**** BEGIN LOGGING AT Wed Sep 21 02:59:58 2016 Sep 21 08:42:18 Hi, I do not know if this is a bug or expected behaviour :). I have four packages rproviding java2-runtime. I have a statement selecting one. PREFERRED_RPROVIDER_java2-runtime = "openjre-8". I I just build my image, only my openjre is build and therefore is the one in the image. On the other hand if I build world, and them I bould my image, a different pkg than my rpreferred is installed. Is this a bug? Sep 21 15:17:51 hrw, who would we talk to about dropping the bitbake rpm from Fedora? Sep 21 15:27:31 hmm, wonder if anyone ever played with dolt for libtool in an oe context. possibly i did, but i can't remember the result Sep 21 15:30:11 i seem to recall a recipe patching dolt out ... Sep 21 15:30:48 avoiding the overhead of libtool for the common case where we can assume a sane build environment always seemed like a nice way to go Sep 21 15:30:56 hmm Sep 21 15:32:15 absolutely Sep 21 15:32:36 i keep on meaning to bring back dolt as a drop-in replacement that needs no fiddling Sep 21 15:33:11 also ran across https://github.com/midipix-project/slibtool at one point, but not sure it's a complete drop-in replacement, and dolt would simplify further anyway Sep 21 15:33:27 yeah Sep 21 15:33:55 the trick will be swapping between dolt and libtool transparently so you can use dolt where it works and libtool where something actually needs it Sep 21 15:34:18 indeed Sep 21 15:34:36 so we can't just mdify the libtool recipe to inject it by default for the libtool.m4 macros Sep 21 15:35:28 i guess ideally would be what you're talking about, drop-in replacement, so we can just override LIBTOOL= Sep 21 15:35:33 but if we have a minilibtool.m4 which provides the same macros then the magic aclocal copy/prune can be used to only provide minilibtool.m4 instead of libtool.m4 if depends are right Sep 21 16:19:32 ANy qt5 wiards have insight http://pastebin.com/1KFnvrNF Sep 21 16:19:45 Seems realted to qt session manager being disabled Sep 21 17:29:29 Crofton|work: will take care of it Sep 21 17:29:37 thanks Sep 21 17:30:02 everytime I type bitbake in the wrong window, it asks if I want to install it Sep 21 17:31:23 * kergoth chuckles Sep 21 17:31:58 good thing: bitbake is orphaned in Fedora. should be easier to get it removed Sep 21 17:32:46 wew Sep 21 17:33:06 btw - what is current version? Sep 21 17:36:37 dunno, never installed Sep 21 17:36:44 commit ffca664c0ae394e798a3da0b5707f38a0b099d6b Sep 21 17:36:44 Author: Till Maas Sep 21 17:36:44 Date: Thu May 19 22:08:07 2016 +0200 Sep 21 17:36:44 2016-05-19: Retired orphaned package, because it was orphaned for Sep 21 17:36:44 more than six weeks. Sep 21 17:36:52 so bitbake is retired already Sep 21 17:36:59 will not happen in F25+ Sep 21 17:37:08 bitbake noarch 1.17.0-2.fc23 fedora 800 k Sep 21 17:37:17 Crofton|work: I meant upstream version ;d Sep 21 17:37:30 lol Sep 21 17:37:45 BitBake Build Tool Core version 1.31.1 Sep 21 17:38:09 Crofton|work: ~3.5y since I used OE for something probably Sep 21 17:38:24 not counting 'lets run OE on AArch64' Sep 21 17:39:45 19:39 hrw@puchatek:~$ LANGUAGE=C sudo dnf info bitbake Sep 21 17:39:45 Last metadata expiration check: 2 days, 5:19:07 ago on Mon Sep 19 14:20:31 2016. Sep 21 17:39:45 Error: No matching Packages to list Sep 21 17:39:53 that's on F24 Sep 21 17:39:55 CMAKE_ALIGN_SYSROOT[1] = "dbusmenu-qt5, -S${libdir}/lib, -S${STAGING_LIBDIR}/lib" Sep 21 17:39:55 CMAKE_ALIGN_SYSROOT[2] = "dbusmenu-qt5, -S${includedir}, -S${STAGING_INCDIR}" Sep 21 17:40:18 even better Sep 21 17:40:19 Crofton|work: so F23 was last Fedora with bitbake package Sep 21 17:40:40 yeah, looks this is last machien I have at that Sep 21 17:40:53 thanks for checking Sep 21 17:41:00 and sorry for the noise Sep 21 17:41:33 Crofton|work: no problem Sep 21 17:42:26 anyone understand the CMAKE_ALIGN_SYSROOT need? Sep 21 17:53:20 can someone else run: rm -rf tmp*; rm -rf cache conf/sanity_info; bitbake -p; bitbake -p a few times in a row? i'm seeing the second -p parsing rather than loading from cache *sometimes*, but not always Sep 21 18:15:04 just did 3 bitbake -p's in a row, the first was parsed, the second was cached, the third was parsed again. Sep 21 18:15:29 at least i know it wasn't my cooker changes, stock bitbake does this Sep 21 18:23:39 kergoth: maybe related to the download of the uninative tarball? Sep 21 18:23:55 perhaps, but why did the third one re-parse, in that case? Sep 21 18:24:03 that's a very good question Sep 21 18:24:32 * kergoth 'll crank up the debugging and try to sift through it Sep 21 18:24:51 I did file this bug: https://bugzilla.yoctoproject.org/show_bug.cgi?id=9643 Sep 21 18:25:58 that sort of functionality could be useful in this kind of situation Sep 21 18:29:18 indeed, will try that Sep 21 20:03:00 bluelightning: that patch was quite helpful, thanks Sep 21 20:46:52 kergoth: looking at it I may have a local version that I developed further, at the moment I can't find it but I'll keep looking Sep 21 20:47:08 really time I cleaned up my local branches... Sep 21 20:47:35 that's what i've been working on lately, branch cleanup, figuring out where various bits stand, polishing incomplete ibts, etc **** ENDING LOGGING AT Thu Sep 22 02:59:59 2016