**** BEGIN LOGGING AT Tue Mar 12 02:59:58 2013 Mar 12 05:51:25 pwd Mar 12 06:06:25 /root Mar 12 09:11:10 good morning all Mar 12 09:38:24 * tasslehoff has just watched the getting-started and suspects he will move to yocto Mar 12 14:46:53 i am working to create a smaller filesystem image than Yocto's default core-image-minimal and one of the things i would like to remove is the 'ltp' package from packagegroup-core-tools-testapps.bb Mar 12 14:47:28 but i am kinda confused as to how packagegroup-core-tools-testapps even gets included in core-image-minimal Mar 12 14:48:05 I see it referenced in the PACKAGE_GROUP_tools-testapps variable in core-image.bbclass Mar 12 14:48:56 but I am not sure how it gets selected based on the inclusion of the "tools-testapps" string in EXTRA_IMAGE_FEATURES in local.conf Mar 12 14:50:40 I suppose I could just remove "tools-testapps" and create an image that manually includes all the other packages that packagegroup-core-tools-testapps RDEPENDS on Mar 12 15:01:20 YPTM: Welcome to the technical meeting, please let me know who's on the bridge Mar 12 15:01:39 YPTM: Laurentiu Palcu joined Mar 12 15:01:42 YPTM: Polk is here Mar 12 15:01:55 YPTM: Scott Rifenbark joined Mar 12 15:02:27 YPTM: Paul Eggleton is here Mar 12 15:03:17 YPTM: Nitin is on the call Mar 12 15:03:34 YPTM: jzhang's on Mar 12 15:03:41 YPTM ross just joind Mar 12 15:03:49 YPTM: Cristian joined Mar 12 15:03:53 YPTM: Opens from anyone Mar 12 15:05:26 YPTM: I got a cancellation? Mar 12 15:05:45 YPTM: will join telecon momentarily Mar 12 15:06:17 i got that, probably someone just removing themselves and it going all weird Mar 12 15:06:18 We got a cancellation, but it was not from Song Mar 12 15:07:20 could be the famous iPhone calendar issue? Mar 12 15:07:21 :-D Mar 12 15:08:50 YPTM: Corneliu joined Mar 12 15:09:31 Song_Liu: YPTM: Sean has joined the call Mar 12 15:10:25 cristianiorga: actually, now that I look closer, it looks like google notifications are to blame... Mar 12 15:18:48 quiet call today Mar 12 15:19:39 darknighte: intel-internal conference this week, which might be having an effect Mar 12 15:19:49 halstead: what changed that caused the delays? Mar 12 15:22:28 halstead: it was me :) (Paul) Mar 12 15:23:49 rburton__: thx Mar 12 15:24:28 halstead: makes sense. thx Mar 12 15:25:21 * darknighte hears crickets chirping Mar 12 15:25:25 bluelightning, Thaknks! And sorry to get the name wrong. Mar 12 15:25:34 halstead: no worries Mar 12 15:25:40 bye all Mar 12 15:25:40 I would share but my connection is so bad I don't thin Mar 12 15:25:53 ... I don't think you'd be able to hear me :) Mar 12 15:26:03 bluelightning: was i audible? this was hotel room on speaker Mar 12 15:26:16 rburton__: yeah Mar 12 15:26:41 rburton__: I'm just going by my attempt to use skype yesterday; I could hear the other end fine but they couldn't hear me Mar 12 15:27:05 yeah i wouldn't bother with skype with this wifi Mar 12 15:27:31 i just dialed the toll-free on the hotel room and if they charge, they have my corporate amex already. Mar 12 15:29:05 rp1: hi! with "stubbing" i mean not really running git, just producing a known result (mocking) Mar 12 16:12:02 hmm, I'm a bit confused. what's the point of all the virtual/libgl, etc provides if xserver-xorg explicitly depends on mesa-dri? Mar 12 16:12:14 * kergoth scratches head Mar 12 16:12:18 * kergoth searches old mailing list threads about this stuff Mar 12 16:13:42 xserver dri extension was depending on some headers only available with mesa-dir Mar 12 16:14:02 but that was in times when xserver was also separate with and without dri Mar 12 16:14:13 ah Mar 12 16:14:34 now with only one xserver and one mesa implementation it does not make much sense Mar 12 18:48:19 