**** BEGIN LOGGING AT Mon Mar 16 02:59:57 2009 Mar 16 08:14:46 morning Mar 16 09:30:18 good morning Mar 16 09:32:07 morning Mar 16 09:36:12 morning all Mar 16 10:05:38 good morning Mar 16 10:10:19 why doesn't show the gitweb interface any commits between 2009-01-29 and 40 hours anymore, http://cgit.openembedded.net/cgit.cgi?url=openembedded/log/&h=org.openembedded.dev ? Mar 16 10:11:17 * florian looks at mickey|zzZZzz Mar 16 10:36:54 hm Mar 16 10:51:23 some knows what happend to the dev tree? Mar 16 10:51:47 i ate it Mar 16 10:51:55 haha Mar 16 10:52:03 *g* must be have a bad taste Mar 16 10:52:17 it was a bit crunchy and sour Mar 16 11:22:37 how does sd/mmc hot plug work Mar 16 11:22:57 what do i need to make it to mount whenever sd card is inserted and umount whenever it is removed Mar 16 11:39:23 re Mar 16 11:39:37 sh1: in OE udev do that Mar 16 11:40:12 do we need udev? Mar 16 11:40:18 or can we do it without that Mar 16 11:40:27 you need some kidn of hotplug manager Mar 16 11:40:29 udev/mdev Mar 16 11:40:41 udev is a hot plug manager? Mar 16 11:41:11 sd/mmc card should report the insertion/removal to user space. isnt it? Mar 16 11:41:17 i mean sd/mmc driver Mar 16 11:41:25 Your branch and 'origin/org.openembedded.dev' have diverged, Mar 16 11:41:25 and have 1075 and 1 different commit(s) each, respectively. Mar 16 11:41:29 o fsck... Mar 16 11:41:34 something is broken badly Mar 16 11:42:30 hrw: i think you didnot understand me Mar 16 11:42:51 03Robert Schuster  07hrw/poky-merge/add-rpm-packages-support * r472883c1b8 10openembedded.git/classes/java.bbclass: Mar 16 11:42:51 java.bbclass: Make oe_makeclasspath more flexible (partly by Marcin Juszkiewicz ) Mar 16 11:42:51 - converted tabs to spaces Mar 16 11:42:51 - allow entries which have no corresponding jar file (just skip) Mar 16 11:42:51 - document that behavior Mar 16 11:42:51 possible Mar 16 11:42:54 03Denys Dmytriyenko  07hrw/poky-merge/add-rpm-packages-support * raa80b60475 10openembedded.git/conf/distro/minimal.conf: minimal.conf: make IMAGE_FSTYPES a weak assignment to allow overwriting Mar 16 11:42:57 03Marcin Juszkiewicz  07hrw/poky-merge/add-rpm-packages-support * r16d71eb2f5 10openembedded.git/: Merge commit 'origin/org.openembedded.dev' into hrw/poky-merge/add-rpm-packages-support Mar 16 11:43:00 03Marcin Juszkiewicz  07hrw/poky-merge/add-rpm-packages-support * r1ed68c0f56 10openembedded.git/packages/python/ (3 files): python-pysqlite2-native: added for rpm Mar 16 11:43:15 * hrw pushed own org.openembedded.dev - 4 days old but have more history Mar 16 11:43:26 okay, im not talking about oe right now. i am asking if we can do in a regular filesystem without udev Mar 16 11:43:47 sh1: what does kernel do on hotplug? it calls hotplug manager Mar 16 11:43:55 and should our driver support that Mar 16 11:43:59 ~logs Mar 16 11:44:00 All conversations are logged to http://ibot.rikers.org/channel, where "channel" is replaced by the URL-encoded channel name, such as %23freenode for #freenode. Lines starting with spaces are not logged. Mar 16 11:44:23 okay i will look at those logs. Mar 16 11:44:38 no - I used logs for my own Mar 16 11:45:40 okay hrw, should mmc/sd driver report the hotplug to user space through sysfs. correct and this event is caught by the hotplug manager which mounts/unmounts Mar 16 11:45:49 am i correct in my understand Mar 16 11:46:18 hard to tell for me - I am not kernel hacker Mar 16 11:46:22 is this sysfs only for mmc/sd or in general for all removal mediums like usb Mar 16 11:46:25 ok Mar 16 11:59:52 hrw you fixed the dev tree now? Mar 16 12:04:38 woglinde: not quite - I lacked 4 days of changes Mar 16 12:04:52 hrw I realised it yesterday evening Mar 16 12:04:56 but only after a pull Mar 16 12:05:04 I wondered my self Mar 16 12:06:27 * hrw -> reboot Mar 16 12:14:46 sh1: not the right channel, but yes for all devices Mar 16 12:14:49 hrw hm strange is too if