files in 3 dirs): contacts: update to 0.3, drop older (from poky) Feb 22 15:06:41 03hrw 07org.oe.dev * r376e54b5... 10/ (1 packages/dates/dates_0.1.bb packages/dates/dates_0.2.bb): dates: dropped 0.1 and 0.2 Feb 22 15:06:45 03hrw 07org.oe.dev * r518bf00c... 10/ (1 packages/xorg-util/util-macros_X11R7.1-1.1.5.bb): util-macros: added 1.1.5 from poky Feb 22 15:06:49 03hrw 07org.oe.dev * reb521106... 10/ (3 files in 3 dirs): util-macros: added patch to get it cross compiled (from poky) Feb 22 15:06:55 03ifaistos 07org.oe.dev * r59341f60... 10/ (1 packages/nmap/nmap_3.81.bb): packages/nmap/nmap_3.81.bb: Change URI to point to new download location Feb 22 15:06:57 03hrw 07org.oe.dev * rd93cbba5... 10/ (1 packages/xorg-proto/inputproto_X11R7.1-1.4.bb): inputproto: added 1.4 from poky Feb 22 15:07:01 03hrw 07org.oe.dev * r68afdf59... 10/ (1 packages/xorg-proto/xproto_X11R7.1-7.0.8.bb): xproto: added 7.0.8 from poky Feb 22 15:07:05 03hrw 07org.oe.dev * ra1cb2529... 10/ (1 conf/bitbake.conf): bitbake.conf: add ext3 images support from poky Feb 22 15:07:39 hrw: ext4 support is coming as well? Feb 22 15:07:44 is uclib also faster or just smaller ? Feb 22 15:08:22 that's an interesting question Feb 22 15:08:27 zecke: when ext5 will be Feb 22 15:09:11 rob_w: for humans it is faster to read all uclibc code than to read everything inside the glibc Feb 22 15:09:46 03hrw 07org.oe.dev * rd73a99d1... 10/ (1 packages/dhcp/dhcp_3.0.2.bb): dhcp: Fix dbg packaging and clean up hardcoded paths (from poky) Feb 22 15:09:50 03hrw 07org.oe.dev * r3eb35e8e... 10/ (1 packages/zaurus-updater/encdec-updater.bb): zaurus-updater: Fix typo in description (from poky) Feb 22 15:10:01 zecke, ah seems so Feb 22 15:11:51 zecke: I know how the path gets clobbered, but what has me confused is why is this breaking now and why '${datadir}' has a leading / Feb 22 15:13:28 gi hrw :) Feb 22 15:13:38 *go, i mean :) Feb 22 15:14:24 koen|uni: any advice on my problem above? Feb 22 15:15:41 robtaylor: I'm afraid I missed the problem Feb 22 15:15:57 * koen|uni is is running windoze full screen using virtualization Feb 22 15:16:10 * koen|uni stabs matlab on MacIntel Feb 22 15:16:23 koen|uni: oh, its ok, i just need to set S Feb 22 15:16:29 koen|uni: using vmware, Xen, KVM? Feb 22 15:16:44 koen|uni: i was justhaving some problems with pkgconfig being renamed to pkg-config upstream Feb 22 15:16:55 hvontres|poodle: because it has an "/" (${prefix}) Feb 22 15:17:57 VoodooZ: Parallels desktop Feb 22 15:18:18 koen|uni: i'm a little rusty, been too long since i last wrote a bb :) Feb 22 15:18:31 hi zecke Feb 22 15:19:07 pb_: hey master, will you make it to the FOSDEM? Feb 22 15:19:40 zecke: barring unforeseen circumstances, yes. Feb 22 15:19:58 my train ticket is booked, so I need only to survive the next 24 hours without debilitating injury. Feb 22 15:20:25 * XorA packages pb_ in bubble wrap Feb 22 15:20:27 heh Feb 22 15:20:42 * robtaylor sends the local goons round Feb 22 15:21:10 robtaylor: attending fosdem? Feb 22 15:21:22 my memory is emulating a sieve quite well nowadays Feb 22 15:21:25 koen|uni: yep :) in the samee hotel too Feb 22 15:21:30 cool Feb 22 15:21:38 need.... more... coffee... Feb 22 15:21:47 koen|uni: have we actually ever met in person before? Feb 22 15:21:50 koen|uni: any special sieve? Feb 22 15:21:54 When do people arrive tomorrow? I arrive quite early on... Feb 22 15:22:09 RP: i'll be there round 3ish Feb 22 15:22:22 RP: IIRC I will arrive at three o'clock Feb 22 15:22:22 I expect