**** BEGIN LOGGING AT Fri Mar 02 03:00:02 2018
Mar 02 03:18:16 <yocti> New news from stackoverflow: Yocto procedure to install generated image to HDD from a USB-pendrive [on hold] <https://stackoverflow.com/questions/49048197/yocto-procedure-to-install-generated-image-to-hdd-from-a-usb-pendrive>
Mar 02 07:49:09 <yocti> New news from stackoverflow: Yocto - Try to use Realtime Kernel Version <https://stackoverflow.com/questions/47185572/yocto-try-to-use-realtime-kernel-version>
Mar 02 09:02:52 <mckoan> good morning
Mar 02 09:22:28 <viky> I am trying to add pyAudio package https://pastebin.com/raw/Cntyrt2E.
Mar 02 09:22:42 <viky> Its compiling but not working.
Mar 02 09:22:56 <viky> can someone help to add this package?
Mar 02 10:03:50 <fabo> fatal: unable to access 'http://git.yoctoproject.org/git/meta-virtualization/': Failed to connect to git.yoctoproject.org port 80: Connection timed out
Mar 02 10:04:07 <fabo> looks like it's down
Mar 02 10:07:26 <welhm> <fabo>  its up if i check from here
Mar 02 10:09:46 <fabo> welhm: ack. maybe geolocalized. it's still down for me
Mar 02 10:10:26 <welhm> fabo are you tying to git clone ?
Mar 02 10:10:38 <welhm> ( i only dit a browser test
Mar 02 10:11:17 <welhm> <viky>: the pyAudio package is running correct true bitbake ?
Mar 02 10:13:15 <fabo> welhm: it's now back. seemed temporary
Mar 02 10:14:46 <welhm> :D
Mar 02 10:46:41 <pn> I have created a custom group e.g "grp1" in my application recipe say "app.bb" . I am trying add my custom user e.g: "chrome" to this group "grp1" in "chrome.bb". The useradd command failed saying the "group does not exist". How can I add my custom user to my custom group?
Mar 02 11:11:50 <viky> welhm: Yes It's detected by bitbake and output bitbake is $ bitbake -v python-pyaudio 2>&1 | tee build.log https://pastebin.com/raw/34p5jGyy
Mar 02 11:12:24 <welhm> are the files installed on the img?
Mar 02 11:22:47 <viky> welhm: Aah  No output image updated after this bitbake
Mar 02 11:24:53 <welhm> <viky> search on do_install () and oe_runmake INSTALL_ROOT=${D} install
Mar 02 11:35:45 <viky> welhm: i am referring this bb file there is no do install. https://pastebin.com/raw/5u8BQ7yS
Mar 02 11:49:58 <yocti> New news from stackoverflow: Yocto: Add custom user to custom group <https://stackoverflow.com/questions/49068076/yocto-add-custom-user-to-custom-group>
Mar 02 12:00:08 <welhm> <viky> http://www.embeddedlinux.org.cn/OEManual/recipes_examples.html  and look at the do_install
Mar 02 12:56:38 <zeddii_home> rburton: RP: I’m trying to test my kernel-devsrc changes with mulsc after the first run in the AB showed a failure on one of my RDEPENDS. I ran into the new binutils breaking the musl build, which I fixed. But now I’m seeing an error that makes no sense for qemux86-64
Mar 02 12:56:44 <zeddii_home> RP: rburton: https://pastebin.com/DrkWsiMj
Mar 02 12:57:37 <zeddii_home> why on earth would a qemux86-64 build be picking up qemuarm, qemumips64 qemuppc devsrc stuff ? Does that mean my package arch is wrong ? or something like that ?
Mar 02 13:02:39 <RP> zeddii_home: dnf's filtering is a bit nuts, you can ignore those, the issue is the /usr/bin/python dep
Mar 02 13:02:56 <RP> zeddii_home: we just merged a patch which stops the other arches interfering fwiw
Mar 02 13:03:23 <RP> zeddii_home: http://git.yoctoproject.org/cgit.cgi/poky/commit/?id=6a07697a6dcb5e77aed4b01df4b30024c4b3d0fe
Mar 02 13:03:44 <zeddii_home> RP: I did notice your filtering patch, which is why I pinged ot see if you thought it might be related. and yah, I’ll also go look at the python part.
Mar 02 13:03:57 <zeddii_home> I have no idea why just switching to musl would trigger that.
Mar 02 13:04:07 <RP> zeddii_home: it was written to fix that kind of problem
Mar 02 13:04:29 <RP> zeddii_home: Just switching between MACHINEs is enough
Mar 02 13:04:40 <zeddii_home> assuming I fix python, will the dnf thing become a warning ?
