**** BEGIN LOGGING AT Sat Mar 26 02:59:58 2011 Mar 26 11:14:59 03Paul Menzel  07master * r12270185fa 10openembedded.git/recipes/xbmc/xbmc_svn.bb: (log message trimmed) Mar 26 11:15:00 xbmc: add `libass` to `DEPENDS` Mar 26 11:15:00 Task compile of `xbmc_svn.bb` fails with the following error message when using `angstrom-2010.x`. `angstrom-2008.1` does/did not have this problem probably because of using Libtool 2.2. Mar 26 12:13:50 he mickeyl Mar 26 12:14:32 morning woglinde Mar 26 12:15:47 Hi guys! Mar 26 12:15:59 ericben: n_gsm works like a charm on my Cinterion MC75i, thanks for the docs! Applying this to the Calypso will make some problems, so I'm not sure whether it's worth the pain, but anyways, it's very good to see that this is now working, especially given that we talked about it for about 4 years now :) Mar 26 12:16:17 Quick question. Trying to build e-image-core and it fails not being able to find a provider for xserver-xorg-extension-dri Mar 26 12:16:52 However it is built, and provided by xserver-xorg (actually the xorg-xserver-common.inc) Mar 26 12:17:00 Anybody an idea what might be wrong there? Mar 26 12:18:15 phdeswer: which MACHINE? Mar 26 12:18:18 pastebin complete error Mar 26 12:18:32 omap4430 Mar 26 12:18:33 all other is crystall ball magic Mar 26 12:18:48 omap4430? Mar 26 12:18:51 thats not a machine Mar 26 12:18:57 thats a processor Mar 26 12:18:59 or soc Mar 26 12:19:21 woglinde, http://pastebin.com/q0GdMML1 Mar 26 12:19:38 woglinde, that is the machine conf name (pandaboard here) Mar 26 12:19:58 omap4430-panda Mar 26 12:20:00 openembedded/conf/machine/omap4430-panda.conf Mar 26 12:20:02 yep Mar 26 12:20:16 hm Mar 26 12:20:25 intressting that E is depending on dri Mar 26 12:21:04 E is not immediately. Only needs glx (which fails in the same way) Mar 26 12:21:17 The omap4430-panda requires it Mar 26 12:21:31 "The giant panda, or panda (Ailuropoda melanoleuca, literally meaning "black and white cat-foot")[2] is a bear[3] native to central-western and south western China." from wikipedia Mar 26 12:21:35 woglinde: that's XSERVER var from panda config Mar 26 12:21:38 I commented them out to try, but then the e build fails because of the missing glx. Mar 26 12:22:05 so xserver var is wrong Mar 26 12:22:14 intressting that koen didnt fixed it Mar 26 12:22:32 phdeswer: can you try to build only xserver-xorg and check it builds dri for you? Mar 26 12:22:37 woglinde, tought so too but could not find anything wrong... Mar 26 12:22:42 JaMa|Off, I did. Mar 26 12:22:55 yeah dri and dri2 is wrong Mar 26 12:23:14 omap stuff has no dri Mar 26 12:23:45 woglinde, if it does or not is irrelevant. dri, dri2 and glx packages are built. Mar 26 12:23:54 Somehow the dep resolving does not notice they are there. Mar 26 12:25:48 Hmm, just noticed xserver is defined in the panda AND angstrom config... Mar 26 12:25:59 phdeswer try bitbake -g too Mar 26 12:26:38 * phdeswer is so freakishly rusty at OE after being out for 1,5 years Mar 26 12:33:45 woglinde, where am I supposed to find that dependency graph? bitbake -g e-image-core gives me the exact same output... Mar 26 12:44:04 frak, autom4te: need GNU m4 1.4 or later: /home/balister/oe/tmp/sysroots/x86_64-linux/usr/bin/m4 Mar 26 12:44:22 running an autotools configure script on an omap3 Mar 26 12:44:31 this bug keeps coming back Mar 26 12:45:54 Worked around it... DIRTY... edited the PROVIDES line in xorg-xserver-common.inc to have the ones I need.... Better