Ill arrive about 17:00 I land at Charloroi at 14:35 Feb 22 15:22:37 * RP lands at 10:35am Feb 22 15:22:55 an, i land at 16:07 actually Feb 22 15:23:07 * XorA is going by Ryanair so expect to find me flapping my arms Feb 22 15:23:16 RP: that sounds like waking up far too early Feb 22 15:23:46 robtaylor: No other choice sadly. Flight options were awful this year... Feb 22 15:23:56 RP: eurostar ftw Feb 22 15:24:00 * koen|uni will arrive ~18.00 at the central station Feb 22 15:24:15 * XorA has to find his way from Midi station to Astrid Feb 22 15:24:16 RP: you get to go to the bar and drink good beer as well Feb 22 15:24:59 18:25 @BRU Feb 22 15:30:23 XorA: that's pretty easy, you can ride the tram from gare du midi to the bourse (or de brouckere) and then walk to the Astrid. Feb 22 15:31:11 pb_: wow, you just told me what 2 days searching took to find :-) they dont like tourists in brussels it seems Feb 22 15:32:15 heh Feb 22 15:32:37 you should have asked in #oe first :-) Feb 22 15:33:11 XorA : It's reserved for EU staff basically :) Feb 22 15:33:17 hi all Feb 22 15:34:02 damn Ryanair flying to middle of nowwhere doesnt help Feb 22 15:34:09 koen|uni: have you ever tried scilab? Feb 22 15:36:47 03pfalcon 07org.oe.dev * r78dff4af... 10/ (3 files in 2 dirs): Feb 22 15:36:47 alsa-lib: Fix packaging of alsa.m4 Feb 22 15:36:47 * TODO: alsa-conf packages lots of unneeded stuff (like bunch of card definitions, Feb 22 15:36:47 not needed on handheld). Figure out and throw away. Feb 22 15:39:40 psokolovsky_: I will fix your todo then ;D Feb 22 15:40:20 hrw, please do. I just shy to leave alsa.conf only w/o comprehensive testing ;-) Feb 22 15:41:41 psokolovsky_: alsa-conf, alsa-conf-base Feb 22 15:41:52 or -extra Feb 22 15:41:56 pb_: Any idea offhand how to get from the main airport to the Astrid? :) Feb 22 15:42:18 psokolovsky_: conf is full, base is limited Feb 22 15:42:28 ok Feb 22 15:44:16 RP: you can take a train from the airport to Central or Midi, then the tram. There might be a direct bus link from the airport to somewhere near De Brouckere as well but I'm not sure. Feb 22 15:45:25 pb_: handy to know, thanks. Looking that up was on this afternoon's todo list :) Feb 22 15:46:14 * koen heads home Feb 22 15:48:52 we should all take pb_ mobile number and he can talk us all in :-) Feb 22 15:49:08 3-way merges... thats the moment when I like monotone and hate subversion Feb 22 15:51:03 psokolovsky_: pushed to ewi Feb 22 15:51:30 hrw, thanks! Feb 22 15:52:36 will have to push it to poky too Feb 22 16:06:40 zecke: do you think I should wait with the pkgconfig change? Feb 22 16:07:20 koen: no, I just wanted to avoid the forever.... :) Feb 22 16:07:22 :) Feb 22 16:07:23 ok Feb 22 16:07:36 koen: feel free to do it, feel free to disable the check inside the insane.bbclass Feb 22 16:07:51 * zecke wrestles with precompiled headers on Windows CE :} Feb 22 16:08:10 koen: Windows CE sucks bad ass, but it has a slow emulator :) Feb 22 16:09:23 03hrw 07org.oe.dev * rdcb9bef3... 10/ (3 files in 2 dirs): alsa-lib: ship needed configs in alsa-conf-base and rest of them in alsa-conf (backward compatibility), also added dependency on configs to library Feb 22 16:09:30 03hrw 07org.oe.dev * r7cd97172... 10/ (1 packages/tasks/task-base.bb): task-base: do not ship alsa-conf as this will be added as rdependency of alsa-lib Feb 22 16:10:39 03koen 07org.oe.dev * r75bd0ba9... 