Mar 02 13:04:57 <zeddii_home> maybe I should just rm -rf tmp
Mar 02 13:04:59 <RP> zeddii_home: it will simply ignore the "bad" packages
Mar 02 13:05:06 <zeddii_home> gotcha
Mar 02 13:05:26 <RP> zeddii_home: its only looking at them since the one it should be using has a dependency problem
Mar 02 13:05:58 <zeddii_home> aha. I’ll go do battle with it then. the other archs were throwing me for a loopl
Mar 02 13:06:00 <zeddii_home> thanks
Mar 02 13:06:18 <RP> zeddii_home: I don't like the way it behaves at all :(
Mar 02 13:11:40 <rburton> RP: i wonder if other bits you merged broke my locale code
Mar 02 13:11:55 <rburton> hm
Mar 02 13:12:00 <rburton> yeah i bet its that package filtering!
Mar 02 13:12:31 <RP> rburton: it was things in -next...
Mar 02 13:12:46 <RP> rburton: more like builds from sstate being bust?
Mar 02 13:13:06 <rburton> glibc was based on master
Mar 02 13:13:26 <rburton> it optimistically installs all locales it can find, but isn't handling 'no locales' and there isn't a dependency on the locale packages
Mar 02 13:13:29 <rburton> hmhmhmhmhmhm
Mar 02 13:13:51 <RP> rburton: that is hardly deterministic
Mar 02 17:45:36 <halstead> RP, rburton armpit: https://autobuilder.yocto.io/ is ready for work now.
Mar 02 17:47:36 <Crofton|work> the beatings will continue until the autobuilds are successful!
Mar 02 17:48:09 <RP> halstead: thanks
Mar 02 17:48:45 <halstead> :)
Mar 02 17:49:04 <rburton> thanks halstead
Mar 02 17:49:18 <Crofton|work> Mayb ein the kinder gentler world, you will jsut need to clean RP's biccyle if there is autobuilder trobule
Mar 02 17:51:07 <yocti> New news from stackoverflow: Is it possible to use Yocto Project to build an image for a Python back-end/C++(Qt5) front-end application? [closed] <https://stackoverflow.com/questions/48942330/is-it-possible-to-use-yocto-project-to-build-an-image-for-a-python-back-end-c>
Mar 02 17:54:37 <mckoan> halstead: impressive
Mar 02 18:39:02 <Schlong> hello
Mar 02 18:39:38 <Schlong> how is it precisely that bitbake gets away with sucking so much dick?
Mar 02 18:40:44 <kergoth> ah, more morons for the channel, lovely
Mar 02 18:46:07 <Crofton|work> some cisco box
Mar 02 20:14:02 <seebs> So I have this great idea for a performance improvement to pseudo, which might be anywhere from negligible to somewhat significant, and would be a large and intrusive change and require at least a couple of weeks of work to get it implemented enough to even have a way to measure the impact.
Mar 02 20:14:18 <seebs> long story short: i think pseudo is spending a lot of time iterating over strings when it doesn't need to.
Mar 02 20:14:40 <seebs> so the basic idea would be "in a whole lot of places, switch to using length-plus-pointer structures for strings".
Mar 02 20:29:15 <fray> calculare the strlen once instead of many times?
Mar 02 20:33:18 <JPEWhacker> iterating for what purpose?
Mar 02 20:34:03 <fray> usually adjusting or copying the string
Mar 02 20:34:25 <fray> hard to copy it if you don't know the length (w/ say strncpy -- since you shouldn't trust strcpy)
Mar 02 20:34:42 <fray> (strlen was an example, some functions it's passed it as well)
Mar 02 20:36:46 <JPEWhacker> Sure. Do you have any idea how much unnecessary iteration is being done (or an idea how to figure it out)?
Mar 02 20:37:00 <JPEWhacker> premature optimization, evil, etc. etc.
Mar 02 22:33:38 <staroselskii> ?
Mar 02 22:37:24 <RP> kergoth: around?
Mar 02 22:38:01 <RP> kergoth: contemplating http://git.yoctoproject.org/cgit.cgi/poky-contrib/commit/?h=rpurdie/wip2&id=0bc6d816bf934f25ef5a73bd05ab067800301714 (which depends on the preceding commit)
Mar 02 22:57:40 <kergoth> RP: seems reasonable. would certainly help with a mismatch between final values and bitbake -e output for configuration metadata. anything we can do to ensure the inspection tools are as effective and accurate as possible would be a good thing. I'm sure there are some cases where some work could be shifted from event handlers / anonymous python to inline python in certain situations which would help as well, since it'd be applied at expansion time
Mar 02 22:57:40 <kergoth>  rather than specific points of time
Mar 02 23:10:26 <rburton> RP: definitely seeing a 30s increase in build time from the latest merge to master
**** ENDING LOGGING AT Sat Mar 03 03:00:00 2018