you check the commit ml, the revisions are shown in the git-webinterface Mar 16 12:24:56 re Mar 16 12:26:47 hrw hm strange is too if you check the commit ml, the revisions are shown in the git-webinterface Mar 16 12:27:19 I think that we need to pester Robert to make him 'git pull/push' always Mar 16 12:27:30 as it looks like his push was last Mar 16 12:27:49 hrw hm Mar 16 12:27:51 nope Mar 16 12:28:01 I checked the history Mar 16 12:28:06 there was no merge Mar 16 12:28:23 which delete the branch till back to 29.1 Mar 16 13:37:20 03Frans Meulenbroeks  07org.openembedded.dev * r5ad2cf5ec3 10openembedded.git/packages/proftpd/files/make.patch: proftpd: added missing make.patch Mar 16 13:37:20 03Frans Meulenbroeks  07org.openembedded.dev * r37dbb4f357 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 13:37:21 03Frans Meulenbroeks  07org.openembedded.dev * r971a08f557 10openembedded.git/packages/proftpd/proftpd_1.3.2.bb: proftpd: fixed paths Mar 16 13:37:22 03Stefan Schmidt  07org.openembedded.dev * r0bd2004ded 10openembedded.git/conf/distro/include/preferred-om-2008-versions.inc: preferred-om-2008-versions.inc: Update to version that are available Mar 16 13:37:25 03Koen Kooi  07org.openembedded.dev * r5422d5ba54 10openembedded.git/conf/machine/dm355-leopard.conf: dm355-leopard: add board file for the dm355 leopard board Mar 16 13:37:28 03Koen Kooi  07org.openembedded.dev * rca12f35c06 10openembedded.git/packages/u-boot/ (u-boot-1.2.0/dm355-leopard.diff u-boot_1.2.0.bb): u-boot 1.2.0: add support for dm355-leopard Mar 16 13:37:31 03Koen Kooi  07org.openembedded.dev * re49b459065 10openembedded.git/conf/checksums.ini: checksums.ini: merge checksums from mamona Mar 16 13:37:36 03Mike Westerhof  07org.openembedded.dev * r8745858efa 10openembedded.git/packages/thttpd/thttpd_2.25b.bb: thttpd: disable PARALLEL_MAKE due to build failures Mar 16 13:37:41 03Denys Dmytriyenko  07org.openembedded.dev * rbff08d17df 10openembedded.git/packages/meta/meta-toolchain.bb: meta-toolchain: use DEPLOY_DIR instead of TMPDIR/deploy for consistency Mar 16 13:37:44 03Stefan Schmidt  07org.openembedded.dev * r1704bef15d 10openembedded.git/conf/checksums.ini: checksums.ini: Add sums for connman packages. (Changed location) Mar 16 13:37:47 03Chris Larson  07org.openembedded.dev * r5ee794ae6c 10openembedded.git/conf/enterprise.conf: enterprise.conf: combine with user-specified BBMASK instead of overwriting it Mar 16 13:37:50 03Graeme Gregory  07org.openembedded.dev * r6cbecaa19d 10openembedded.git/conf/checksums.ini: Mar 16 13:37:55 checksums.ini : add new checksums and fix checksum for refreshed busybox Mar 16 13:37:57 patch Mar 16 13:37:59 03Stefan Schmidt  07org.openembedded.dev * ra1c86eeaca 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 13:38:02 03Stefan Schmidt  07org.openembedded.dev * r84eecfff21 10openembedded.git/conf/distro/include/preferred-om-2008-versions.inc: preferred-om-2008-versions.inc: Update more versions to available ones Mar 16 13:38:07 03Denys Dmytriyenko  07org.openembedded.dev * r8633d77524 10openembedded.git/conf/distro/minimal.conf: minimal.conf: make IMAGE_FSTYPES a weak assignment to allow overwriting Mar 16 13:38:10 03Stefan Schmidt  07org.openembedded.dev * r82e5731e86 10openembedded.git/packages/connman/ (5 files): connman: Moved from moblin.org to kernel.org Mar 16 13:38:15 03Roger Monk  07org.openembedded.dev * r45362bc05d 10openembedded.git/packages/dsplink/ti-codec-combos_3.16.bb: Mar 16 13:38:20 ti-codec-combos: Combo Build Cleanup and additional html docs in package Mar 16 13:38:22 * Updated recipe to correct build warnings where combos weren't properly cleaned before building, and hence gave confusing warnings. Mar 16 13:38:25 * Removed some unused paths from XDCPATH Mar 16 13:38:27 * Added top level html doc to install target and qualiTI reports Mar 16 13:38:31 03Koen Kooi  07org.openembedded.dev * r726bc3af56 