**** BEGIN LOGGING AT Mon Mar 07 02:59:58 2016 Mar 07 07:52:28 bluelightning: hello Mar 07 07:52:40 hi ant_work Mar 07 07:53:13 ka6sox already uploaded the gnu-tar into oe mirror Mar 07 07:53:18 pls discard that patch Mar 07 07:59:03 btw discard also the second one for kernel 4.4. MMC fix went into stable 4.4 Mar 07 07:59:23 ok, will do Mar 07 07:59:24 (I'm marking now both as superseded in patchwork) Mar 07 07:59:28 thanks Mar 07 08:03:46 I'm still in touch with Bruce battling with linux-yocto-tiny-kexecboot Mar 07 08:04:08 build fails from scratch Mar 07 08:04:28 and also if linux-yocto is not built before/removed with cleansstate Mar 07 08:05:19 I'm not so convinved about the shared-workdir...it should maybe be per-machine...just wondering Mar 07 21:20:22 bluelightning: I'm a bit lost following the flow of oe-core commits Mar 07 21:20:33 hi ant_home Mar 07 21:20:35 how so? Mar 07 21:21:02 well, now the one about formfactor is strange... Mar 07 21:21:20 if [ -z "$HAVE_KEYBOARD" ]; then Mar 07 21:21:20 - HAVE_KEYBOARD=0 Mar 07 21:21:20 + HAVE_KEYBOARD=1 Mar 07 21:21:50 what's the point declaring the var then.. Mar 07 21:21:51 which commit is that in? Mar 07 21:22:31 a82ce3e477a47 of yesterday Mar 07 21:24:06 biggest changes are: devtool (I'll ask you someday) and the recent uninative stuff Mar 07 21:24:17 uninative = prebuilt sstate? Mar 07 21:25:11 well the comment explains why that change was made... it just specifies the default, your formfactor file can still specify HAVE_KEYBOARD=0 Mar 07 21:25:26 not entirely sure I agree with the change but I don't think it makes all that much difference Mar 07 21:25:51 uninative is a mechanism to remove the dependency on the host libc, which allows us to have native sstate that works across distros Mar 07 21:27:19 hmm. I've read about tarballs iirc Mar 07 21:31:43 yep Mar 07 21:31:44 +# This include points at uninative tarballs generated and maintained Mar 07 21:31:44 +# by the Yocto Project Mar 07 21:32:56 yes, the tarball contains the built libc Mar 07 21:33:53 then there is that --disable-static thing Mar 07 21:34:12 I have tounderstand whether it is included in nodistro. apparently not Mar 07 21:34:50 I bet this will break static recipes built against klibc Mar 07 21:34:59 I don't believe so, I think the idea is it's opt-in at the distro level; poky has it on Mar 07 21:35:31 I should add some guard into the recipes I think Mar 07 21:35:40 do you need to though? Mar 07 21:36:07 I should try it first... Mar 07 21:36:19 surely it'll only be a problem for recipes that try to link to klibc *and* one of the recipes that has static linking disabled Mar 07 21:36:35 (well, a library built by one of those recipes, to be precise) Mar 07 21:40:26 added to my to-do :) Mar 07 21:42:07 hopefully in the next days the commits will stop so that I Mar 07 21:42:21 shall have the time to digest the changes Mar 07 21:45:51 well, back to the little formfactor thingie, it doesn't impact meta-handheld Mar 07 21:46:25 neverthless, soucing machconfig and the setting the var true is the same as forcing it Mar 07 21:46:31 *sourcing Mar 07 21:46:44 argh this kb is dying... Mar 07 21:49:24 bs...well, you can explicitely opt-out defining it =0 Mar 07 22:49:14 okay, I added include conf/distro/include/world-broken.inc to defaultsetup.conf Mar 07 22:49:35 all -klibc binaries build correctly Mar 07 23:02:08 gn **** ENDING LOGGING AT Tue Mar 08 02:59:59 2016