**** BEGIN LOGGING AT Fri Sep 16 02:59:59 2016 Sep 16 04:56:26 allah is doing Sep 16 04:56:39 sun is not doing allah is doing Sep 16 04:56:47 moon is not doing allah is doing Sep 16 04:56:59 stars are not doing allah is doing Sep 16 04:57:28 planets are not doing allah is doing Sep 16 04:57:36 galaxies are not doing allah is doing Sep 16 04:57:46 oceans are not doing allah is doing Sep 16 04:59:26 mountains are not doing allah is doing Sep 16 04:59:33 trees are not doing allah is doing Sep 16 05:01:10 mom is not doing allah is doing Sep 16 05:01:19 dad is not doing allah is doing Sep 16 05:01:26 boss is not doing allah is doing Sep 16 05:01:33 job is not doing allah is doing Sep 16 05:01:41 dollar is not doing allah is doing Sep 16 05:04:29 degree is not doing allah is doing Sep 16 05:04:39 medicine is not doing allah is doing Sep 16 05:04:48 customers are not doing allah is doing Sep 16 05:05:06 you can not get a job without the permission of allah Sep 16 05:07:20 you can not get married without the permission of allah Sep 16 05:09:56 nobody can get angry at you without the permission of allah Sep 16 05:10:04 light is not doing allah is doing Sep 16 05:10:12 fan is not doing allah is doing Sep 16 05:10:20 businessess are not doing allah is doing Sep 16 05:10:29 america is not doing allah is doing Sep 16 05:10:54 fire can not burn without the permission of allah Sep 16 05:11:16 knife can not cut without the permission of allah Sep 16 05:11:24 rulers are not doing allah is doing Sep 16 05:11:34 governments are not doing allah is doing Sep 16 05:11:44 sleep is not doing allah is doing Sep 16 05:11:54 hunger is not doing allah is doing Sep 16 05:12:16 food does not take away the hunger allah takes away the hunger Sep 16 05:12:37 water does not take away the thirst allah takes away the thirst Sep 16 05:12:55 seeing is not doing allah is doing Sep 16 05:13:05 hearing is not doing allah is doing Sep 16 05:13:12 seasons are not doing allah is doing Sep 16 05:13:31 weather is not doing allah is doing Sep 16 05:13:53 humans are not doing allah is doing Sep 16 05:14:11 animals are not doing allah is doing Sep 16 05:14:37 the best amongst you are those who learn and then teach quran Sep 16 05:15:22 one letter read from book of allah amounts to one good deed and allah multiplies one good deed ten times Sep 16 05:16:21 hearts get rusted as does iron with water to remove rust from heart recitation of quran and rememberance of death Sep 16 05:16:39 heart is likened to a mirror Sep 16 05:17:19 when a person commits one sin a black dot sustains the heart Sep 16 05:18:09 to accept islam say that i bear witness that there is no deity worthy of worship except allah and muhammad peace be upon him is his slave and messenger Sep 16 05:18:44 read book http://www.fazaileamaal.com Sep 16 05:53:26 allah is doing Sep 16 05:53:34 sun is not doing allah is doing Sep 16 05:53:39 moon is not doing allah is doing Sep 16 05:53:58 stars are not doing allah is doing Sep 16 05:54:10 planets are not doing allah is doing Sep 16 05:54:22 galaxies are not doing allah is doing Sep 16 05:54:34 oceans are not doing allah is doing Sep 16 05:54:45 mountains are not doing allah is doing Sep 16 05:54:55 trees are not doing allah is doing Sep 16 05:55:04 mom is not doing allah is doing Sep 16 05:55:10 dad is not doing allah is doing Sep 16 05:55:22 boss is not doing allah is doing Sep 16 05:55:29 job is not doing allah is doing Sep 16 05:56:18 dollar is not doing allah is doing Sep 16 05:56:24 degree is not doing allah is doing Sep 16 05:56:33 medicine is not doing allah is doing Sep 16 05:56:44 customers are not doing allah is doing Sep 16 05:57:04 you can not get a job without the permission of allah Sep 16 05:57:19 you can not