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 13:38:34 03Stefan Schmidt  07org.openembedded.dev * rd704e9b027 10openembedded.git/packages/connman/connman_0.12.bb: connman: Add latest release Mar 16 14:00:10 morning Mar 16 14:06:20 http://www.stevechamberlin.com/cpu/about/ Mar 16 14:07:18 http://blog.stackoverflow.com/wp-content/uploads/nice-server-rack.jpg Mar 16 14:09:58 hi Mar 16 14:11:18 i have a slightly dumb question .. i'm trying to installing bitbake-1.8.10 on a system where previously 1.8.12 was installed (don't ask..) .. however I seem to be missing something .. installing bitbake via ./setup.py install yields core version: 1.8.12 / bitbake version: 1.8.10 -- what am I missing here ? Mar 16 14:14:24 * hrw never installed bitbake Mar 16 14:29:14 hrw: yeah I just figured that would have been better as well hehe Mar 16 14:40:52 gilligan-: you realise that 1.8.10 is not really recomended with the trunk oe.dev metadata? Mar 16 14:42:41 djw: it won't work at all.. it's for something way older that I have to work on here .. Mar 16 14:44:23 s/won't/wouldn't/ Mar 16 14:48:23 gilligan-: oe installs quite some files apart from the actual binary. you need to get rid of these too Mar 16 14:48:29 s/oe/bitbake Mar 16 14:54:52 florian: yep.. took me a bit to figure out that there was some dir somewhere in /usr/lib/python-support/yadayada/bb/*.py .. I was just grep'ing for bitbake Mar 16 14:56:45 bah.. company needs a faster build server Mar 16 14:57:08 hi, all! do anybody knows of some automation tool to produce reports from git repositories to satisfy project managers enjoying developers doing paperwork instead of actual work? Mar 16 14:59:13 * slapin is really sorry for offtopic :( Mar 16 14:59:29 morning Mar 16 14:59:40 kergoth: hi Mar 16 15:00:48 morning all Mar 16 15:00:59 ping cbrake Mar 16 15:02:02 garren: good morning Mar 16 15:02:28 garren: I've not had time to work on the GTK-sharp issue yet Mar 16 15:02:51 cbrake: morning... any progress on gtk-sharp? I'm planning on looking at it tonight and try and determine the missing dependancy. Mar 16 15:09:06 cbrake: happy birthday! Mar 16 15:10:21 hrw: thanks :-) Mar 16 15:10:55 garren: its on my long list :-\ Mar 16 15:12:28 birthday? happy bday cbrake. how old? Mar 16 15:12:33 cbrake: :-) I know what you mean Mar 16 15:13:45 * cbrake tries to remember how old he is without asking his wife ... the math tells me 37 years old Mar 16 15:13:52 03Stanislav Brabec  07org.openembedded.dev * rf0d34a2170 10openembedded.git/ (3 files in 2 dirs): m4: Updated to version 1.4.12. Mar 16 15:14:51 cbrake: happy birthday :-) Mar 16 15:15:41 cbrake happy birthday :) Mar 16 15:17:48 thanks all! Mar 16 15:29:02 So, is the OE git repo fixed, then? Did we lose any history? Mar 16 15:29:21 mwester: probably last few days - unless someone pushed them Mar 16 15:30:04 :( Ok. Mar 16 15:30:54 There were no commits for the past 36 hours, I have been watching (to see if this got fixed, because I didn't want to commit my attempt to merge it) Mar 16 15:31:07 So anything we lost would have been earlier than that. Mar 16 15:35:11 I pushed my 4 days old tree today in the morning Mar 16 15:39:49 hrw, better chance rpm will work now ? Mar 16 15:40:23 gremlin[it]: I got base-image built with ubuntu 7.10 - will rebase patches and push into OE this week Mar 16 15:40:58 good ! Mar 16 15:43:32 i have a little trouble with x11-image and qemux86 ... basically no input (keyb and mouse) work on x11 ... you have any clue ... (now i'm trying to build with angstrom to see if happen the same thing) Mar 16 15:47:51 cbrake: Happy Birthday!! Mar 16 15:47:52 cbrake: Happy Birthday!!! Mar 16 16:03:31 hrw: i'm pretty surprised the repository stayed in a corrupted state for as long as it did. It sounds like OpenEmbedded needs a specific person or persons who are assigned CM type responsibilities. Of course, this doesn't seem like it's the only problem