**** BEGIN LOGGING AT Wed Dec 16 02:59:59 2015 Dec 16 09:26:05 Hi everyone, Since a few days we have a issue to build openssl_1.0.2d.bb on Fido, the url of tarball has change from /source to /source/old/. [#oe] Hi everyone, Since a few days we have a issue to build openssl_1.0.2d.bb on Fido, the url of tarball has change from /source to /source/old/. Someone you already submitted a patch to fix it ? Dec 16 10:11:17 I just submitted a patch to fix it: http://patchwork.openembedded.org/patch/110017/ Dec 16 10:23:09 t0mmy: that seems like a big patch just to add a SRC_URI entry? Dec 16 10:23:11 recipes-connectivity/openssl/openssl_1.0.2d.bb | 51 +++++++++++----------- 1 file changed, 26 insertions(+), 25 deletions(-) Dec 16 10:32:09 yeah that's wrong Dec 16 10:32:41 upstreams moving tarballs around is also wrong Dec 16 10:33:50 joshuagl: It is because I remove spaces to align the list of patches with the URI. Dec 16 10:45:51 t0mmy: hrmm, OK — but the 2nd entry isn't aligned Dec 16 10:46:12 file://configure-targets.patch \ is aligned with the " whereas the following entries are aligned with the h of the first entry Dec 16 13:18:47 hello giys Dec 16 13:18:50 giys Dec 16 13:18:53 guys Dec 16 13:19:13 I have 2 layers which have bbappend for poky recipe Dec 16 13:19:28 these 2 bbaappend are adding patches Dec 16 13:19:44 may question is in which order these patches are applied Dec 16 13:20:00 is it dependent on Layer priority? Dec 16 13:20:51 Noor: i think order of mention in bblayers.conf Dec 16 13:21:35 mago__: really Dec 16 13:21:45 Noor: but it's best if you design your layer so that it doesn't matter in what order it is appended Dec 16 13:21:47 I have doubt but I'll check Dec 16 13:22:22 Yeah but the issue we have is we have to add a patch on top of patch being applied in one layer Dec 16 13:22:34 so ordering matter to us Dec 16 13:24:16 Noor: i think maybe it's the order of which each dir is mentioned in BBFILES. This gets updated by your conf/layer.conf, so the order of parsing for this file is important (i think) Dec 16 13:24:42 BBFILES = "layer1/recipes-*/*.bbappend layer2/recipes-*/*.bbappend" Dec 16 13:25:06 .. will probably give priorty to layer1 Dec 16 13:28:51 mago__: yeah in the end it comes to priority Dec 16 13:41:37 Noor: BBFILES should control it Dec 16 13:42:14 RP: and I think BBFILES is controlled by Layer priority ? Dec 16 14:00:14 bbfiles depending from layer priority and recipe version. Dec 16 14:36:30 Noor: BBFILES is constructed by layer.conf Dec 16 14:37:16 RP: yeah but layer are selected based on layer priority Dec 16 14:37:30 RP: thanks for clearing it up Dec 16 14:37:41 we will pick it from here on our side Dec 16 15:13:53 Noor: to clarify, bbfiles order from layer.conf is controlled by BBLAYERS order, since that controls layer.conf parse order. in the *mel* case BBLAYERS order is sorted by defined layer priorities, but thats not the case elsewhere. i explained to the person whose problem you're working on about the priority controlling this last night.. Dec 16 15:20:03 kergoth: ok Dec 16 17:46:00 when submitting a patch, what branch should you submit for? Dec 16 17:50:59 CipherWizard: patches should be for master Dec 16 17:58:36 What do I do if I want a patch back-ported to another branch like jethro? Dec 16 18:01:54 CipherWizard: when the change is in master, you can send it for jethro (with "[jethro]" in the email title) Dec 16 18:05:41 hmm, anyone know anything about the edison bsp? Dec 16 18:05:50 WARNING: Unable to get checksum for u-boot-edison SRC_URI entry