**** BEGIN LOGGING AT Thu Jan 08 02:59:59 2015 Jan 08 04:57:31 hi guys Jan 08 04:57:35 I am in a dillemma. I have a board with 32M nor flash(no partition) and currently the uboot and uimage are sitting in the flash. My file system - rootfs also very small. May I know how can I achieve(put rootfs into flash) this? Jan 08 06:57:31 hi if a recipe, say a yocto/meta-intel/meta-isg/meta-valleyisland/recipes-multimedia/helloworld/helloworld_1.0.bb Jan 08 06:57:37 when i do a Jan 08 06:57:59 bitbake -b ../meta-intel/meta-isg/meta-valleyisland/recipes-multimedia/helloworld/helloworld_1.0.bb Jan 08 06:58:01 it works Jan 08 06:58:27 but when i do a core-img-sato Jan 08 06:58:36 it doesn't seem to pick it up Jan 08 06:58:54 what should i pay attention to, in order to find the possible mistake? Jan 08 07:21:51 yctuser: well if the image you are building does not want to install the package from your new recipe, then it does not get "picked up" Jan 08 07:22:25 yctuser: IMAGE_INSTALL is your friend here. respectively, creating a new (derived) image that installs what you want Jan 08 07:27:55 LetoThe2nd: how can I check to make sure my recipe is included in my image build? Jan 08 07:29:24 yctuser_: like i said, make sure the IMAGE_INSTALL variable is set according to your wishes Jan 08 07:29:37 as a graphical representation, you can look up http://www.yoctoproject.org/docs/current/ref-manual/ref-manual.html#usingpoky-debugging-dependencies Jan 08 07:30:06 thanks! sorry i was disconnected, didn't see your earlier message Jan 08 08:16:21 Hi, i'm trying to put the image on usb using mkefidisk.sh but it always complain with ERROR: Device /dev/sdb does not exist or is not writable Jan 08 08:16:54 the usb stick is fine and i tried with both mounted/unmounted/fat32/ext3 Jan 08 08:25:36 ah, i needed to be sudo Jan 08 08:25:37 good morning Jan 08 08:28:25 good morning Jan 08 08:49:24 i flashed the image on usb, but my bootloader complains "this is not a bootable disk". the board i have has atome3825 Jan 08 08:49:32 with BIOS of American Megatrends Jan 08 09:00:02 i'm wondering if having a efi image on usb is to playing well with the BIOS Jan 08 09:09:59 it has Aptio® UEFI 2.x firmware. Jan 08 09:32:43 I build an image with x11-base and no dm, I want to calibrate my touchscreen but the matchbox-terminal pops up in front of the calibration wizard screen. Running xinput_calibrator from tty shows me "Unable to connect to X server", how can I calibrate my touchscreen? Jan 08 09:42:43 morning all Jan 08 10:19:36 Ugh, I can't seem to get my touchscreen to work with xinput_calibrator, using tslib on a linuxfb works perfectly. What am I missing? Jan 08 10:27:20 JaMa: is that a yocto-powered LG smartwatch? Jan 08 10:28:26 the webos smartwatch? Jan 08 10:29:00 http://www.theverge.com/2015/1/7/7511603/lg-putting-web-os-on-new-smartwatch-ces-2015 Jan 08 10:31:31 the web is full of how to start docker, noone says anthing about how to stop it. Jan 08 10:58:54 ndec: "open webos" is the reboot which is yocto instead of oe-classic based,iirc Jan 08 10:59:10 jama is seriously failing at self-promotion :) Jan 08 11:00:20 rburton: are all open webos builds based on oe? i thought the webos-port stuff was itself a reboot of open webos, built with oe Jan 08 11:00:40 ndec: see, this is why i wanted jama to confirm/deny :) Jan 08 11:02:02 JaMa: ^ Jan 08 11:03:27 Open webOS as well as webos-ports (renamed to LuneOS) are all built with OE.. then there are some proprietary layers on top of that for other flavors of webOS like systems Jan 08 11:03:42 ah. nice. Jan 08 11:03:56 keep in mind that it intentionally doesn't answer Ross's original question :) Jan 08 11:04:54 JaMa: bah! Jan 08 11:05:36 luneos is the "community" one? Jan 08 11:08:24 yes Jan 08 11:08:51 there is also