of that sort. Things seem too fuzzy nowadays. everything handled by a pile of people means no individual responsibility, and inconsistent direction Mar 16 16:03:33 * kergoth shrugs Mar 16 16:03:59 * mwester agrees Mar 16 16:05:37 * hrw agrees too Mar 16 16:05:54 I'd nominate myself, but I'm a newcomer to the community and have way too much on my plate already. Mar 16 16:06:15 MostAwesomeDude, has just described most of us Mar 16 16:06:25 welll, the too much on the plate Mar 16 16:06:45 I plan to add more to my plate Mar 16 16:06:49 what was the problem and how was it fixed? Mar 16 16:07:03 * hrw starts 'let create .stable branch' Mar 16 16:09:17 I'd consider volunteering for aspects, but I'm still getting back into the swing of things Mar 16 16:09:26 and i don't have nearly the time i used to Mar 16 16:09:38 when i was at digi i think i was spending like 20 hours a week on OZ & later OpenEmbedded Mar 16 16:09:43 outside of work Mar 16 16:09:46 thats just not viable now Mar 16 16:09:46 heh Mar 16 16:11:45 Can anyone point me how can I build "uclibc-initial" using "uclibc-initial_svn" ? PREFERRED_VERSION_uclibc-initial = "svn" didn't work =/ Mar 16 16:12:12 vo5, the 'svn' recipe has a different version Mar 16 16:12:14 look at the bb file Mar 16 16:12:14 "svn" isn't the version of _svn. see the PV for that package Mar 16 16:12:24 question: does oe use a .so style library nomenclature? Mar 16 16:12:26 ... for the PV variable :) Mar 16 16:12:38 garren: debian.bbclass does debian style package renaming based on library soname, if that's what you're asking Mar 16 16:12:41 probably better PREFERRED_PROVIDER Mar 16 16:12:45 you'll possibly need to expand SRCREV from conf/distro/include/sane-srcrevs.inc Mar 16 16:12:48 garren: its up to the distro whether to INHERIT it Mar 16 16:13:19 great thanks Mar 16 16:13:52 kergoth: what do I inherit to use it? Mar 16 16:14:15 i just said. debian.bbclass is the filename, so its INHERIT += "debian". but most distros already pull it in Mar 16 16:14:27 not sure if minimal does, but i know angstromd oes Mar 16 16:14:33 I need it because it contains i586/uClibc.config, i'm trying to build for i586 but current uclibc do not have a config for it Mar 16 16:14:39 so it tries to build for alpha T_T Mar 16 16:15:47 kergoth: thanks Mar 16 16:15:51 np Mar 16 16:57:31 anyone remember why bb.fetch assumes DL_DIR rather than just letting you tell it where you want it? Mar 16 16:57:42 if anyone, i suppose it should be me that'd remember, since i wrote that pile of crap to begin with Mar 16 16:57:44 * kergoth grumbles Mar 16 17:00:38 dangit, i think i forgot my usb key at home that has my private key for pushing to OpenEmbedded on it Mar 16 17:00:39 hrmph Mar 16 17:02:30 03Ulf Samuelsson  07ulf/linux * rb1e4235026 10openembedded.git/packages/u-boot/u-boot_2009.01.bb: Remove u-boot 2008.10 patchset already present in 2009.01 Mar 16 17:05:48 cbrake: spoke to some guys in the monodev channel they think that the runtime cant find the machine.config... Its located in etc/mono/2.0/ on my ubuntu machine. Any ideas hot to get ubuntu to point to the correct one for the cross compile? Mar 16 17:06:03 cbrake: I mean openembedded not ubuntu :-) Mar 16 17:27:53 hi, i am working on a package that does not install everything, but leaves some files in the src dir, what is the best way to package them? Mar 16 17:29:24 ? Mar 16 17:29:45 eFfeM: by 'src dir' you mean ${S}? Mar 16 17:30:23 what package needs to be added to a toolchain for kernel sources? Mar 16 17:30:37 hrw, yes, it is perl stuff so not even sure if it is build or was already in the src Mar 16 17:30:55 eFfeM: rm_work.bbclass removes S Mar 16 17:31:39 hrw ? don't understand that Mar 16 17:31:40 but let me clarify Mar 16 17:32:22 i've checked in a recipe for xmltv, and after that I noticed that debian had 4 xmltv related packages. No problem, i know how to handle that, but: Mar 16 