options are welcome. Preferably the "right way tm" Mar 26 12:47:28 crfoto what the hell you are doing? Mar 26 12:57:00 phdeswer ask koen Mar 26 12:57:14 on #panda Mar 26 12:57:23 or was it #pandaboard Mar 26 13:03:56 woglinde, I will ask koen indeed. Mar 26 13:04:03 #pandaboard btw indeed Mar 26 13:04:19 atm it works anyway :D Mar 26 13:04:43 you are not the only one with panda Mar 26 13:05:51 woglinde, I know :) Just want to build a working image now because Ubuntu is so painfully slow and heavy (after boot it uses 3,5x more memory than my default Debian laptop install???) Mar 26 13:06:22 I will sort out the details later. Mar 26 13:09:06 phdeswer dont use gnome Mar 26 13:09:11 thats the root of evil Mar 26 13:09:17 dont use networkmanager Mar 26 13:09:24 use kde4! :) Mar 26 13:09:25 and a lot of other shit Mar 26 13:09:52 ericben: ping Mar 26 13:09:52 hm my javac script seems to work now Mar 26 13:10:13 shitty workedaround the aborts from jamvm and cacao Mar 26 13:10:42 woglinde, gnome is not that bad. My tweaked debian install after booting into gnome uses 170Mb ram. Mar 26 13:10:51 tries now to execute javac 5 times when its failing Mar 26 13:11:02 but agree on the network manager... Jeez just pulls in 4 daemons etc to do network settings... Mar 26 13:11:10 gnome3 will be your nitemare Mar 26 13:11:22 agree there.... Mar 26 13:11:39 But a gnome2 install without network manager, hal or mono is pretty ok Mar 26 13:11:52 hm no Mar 26 13:12:00 to many applets running Mar 26 13:12:19 lunch now Mar 26 13:12:21 till later Mar 26 13:12:25 woglinde, I only have few Mar 26 13:12:28 enjoy lunch! Mar 26 14:22:24 re Mar 26 14:30:15 woglinde, native compiles Mar 26 14:34:02 crofton pyqt? Mar 26 14:34:31 no Mar 26 14:34:35 gnuradio Mar 26 14:34:43 but it uses autotools Mar 26 14:34:55 hm? Mar 26 14:34:57 and autotools is looking for something in the build system path Mar 26 14:35:03 yes Mar 26 14:35:06 need to talk with Tartarus /khem Mar 26 14:35:12 they last poked at it Mar 26 14:35:14 show me the log Mar 26 14:35:28 autom4te: need GNU m4 1.4 or later: /home/balister/oe/tmp/sysroots/x86_64-linux/usr/bin/m4 Mar 26 14:35:42 is the error Mar 26 14:35:45 and we dont have m 1.4 Mar 26 14:35:47 this has cropped up b4 Mar 26 14:35:58 hm lets see Mar 26 14:36:02 root@usrp-e1xx:~/gnuradio# m4 --version Mar 26 14:36:02 m4 (GNU M4) 1.4.14 Mar 26 14:36:13 :) Mar 26 14:36:20 can you pastebin the the recipe? Mar 26 14:36:28 and the configure.ac? Mar 26 14:36:47 this is running configure on the omap3 Mar 26 14:36:50 no recipe Mar 26 14:37:08 I am pretty sure I can work it out with Tartarus Mar 26 14:37:15 ah okay sdk Mar 26 14:37:21 he mde the change that made this happen last time :) Mar 26 14:37:23 yeah Mar 26 15:41:39 Some amount of additional mangling needed somewhere I think, heh Mar 26 15:47:35 Tartarus, yeah Mar 26 15:47:49 I think this is the same problem we had before Mar 26 15:47:55 and you reverted something Mar 26 15:48:06 be nice to fix it once and for all ... Mar 26 15:48:30 I see to be the native sdk test subject .... Mar 26 17:03:11 hi DJWillis Mar 26 17:03:19 I'll push slim now if you don't mind Mar 26 17:03:24 because else I forgett Mar 26 17:03:52 forget means keep it in my head in the todo list but never do it because of the lack of time Mar 26 17:23:49 hmmm Mar 26 17:23:55 DEPEND="virtual/x11 libxmu libpng libjpeg libpam freetype sessreg" Mar 26 17:23:55 RDEPEND="${DEPEND} perl libpam-meta xauth" Mar 