10/ (1 conf/bitbake.conf): Feb 22 16:10:39 bitbake.conf: fix PKG_CONFIG_PATH as per RFC on the mailinglist Feb 22 16:10:39 * rebuild all your pkgconfig using packages Feb 22 16:10:39 AND/OR Feb 22 16:10:40 * cd ; mv share/pkgconfig/* lib/pkgconfig Feb 22 16:15:22 I wanted to get soem feedback on this: http://www.pastebin.ca/367757 Feb 22 16:15:52 Seems that MakeMaker isn't accepting the argument --host, or isn't parsing it or something Feb 22 16:16:26 http://img67.imageshack.us/img67/8276/veryfreshgreens1us3.jpg Feb 22 16:17:02 hrw: yummies Feb 22 16:17:32 mmmm Feb 22 16:17:41 * hillct is ghetting hungry alreadyu Feb 22 16:18:19 * hrw got pizza hour ago Feb 22 16:18:20 a frog for free, without extra cost? Feb 22 16:19:20 Protein Fortified Feb 22 16:20:17 03koen 07org.oe.dev * r826367bb... 10/ (1 classes/insane.bbclass): insane.bbclass: only check for workdir in .pc files, we already forcefully sed out /usr/include references Feb 22 16:20:47 hrw: hehe...i wonder it the frog is "fresh and washed" too Feb 22 16:21:02 cu - have to pack for fosdem Feb 22 16:21:37 ok see everyone tomorrow somewhere in Brussels :-) Feb 22 16:21:53 hillct: your analisys seems correct Feb 22 16:21:56 ~spell analisys Feb 22 16:21:58 possible spellings for analisys: Analise's analysis annalists Analiese's Anallise's Annalise's analysts analysis's Analise annalist's Annaliese's analyst's Anallese's Annelise's Feb 22 16:22:17 analysis Feb 22 16:24:58 koen: any thoughts on why that might occur? Feb 22 16:25:37 hillct: broken configure.ac, broken m4 files, full moon Feb 22 16:25:44 hillct: it's hard to pinpoint :( Feb 22 16:25:52 I haven't yet dug into the configure script. that's my next move. Just thougt someone might have seen this before with other CPAN builds Feb 22 16:42:16 heh Feb 22 16:42:23 buildroot is applying patches from OE now :) Feb 22 16:43:58 it seems our unbreak-armv4t patch is da bomb Feb 22 16:46:24 ade|desk did you compile the last uclibc svn ? Feb 22 16:46:45 since the last update i think its broken Feb 22 16:46:49 at least for me Feb 22 16:48:05 saladino: not lately, svn broken on includes Feb 22 16:48:15 koen: nice to be able to give back :) Feb 22 16:48:53 undefined reference to `_errno' Feb 22 16:48:57 yep Feb 22 16:49:17 saladino: best to track http://uclibc.org:8010/ Feb 22 16:50:02 the last commit is at 13:00, but i dont know when is that at my local timezone, xD Feb 22 16:50:15 ade|desk: .29 after fosdem! Feb 22 16:53:33 koen: yeah i know, i pointed that out on here a few days ago ;) Feb 22 16:54:05 ade|desk: what do you think of the idea to get .29 + arm/mips ntpl patch into OE? Feb 22 16:55:14 what sjhill+cs patch sets ? Feb 22 16:55:21 hmmm maybe Feb 22 16:56:02 it would give use the ueber-uclibc Feb 22 16:57:25 but a non compatible uclibc with any future uclibc revision Feb 22 16:58:22 i know there is not any compat between any release, but there has been a mainly stable abi for a while, (2 years or more) Feb 22 16:59:30 after .29 release svn will have the patches merged in head soon anyway (i think) Feb 22 16:59:49 could be fun to play with though Feb 22 17:02:19 * koen has a feeling uclibc is stil a toy Feb 22 17:07:15 any news on mickey|defence ? Feb 22 17:08:33 in package edje-native, there is binary file called edje-config, and later bb it fails on not finding edje-config-native - it's probably the same file and BUG in OE, am I right? Feb 22 17:08:53 koen: ok pkgconfig 0.21 is working now Feb 22 17:09:40 koen: buzilla? commit to mtn? Feb 22 17:10:15 bugzilla please Feb 22 17:12:14 hrw|gone: xserver-nodm-init might do the job for you Feb 22 