17:33:45 lenny/'s xmltv-gui package has a file /usr/bin/tv_check, this one is not in my install dir but is in mltv-0.5.54/choose/tv_check (where the .54 dir is ${S} ) Mar 16 17:34:45 should I add a copy line in do_install and add it to FILES_${PN}-gui ? Mar 16 17:35:09 or is there a better way? (or evne keep one package xmltv with all files) Mar 16 17:35:31 03Ulf Samuelsson  07refs/remotes/origin/ulf/linux * rb1e4235026 10openembedded.git/packages/u-boot/u-boot_2009.01.bb: Remove u-boot 2008.10 patchset already present in 2009.01 Mar 16 17:39:03 garren: its probably something in the mono.bbclass where paths are not set up correctly Mar 16 17:39:29 garren: I would probably start by looking at the output of bitbake -DDDD and then start tracing through the mono.bbclass file Mar 16 17:41:11 hrw any idea? Mar 16 17:42:13 moment Mar 16 17:42:34 ah Mar 16 17:42:47 cbrake: ok cool I'll give it a try Mar 16 17:43:05 eFfeM: I would go that way Mar 16 17:43:59 hrw, ok, seemed the best way to me as well, but wanted some feedback to be sure that I am not unintenionally doing things in a way that was not desired Mar 16 17:45:13 * hrw -> off Mar 16 17:45:18 have a nice rest of day Mar 16 17:45:30 have a nice evening too! Mar 16 17:48:22 03Ulf Samuelsson  07ulf/linux * r55a31b3d72 10openembedded.git/packages/libogg/ (libtheora-1.0/libtheora-1.0-no-docs.patch libtheora_1.0.bb): Mar 16 17:48:22 Do not build libtheora-1.0 docs or examples Mar 16 17:48:22 since this will run a cross-compiled utility Mar 16 17:48:22 (vp3huff) on the host Mar 16 17:48:53 is anyone looking into making OpenEmbedded packages relocatable? Mar 16 17:57:52 kergoth, in what way? Mar 16 17:58:15 03Ulf Samuelsson  07ulf/linux * r8ef8f32bf4 10openembedded.git/packages/perl/perl-native_5.8.8.bb: Do not build perl package ODBM_File in perl-native, since it breaks the build Mar 16 17:58:17 as in, being able to take a binary tarball emitted by oe, unpack it anywhere, an dhav eit still work Mar 16 17:58:23 because it finds things relative to itself Mar 16 17:58:29 rather than hardcoded --prefix Mar 16 17:58:31 not sure Mar 16 17:58:40 you know, like rpm's relocation feature Mar 16 18:05:10 In'ts the rfc that koen emitted appropriate to that, Crofton? Mar 16 18:05:45 I think that would be a good place to talk about it Mar 16 18:06:06 kergoth: are you talking about sdk? Mar 16 18:06:12 no Mar 16 18:06:35 i'm talking about packages built by OpenEmbedded in general, native, cross, and target Mar 16 18:06:49 robtow: link to the rfc in question? Mar 16 18:07:01 target packages are supposed to be relocatable... Mar 16 18:07:02 let me dig that out - I have it... BRB Mar 16 18:07:05 k Mar 16 18:07:17 denix: i'd bet a very, very small % actually are. Mar 16 18:07:29 autoconf has hardcoded paths to the prefix in its scripts Mar 16 18:07:31 so does quilt Mar 16 18:07:33 and many others Mar 16 18:07:43 It's the closest match I've found: http://thread.gmane.org/gmane.comp.handhelds.openembedded/21326/focus=21488 Mar 16 18:07:49 I know what you are talking about - but those are native and cross Mar 16 18:07:59 eh? Mar 16 18:08:03 And in talking with koen he says "early april" for getting everyone to work it up. Mar 16 18:08:04 both, final ipks and pstage as well Mar 16 18:08:06 autoconf can be built for the target too Mar 16 18:08:08 as can quilt Mar 16 18:08:12 So we can contribute - even get it started. Mar 16 18:08:25 i'm talking about all packages built by OpenEmbedded period, regardless of what prefix and target arch they happen to be built with Mar 16 18:08:32 robtow: anxious? Mar 16 18:08:45 Anxious? eager! Mar 16 18:09:10 robtow: I was about to move packages to recipes, but git got broken over the weekend Mar 16 18:09:12 And don't want to build an oblong wheel while others built a round one :-) Mar 16 18:09:21 that thread isn't about the same sort of thing i'm talking about. that split is necessary, but