**** BEGIN LOGGING AT Wed Jun 17 02:59:59 2015 Jun 17 08:34:53 morning all Jun 17 08:49:20 rburton: any idea why [PATCH 1/2] glibc: CVE-2015-1472: wscanf allocates too little memory didn't get merged but the second patch in that series by Haris Okanovic did? Jun 17 08:49:30 * joshuagl has that on his fido watchlist Jun 17 08:51:53 erm Jun 17 08:52:51 because the log says "fixed in glibc 2.21" and master has 2.21 Jun 17 08:56:09 rburton: aha! Jun 17 08:57:19 :) Jun 17 08:57:33 hmm, we have 2.21 in fido too Jun 17 08:57:38 carry on, nothing to see here... Jun 17 09:07:20 rburton is my python-netserver patch now okay? Jun 17 09:12:39 rburton: ohman, we have CVE fixes languishing unmerged because of a missing sob :-/ Jun 17 09:21:03 rburton: oh, looks like the patch proposer sent a v2 with sob on Monday Jun 17 10:13:35 joshuagl: what one? Jun 17 10:19:50 rburton: libxml2 Jun 17 10:23:32 Commit: Ross Burton Jun 17 10:23:32 CommitDate: Tue Jun 16 14:34:11 2015 +0100 Jun 17 10:23:37 (in mut) Jun 17 10:23:48 cool :-) Jun 17 10:24:05 should probably have two streams, urgent fixes and normal fixes Jun 17 10:24:08 so cves get fast-tracked Jun 17 10:24:25 aye, that sounds like something worth exploring Jun 17 10:42:28 rburton, the remove oprofile patch is going in? Jun 17 10:58:34 yes Jun 17 11:10:22 how about the distcc one? Jun 17 11:12:49 Crofton|work: yes. check ross/mut in poky-contrib for my pending branch Jun 17 11:13:08 about to get v2 of the gnutls patches dropped after they exploded on the AB Jun 17 11:13:09 ah. silly me, I looked in master-next :) Jun 17 12:53:29 Anyone have any thoughts on cherry-picking this grub gcc-5 fix to fido: http://cgit.openembedded.org/openembedded-core/commit/meta/recipes-bsp/grub?id=7d79a7bfffbc39090b22bb7696cc5dbc832e49b6 Jun 17 12:59:42 maybe the git.yoctoproject.org cgitrc should set enable-index-owner=0 since that's not very informative in the yocto repos Jun 17 13:12:44 georgem: yeah, I'd like to have fido build on fedora 22 Jun 17 13:12:57 georgem: I already have some gcc5 fixes in my fido/next branch Jun 17 13:13:14 in fact, I can just go ahead and cherry-pick that now Jun 17 13:14:55 joshuagl: sounds good Jun 17 13:23:35 hello, I'm trying to create a recipe rdesktop_1.8.3.bb but I get an error http://pastebin.com/nXCXauZX Jun 17 13:23:45 ['endian-little', 'bit-32', 'arm-common', 'common-linux', 'common-glibc', 'arm-linux', 'arm-linux-gnueabi', 'common'] Jun 17 13:23:50 any clue? Jun 17 13:25:49 mckoan: are you building from a tar ball or version control? Jun 17 13:27:32 I'm not familiar with building rdesktop but looks like maybe it's using automake and needs to generate Makefile.in Jun 17 13:28:13 georgem: I pastebin-ed the recipe, however is SRC_URI = "${SOURCEFORGE_MIRROR}/rdesktop/rdesktop-${PV}.tar.gz" Jun 17 13:30:16 georgem: inherit autotools was enough time ago Jun 17 13:30:46 I can'r figure out what is changed in the classes (maybe) Jun 17 13:34:22 mckoan: hrm... I don't see any obvious reason why it wouldn't work as is. Jun 17 15:26:07 Hi all, I'm switching from git submodules to combo-layer for layer management. But I can't find some convenient way how to push changes done in combo layer into the upstream repo. I tried 'combo-layer splitpatch', but next 'combo-layer update' fails due to conflict. Is there a different way to push changes back? Jun 17 15:40:07 tomasnovotny: there's nothing automatic to handle that situation - either you would not push the change to the combo repository in the first place (i.e. do it on a local branch, then split and send for merging into the component repos); or alternatively you just skip the patch when it comes in from the component during update Jun 17 15:43:10 bluelightning: ok, thanks for the info...I was thinking that combo-layer can somehow handle that situation (semi)automatically. For me it is usual work scenario - play around with tha layer, then push it and later update the combo layer. Jun 17 15:44:09 it's possible we could add some handling for that I agree, there isn't anything right now though Jun 17 15:45:18 Is there any other recommended way (other than combo-layer) for layer management? I was using git submodules and especially for OE there are some drawbacks which are perfectly handled by combo-layer Jun 17 15:46:45 submodules, repo, git-subtree, and a number of home-rolled options, and some others also (mr, etc) Jun 17 15:46:53 depends on your requirements Jun 17 15:49:34 yes, I read all the options on combo-layer wiki. I will try some other option. Thanks to all for your answers. Jun 17 16:46:28 I sent a patch (lttng-modules: drop makefile patch) to oe-core on 15 June but it looks like patchwork didn't pick it up, any help? Jun 17 17:05:10 ntl: folks maintaining OE-Core don't currently use patchwork (although the intention is to do so in future) - for now if you haven't received a reply to your patch and it hasn't been merged then you can reply with a "ping" to find out what's going on Jun 17 17:06:42 bluelightning: oh, I see. I'll just ping in a week or so if there's no response. thanks. Jun 17 17:13:10 hello everybody, after inheriting extrausers class, i am able to create user in yocto..but it is not login in user mode..what are changes i need to do to login as user mode? Jun 17 17:16:01 ??? Jun 17 17:17:35 bryan: what does "is not login in user mode" mean? Jun 17 17:20:36 i meant when i reboot, it is login as root only not the one which i created..it seems i missed something while creating the new user...is it correct? Jun 17 17:21:44 I'm still not quite following Jun 17 17:21:58 do you mean if you attempt to log in as the extra user, it gives you an error ? Jun 17 17:22:18 if so, what error do you receive? Jun 17 20:35:55 do we have python vcversioner Jun 17 20:52:47 wtf license is this? Jun 17 20:52:48 https://github.com/habnabit/vcversioner/blob/master/COPYING Jun 17 20:53:40 Crofton|work: looks like MIT or a variant to me, but check opensource.org Jun 17 20:54:48 the I am too lazy to chose a real license file license **** ENDING LOGGING AT Thu Jun 18 02:59:59 2015