meta-luneui now, for people who need just simple qt5 based UI for other projects Jan 08 11:09:51 JaMa: looking at the openwebos "building for desktop" page, i really hope the watch isn't running webos on ubuntu :) Jan 08 11:09:54 JaMa: cool... is that by chance in the layer index? ;) Jan 08 11:24:33 rburton, what is Yocto? Jan 08 11:24:45 Crofton|work: don't start that again Jan 08 11:24:56 also, I do not think open webos could be described as oe-classic :) Jan 08 11:24:57 "powered by tools developed by the yocto project community" Jan 08 11:25:03 much better Jan 08 11:25:15 the last time i looked at the webos layers it was pretty much old-school oe Jan 08 11:25:19 but that was a loooong time ago Jan 08 11:25:46 apparently it has some interesting hacks :) Jan 08 11:25:53 https://fosdem.org/2015/schedule/event/tizen_iot_security/ Jan 08 11:26:05 embedded track at fosdem slowly appearing Jan 08 11:30:09 Crofton|work: that talk is basically "smack rocks" Jan 08 11:31:48 I can see the talks that might be on the schedule, here are a few more that use the word "Yocto" Jan 08 11:31:55 heh Jan 08 11:32:02 I read "smart rocks" Jan 08 11:32:19 haha Jan 08 11:32:24 connected rocks! Jan 08 11:39:40 bluelightning: not yet, because it was merged to dizzy just today :) Jan 08 11:39:58 JaMa: ah ok, fair enough :) Jan 08 11:58:07 Hi folks Jan 08 11:58:14 how are you all doing ? Jan 08 11:58:32 I would like to get involved in yocto design. Jan 08 11:58:44 Could someone please assist on how to get stated Jan 08 11:59:37 design of what? Jan 08 12:00:16 rburton:any of the yocto sub projects Jan 08 12:01:53 i guess pick a project and contain the people working on it Jan 08 12:02:23 presumably you mean visual design for web pages / applications? Jan 08 12:02:48 Am I doing something wrong or is openembedded-core/scripts/buildhistory-diff very slow for big diffs? Jan 08 12:03:13 JaMa: i've certainly see it sit for a while, but how long is very slow? Jan 08 12:03:17 JaMa: is your repo large? Jan 08 12:03:18 git diff size: 439777 buildhistory-diff Jan 08 12:03:32 but buildhistory-diff doesn't show any output in 30 minutes Jan 08 12:03:50 980M buildhistory Jan 08 12:04:05 relatively small comared to my other buildhistory repos :) Jan 08 12:04:27 openembedded-core/scripts/buildhistory-diff -v 8cae983119a7eda5473711d545769b92f71c739f 7f3fba215f6d72dd5a0505882973a0a5c8a917c6 Jan 08 12:04:29 hmm... I can't immediately explain that Jan 08 12:04:39 certainly 30 minutes is egregiously long Jan 08 12:04:43 ok, will try to debug Jan 08 12:10:25 bluelightning: is it possible to edit layer in layer index to say that it's meta-webos-ports for dizzy release and older and for master it's meta-luneui + meta-luneos? Jan 08 12:10:53 or do I have to rename/delete meta-webos-ports if I don't want to show it for master branch? Jan 08 12:11:23 yesterday I was searching for meta-cedartrail and I guess it was also lost because of some changes like this Jan 08 12:11:32 JaMa: well, a layer can disappear for a branch if there's no branch record for that branch, but at one time at least I assumed all layers would have a master branch Jan 08 12:12:03 I did do some fixes a while back to try to make it possible to not have one, I don't remember how far I got with it Jan 08 12:12:18 ok, so it's even more complicated case, because the repository and branch still exists Jan 08 12:12:20 certainly the deleting needs to be done from the admin interface, but you may have rights to do that actually Jan 08 12:12:30 yes Jan 08 12:12:34 but newer dizzy revision and master branch contain 2 layers in it Jan 08 12:13:20 ok, they would need to be created as new layers, but that in itself should be OK since they would not have branch records for the old branches they aren't present in Jan 