has nothing to do with relocatability for build output. that restructures it, it doesn't require that you be able to move things around without it exploding Mar 16 18:09:39 kergoth: hmm, but those paths are absolute relative to target root, not host Mar 16 18:09:39 Right. It's not directly related. Mar 16 18:09:52 denix: and? i dont htink you're understanding what relocation is in this context Mar 16 18:09:59 rpm and ipkg let you install a package somewhere other than / Mar 16 18:10:09 most packages blow up if you do that, because they hardcode the paths to their files Mar 16 18:10:16 kergoth: tell me :) Mar 16 18:10:26 the correct way to fix it is for packages to find their files relative to themselves Mar 16 18:10:39 so they get their libs in PATHTOBINARY/../lib, assuming lib is a sibling to the bin dir Mar 16 18:10:43 etc Mar 16 18:10:50 so no matter where it goes, it still functions Mar 16 18:11:07 this sort of fix is needed for projects like autopackage, klik, zeroinstall which want to put an app in a directory, rather than installing it in / Mar 16 18:11:30 not sure if its needed for nixos, i think it installs in the same place all the time, even though the filesystems for each package are independent Mar 16 18:12:13 denix: http://autopackage.org/docs/binreloc/ Mar 16 18:14:46 * denix is reading on binreloc... Mar 16 18:17:38 kergoth: interesting. it maybe a noble goal to make target packages relocatable, but not sure it is such a high demand feature... Mar 16 18:18:01 denix: making the native and cross packages relocatable would allow moving TMPDIR without exploding Mar 16 18:18:24 kergoth: THAT one is a very desirable one! :) Mar 16 18:18:30 yep Mar 16 18:18:30 Exactly. Mar 16 18:18:33 what arch, from local.conf's perspective, is the omap35x series of ARMv7? Mar 16 18:18:48 armv7a? Mar 16 18:18:49 (i'm targetting the beagle board in particular) Mar 16 18:19:01 Crofton|work: cheers. Mar 16 18:19:05 MACHINE=beagleboard Mar 16 18:19:12 ahh super. Mar 16 18:19:16 you don't specify arch in local.conf Mar 16 18:19:23 03Ulf Samuelsson  07ulf/linux * r0b91593c03 10openembedded.git/ (14 files in 7 dirs): Add support for busybox-1.13.3 Mar 16 18:19:32 I am hoping he meant MACHINE :) Mar 16 18:19:33 03Ulf Samuelsson  07ulf/linux * rcd6e0709f2 10openembedded.git/conf/machine/ (6 files): Set default at91bootstrap version to 2.10 in at91 board files Mar 16 18:19:43 denix - so you don't know of anyone concerned with this sort of relocation right now? Mar 16 18:20:49 I talked with koen about it; he said he "had some ideas", and that he wanted to make a push for it in early april. Mar 16 18:21:02 denix: Crofton|work yes, i did. i was looking at a line for TARGET_ARCH when i wrote this. i meant MACHINE.. Mar 16 18:21:07 robtow: can't say for others, but I can't see big benefits of target relocation for me Mar 16 18:21:36 But you wouldn't OBJECT to such a change, would you? Mar 16 18:22:37 TARGET_ARCH will get through machine I think Mar 16 18:22:38 OBJECT? not really. kergoth knows what he's doing :) Mar 16 18:22:49 IYes :-) Mar 16 18:23:26 denix: if you fix relocation for a package, it'll be relocatable whether its native, cross, or target. Mar 16 18:24:15 kergoth: ok, what about everything else, which is not a package? like files in staging, cross, work and other directories. Mar 16 18:24:22 what do you mean? Mar 16 18:24:44 emitted text files and such? there are very few of those. particularly using packaged staging Mar 16 18:25:18 03Michael 'Mickey' Lauer  07org.openembedded.dev * r8c1892e90a 10openembedded.git/packages/vala/vala-native_git.bb: Mar 16 18:25:18 vala-native: build git version from freesmartphone.org which has some additional patches on top Mar 16 18:25:18 (always sent upstream for inclusion) Mar 16 18:25:48 yeah, pstage is another issue Mar 16 18:26:41 well, the packaged-staging class has some assumptions it makes about TMPDIR