26 17:24:04 what is autodetected in all that? Mar 26 17:24:12 should I leave the recipe as is? Mar 26 17:24:24 it comes from slim which comes from openpandora Mar 26 17:24:30 I'll commit and push soon Mar 26 17:25:42 maybe I should ldd slim to check? Mar 26 17:26:25 hmmm theses DEPENDS and RDEPENDS seem wrong Mar 26 17:29:18 hi JaMa|Off Mar 26 17:29:51 a quick comment on that: Mar 26 17:29:55 DEPEND="virtual/x11 libxmu libpng libjpeg libpam freetype sessreg" Mar 26 17:29:59 RDEPEND="${DEPEND} perl libpam-meta xauth" Mar 26 17:30:00 ? Mar 26 17:30:17 for me it seems bogus at fist sight Mar 26 17:31:49 I'm very tempted to push Mar 26 17:31:54 as is Mar 26 17:32:04 since I fixed all the rest in this recipe Mar 26 17:32:15 GNUtoo|laptop: 1) it should be RDEPENDS_${PN} Mar 26 17:32:22 ok Mar 26 17:32:41 2) most of it is bogus, shlibs should generate right RDEPENDS for all libs Mar 26 17:33:21 ok Mar 26 17:33:35 if it's not recipe like task/task-devlibs-for-something :) Mar 26 17:33:53 so... Mar 26 17:34:07 libpam-meta and xath are not needed for compilation? Mar 26 17:34:15 *xaht Mar 26 17:34:17 *xauth Mar 26 17:35:01 freetype are fonts Mar 26 17:35:03 so they should not Mar 26 17:35:10 I'll look what sessreg is Mar 26 17:36:11 sesreg seem to be a program Mar 26 17:36:14 an xorg program Mar 26 17:36:24 I'm a bit lost with all theses deps beside the obvious libs Mar 26 17:40:43 in other words: Mar 26 17:40:45 DEPEND = "virtual/x11 libxmu libpng libjpeg libpam freetype sessreg" Mar 26 17:40:45 RDEPEND_${PN} = "perl libpam-meta xauth virtual/x11 freetype sessreg" Mar 26 17:40:46 is fine? Mar 26 17:41:09 or should xauth go in DEPENDS to? Mar 26 17:41:36 I'll keep like that Mar 26 17:42:14 (the ${RDEPENDS} was so obvious that I forgott to mention it Mar 26 17:42:24 s/RDEPENDS/DEPENDS Mar 26 17:43:36 re Mar 26 17:44:51 hi Mar 26 17:45:57 woglinde, hi do you know if freetype, sessreg and virtual/x11 are correct in DEPENDS(are their header and libs usually needed), what about xauth? Mar 26 17:46:15 because I'm fixing a recipe Mar 26 17:46:22 that is not in oe yet Mar 26 17:46:30 because it was considered too hackish Mar 26 17:46:40 so I'm standardizing it to the oe standards Mar 26 17:46:54 here's my current DEPENDS and RDEPENDS: Mar 26 17:46:58 DEPEND = "virtual/x11 libxmu libpng libjpeg libpam freetype sessreg" Mar 26 17:47:04 RDEPEND_${PN} = "perl libpam-meta xauth virtual/x11 freetype sessreg" Mar 26 17:47:19 basically it's the last thing that I need to fix before pushing I think Mar 26 19:03:19 well.. time to install second HDD Mar 26 21:02:06 This change from a couple weeks ago seems to have broken docbook-sgml-3.1-native for me: http://git.openembedded.org/cgit.cgi/openembedded/commit/recipes/docbook-sgml-dtd/docbook-sgml-dtd-native.inc?id=a1a3a895663ac23c17bdfb9c0ab625dfd75b2717 Mar 26 21:02:25 All it did was add subdir to the SRC_URI. Mar 26 21:03:33 If I back it out, it builds fine. I'm using up to date bitbake, I believe. Mar 26 21:04:24 The error that I get is because the current directory is set wrong, and it can't find a file, which is actually there, just in the parent of where it's looking. Mar 26 21:04:41 sed: can't read docbook.cat: No such file or directory Mar 26 21:05:27 Anyone have a suggestion for me? Mar 26 21:07:59 check the dir in oetmp/work/arch/docbook Mar 26 21:15:19 yes, docbook.cat is in overo-oe/tmp/work/x86_64-linux/docbook-sgml-dtd-3.1-native-1.0-r2.0 Mar 26 21:15:44 but