08 12:15:05 cannot be renamed? Jan 08 12:15:10 rburton : yes I would love to start with web pages and proceed to application Jan 08 12:15:19 just depends on what needs most work so I can help Jan 08 12:15:38 http://layers.openembedded.org/layerindex/branch/master/layer/meta-luneos/ Jan 08 12:16:03 ah it renames it for all versions including e.g. http://layers.openembedded.org/layerindex/branch/daisy/layer/meta-luneos/ Jan 08 12:24:25 ok, created 2 new layers and published them Jan 08 12:26:17 hi guys, does yocto have a policy for updates on stable releases? I mean, I'm on the dizzy branch and it continues to receive updates. Jan 08 12:26:29 is there any page that explains that? Jan 08 12:56:35 jwendell: there is indeed: https://wiki.yoctoproject.org/wiki/Stable_branch_maintenance Jan 08 13:00:01 JaMa: FWIW I do have a long-term todo list item to add this kind of splitting functionality, but there's nothing like that atm Jan 08 13:07:52 bluelightning, thanks! Jan 08 13:12:42 Hello, I'm having some trouble trying to build meta-ti with meta-ivi. The meta-ivi is my problem : I downloaded the layer (and dependencies); added the layer folder into bblayers.conf and path to recipes in layres.conf. When i try to build (bitbake intrepid-image) i'm getting the following error : ERROR: Nothing PROVIDES 'intrepid-image'. Any help would be appreciated. Thank you. Jan 08 13:14:54 dpavel: because exist intrepid-image.bb in meta-ivi/recipes-yocto-ivi/images the problem must be your bblayers.conf Jan 08 13:18:29 Line added in bblayers.conf : path_to_poky/meta-ivi/meta-ivi Jan 08 13:19:04 dpavel: I guess it should be path_to_poky/meta-ivi Jan 08 13:19:52 dpavel: no, path_to_poky/meta-ivi/meta-ivi is ok Jan 08 13:20:18 mckoan: and line added in layres.conf Jan 08 13:20:36 dpavel: which line? Jan 08 13:20:37 dpavel: so you git cloned meta-ivi while the current dir was poky? Jan 08 13:21:03 mckoan : local.conf * (sorry) : BBMASK="meta-ivi/meta-ivi/recipes-yocto-ivi" Jan 08 13:21:13 void-dev: yes Jan 08 13:21:46 dpavel: ok, so I guess you found the issue Jan 08 13:22:27 void-dev: no, I was just posting the line I added in local.conf and still doesn't work Jan 08 13:24:09 dpavel: the BBMASK most do the opposite you want Jan 08 13:24:23 dpavel: remove BBMASK line Jan 08 13:25:33 After removing : ERROR: Please ensure that your setting of VIRTUAL-RUNTIME_init_manager (systemd) matches the entries enabled in DISTRO_FEATURES Jan 08 13:25:33 ERROR: Failed to parse recipe: /home/daniel/Karanka/yocto/poky/meta-ivi/meta-ivi/recipes-yocto-ivi/packagegroups/packagegroup-core-boot-genivi.bb Jan 08 13:26:38 dpavel: did you set the DISTRO to poky-ivi-systemd Jan 08 13:27:23 void-dev: nope, how can I do that ? Jan 08 13:27:36 dpavel: in your local.conf Jan 08 13:28:43 dpavel: if you would have clones the layers side by side, and not inside poky, you could follow the README file Jan 08 13:29:10 dpavel: but in your setup the provided TEMPLATECONF won't work Jan 08 13:29:31 it started building, thank you for the help Jan 08 13:29:48 just added distro and it's all good. Jan 08 13:30:38 dpavel: you may want to add INCOMPATIBLE_LICENSE ?= "GPLv3" too to your local.conf as GENIVI does not like GPLv3 licensed stuff Jan 08 13:32:39 void-dev: ok, thank you. Jan 08 13:32:47 dpavel: np Jan 08 14:51:32 good morning Jan 08 14:51:43 https://www.yoctoproject.org/downloads -> YP Core - Daisy 1.5.4 -> Do you report such bugs? Jan 08 16:35:09 hello:) Jan 08 16:35:22 there is a question I have. Jan 08 16:36:03 Which script is responsible for creating/maintaining GRUB entry after sato installation from USB to HDD? Jan 08 16:38:40 arcomivi: probably ./meta/recipes-core/initrdscripts/files/init-install-efi.sh or ./meta/recipes-core/initrdscripts/files/init-install.sh are what you