get married without the permission of allah Sep 16 05:57:44 nobody can get angry at you without the permission of allah Sep 16 05:57:51 light is not doing allah is doing Sep 16 05:57:57 fan is not doing allah is doing Sep 16 05:58:08 businessess are not doing allah is doing Sep 16 05:58:18 america is not doing allah is doing Sep 16 05:58:31 fire can not burn without the permission of allah Sep 16 05:58:42 knife can not cut without the permission of allah Sep 16 05:58:49 rulers are not doing allah is doing Sep 16 05:58:56 governments are not doing allah is doing Sep 16 05:59:04 sleep is not doing allah is doing Sep 16 05:59:12 hunger is not doing allah is doing Sep 16 05:59:28 food does not take away the hunger allah takes away the hunger Sep 16 05:59:44 water does not take away the thirst allah takes away the thirst Sep 16 05:59:51 seeing is not doing allah is doing Sep 16 05:59:59 hearing is not doing allah is doing Sep 16 06:00:07 seasons are not doing allah is doing Sep 16 06:00:15 weather is not doing allah is doing Sep 16 06:00:31 humans are not doing allah is doing Sep 16 06:00:40 animals are not doing allah is doing Sep 16 06:01:00 the best amongst you are those who learn and then teach quran Sep 16 06:01:26 one letter read from book of allah amounts to one good deed and allah multiplies one good deed ten times Sep 16 06:01:50 * gtristan is doing... Sep 16 06:02:36 * gtristan is doing the sun, the moon, stars, planets, galexies, oceans, mountains, trees, mom, dad, boss, job, dollar, degree, medicine, customers... Sep 16 06:02:41 * gtristan is doing allah ! Sep 16 06:03:13 hearts get rusted as does iron with water to remove rust from heart recitation of quran and rememberance of death Sep 16 06:03:23 heart is likened to a mirror Sep 16 06:03:42 when a person commits one sin a black dot sustains the heart Sep 16 06:04:41 to accept islam say that i bear witness that there is no deity worthy of worship except allah and muhammad peace be upon him is his slave and messenger Sep 16 06:05:17 * gtristan pulls out handy catch-all retort: https://paste.fedoraproject.org/428721/74005876/raw/ Sep 16 06:06:52 read book http://www.fazaileamaal.com Sep 16 06:08:15 read book http://www.muntakhabahadith.com Sep 16 06:08:40 Guest_84857: if you don't quit it now, you know what will happen Sep 16 06:14:49 need spiritual teacher visit http://www.alhaadi.org.za Sep 16 06:15:09 allah created the sky without any pillars Sep 16 06:26:30 allah up your fucking ass Sep 16 06:26:39 fucking raghead Sep 16 06:49:55 hi everyone Sep 16 06:52:10 hey! im building images with autobuilder, one recipe adding a group, another adding a user and includes that user into the group. My problem is that when building an image, the useradding recipe complains about that the group does not exists, so its like the group is not added to the sstate cache or something because it fails on populate_sysroot_setscene, anyone had something similar? Sep 16 09:07:58 this channel gets weird at night Sep 16 09:09:47 7 AM isn't exactly "night" Sep 16 09:10:16 normal sleep schedules are for casuals Sep 16 09:12:48 i think i'm on US west coast time atm :P Sep 16 09:32:55 Hi, I would like to change how the device trees are generated in Yocto, but not sure how to do it Sep 16 09:33:14 usually I would extend the recipe for a package, but this seems to be implemented in a 'class' Sep 16 09:33:27 is there a way to extend a class like a bbappend for a recipe? Sep 16 09:33:42 i just want to introduce a post-compilation pre-deployment step (sign or encrypt) Sep 16 09:37:53 You can't bbappend a class, but you can add a class with the same name in a layer with higher priority and explicitly require ... the bbclass file from the other layer Sep 