**** BEGIN LOGGING AT Mon Dec 18 03:00:02 2017 Dec 18 03:08:48 I've got a recipe with a long SRC_URI, is it possible to break this up into multiple lines? Dec 18 03:17:32 dushara: certainly, just end lines before the last with \ Dec 18 03:17:53 it's very commonly used, if you look at other recipes you're sure to see examples Dec 18 03:18:45 like "part1" \ "part2" Dec 18 03:18:50 pr "part 1\ Dec 18 03:18:53 part2" Dec 18 03:18:59 pr=or Dec 18 03:39:43 dushara: the second Dec 18 04:29:17 Got it. Thanks Dec 18 07:40:23 New news from stackoverflow: How to put an extra file in the kernel image by yocto Dec 18 09:54:14 hello Dec 18 09:54:57 what is this line doing? "addtask T1 after T2 T3 T4" Dec 18 09:55:16 i've found online only examples with "... after T2" Dec 18 10:01:56 for example: are T3 and T4 even considered? or are they just ignored? Dec 18 10:13:31 Hi guys Dec 18 10:13:56 I'd like to check if uboot env partitions are properly initialized and if not, perform a saveenv Dec 18 10:14:02 How can I do that from uboot itself Dec 18 10:14:07 At boot Dec 18 10:14:14 Automatically Dec 18 10:14:21 Programmatically* Dec 18 10:23:47 ? Dec 18 10:32:40 MajorGrub: maybe ask the uboot channel? Dec 18 10:36:31 LetoThe2nd: Shit I'm such an idiot lol Dec 18 10:36:32 Thx Dec 18 10:40:20 cornel: they're considered Dec 18 10:40:45 cornel: means after T2, after T3 and after T4 Dec 18 11:21:35 RP, that is, they all must be completed in this order before T1 is executed? Dec 18 11:25:18 cornel: yes Dec 18 11:39:21 RP, thank you very much Dec 18 11:41:15 New news from stackoverflow: bitbake recipe for copying folder, subfolders for yocto || Bitbake gets stuck at do_fetch? Dec 18 14:47:06 zeddii: did you get anywhere with the kernel issue? Dec 18 15:23:18 halstead: I saw we suffered the inode issue again. Did we have any monitoring in place? Dec 18 15:45:20 RP, I need to add more but I'll check what there is. Dec 18 15:56:46 I am missing drivers headers to build on the target board made by yocto. How do I get those headers? Dec 18 15:59:31 learningc: probably by adding the kernel-headers package Dec 18 16:00:02 ok thanks Dec 18 16:00:55 kanavin: thanks for the rpm turnaround, queued in mut2 now Dec 18 16:04:00 LetoThe2nd, I did the image append, but no kernel-headers Dec 18 16:05:46 RP: nothing obvious for that missing character on the serial. We had a missing feature port as port for the 4.12 updates. So I’m going to queue new 4.12 stable updates and see if the udp fixes are somehow causing this issue with the tcp/ip serial connection to qemu. Dec 18 16:07:57 rburton: cheers Dec 18 16:08:12 rburton: sadly, the whole signing thing needs to be looked at properly Dec 18 16:08:30 rburton: it's pretty insane that making core-image-minimal out of signed packages takes over an hour Dec 18 16:14:19 learningc> you want to be able to build on the target ? Dec 18 16:18:01 nayfe, yes, I have gcc compiler on the target already, but when building, some drivers headers are missing, for example the accelerometer Dec 18 16:21:15 learningc: maybe kernel-dev or kernel-devsrc ? Dec 18 16:21:49 ok Dec 18 16:29:33 kernel-dev compiles Dec 18 16:39:22 alimon, do you have time to answer questions about auh? It looks like it isn't using the /home/auh/build we set up for it. Dec 18 16:39:28 zeddii_home: thanks, anything I can help test or anything like that? I asked on Friday " I'm also puzzled, you updated to 4.12.16 but https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git/log/?h=linux-4.12.y has only 4.12.14 ?" - any pointers to where .16 comes from? Dec 18 16:40:57 Yep. PaulG @ Wind River is maintaing the 4.12.x -stable backports for us, and is queuing and sending them to linux-yocto. Dec 18 16:41:03 he