bitbake seems to be looking for it in overo-oe/tmp/work/x86_64-linux/docbook-sgml-dtd-3.1-native-1.0-r2.0/docbook-sgml-dtd-3.1-1.0 Mar 26 21:15:44 What's the proper SRC_URI to copy an entire folder tree? I have file://* but I get a "cp" error about the subdirectories Mar 26 21:17:55 sr105 there isnt something Mar 26 21:18:04 either archive or files Mar 26 21:18:09 no folders Mar 26 21:18:11 ok Mar 26 21:18:15 why you need folders? Mar 26 21:18:53 I'm building my own app and using bb to compile it. I have my code separated into sub-dirs Mar 26 21:19:14 ? Mar 26 21:19:29 how this should work when others will compile your app? Mar 26 21:19:50 use a sane code repository Mar 26 21:19:55 as everybody Mar 26 21:20:14 I have a bitbake recipe for my own qt app. and would I then need to check in the code every single time I want to compile it? Mar 26 21:20:26 ? Mar 26 21:20:38 get familar with git Mar 26 21:21:18 setup up your repos Mar 26 21:21:19 can the git fetcher pull a working revision? Mar 26 21:21:26 o.O Mar 26 21:21:35 fun question? Mar 26 21:22:12 The recipe is in a separate tree of recipes and that entire tree is under source control Mar 26 21:22:44 as soon as my dev is complete, I'll move the code into a tar ball or another source control repo for bitbake Mar 26 21:22:47 please! look at other recipes ending with _git _svn or _hg Mar 26 21:22:58 it does mercurial? Mar 26 21:23:35 is there a better source of info than the bitbake manual? Mar 26 21:23:40 the online one Mar 26 21:24:10 mercurial's not even listed in the list of fetchers Mar 26 21:24:20 hm seems I speaking spanish Mar 26 21:24:37 find -name "*_hg*" inside recipe dir Mar 26 21:26:28 found some, thanks. But of course, they can still only pull from checked in revs. I guess I can work with that. Mar 26 21:26:36 I'll create a throw-away repo Mar 26 21:28:03 ? Mar 26 21:28:18 please look at the cacao-native_hg.bb Mar 26 21:35:25 back on docbook-- bitbake does a chdir to the "subdir", then does the unpacking of the zip to the parent, then it tries to sed on docbook.cat, which is can't find, because it's in the parent, not the cwd. Mar 26 21:36:17 I assume this works for others? Mar 26 21:38:54 bluelightning: hi, are you here? Mar 26 21:39:03 hi anarsoul Mar 26 21:39:44 bluelightning: is it possible to run opie on device without touchscreen/mouse? Mar 26 21:40:26 anarsoul: it would be difficult I think, most of the UI is designed assuming there is some kind of pointing device Mar 26 21:40:48 and there's no some kind of mouse emulation? Mar 26 21:45:32 bluelightning: ? Mar 26 21:45:38 bluelightning: as I remember from my opie on Zaurus experience it was possible Mar 26 21:51:39 hii! can anyone help me with this error, got during setting up OE "bitbake helloworld" http://pastebin.com/bBWcGzXf Mar 26 21:52:24 anarsoul: I'm not aware of any emulation unfortunately Mar 26 21:52:58 anarsoul: you'd have to test to see if you could use it comfortably with keys only Mar 26 22:03:17 anyone?? Mar 26 22:04:05 vp: /usr/bin/env: svn: No such file or directory Mar 26 22:04:39 vp: I'd make sure you have all the stuff listed here first http://openembedded.org/index.php/Required_software Mar 26 22:04:47 vb, it's a fetch error Mar 26 22:04:48 retry Mar 26 22:05:11 GNUtoo|bug20: fetch error because he does not have svn installed.. at least from what I see from the log Mar 26 22:05:22 ok Mar 26 22:05:30 * GNUtoo|bug20 goes back to fixing abuse Mar 26 22:06:20 i tried