16 10:17:38 is it possible to have the rootfs manifest more verbose? Sep 16 10:20:20 ? Sep 16 10:25:56 or, make the match licence.manifest more verbose Sep 16 11:32:27 rburton: just noticed my patch in master, thx. Sep 16 11:32:42 i'll look into a krogoth/jethro backport next Sep 16 11:51:15 neverpanic: done for the device trees =D do you know how could I do something similar for the initramfs? Sep 16 11:51:43 If it's a bbclass, the same approach should work? Sep 16 11:51:45 AFAIK, the ramdisk is magically generated when you just add the correct value to IMAGE_FSTYPES Sep 16 11:52:06 Sorry, no idea then. Sep 16 11:52:10 ok Sep 16 11:52:16 thanks anyway for the other piece of advice Sep 16 13:31:45 Should a source archive created by the archiver class contain native package directories Sep 16 13:33:36 I'd presume not, but it seems to Sep 16 14:16:24 CTtpollard: There are no native packages, so what do you mean with "native package directories"? Sep 16 14:17:05 I have quite a few package directories (produced using archiver) that have the -native suffix Sep 16 14:17:47 elfutils-natic, etc Sep 16 14:17:50 *native Sep 16 14:18:53 You mean build directories? Sep 16 14:19:02 (the whole of the x86_64-linux directory) Sep 16 14:19:33 No, the source directories that the archiver class produces, which 'should' be every packages deployed to the system Sep 16 14:26:22 Hi, I'm having issues with youtube and python, I cant verify the authenticity of the https://www.youtube.com Sep 16 14:28:04 I have selected the certificate-ca package Sep 16 14:29:58 and I have some certificates in: /usr/share/ca-certificates and /etc/ssl/certs Sep 16 14:32:05 Make sure your OpenSSL is new enough, since youtube's certificate is cross-signed IIRC and older versions of OpenSSL didn't deal with that correctly Sep 16 14:32:32 They'd allways follow to one end of the chain, even if an intermediate was already part of the CA store Sep 16 14:33:16 OpenSSL added a flag and Python now passes that flag, so both must be recent enough Sep 16 14:37:34 neverpanic: recent like in build from yesterday in krogorth Sep 16 14:37:36 ? Sep 16 14:37:46 I'd argue thats quite recent Sep 16 14:40:45 Though, wget is able to download https://www.youtube.com Sep 16 14:40:54 fenrig: that should be recent enough Sep 16 14:41:01 and I do think wget does validate https sources :/ Sep 16 14:42:46 wget uses gnutls which uses openssl Sep 16 14:42:54 and python uses openssl too? Sep 16 14:42:59 shouldnt both fail then> Sep 16 14:46:35 https://lists.yoctoproject.org/pipermail/yocto/2015-July/025774.html Sep 16 14:46:39 I think thats the isue Sep 16 14:52:29 there is definitely something broken in the certificates area :/ Sep 16 15:06:27 hello Sep 16 15:08:11 I'm trying to get a recipe to use libpcap, depends=libpcap does not work. I can see openembedded-core/meta/recipes-connectivity has libpcap. However this layer is not included in my BASELAYERS. Sep 16 15:08:50 i tried to add the layer, but its missing a layer.conf file. Sep 16 15:20:49 hi! Sep 16 15:22:07 when running a test suite, pexpect is missing from my python env. shall it be installed manually ? Sep 16 15:23:06 yes you should install it on your host Sep 16 15:23:17 ok, thanks! Sep 16 15:27:11 hi Sep 16 15:27:44 Is it a good idea to use sysroot's pkgconfig script instead of system's one (Default) Sep 16 15:28:58 ? Sep 16 15:29:20 every recipe running pkgconfig needs to inherit pkgconfig, which will build pkgconfjig-native and use the one in the sysroot automatically Sep 16 15:29:24 just add the inherit Sep 16 15:29:47 I did this but still not enough Sep 16 15:29:50 yes, it is Sep 16 15:29:54 it is enough Sep 16 15:30:07 if it doesn't work then thats the recipe and/or your confgure's fault