structure that need fixing, but other than that, its not really a factor in the relocation issue Mar 16 18:27:19 pkgdata/runtime have files with TMPDIR embedded Mar 16 18:27:42 TMPDIR is a problem. Mar 16 18:28:12 many builds/installs in staging and work do have TMPDIR embedded, hence no way to easily move TMPDIR around Mar 16 18:31:35 denix: if you're using packaged staging, making that package relocatable fixes that issue, as i've said like 3 times now Mar 16 18:31:36 BTW, koen says he has a stomach bug, and that he trusts us to work on relocation. Mar 16 18:31:46 workdir i expect you're just hosed on Mar 16 18:33:37 He says he talked to denix and trusts him to know his thoughts and issues.... and that he is sick now. Mar 16 18:36:44 kergoth: please bear with me :) don't get mad :) so, you are saying everything will be restored from pstage packages. what about unfinished builds? will they be hosed and start from scratch? Mar 16 18:37:23 unfinished builds might have to be restarted, depending on workdir contents. haven't looked into that yet. Mar 16 18:37:34 Assume they are. Mar 16 18:37:43 NOt a big problem, is it? Mar 16 18:37:57 might still be fixable, would have to give it more thought Mar 16 18:39:03 robtow: I still don't get what's so important in renaming packages directory to recipes? that's so far is the only item approved... Mar 16 18:40:09 Well, *that* is mostly about "clarity of understandin", as koen originally put it. Mar 16 18:40:13 It makes sense to me. Mar 16 18:40:50 indeed. "package" is too overloaded a term Mar 16 18:40:54 confusing for the user Mar 16 18:41:03 PYes. Mar 16 18:41:30 kergoth: you know OE inside-out better than any/most of us here. I'm not trying to question you, just curios... :) Mar 16 18:42:23 so, is dev tree now completely fixed? is it safe to pull/commit? Mar 16 19:08:10 denix: it seems to be fixed. Mar 16 19:08:42 denix: pull, and if you find anything missing, let someone know and I expect it will have to be cherry-picked and re-applied at this point. Mar 16 19:11:47 hi Mar 16 19:12:01 jo hrw Mar 16 19:12:02 mwester: thanks! Mar 16 19:12:11 re Mar 16 19:12:12 hrw do you know now what happend? Mar 16 19:12:17 no Mar 16 19:47:29 mickeyl: i noticed that we build zope-interface as part of zope, but it seems to be a separate src package: see http://pypi.python.org/pypi/zope.interface, shouldn't we have a separate zope.interface package build from this in oe ? Mar 16 19:47:56 I am a bit confused about the opkg and or ipkg. Mar 16 19:49:20 opkg is apparently the furute but the packages and directories are still called ipk. is that right? Mar 16 19:49:26 right Mar 16 19:52:49 http://wiki.openmoko.org/wiki/Opkg#FAQ Mar 16 19:53:50 so only the docs remain to be ported :( Mar 16 19:55:00 the extension is irrelevant anyway Mar 16 19:55:55 other then confusing me yes, same for the configs /etc/ipkg btw Mar 16 19:56:28 extension means nothing on *nix anyway Mar 16 19:56:43 and the packages are the same format as .dev :-D Mar 16 19:56:46 .deb Mar 16 19:56:57 off by one error :-D Mar 16 19:58:37 how do I update the Packages.gz of opkg repository ? Mar 16 20:00:05 http://www.theregister.co.uk/2009/03/13/miniature_computer_field_guide/ Mar 16 20:00:10 bitbake package-index Mar 16 20:00:30 thanks Mar 16 20:01:07 hrw: LOL Mar 16 20:10:24 cbrake: still not sure what the problem is from what I gather its not finding the machine.config, not sure how to point it towards the mono package and the location of machine.config Mar 16 20:10:54 cbrake: the machine.config is in the mono package so how can I make gtk-sharp see that? Mar 16 20:19:51 eFfeM: yeah, why not Mar 16 20:21:06 mickeyl: will you do it or do I need to see if I can make the recipe ? Mar 16 20:21:19 (not too much python skills here) Mar 16 20:21:52 eFfeM: i can take a look if it's not too urgent. need to clean my head about some things but will have some