17:15:59 JustinP: using pulseaudio in mpd is making trouble over here, it is unusable slow on my shark Feb 22 17:16:20 zecke: did enable the fixedpoint resampler? Feb 22 17:16:47 ~seen doku Feb 22 17:17:09 koen: whatever is default, will forward the question to my sith lord doku Feb 22 17:17:26 zecke: /etc/pulse/daemon.conf -> search for resample-method, uncomment, set to 'trivial' Feb 22 17:17:46 doku was last seen on IRC in channel #oe, 1d 20m 6s ago, saying: 'yep, that's what i'll do. but i have to stop for today.... bye'. Feb 22 17:17:56 zecke: I still have't found a clean way to automagically do that for TARGET_FPU = "soft" Feb 22 17:19:21 koen: hmm a postinst script that will the daemon.conf? Feb 22 17:19:39 DoKu: you are supposed to work?! Feb 22 17:20:04 zecke: something like that Feb 22 17:20:47 do_install_append() { if [ x${TARGET_FPU} == xsoft] ; then foo} ? Feb 22 17:21:23 koen: alternatively write a daemon.conf.in and let autoheader fix it/generate it Feb 22 17:27:27 Actually, I do work ;) Feb 22 17:28:12 DoKu: prepare for my hurricane Feb 22 17:28:25 does someone have a recommendation for a VERY minimal recipe? Feb 22 17:28:34 just busybox, basically. Feb 22 17:30:11 and/or how do I have bitbake generate .tgz files instead of .ipk files? Feb 22 17:30:22 and yes, I know I can extract the actual files from the .ipk Feb 22 17:30:42 elliotf__: RTFM? or FTFM? Feb 22 17:31:13 FFT? Feb 22 17:31:18 FTFM would be nice.. Feb 22 17:31:25 TFM stops at "setting up OE" Feb 22 17:32:08 zecke: point me to documentation that you would suggest. the bulk of the tutorials/docs are "how to set up" Feb 22 17:32:11 rather than "how to use" Feb 22 17:33:05 elliotf__: http://www.openembedded.org/user-manual I bet something might be there (and the feel free to make a difference and send a patch) Feb 22 17:33:53 I'll look over it again, but after reading through it I had the feeling of "well, what next?" Feb 22 17:34:04 it was very good for setting up OE, but kind of stops. Feb 22 17:34:28 INHERIT = "package_tar" Feb 22 17:35:02 thank you for the hand holding, I appreciate it. Feb 22 17:35:12 I'll RTFM now. Feb 22 17:35:15 elliotf__: yes, what's next totally depends on you. What do you want to do? add packages, add new package formats, new rootfs format :) Feb 22 17:35:36 :) Feb 22 17:35:41 elliotf__: I'm not sure that package formats are described in the manual, they should be there (so FTFM FixThe...) Feb 22 17:35:57 ah, I thought FTFM = "Find the.." Feb 22 17:35:58 :) Feb 22 17:36:10 thanks again. Feb 22 17:36:58 lrg: XorA|gone : are you coming to the LAC this year? Feb 22 17:38:52 zecke: i should be :) Feb 22 17:39:07 zecke: are you going ? Feb 22 17:39:37 lrg: no, but I'm a local. So make sure to drop by and get some tea+biscuits ;) Feb 22 17:39:58 zecke: cool, will do, thanks :-D Feb 22 17:45:38 zecke: http://rafb.net/p/Dj2XSE70.html ? Feb 22 17:48:23 koen: http://bugs.openembedded.org/show_bug.cgi?id=1903 Feb 22 17:48:35 robtaylor: cool, thanks? Feb 22 17:49:08 koen: why would you want to do that? where should TARGET_FPU be read? Feb 22 17:49:08 s/?