Sep 16 15:30:24 kergoth: that class is just a DEPENDS assignment, i'd love to call it deprecated :) Sep 16 15:30:33 ah like configure not doeing cross compilation Sep 16 15:30:39 yeah, agreed. i think it used to do sed mangling or something back in the day Sep 16 15:30:52 RzR: as long as it just runs pkg-config from the path, it'll work fine. if it's hardcoding /usr/bin or something, it's broken and needs fixing Sep 16 15:32:40 well I am using scons that does : ParseConfig("pkg-config ... Sep 16 15:32:51 I am digging Sep 16 15:34:16 are you inheriting the scons class? Sep 16 15:34:24 yea and I have PKG_CONFIG_PATH exported too Sep 16 15:34:42 that happens for you, no need Sep 16 15:35:08 well that might be the breakage actually. our pkgconfig knows where to look. Sep 16 15:35:33 yeah, do not set or modify PKG_CONFIG_PATH Sep 16 15:36:21 I think I needed them Sep 16 15:36:29 Idont modify them Sep 16 15:36:37 just Sep 16 15:36:39 env["ENV"]["PKG_CONFIG"] = os.environ.get("PKG_CONFIG") Sep 16 15:36:48 we set PKG_CONFIG_PATH correctly. don't touch it Sep 16 15:36:52 so it's picking the wrong one Sep 16 15:37:01 OK makes sense now Sep 16 15:38:43 I had to do that to prevent looking into /usr/include Sep 16 15:39:11 I mean picking headers in .. Sep 16 15:40:18 why oh why oh why can't people just use autotools Sep 16 15:40:24 hehe Sep 16 15:40:29 scons being stupid in this channel, cmake in the other Sep 16 15:40:50 just wait till we start seeing more use of some other up and coming buildsystems, like bazel Sep 16 15:41:23 i should push my meson recipe Sep 16 15:41:45 but it only does a hello world build, haven't been brave/stupid enough to test it with gstreamer yet Sep 16 15:41:49 i haven't looked at meson much yet, how is it? Sep 16 15:42:45 haven't really played with it much Sep 16 15:42:47 lots of "magic" Sep 16 15:54:59 humm Sep 16 15:55:04 from subprocess import call Sep 16 15:55:05 call(["which", "pkg-config"]) Sep 16 15:55:09 right place Sep 16 15:55:20 env.ParseConfig("pkg-config Sep 16 15:55:27 | /bin/sh: 1: pkg-config: not found Sep 16 15:57:45 * RzR jump in #scons Sep 16 16:07:36 rburton: trust me. i'm not a fan of cmake. its just the hand ive been currently dealt which making me use it. Sep 16 16:08:38 i had to remove lots of code in macros to put build dependencies in proper subdirs. Sep 16 16:09:02 now, other than one problem I can actually build the entire source with multiple cores. Sep 16 16:27:42 ok, i've confirmed. pcmx-native builds using the host /usr/include/pcap.h. Sep 16 16:28:32 DEPENDS_class-target = "libpcap" Sep 16 16:29:07 does not create a pcap.h in my bitbake build fs. Sep 16 16:29:33 it does not generate a bitbake error Sep 16 16:29:59 the closest I have is a kernel header file in my target fs Sep 16 16:30:01 ./build/tmp-glibc/work/corei7-64-oe-linux/linux-libc-headers/4.4-r0/linux-4.4/arch/um/drivers/pcap_user.h Sep 16 16:38:12 davis: as i said, if you don't want pcmx-native to rely on the host's pcap, add native to the libpcap recipe's BBCLASSEXTEND and remove the _class-target from your DEPENDS Sep 16 16:40:48 I sorted out w/ this workaround Sep 16 16:40:49 if os.environ.get('PKG_CONFIG') != None: Sep 16 16:40:49 env["ENV"]["PKG_CONFIG"] = os.environ.get("PKG_CONFIG") Sep 16 16:40:53 err Sep 16 16:40:58 if os.environ.get('PATH') != None: Sep 16 16:40:58 env["ENV"]["PATH"] = os.environ.get("PATH") Sep 16 16:45:56 not the best option Sep 16 17:02:50 kergoth: yeah i screwed up. i'm ok with host-native using host. I had overwriten the depnds pcap when i had the target version of pcmx relying on the pcmx-native. Sep 16 17:03:17 that is working now. it has another error in cmake at the moment. Sep 16 17:03:40 ah Sep 16 17:04:43 lol, its a scafolding problem. fix