time later this week Mar 16 20:23:35 guess that is best, there is no rush Mar 16 20:23:38 thanks alot Mar 16 20:25:48 see you guys Mar 16 20:29:47 03Khem Raj  07org.openembedded.dev * r9e83f3f1b9 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:29:49 03Khem Raj  07org.openembedded.dev * re7ec320c3d 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:29:49 03Khem Raj  07org.openembedded.dev * r3992e7dd43 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:29:50 03Khem Raj  07org.openembedded.dev * r31abc4f834 10openembedded.git/classes/autotools.bbclass: (log message trimmed) Mar 16 20:29:53 autotools.bbclass: Add new pattern to match -L in dependency_libs Mar 16 20:29:55 * Current pattern match does one think wrong consider Mar 16 20:29:57 something like this Mar 16 20:29:59 dependency_libs=' -L/home/kraj/work/oe/build/eglibc/omap5912osk/work/x86_64-linux/ncurses-native-5.4-r17/ncurses-5.4/lib' Mar 16 20:30:02 this pattern when matched gets converted to Mar 16 20:30:04 dependency_libs=' -L/home/kraj/work/oe/build/eglibc/omap5912osk/staging/x86_64-linux/usr/lib/lib' Mar 16 20:30:08 03Khem Raj  07org.openembedded.dev * rcf14de067d 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:30:11 03Khem Raj  07org.openembedded.dev * r151270f47b 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:30:14 03Khem Raj  07org.openembedded.dev * r0c7cbe92da 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:30:19 03Khem Raj  07org.openembedded.dev * r9b18402de1 10openembedded.git/: Merge branch 'org.openembedded.dev' of git@git.openembedded.net:openembedded into org.openembedded.dev Mar 16 20:30:53 night all Mar 16 20:32:36 03Michael 'Mickey' Lauer  07org.openembedded.dev * r16c5d6f1b9 10openembedded.git/packages/freesmartphone/libfsoframework_git.bb: libfsoframework: new recipe; support library for freesmartphone.org Mar 16 20:32:43 03Michael 'Mickey' Lauer  07org.openembedded.dev * re72e5da270 10openembedded.git/packages/freesmartphone/fsodeviced_git.bb: fsodeviced: new recipe; C reference implementation of org.freesmartphone.Device.* APIs Mar 16 20:41:44 03Michael 'Mickey' Lauer  07org.openembedded.dev * r551eb3b528 10openembedded.git/classes/ (fso-plugin.bbclass vala.bbclass): Mar 16 20:41:44 fso-plugin.bbclass: new class for fso plugin packaging Mar 16 20:41:44 vala.bbclass: new class for vala-based packages Mar 16 21:05:34 hi jay7 Mar 16 21:05:49 hello all :) Mar 16 21:05:52 woglinde: :) Mar 16 21:10:51 Jay7: hi Mar 16 21:11:03 denix: :) Mar 16 21:11:29 * Jay7 is hardly php'ing.. Mar 16 21:11:49 Jay7: I dug out the guy behind oe.ru... :) Mar 16 21:12:01 denix: who is it? :) Mar 16 21:13:31 Jay7: his name is Anatoly Shipitsin and he had few posts to oe-users and oe-dev Mar 16 21:13:54 jay7 I hope you get a lot of money for this pain Mar 16 21:14:09 woglinde: yeah.. :) Mar 16 21:15:07 denix: well, now we can create russian community around ;) Mar 16 21:16:21 Jay7: I guess so. For now oe.ru looks like a personal blog with series of articles on how to use OE... Mar 16 21:17:02 denix: I'll translate OE manual... but not sure about time for this :) Mar 16 21:17:46 seems we should contact with Anatoly about this :) Mar 16 21:17:54 they say time is money, but I'd say time is priceless :) Mar 16 21:17:58 I hope this is not commercial project.. Mar 16 21:19:34 doesn't look like, but who knows... Mar 16 21:22:17 bbl, need to eat something besides coffee... Mar 16 21:41:31 03Ulf Samuelsson  07ulf/linux * rab2fe0712a 10openembedded.git/packages/busybox/busybox_1.13.3.bb: Fix broken recipe for busybox-1.13.3 Mar 16 21:41:41 03Ulf Samuelsson  07ulf/linux * r3979740c9d 10openembedded.git/packages/gpsd/gpsd.inc: Avoid parallel make for gpsd Mar 16 22:21:05 mickeyl: ping Mar 16 22:37:57 mirko I think mickeyl will not come here until tomorrow Mar 16 22:56:41 woglinde: ok, thanks Mar 16 23:12:13 * Jay7 -> sleep **** ENDING LOGGING AT Tue Mar 17 02:59:57 2009