// Feb 22 17:49:09 koen: i might try sort out some of these pkg-config issues now ;) Feb 22 17:49:12 heh Feb 22 17:49:46 zecke: http://rafb.net/p/9Iduq651.html Feb 22 17:50:31 koen: are you sure this export is needed Feb 22 17:50:40 it doesn't work without :( Feb 22 17:53:55 koen: you can export it localy, or embed python code to get this var Feb 22 17:54:02 koen: have they not released the new sampling lib yet? Feb 22 17:57:52 robtaylor: I'm going to wait a bit with adding pkgconfig, I don't want to accidentilly break stuff just before going away to FOSDEM Feb 22 17:59:16 koen: sure, i'll just work in my local tree for now Feb 22 18:06:47 zecke: ok, I export it locally now Feb 22 18:06:49 03koen 07org.oe.dev * rde732b9e... 10/ (1 packages/pulseaudio/pulseaudio_0.9.5.bb): pulseaudio: use integer resampling on FPU-less devices Feb 22 18:07:37 zecke: I never claimed that it worked :-P I never even got pulse to run correctly myself Feb 22 18:07:53 zecke: so who is "we"? (we have decided to fork mpd) Feb 22 18:09:33 Hello Feb 22 18:09:42 hey sirfred Feb 22 18:09:45 JustinP: DoKu and me, we have not done much in regard to forking though Feb 22 18:14:47 * koen hands zecke and DoKu a spoon Feb 22 18:15:28 DoKu: clockwork orange + alcohol? or fahrenheit 451 + alcohol? Feb 22 18:15:48 fahrenheit orange + alcohol? Feb 22 18:16:06 koen: fahrenheit alcohol + 451*organge Feb 22 18:19:02 zecke: clockwork orange + THC ? Feb 22 18:19:34 hvontres|poodle: hehehe Feb 22 18:22:10 hi all Feb 22 18:23:08 psokolovsky_, bug 1904, any additional info is needed? Feb 22 18:33:40 I added bug 1869 some days ago, is there any problem with it? Feb 22 18:39:25 OMG,i hadn't realised how many packages have fucked up pkgconfig Feb 22 18:39:40 /usr/lib/pkgconfig/gnome-system-tools.pc:backenddir=@scriptsdir@ Feb 22 18:39:42 nie Feb 22 18:40:03 /usr/lib/pkgconfig/xft.pc:freetypecflags="-I/usr/include/freetype2 " Feb 22 18:40:06 grag Feb 22 18:40:11 (these are installed) Feb 22 18:45:21 robtaylor: see what I meant with "It's not that simple"? Feb 22 18:45:31 koen: yeah :) Feb 22 18:45:33 robtaylor: crosscompiling makes people sad Feb 22 18:45:50 koen: strangely only mono people get it right Feb 22 18:45:54 go figure =) Feb 22 18:47:02 heh Feb 22 18:48:59 Hello all Feb 22 18:50:09 HopsNBarley: ping Feb 22 18:51:08 zecke: and DoKu is? why exactly are you forking? are the devs not responding? Feb 22 18:52:04 morning likewise Feb 22 18:52:09 morning NAiL Feb 22 18:52:19 JustinP: we did not fork yet, the code is disgusting and makes me cry :) Feb 22 18:52:31 hi likewise, been a bit, how are ya? Feb 22 18:52:41 JustinP: and it is unlikely that we will really fork Feb 22 18:52:47 zecke: so your saying it's all forked up then? Feb 22 18:53:16 * hvontres|poodle is an incorrigible punster...do not incorrage Feb 22 18:53:31 JustinP: doku so far has cleaned up some autoconf stuff, and has a set of patches he is too shy to push to OE Feb 22 18:54:35 off to watch some movies Feb 22 18:54:58 JustinP: not forked, patches were sent upstream, positive reply came back :) Feb 22 18:55:07 HopsNBarley: yeah been busy with non-dev stuff lately. I am in Silicon Valley for 12 days from next tuesday Feb 22 18:55:32 HopsNBarley: you? how's s/w and h/w coming along? Feb 22 18:56:41 hey fantastic! Lets meet! where will you be in SV? Feb 22 18:57:04 HopsNBarley: Sunnyvale. I planned on going to SF during the weekend. Feb 22 18:58:51 hi all ! Feb 22 18:58:51 likewise: report nice stuff you see, I'll be in santa clara in ~8 weeks from now :) Feb 22 18:58:58 hey gremlin[it] Feb 22 18:59:18 koen: damn, we could have had a OE meeting if we timed up properly :-) Feb 22 18:59:28 likewise: blame CELF ;) Feb 22 19:00:08 koen: ooh, youre heading to ELC? Feb 22 19:00:19 robtaylor: that's the plan Feb 22 19:00:35 koen: cool, i'll probably be there too Feb 22 19:01:02 robtaylor: am I stalking you, or are you stalking me? Feb 22 