annouces them on linux-yocto Dec 18 16:41:04 [linux-yocto] v4.12.x - stable updates comprising v4.12.17 Dec 18 16:41:16 in the repo in my kernel.org directory here: Dec 18 16:41:17 https://git.kernel.org/cgit/linux/kernel/git/paulg/linux-4.12.y.git/ Dec 18 16:46:17 RP, https://lists.yoctoproject.org/pipermail/linux-yocto/2017-December/006763.html Dec 18 16:46:38 * paulg is putting 4.12.18 through finalization right now. Dec 18 16:48:05 paulg: ah, I didn't realise you were the 4.12 stable maintainer :) Dec 18 16:48:47 well, maintaining a specific 4.12 stable Dec 18 16:49:06 I'm only really doing it for yocto and just put it on kernel.org for transparency, and in case any one else wants to make use of it. Dec 18 16:49:23 same was done for 4.8.x Dec 18 16:49:33 paulg: I'm just a bit worried about the numbering overlap with linux-stable :/ Dec 18 16:50:03 4.12 is EOL upstream, so its numbers have stopped Dec 18 16:50:07 not a concern. Greg explicitly marked 4.8 and 4.12 End of life. Dec 18 16:50:10 jinx Dec 18 16:50:19 * zeddii_home has fast fingers Dec 18 16:50:25 ok, thanks :) Dec 18 16:51:10 anyone know why we do ac_cv_sizeof_bool=${ac_cv_sizeof_bool=1} in meta/site/arm-linux? Dec 18 16:51:58 aratiu: it could have been in there for years from when we couldn't autocalculate it. Can it be calculated correctly when cross compiling now? Dec 18 16:52:54 zeddii_home, paulg: it helps to know that answer as its the kind of thing I get asked :) Dec 18 16:54:52 RP: I search the git log for info on it to the initial population commit back in 2005, it's the same since then Dec 18 16:55:25 I have a recipe which tries to calculate it and fails because it's defined as 1 on arm Dec 18 16:56:56 RP: yes, sizeof can always be calculated now Dec 18 16:57:07 if its wrong then that's "interesting" Dec 18 16:57:15 I'm running some tests now to verify but it looks like the answer is yes Dec 18 16:57:47 rburton: I thought we could... Dec 18 16:58:23 aratiu: though a test failing makes no sense Dec 18 16:58:25 aratiu: things have changed in the 12 years since and we could really do with clearing a few things out the site files now... Dec 18 16:58:36 the failing test does sound odd Dec 18 16:58:45 normally it just uses the cached value and continues Dec 18 16:59:01 pastebin? Dec 18 16:59:10 that code is weird, it's in the old iperf2 codebase, I'll send you a link Dec 18 17:00:13 https://sourceforge.net/p/iperf2/code/ci/master/tree/configure Dec 18 17:00:46 there's that "#define bool int" code in there which doesn't happen on arm Dec 18 17:01:33 if test "$ac_cv_sizeof_bool" = 0 ; then Dec 18 17:01:33 AC_DEFINE(bool, int) Dec 18 17:01:34 fi Dec 18 17:02:21 that would be why Dec 18 17:02:26 (m4/dast.m4) Dec 18 17:03:01 thanks, my autotools fu is not that strong Dec 18 17:10:11 aratiu: the internet says bool only exists if you include stdbool.h so there's an argument that we shouldn't cache it Dec 18 17:10:23 so patches welcome to remove it from the site files that define it Dec 18 17:10:59 the site files pretty much could be deleted and rewritten from scratch Dec 18 17:11:11 lots of duplication for 'linux' behaviour in the arch files Dec 18 17:11:56 great, again thanks, I'll send a patch to remove it Dec 18 17:12:25 New news from stackoverflow: linked a dynamic library in C++ cross platform project Dec 18 21:02:14 Hello there. I know this is a simple thing once you know how to do it... but I am not there yet. What is the easiest way to apply the follwing 3 commands to my rootfs in bitbake: groupadd gpio, chgrp gpio /dev/gpiomem, chmod g+rw /dev/gpiomem? Dec 18 21:03:53 Tried "extrausers" thingy, "roofs_postprocess_command" but didn't quite make it. **** ENDING LOGGING AT Tue Dec 19 03:00:01 2017