**** BEGIN LOGGING AT Fri Mar 13 02:59:59 2015 Mar 13 08:38:54 morning all Mar 13 08:56:48 hi bluelightning Mar 13 08:56:56 hi woglinde Mar 13 10:58:24 I am hating PyQwt atm Mar 13 10:58:41 Crofton|work: hey! Mar 13 10:58:47 hi Mar 13 10:58:58 gm crofton Mar 13 10:59:08 yes its ugly to build Mar 13 10:59:21 Crofton|work: you never told me how fosdem went, and if your booth was a great success Mar 13 10:59:27 PyQt or really PyQwt? Mar 13 10:59:50 mag0, It went very well Mar 13 10:59:56 crofton has it ons session room Mar 13 11:00:06 PyQt I just need to get a set of fixes into dizzy and that will be good Mar 13 11:00:18 SDR devroom also went good Mar 13 11:00:36 pyqwt I need to remove from gnuradio Mar 13 11:02:59 just getting noise over that situation Mar 13 11:03:20 Getting so close to fixing that pain forever Mar 13 12:34:12 what am I doing wrong if bitbake warns me about multiple providers for a runtime package, even though I have specified a PREFERRED_PROVIDER in my distro config? Mar 13 12:34:22 multiple providers are available for runtime java2-runtime (cacao, jamvm, openjdk-7-jre, openjdk-6-jre) Mar 13 12:34:32 PREFERRED_PROVIDER_java2-runtime = "jamvm" Mar 13 12:35:57 and why does bitbake still warn about multiple providers if I build a specific recipe? for example, if I build jamvm (which RPROVIDES java2-runtime), I still get the warning that java2-runtime has multiple providers. I'd say explicitly building one of the providers should make the choice, right? Mar 13 12:40:29 mag0 hm its a bit compilcated Mar 13 12:41:04 try to set PREFERRED_PROVIDER_openjdk-7-jre = "openjdk-7-jre" and PREFERRED_PROVIDER_openjdk-6-jre = "openjdk-6-jre" in local.conf and see if it helps Mar 13 12:46:18 woglinde: okay, so that solved the problem. could you give a brief explaination of what just happened? :) Mar 13 12:47:12 nope Mar 13 12:49:21 i see, well thanks for the fix anyway! Mar 13 12:50:08 isn't it in the Readme anymore? Mar 13 12:50:29 it just says "Sometimes you need this" and list the lines you just gave me Mar 13 13:12:51 khem: glibc.inc, get_optimization() spits a bb.note on me even for TCLIBC != glibc :/ Mar 13 13:22:30 Hello does anyone know if there is a way to configure EFI to autoboot grub instead of running bootx64 or other command on EFI shell ? Mar 13 13:34:02 bluelightning, hi :) .. do you know if there is a way to configure EFI to autoboot grub instead of opening EFI shell and manually enter bootx64 or whatever the command is ? Mar 13 13:34:25 realBigfoot: I'm not sure, though I would assume that sort of thing would be controlled via an EFI variable Mar 13 13:34:54 (I feel like it's something I ought to know...) Mar 13 13:34:55 is there any variable like this on yocto recipes ? Mar 13 13:35:41 alright, i used the script mkefidisk.sh and I have generate an image on a sd card. Mar 13 13:36:42 Now i want to boot the image but, it opens EFI Shell and I have to type bootx64 to enter on grub and then I can boot the OS Mar 13 13:37:05 do you know if there is a way to boot it automatically ? Mar 13 13:37:07 bluelightning, ^ Mar 13 13:37:46 well, the EFI environment is controlled within EFI as far as I know, it's not something our system would be involved with Mar 13 13:40:03 bluelightning, i see... i have tried channel #efi and #uefi but there is nobody there. Would you happen to know how to configure this ? Mar 13 13:57:26 realBigfoot: I have a MBM that runs efiShell, pauses, and then runs grub. I haven't taken the time to RTFM enough to know if efiShell is calling bootx64 which then calls grub. Do you know how to query efiShell to find out? Mar 13 13:59:11 vmeson, unfortunately I don't know. What is MBM? Mar 13 14:01:24 realBigfoot: MinnowboardMax : http://elinux.org/Minnowboard:MinnowBoard Mar 13 14:02:36 vmeson, ok! yes it is the same as mine :) sorry I always call it Minnow or something ;) Mar 13 14:03:10 realBigfoot: I'd like to help you out and learn more about the MBM & EFI but I don't have much time for that today... I can reboot and type a few queries if you wnat to compare configs. Mar 13 14:03:16 vmeson, to clarify ... don't you have