19:01:15 koen: we just don't know Feb 22 19:01:19 =) Feb 22 19:03:59 hi psokolovsky ! Feb 22 19:04:47 gremlin[it], hi! Feb 22 19:21:08 psokolovsky i still continue haveig a big image ;( ... next week i try to do a complete clean and rebuild from scratch :( Feb 22 19:21:50 gremlin[it], current size I have is again back to 16Mb. I'm working to cut it down again. Feb 22 19:22:10 mine is still 17+Mbyte :( Feb 22 19:24:13 psokolovsky: I have a few ideas on how to get it smaller, but I won't have much time next week to implement those Feb 22 19:25:27 koen u'll tell me and i'll do :) Feb 22 19:25:37 koen, nice, would need to discuss them a bit later. in the meantime, I'm working to add @base_less_or_equal() Feb 22 19:45:11 hmmm Feb 22 19:45:26 * koen tries to find some spare extension cords Feb 22 19:46:04 I get this feeling we'll have a socket shortage at the OE booth Feb 22 19:46:13 re Feb 22 19:46:19 hrw: wb Feb 22 19:46:23 koen: get some if you can - I lack such stuff Feb 22 19:46:39 trying to do a new distro (minimal x86) and I get an following error when trying to do `bitbake bootstrap-image` that says 'Cannot find package task-base.' Feb 22 19:47:04 argh.. 'bitbake task-gpe' does not build stuff Feb 22 19:47:16 I try bitbake task-base, and nothing is attempted to be built. Feb 22 19:47:23 elliotf__: x86 machine? Feb 22 19:47:35 hrw: yes Feb 22 19:47:52 elliotf__: how old metadata? and which distro? Feb 22 19:48:06 hrw: trying to define a new distro (for testing and use as a flash-based distro just for iptables/iproute2) Feb 22 19:48:22 hrw: updated oe an hour or so ago.. should I try to backrev? Feb 22 19:48:31 nope. it should work Feb 22 19:48:47 hrw: some logs I should/could look at for more info? Feb 22 19:49:00 * elliotf__ is learning oe Feb 22 19:49:28 elliotf__: MACHINE="x86" DISTRO="generic"? Feb 22 19:49:45 hrw: DISTRO="elliotf-test01" Feb 22 19:50:08 I'll try the generic and see how that goes. Feb 22 19:50:34 elliotf__: try 'angstrom-2007.1' as this works for sure Feb 22 19:50:45 hrw: angstrom is x86? Feb 22 19:50:45 elliotf__: I'm running angstrom on progear (x86 webpad) Feb 22 19:50:50 ah. Feb 22 19:50:50 k. Feb 22 19:50:52 elliotf__: angstrom is anything Feb 22 19:50:56 I'll give it a go. Feb 22 19:51:27 Hello, I want to build a kernel+rootfs for my hx4700. I am about to pull a local copy of the oe db. should I use the defaults specified in the GettingStarted Wiki? Feb 22 19:51:51 justin__: MACHINE="hx4700" DISTRO="angstrom-2007.1" Feb 22 19:59:28 are bugs 1894 and 1895 setup properly? Should they be assigned to someone in particular or left as new? Feb 22 20:02:16 Hello Feb 22 20:02:24 I'd like to ask a question Feb 22 20:02:29 ... Feb 22 20:02:41 B_Lizzard: I'd like to ask a question too Feb 22 20:02:51 Is it personal? Feb 22 20:03:06 you do not like personal questions? Feb 22 20:03:17 Maybe a question of sexual prefference? Feb 22 20:03:33 Is that a personal question you just made? Feb 22 20:04:02 ok. so I asked one question. you asked 3 but wanted 1. so you reached limit. Feb 22 20:04:03 ;D Feb 22 20:04:12 Game Over Feb 22 20:04:14 :D Feb 22 20:04:49 So, I was wondering if I can refer to "Packages" files through another "Packages" file Feb 22 20:04:59 no Feb 22 20:05:47 Say I've made a feed where I've made alphabetical folders, packages inside, and the "Packages" file split for each letter, do I still need the big central one? Feb 22 20:06:04 Which refers to all packages? Feb 22 20:06:21 no need - you just got lot of feeds instead