bitbake, fix cmake; fix bitbake. fix native; fix target; yadda yadda Sep 16 17:04:48 kergoth: I'm very very pleased pkgconfig.bbclass is near empty now. It used to be horrific! :) Sep 16 17:04:53 agreed. Sep 16 17:05:38 kergoth: nice exception patches btw Sep 16 17:05:47 heh, every little bit helps Sep 16 17:05:52 exception handlign is an ongoing battle :) Sep 16 17:05:55 kergoth: I think there are some small tweaks like that elsewhere which might also help Sep 16 17:05:59 kergoth: indeed Sep 16 17:06:47 kergoth: https://bugzilla.yoctoproject.org/show_bug.cgi?id=10061 is related to exception handling. I'm thinking the server needs to send a SIGINT to all the server's child processes but worry about how that will get handled Sep 16 17:06:48 Bug 10061: minor, Medium, 2.2, richard.purdie, NEW , Ctrl+C during BB_HASHCHECK_FUNCTION execution does not interrupt processing nicely Sep 16 17:07:43 kergoth: its a worry that previously, we blocked it as having it at random points in bitbake was a big problem :/ Sep 16 17:08:28 i have a do_install{} stanza. how can I vary that between native and target? Sep 16 17:08:50 do_install_target {} fails Sep 16 17:09:04 davis: do_install_class-native and/or do_install_class-target Sep 16 17:09:26 hm. just like the variable. Sep 16 17:09:31 many thanks Sep 16 17:11:04 RP: i wonder if we could do something to let us get the stop command sooner, instead, like subprocess the function we're waiting on and loop / wait there Sep 16 17:11:05 hmm Sep 16 17:12:02 kergoth: subprocesses are one issue but its a deeper problem with things like the length of time some runqueue operations take (e.g. the prepare stage) Sep 16 17:12:31 kergoth: during that time the server won't even see if has a stop command queued :/ Sep 16 17:12:52 ah, true. but if we let the server get an int/term, we'll have to add a bunch of handling of it :\ Sep 16 17:13:17 kergoth: We might need to split it into two threads I guess :/ Sep 16 17:14:06 kergoth: I'm struggling to see how we can solve it easily... Sep 16 17:29:46 Hi is there any yocto documentation about codecs? Sep 16 17:30:11 First of all I get double free's when playing a mp3 file in mpd Sep 16 17:30:22 it is likely minimal due to IP/patent issues.. Sep 16 17:30:46 you are more likely to get help on codec in the oe-core channel, as that is more community focused (less IP concerns) Sep 16 17:31:01 okay so Sep 16 17:31:09 because Its just for myself Sep 16 17:31:18 (#oe is the channel) Sep 16 17:33:01 okay Im asking it there Sep 16 17:37:56 https://autobuilder.yoctoproject.org/main/buildslaves maintenance is done. Sep 16 17:38:42 btw I have a package with srvrev = autorev Sep 16 17:39:05 does it check the git out everytime I try building an image containen that recipe? Sep 16 17:43:43 thanks halstead Sep 16 17:43:57 :) Sep 17 00:22:12 why is ALLOW_EMPTY_${PN}-dev/dgb = "1" the default? won't that cause lots of redundant packages to be generated? Sep 17 00:45:45 Ulfalizer: due to 'depchains' processing. a -dev will depend on the -dev of the DEPENDS of the recipe. i.e. if ncurses needs glibc, then ncurses-dev needs glibc-dev, all the way down. if one wasn't emitted, it'd break the chain and leave bits uninstalled Sep 17 00:47:26 ok, thanks. might be undocumented. Sep 17 02:45:37 kergoth: https://gist.github.com/anonymous/48072864356daf934490dc3102d98266 does this look alright? Sep 17 02:46:48 seems reasonable to me, but best to run it by the technical docs folks on the project to review the wording Sep 17 02:47:12 yeah, i always submit bugs for documentation patches anyway. ty. Sep 17 02:51:49 np Sep 17 02:52:12 thanks for writing it up, i'm sure plenty of folks don't know about it **** ENDING LOGGING AT Sat Sep 17 02:59:58 2016