to type anything on efiShell? Mar 13 14:03:24 realBigfoot: correct Mar 13 14:04:11 vmeson, let me make a test... I always typed enter on efi shell maybe I'm breaking this wait part... Mar 13 14:05:02 realBigfoot: k, also fyi: http://elinux.org/Minnowboard:MinnowMax_HSE_issue -- my board had the plate but seems to be healthy anyway. Mar 13 14:05:46 vmeson, nice... I will see it... yeap I have tested here but it says "Map name not found" and then it drops to efi shell Mar 13 14:07:17 realBigfoot: ok, let me know if you figure out how to query the configuration. Mar 13 14:07:46 vmeson, alright! Thank you Mar 13 14:11:38 Do I need to nuke tmp after switching from ipkg to rpm? Mar 13 14:18:09 georgem: no, that is safe to change without doing that Mar 13 14:20:24 also nuking tmp isn't that expensive since it gets rebuilt from sstate Mar 13 14:21:03 bluelightning: anything else I need to do, I try to build the image now and it's erroring out with "$package not found in the base feeds" Mar 13 14:21:22 Crofton|work: true, maybe I should just try it and see if it fixes the problem Mar 13 14:22:30 georgem: that is odd and should not happen - what package is it complaining about? Mar 13 14:22:36 vmeson, i got this figured out Mar 13 14:22:58 vmeson, It was something king of idiot but however it worked .... Mar 13 14:23:22 realBigfoot: do you know what was wrong? Mar 13 14:23:52 bluelightning: dbus-x11. I'd normally assume I did something else wrong to cause this, but image build I did immediately before changing the package type worked fine. Mar 13 14:23:53 vmeson, yes... it was the boot order :( Mar 13 14:24:38 realBigfoot: ah, right. good luck. Don't forget to take that plate off the back of your board. Mar 13 14:24:59 vmeson, thank you very ! I have removed Mar 13 14:25:10 s/very/very much/ :P Mar 13 15:52:20 is there a way to cause a recipe to always get rebuilt (and not cached, etc) Mar 13 15:52:43 we have a simple version # recipe that we'd like to get built on every bitbake Mar 13 16:17:08 another way to state this is how can I not used shared state cache or invalidate for a particular recipe. The issue we're having is we toggle a locale variable between two values. Once each variation is built, it does not trigger a rebuild of dependent components (image in this case) Mar 13 16:34:49 i'd fix that then. if hte variable feeds into the build, then it should change the checksums, and that should ripple out through the dependency graph Mar 13 16:46:57 sgw_, ah busybox doesn't prvide zip. I was wondering why the zip recipe didn't use u-a Mar 13 16:47:50 Crofton|work: hi there, right, but unzip is there and should use u-a Mar 13 16:47:56 yep Mar 13 16:48:26 thanks for figuring that out, I was having a senior moment Mar 13 16:49:18 Crofton|work: no worries, we all have them from time to time ;-) Mar 13 16:49:49 it never occured to me that zip and unzip would come from differnet recipes Mar 13 17:49:03 pandaboard is magic board Mar 13 18:22:33 * nerdboy points out the logo in the slides => https://ams.confex.com/ams/95Annual/webprogram/Paper261125.html Mar 13 18:24:57 Thanks! Mar 13 19:21:36 Crofton|work: yeah, a not-too-horrible recording even Mar 13 19:25:42 hi, I am trying to create a python-zip package that will PROVIDE all the python standard libraries, but compressed Mar 13 19:26:03 I need python & Salt on a target. so I need to minimize the footprint Mar 13 19:26:57 my first attempt was to add a bbappend where I do all my magic on a do_install_append, copying all the ptyon libs to ${D}/zip, then having FILES_${PN}-zip = "zip" Mar 13 19:27:38 I had 2 problems: one is that QA fails since the split-package dir has .debug (can't remove the .debug on the do_install since it hasn't been created yet) Mar 13 19:28:10 the other is that everything is installed to /zip/usr/local/... , not sure how to tell OE to not use the /zip when packaging things up.... Mar 13 19:28:51 does this sound like a good way to tackle the problem or should I just create a copy of the python recipe and package things up as a separate recipe? **** ENDING LOGGING AT Sat Mar 14 02:59:59 2015