Feb 22 20:06:30 I was thinking I could improve performance by splitting the Packages file... Feb 22 20:06:40 Hmmmm Feb 22 20:06:47 B_Lizzard: ipkg read all anyway at once Feb 22 20:06:57 Aha.... Feb 22 20:07:18 So, no performance gain, eh.... Feb 22 20:07:28 Thanks for the quick answers, hrw Feb 22 20:07:33 np Feb 22 20:07:37 They're very much appreciated Feb 22 20:10:41 gpe-dm on XGA looks weird Feb 22 20:11:34 YEAH! gpe on progear! Feb 22 20:11:37 hrw: angstrom fails to build, complaining that 'kernel 2.6.17 is too old' during package glibc-intermediate-2.5-r4: task do_configure: failed Feb 22 20:11:50 elliotf__: update metadata Feb 22 20:11:55 mokay. Feb 22 20:12:03 as in, update oe (via mtn)? Feb 22 20:12:14 yes Feb 22 20:12:26 already up to date Feb 22 20:13:16 elliotf__: mtn pull;mtn update Feb 22 20:13:19 yeah, doing a pull now. Feb 22 20:14:26 I'm also building a generic x86 task-base Feb 22 20:17:24 is anyone else having trouble building the current e2fsprogs-native? Something is pulling it in after my latest mtn update and it won't build because it can't find fsck Feb 22 20:18:09 hmm thats probably be me Feb 22 20:18:24 I have an -r6 and -r7 that built in the past, but -r8 fails Feb 22 20:18:49 and the ViewMTN repository is locked again so I can't see what changed Feb 22 20:21:16 mtn log --no-merges | more ? Feb 22 20:24:09 Koen: Hoe ga jij naar de fosdem ? Feb 22 20:24:30 gerwinin: trein Feb 22 20:30:47 koen: can you upload to scap.ltg? Feb 22 20:31:06 hrw: anyone can with gpe-scap Feb 22 20:31:12 dunno how to do it by hand Feb 22 20:31:35 koen: does not work for me Feb 22 20:31:47 koen: gpe-scap say that he cannot upload Feb 22 20:32:00 I had a similar error on tuesday Feb 22 20:32:08 I though it was my network being messed up Feb 22 20:32:12 'error uploading screenshot: unable to upload screenshot (2)' Feb 22 20:33:15 THERE'S BUTTER ON MY FACE! Feb 22 20:33:25 * B_Lizzard sets mode: +m Feb 22 20:35:03 ok suitcase closed ... :) Feb 22 20:35:10 Looks like e2fsprogs was updated to install a bunch of fsck tools. e2fsprogs-native requires e2fsprogs and tries to install tools that don't exist Feb 22 20:35:21 apaulsen: kind of Feb 22 20:35:32 ok, any idea on how to fix it? Feb 22 20:36:46 looking at it Feb 22 20:39:42 working on it Feb 22 20:41:41 koen: galculator on XGA is funny Feb 22 20:41:48 hrw: did a pull, moved the conf/distro out of the way and restored it via mtn, the build is still b0rked. Feb 22 20:42:10 elliotf__: build linux-libc-headers 2.6.18 Feb 22 20:42:25 hrw: bitbake, or build? Feb 22 20:43:00 the funny thing is that it's looking for >= 2.6.17, and they should be there. Feb 22 20:43:05 with bitbake Feb 22 20:43:23 hrw: thanks Feb 22 20:44:06 fixed, pushing Feb 22 20:44:14 hrw: why not just lower the requirement of the kernel? (I'm building 2.6.18) Feb 22 20:45:37 glibc-intermediate managed to configure Feb 22 20:45:48 still kind of weird. Feb 22 20:46:14 hrw: is there logic behind that, that I'm not understanding? it seems rather weird. Feb 22 20:49:09 03hrw 07org.oe.dev * r78bf60d0... 10/ (4 files in 2 dirs): e2fsprogs: dropped old versions Feb 22 20:49:16 03hrw 07org.oe.dev * re09470e6... 