zc702-zynq7.env: file could not be found Dec 16 18:05:50 WARNING: Unable to get checksum for u-boot-edison-fw-utils SRC_URI entry zc702-zynq7.env: file could not be found Dec 16 18:06:08 I see this, but I am doing a build for a zc702 usin gthe angstrom layer stack Dec 16 19:02:06 just had to install perl-bignum on a build machine to get openssl to build Dec 16 19:02:22 well, let me make sure this was the issue :) Dec 16 19:20:21 Crofton|work: I hope we actually got that one documented Dec 16 19:20:29 it's certainly known Dec 16 19:20:51 nope, we did not :/ Dec 16 19:20:54 rburton: ^ Dec 16 19:21:53 guess I should file a bug Dec 16 19:25:54 always fun times starting a freesh vm Dec 16 19:26:19 filed, https://bugzilla.yoctoproject.org/show_bug.cgi?id=8822 Dec 16 20:16:11 how do I check on an oe core patch status these days? Dec 16 20:16:46 my process is: 1) check the list for replies, 2) check the master branch, 3) check the master-next branch, 4) check ross's mut (master-under-test) Dec 16 20:16:52 not sure what the best method is.. Dec 16 20:16:55 :) Dec 16 20:19:25 this is why we want to improve the patch process ;) Dec 16 20:19:43 yeah Dec 16 20:19:54 MArcus is trying to make OE usable by zsh users Dec 16 20:20:03 sounds good Dec 16 20:21:57 should experiment with some good code review tools like reviewboard or gerritt, but there's so much resistance to using anything but the mailing lists.. Dec 16 20:24:10 I think there's disagreement about "good" there... Dec 16 20:24:23 I've not used gerrit, but I found reviewboard infuriating Dec 16 20:24:39 at least the KDE instance thereof Dec 16 20:25:14 they all require some effort to understand and adapt workflow Dec 16 20:25:46 i think reviewboard's still not where it needs to be for git. i don't want to review a branch as one diff.. Dec 16 20:25:57 github handles that well, but has its own problems Dec 16 20:31:14 right, it was the adulteration of commits that bothered me most about reviewboard Dec 16 20:49:41 hmm, there was a recent ML discussion on whether patches should be added to SRC_URI in the commit that introduces the patch, anyone know if that thread was resolved? Dec 16 20:50:06 * joshuagl is looking through the fido backlog and has a 15 patch series where SRC_URI is updated in the final patch Dec 16 20:58:37 ah, no — it seems there wasn't a resolution Dec 16 21:02:28 i really prefer patch and src_uri update in one Dec 16 21:02:38 yeah, likewise — just writing a mail Dec 16 21:03:19 Crofton|work: patchworks has promise and we've got people actively working on it Dec 16 21:04:02 Crofton|work: re openssl, did you get a moan from the build, or did you have to guess? Dec 16 21:04:28 I had to look at the openssl build log and notove the perl impot fail Dec 16 21:04:42 Crofton|work: jethro release or git branch or something else? Dec 16 21:08:03 jethro I think Dec 16 21:08:16 building Angstrom 2015.12 Dec 16 21:08:34 on brand new F23 vm Dec 16 21:13:15 rburton: I'd like to use the AB if I may? It appears to be idle atm… Dec 16 21:37:52 I guess I just missed him Dec 16 21:52:33 joshuagl: tracy was killing my builds earlier one sec Dec 16 21:53:00 joshuagl: oh right yeah its fine, fire away Dec 16 21:53:37 * rburton also fires a single build on it Dec 16 21:53:49 thanks rburton Dec 16 21:56:16 off it goes, and in true me fashion I'm going to disappear just before anything explodes Dec 16 21:57:59 oh that's the best approach Dec 16 21:58:10 https://wiki.mozilla.org/images/5/56/Land_patch_-_go_home.jpg **** ENDING LOGGING AT Thu Dec 17 02:59:58 2015