JaMa: martin, can you tell me why did you disable xcb and dbus in meta-qt5? Mar 12 18:48:49 you added this: reenable when we switch to "-qt-xcb" and build of src/platformsupport/glxconvenience/ is resolved to be before build of src/plugins/platforms/xcb/ Mar 12 18:51:34 i created a bbappend, "recipes-core/packagegroups/packagegroup-core-tools-testapps.bbappend" to modify RDEPENDS_${PN} but it does not show up when I run "bitbake-layers show-appends" Mar 12 19:48:42 better. Mar 12 20:26:35 question: I know how to replace the inittab for a specific machine (I already do it to remove serial tty from a box). Is there anything similar for a specific image or are machine specific changes more take what is given and sed modify it? Mar 12 20:30:32 inittab goes into a binary package. one package. that gets installed into both images. you'd have to have multiple recipes to emit multiple packages, or generate the inittab in the image recipe from shell Mar 12 20:33:45 afternoon walters Mar 12 20:33:59 heya rburton_ Mar 12 20:34:05 For machine, you can have a machine specific package. I see nothing for image specific package. Mar 12 20:34:22 you can't have image-specific recipes Mar 12 20:34:38 you can have multiple packages, and pick them per-image Mar 12 20:34:46 i.e. inittab-foo and inittab-bar Mar 12 20:34:47 rburton_, btw, do you read desktop-devel-list? It'd be awesome to get some alignment with the hacky "ptest" stuff going on now with what I'm proposing Mar 12 20:35:15 walters: reading ddl is about the extend of my gnome work these days ;) Mar 12 20:35:24 i'd been meaning to read them Mar 12 20:35:50 don't they solve different things at the moment - ptest is all about running it on the target, whereas this is standardised control of running and results? Mar 12 20:37:17 well, they're basically the same thing except i'm advocating a common upstream approach across modules (configure options, etc.) as well as ideally standardized reporting and metadata Mar 12 20:37:54 cool Mar 12 20:38:21 * rburton__ tries to remember bjorn's nick Mar 12 20:47:24 ok, if I am want an image that autologins a user on tty1 and runs a script automatically, would the best thing be to have a package that post-install modifies inittab to implement that or is there a cleaner way? Mar 12 20:50:08 rburton__: sysvinit-inittab is pulled in by sysvinit, I don't see a way to cleanly plug a different inittab into it, and nothing that wouldn't lose machine specific customizations. Mar 12 21:13:55 ftonello: platforms/xcb was using some symbols defined in glxconvenience before glxconvenience was built Mar 12 21:14:22 ftonello: that resulted in build issues, maybe it was resolved in 5.0.1, I haven't tried it with 5.0.1 yet Mar 12 21:23:25 JaMa: I'm trying 5.0.1.. I didn't get this error.. but I'm having some issues with egl support Mar 12 21:23:33 trying to fix those Mar 12 21:23:47 JaMa: do you have plans to work in qt5 again? I'm planning to start working on it Mar 12 21:29:02 yes see jansa/test branch Mar 12 21:29:06 that's the start Mar 12 21:29:17 but target recipes need to be fixed like this too Mar 12 21:29:25 and I wont merge that before all is fixeds Mar 12 21:30:13 but the jansa/test is older stuff Mar 12 21:30:37 hmm Mar 12 21:30:41 maybe not, lemme pull Mar 12 21:32:26 ok, you have 3 patches on top of master Mar 12 21:47:27 yes only first 2 are "finished" Mar 12 21:48:58 ok Mar 12 21:54:29 whenever i boot this yocto fs udev fnords the static dev I put on there Mar 13 00:18:34 is the D variable different in a native recipe? Mar 13 00:21:30 ${D} inside a do_install, in a native recipe, is basically returning ${D}${STAGING_DIR_NATIVE} Mar 13 02:22:31 halstead, ping **** ENDING LOGGING AT Wed Mar 13 02:59:58 2013