10/ (4 files in 2 dirs): e2fsprogs: unbreak -native version, move common code to .inc from 1.38.bb Feb 22 20:49:23 is org.openembedded.dev the branch I should use if my target is an hx4700? Feb 22 20:49:31 yes Feb 22 20:49:47 apaulsen: pull/update/build Feb 22 20:49:49 previously they recommended oz354fam083 for some reason Feb 22 20:50:02 justin__: oz354fam083 is over year dead Feb 22 20:50:31 what exactly is angstrom-2007.1? Feb 22 20:50:35 distro Feb 22 20:50:50 Can i run GPE with it? Feb 22 20:50:54 you can Feb 22 20:51:01 I have gpe on my progear/angstrom now Feb 22 20:51:04 How is it compared to fam084? Feb 22 20:51:17 better, newer, faster Feb 22 20:51:27 awesome. i'll give it a try Feb 22 20:51:28 and working :) Feb 22 20:52:19 At FOSDEM progear will show Angstrom/GPE Feb 22 20:52:35 WOW Feb 22 20:52:48 i'll see ... Feb 22 20:53:09 g'night gotta sleep tomorrow have to wakeup a bit early ... Feb 22 20:53:11 bye bye Feb 22 20:53:33 cu Feb 22 20:53:45 for real this time ;) Feb 22 20:55:21 hrw: seems to work now Feb 22 20:56:59 NAiL: structurally, what's the more sound approach when building a recipie? I have a CPAN Makefile.PL which calls gnu autoconf. Should I: inherit 'autotools cpan' and expect them not to collide? Does the order in which they're inherited make a difference in which overrides the other? can I still get at oe_runconf from the autotools class after the cpan class is inherited? What's the best practice in this case? Feb 22 20:57:54 hillct: order matter iirc Feb 22 20:58:08 hillct: try both? Feb 22 20:58:18 heh Feb 22 20:58:32 Just trying to avoid beating my head against the wall. Feb 22 20:59:17 looking at each of the classes, it looks like I'd be better off inheriting autotools and incorporating the elements i need from the cpan class into my .bb file manually Feb 22 20:59:50 I'm sure reproducing classes like this is frowned upon, but it seems the clearner approach at this point Feb 22 21:44:51 yay! gpe-image is being built now... can't wait to try out this new angstrom distro & hx4700 wifi. thanks for your help hrw Feb 22 21:45:13 np Feb 22 22:08:00 hi Feb 22 22:29:29 for automated builds too Feb 22 22:29:38 oops Feb 22 22:30:29 hey CoreDump|home :) Feb 22 22:30:39 heya hvontres|poodle Feb 22 22:30:47 I guess traffic here is gonna be pretty slow this weekend :) Feb 22 22:30:53 indeed Feb 22 22:30:55 hvontres|poodle: yes :) Feb 22 22:31:55 RP: hey there :) Have you ever tried modifying the kernel command line inside of the zImage file? Feb 22 22:32:18 hvontres|poodle: No, I'd just recompile it ;-) Feb 22 22:33:20 RP: heh. I was just thinkig about a good way to have linux-rp build two zImages for the same machine with different cmd lines in one pass Feb 22 22:37:02 philippe|away: ping Feb 22 23:08:46 bye y'all. time to go home :) Feb 22 23:09:34 nite hvontres Feb 22 23:10:29 Have a good time at FOSDEM. Feb 23 00:04:19 I've been fighting with this recipe all afternoon. I'm down to one last error. Any guidance would be appreciated: http://www.pastebin.ca/368351 Feb 23 00:06:03 hillct: My guess would be that ${STAGING_DIR}/${TARGET_SYS} isn't set when INCLUDEPATH is evaluated Feb 23 00:07:04 ok. actually the behavior is identical when that var is not set at all Feb 23 00:07:14 Lemme try moving it around a bit... Feb 23 00:10:41 that line seems to have no impact no matter where it's placed Feb 23 00:18:14 gn Feb 23 00:18:46 any turbo station folks around? Feb 23 01:07:39 Bah, python is trying to pull in dietx11 Feb 23 01:57:30 ~/linux-ppc/bitbake/bin/bitbake gtk-engines -c rebuild Feb 23 01:57:50 IOError: [Errno 2] No such file or directory: '/home/pfalcon/linux-ppc/build-oe-angstrom/tmp/staging/arm-angstrom-linux-gnueabi/share/pkgconfig/gtk+-2.0.pc' Feb 23 01:58:06 just fyi, have no idea if it's on my side **** ENDING LOGGING AT Fri Feb 23 02:59:57 2007