installing svn http://pastebin.com/3wZjYGVk but got this error Mar 26 22:06:40 vb,what distro are you using? Mar 26 22:06:54 metano Mar 26 22:07:15 metano whats distro is that? Mar 26 22:07:26 no Mar 26 22:07:30 i mean on the pc Mar 26 22:07:40 Metano is a minimal-distro and angstrom-distro based distribution Mar 26 22:07:40 vb: why are you bitbaking svn? I meant - you have to install svn on your build system Mar 26 22:08:01 :O ohkk.. Mar 26 22:08:07 and it would be svn-native Mar 26 22:08:08 yep vb,what's your distro on your PC? (fedora,debian...) Mar 26 22:08:26 ubuntu. Mar 26 22:08:31 he's on ubuntu 10.10 Mar 26 22:08:41 vb: that required software list that I pasted - it refers to your build system, i.e. install those packages before attempting to build anything with OE Mar 26 22:09:21 okay Jin^eLD thanks. Mar 26 22:09:35 vb: "aptitude install sed wget cvs subversion git-core \ Mar 26 22:09:36 coreutils unzip texi2html texinfo docbook-utils \ Mar 26 22:09:36 gawk python-pysqlite2 diffstat help2man make gcc build-essential g++ \ Mar 26 22:09:36 desktop-file-utils chrpath" Mar 26 22:17:46 thanks angelox_123 :) Mar 26 22:17:59 np :) Mar 26 22:40:22 is IMAGE_ROOTFS_SIZE in bytes? somehow it has no effect fo rme Mar 26 22:40:31 hi Mar 26 22:40:51 can I push something with that inside: Mar 26 22:41:40 +DEPEND = "virtual/x11 libxmu libpng libjpeg libpam freetype sessreg" Mar 26 22:41:41 +RDEPEND_${PN} = "perl libpam-meta xauth virtual/x11 freetype sessreg" Mar 26 22:41:49 without the + of course Mar 26 22:42:01 does that seem correct? Mar 26 22:42:30 no RDEPENDS with virtual Mar 26 22:43:34 ok thanks a lot Mar 26 22:43:44 it's such kind of comment that I was looking for Mar 26 22:43:48 I remove the virtual Mar 26 22:44:10 and does some RDEPEND stuff should go in DEPENDS too? Mar 26 22:49:21 offtopic question: do anybody using GPT partitions with linux on PC? Mar 26 22:49:47 ahh off topic Mar 26 22:49:54 becaause there are x86 targets in oe Mar 26 22:50:21 well, semi-offtopic then :) Mar 26 22:50:30 lol Mar 26 22:50:41 because the x86 install are as hard as gentoo Mar 26 22:50:56 and there isn't even grub2 in oe Mar 26 22:51:01 if I remember well Mar 26 22:51:05 only a fixed grub1 Mar 26 22:52:38 unfortunately I use plain BIOS stuff Mar 26 23:00:44 mhm, I created an ext3 image in OE, resized it using qemu-img (since IMAGE_ROOTFS_SIZE_ext3 did not do anything for me), still "no space left on device", did anyone mess around with qemu/OE-built rootfs images? Mar 26 23:04:58 butler did it again Mar 26 23:13:55 * mwester didn't know woglinde has a butler Mar 26 23:22:24 hey. does bitbake 1.10.2 still uses environment variables during the build? Mar 26 23:23:20 through BB_ENV_EXTRAWHITE i mean Mar 27 00:04:40 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * r40225fbb40 10openembedded.git/recipes/ekiga/ekiga_git.bb: (log message trimmed) Mar 27 00:04:40 ekiga_git: fix configure not finding boost signals Mar 27 00:04:40 Without that fix we have that in the configure log: Mar 27 00:04:52 03Denis 'GNUtoo' Carikli  07org.openembedded.dev * r58b8da5fdf 10openembedded.git/recipes/abuse/ (3 files in 2 dirs): (log message trimmed) Mar 27 00:04:52 abuse: update to 0.7.1,fix license,compilation(host libdir),and sdl install Mar 27 00:04:52 Without the sdl install fix() we have: Mar 27 00:14:08 hmm, how do I tell OE to put the kernel to the rootfs? Mar 27 02:24:55 ~ugt **** ENDING LOGGING AT Sun Mar 27 02:59:59 2011