**** BEGIN LOGGING AT Sat Sep 22 03:00:00 2018 Sep 22 05:30:37 armpit: hmm error ? Sep 22 10:33:49 7leave Sep 22 17:02:51 mhm, so I am still trying to figure out how to build multiple kernels for the same rootfs using this new mechanism of KERNEL_PACKAGE_NAME and somehow it does not work out for me, can anyone who is using that give me some hints? Sep 22 17:03:57 when I'm setting KERNEL_PACKAGE_NAME to the alternative kernel recipe, I get an error telling me that it should be set in PREFERRED_PROVIDER_virtual/kernel if I want to have it enabled... but it's the "second" kernel for that rootfs Sep 22 17:04:29 the "stock" one is already set as preferred provider.. so how is this supposed to work? what am I doing wrong? Sep 22 17:11:02 well KERNEL_PACKAGE_NAME is just a prefix Sep 22 17:13:22 I understand that the resulting packages will need to get a different name, that's what it does, right? but I should also be able to build that second package in parallel to the existing preferred virtual/kernel provider? Sep 22 17:20:41 Dont know if that makes difference but it does change providers name so it might be effective Sep 22 17:24:13 but I don't actually want to touch/change my default kerenel provider, its another package anway, like - totally different kernel versions and different kernel recipes Sep 22 17:26:55 understood Sep 22 17:27:33 the example in the commit goes into a direction where it would allow you to build the same kernel, just with a different defconfig Sep 22 17:27:44 so I wonder if building two independent kernel recipes is supported at all? Sep 22 18:34:04 I doubt Sep 22 18:36:45 doh... :( Sep 22 18:37:00 so we're back where we started :( Sep 22 18:38:18 I think in past we have multi uboot recipes Sep 22 18:38:26 maybe thats something to look into Sep 22 18:42:26 well, our multikernel classes worked really well for us, too bad we were not able to upstream them, but we stayed on 2.0.2 for several years and now it's the same story - update results in a lot of work :) the classes in 2.5 changed a lot Sep 22 18:50:26 well you collected technical debt :) Sep 22 19:24:35 :) Sep 22 21:16:36 khem, hi Sep 22 21:17:07 after long I am building with TC glibc Sep 22 21:17:09 heh Sep 22 21:17:15 today's pull Sep 22 21:17:17 | checking for broken __attribute__((alias()))... yes Sep 22 21:17:17 | configure: error: working alias attribute support required Sep 22 21:17:42 glibc-initial_2.28.bb:do_configure) failed with exit code '1' Sep 22 21:27:08 doh, this will hit soon the autobuilders **** ENDING LOGGING AT Sun Sep 23 02:59:59 2018