are looking for. Jan 08 16:38:48 What i mean exactely is: 1) deploy(dd) .hddimg to USB 2) boot device from USB and install image to sda 3) start device to boot from sda, but there has to be a script creating a GRUB entry during installation Jan 08 16:38:49 . Jan 08 16:39:06 dmoseley: yes, are they run on default? Jan 08 16:39:32 dmoseley: at the moment after the installation there is: no bootable device Jan 08 16:40:18 They are run when you select the "install" option from the USB boot. They are included in initrd and install grub to the HDD. Jan 08 16:41:25 hmm, if they are run, what would be another possible problem getting this error? Jan 08 16:41:31 lpapp: I got Beth to fix that, thanks for the report Jan 08 16:42:09 arcomivi: Hard to say. Depending on your target the bootloader may need to be setup to boot from the HDD. Jan 08 16:42:11 bluelightning: thanks. Jan 08 16:43:15 Hi all -- I'm working on a Yocto managed project using the 3.10.17 kernel on a freescale IMX6Q. I'm interested in making changes to the system's internal pin mux. Where is the pin mux configured? Does Yocto/the BSP own that, or is it a general linux matter? Jan 08 16:43:36 dmoseley: ok, well, target is simple atom cpu x86, installation asks for whether to install to /dev/sda1. After reboot - no bootable device. I am trying to figure out where to start. Jan 08 16:44:25 arcomivi: Initially I would suspect something in the target BIOS setup. Jan 08 16:45:10 dmoseley: what exactely do you mean by BIOS setup? Boot sequence maybe? Jan 08 16:46:50 arcomivi: I am not familiar with any of the atom boards but if it's a standard PC style architecture there is likely a BIOS setup utility where you can specify boot devices. Is it a standard BIOS based system or is it using some other boot loader? Jan 08 16:48:20 dmoseley: ok, BIOS already checked. No problems there, booting from HDD is set up. Maybe there is a problem with those init scripts, not beeing run properly. Any other hints from experience? Jan 08 16:49:35 It could well be a bug in the scripts. Nothing specific comes to mind though. Jan 08 16:50:16 dmoseley: thanks a lot, that helped me to sort things out that have to be checked. Jan 08 16:50:26 Good luck. Jan 08 16:50:48 i will need it ;) I am trying to install an image for 3 days now ;) Jan 08 16:55:57 <_jmleo> Hi ! I am using ssh to connect to my board (i.MX6 based) and I need about 5-6 seconds to connect on it. Is this a problem you already saw ? Jan 08 16:58:32 maybe its resolving your ip via dns for the logs, and that times out? Jan 08 18:29:23 is there a reason PREFERRED_PROVIDER_java2-runtime="oracle-jse-jre-x86-64" in my local.conf has no effect? Jan 08 18:38:57 Hi Jan 08 18:39:09 I have added metacity in yocto Jan 08 18:39:38 but it schemas is looking for /opt/gnome/metacity path which is not present in my image Jan 08 19:30:09 if I'm sending a patch for a recipe that includes a backported patch from more recent upstream version, besides an Upstream-Status: Backport, does the patch also need a Signed-off-by? Jan 08 19:47:30 is there a recipe to build all the *-native recipes? Jan 08 19:50:22 Aethenelle: unlikely. you could certainly script it via processing the output of bitbake -s Jan 08 19:50:45 that's what i'm working on now... just hoped there was a better way... Jan 08 19:51:31 what's the purpose of this? Jan 08 19:51:38 prepopulating a native sstate cache? Jan 08 19:55:02 basically... i have a bunch of archs i need to run tests on Jan 08 20:08:58 there used to be a packagegroup that was most native packages, but it wasn't anything like complete Jan 08 20:10:23 plus what 'all' means depends on your configuration and enabled layers. i could see having one covering whats in oe-core, perhaps **** ENDING LOGGING AT Fri Jan 09 02:59:58 2015