**** BEGIN LOGGING AT Sun May 15 02:59:58 2016 May 15 21:01:56 bluelightning: gm May 15 21:02:05 hi ant_home May 15 21:02:23 pls have a look at the pull requests May 15 21:02:46 I'd say you can tag afterwards May 15 21:04:29 about meta-opie, I'm under the impression there might be an underlying bug May 15 21:04:35 let me show you May 15 21:08:49 bluelightning: i't the qt.conf May 15 21:09:03 maybe it's ok, I don't know qmake.. May 15 21:09:17 [Paths] May 15 21:09:18 Prefix = May 15 21:09:18 Binaries = /tmp/build/tmp-glibc/sysroots/x86_64-linux/usr/bin May 15 21:09:18 Headers = /tmp/build/tmp-glibc/sysroots/collie/usr/include/ May 15 21:09:18 Plugins = /tmp/build/tmp-glibc/sysroots/collie/usr/lib//plugins/ May 15 21:09:18 Mkspecs = /tmp/build/tmp-glibc/sysroots/collie/usr/share//mkspecs/ May 15 21:10:52 i.e. Plugins is May 15 21:11:00 /tmp/build/tmp-glibc/sysroots/collie/usr/lib/opie/plugins May 15 21:14:54 I'm not too sure I'm afraid May 15 21:15:18 well, I've seen it when fixing opie-multikey May 15 21:16:02 so there's an "opie" substitution missing there? May 15 21:16:16 well, maybe. see the double slash May 15 21:17:26 then there are some interesting include in the three files I've checked May 15 21:18:13 (opie-multikey, opie-pickboard, opie-unikeyboard) May 15 21:18:36 now opie-pickboard is similar to -unikeyboard: both are plugins May 15 21:19:00 however -pickboard do stages headers (and the .so) for opie-multikey to build May 15 21:19:05 as dependency iirc May 15 21:23:21 bluelightning, you should check the code generating the .pro files in the source May 15 21:38:50 bluelightning: sigh.. I notice now the jffs2 image generation for collie is broken :/ May 15 21:39:08 :( May 15 21:39:23 smthg happened to EXTRA_IMAGECMD_jffs2 probably May 15 21:40:04 I see now opie-ima~s.jffs2│ 16384K May 15 21:40:26 but it was EXTRA_IMAGECMD_jffs2 = "--pad=14680064 -l -e ${JFFS2_ERASEBLOCK}" May 15 21:41:33 this would be the third or fourth time I'm fixing that code for Z :/ May 15 21:54:00 hm.. the run.do_image_jffs2 is ok May 15 21:54:20 maybe the image is simply too big and padding doesn't fit May 15 21:56:08 yep..core-ima~s.jffs2│ 14336K│ May 15 21:56:13 my bad... May 15 21:56:39 bluelightning, iirc there was once a size-check May 15 21:56:58 it must have been revamped recently, must dig the emails May 15 21:58:12 by the way, last hope ismusl-libc for opie on collie... May 15 21:58:16 mtd May 15 22:07:54 if you can build opie against uclibc then musl should be fine (in theory) May 15 22:09:33 https://github.com/LinuxPDA/OPIE-images_2016-05-14/blob/master/collie/opie-image-16mb-collie-20160514213122.rootfs.manifest May 15 22:09:49 ^ at first sight anything extraneous May 15 22:17:41 well there was the thing about the udev cache May 15 22:26:13 which is quite large... until that gets untangled BAD_RECOMMENDATIONS in the image recipe can be used to exclude that May 15 22:26:32 other than that it all looks OK though I'd like to know why expat is in there May 15 22:46:26 note the udev cache data is 0 at build time May 15 22:46:38 the problems ould be eudev hwdb May 15 22:46:51 this is already nuked in zaurus.inc May 15 22:47:08 Ross hinted he could accept patches to revert the defaults, though May 15 22:54:41 oh right I mixed the two up, sorry **** ENDING LOGGING AT Mon May 16 02:59:59 2016