**** BEGIN LOGGING AT Fri Jun 28 02:59:57 2019 Jun 28 03:37:48 does yocto offer bsp for mikrotik routerboards? Jun 28 04:13:18 emfipp: I've never seen a BSP-layer for those :/ Jun 28 04:14:14 but is there a list of sbcs with two or more ethernet ports that yocto offers support? Jun 28 04:19:32 emfipp: So Yocto as a project only has a limited set of boards supported, then the community adds a lot of different machines by providing more meta-layers like meta-raspberrypi, meta-intel, meta-freescale etc. I don't know of such a list, but here's at least a list where you can search for machine names to try to find a layer that adds support: http://layers.openembedded.org/layerindex/branch/master/machines/ Jun 28 04:23:33 lots of imx6 Jun 28 04:23:34 no imx8 Jun 28 04:25:38 and edgerouter did not specify which model Jun 28 04:28:22 About the edgerouter there's more info in README.hardware of the meta-yocto-bsp layer: http://git.yoctoproject.org/cgit/cgit.cgi/meta-yocto/tree/meta-yocto-bsp/README.hardware Jun 28 04:37:05 just the lite variant then. too bad :( Jun 28 06:49:04 Is there a way to disable "update_gio_module_cache"? I want to deploy kernel-devsrc into an SDK but this fails when invoking the previos mentioned function Jun 28 06:59:00 guys ... i am looking for a simple Makefile.am which uses parallel make , so each c file to a .o then link to speed up compilation ? Jun 28 08:58:22 hello, what's the taks responsible of populating DEPLOY_DIR_IMAGE? Jun 28 08:58:55 qschulz: do_image_complete? Jun 28 08:59:28 I have a function in IMAGE_POSTPROCESS_COMMAND which is reading from it but it's missing files. Jun 28 09:03:59 qschulz: from DEPLOY_DIR_IMAGE? That is only written when the task completes due to sstate Jun 28 09:04:57 qschulz: you probably want IMGDEPLOYDIR which is its staging directory Jun 28 09:05:16 (files are staged there and then moved to DEPLOY_DIR_IMAGE when things complete) Jun 28 09:06:29 I've discovered that but I'm still looking for the "when things complete". I want to deepen my understanding of Yocto every time I have the opportunity :) Jun 28 09:09:54 I understand that this is normally done for packages in the deploy.bbclass Jun 28 09:10:45 Dear All, Jun 28 09:10:46 Maybe a bit strange question - I need to execute a script to build some image - the recipe with IMAGE_CMD_fooimage = .... will do the trick Jun 28 09:10:46 The problem is that I would need to set dependencies on the images (put into deploy dir) from two different MACHINES: Jun 28 09:10:46 do_image_fooimage[depends] += " foo-image:do_image_complete foo-image:do_image_complete" Jun 28 09:10:46 I could introduce new machine with MACHINEOVERRIDES to MACHINE1 and MACHINE2, but I'm not sure if this would work properly. Jun 28 09:12:36 qschulz: the key is do_image_complete[sstate-inputdirs] = "${IMGDEPLOYDIR}" do_image_complete[sstate-outputdirs] = "${DEPLOY_DIR_IMAGE}" Jun 28 09:12:51 RP: I was just about to write you that I found it :) Jun 28 09:12:54 RP: thanks! Jun 28 09:12:57 qschulz: that tells sstate to copy from inputdirs to outputdirs Jun 28 09:13:20 and now I can yell at people writing to DEPLOY_DIR_IMAGE :) Jun 28 09:13:30 qschulz: indeed :) Jun 28 09:14:40 RP: last thing, I don't think there is another way than reading from DEPLOY_DIR_IMAGE for getting for example U-Boot binary, or is there? Jun 28 09:15:05 from an image task I mean Jun 28 09:16:31 e.g. we create a fitimage out of kernel, initramfs, u-boot, and some other binaries, so currently we get everything from DEPLOY_DIR_IMAGE Jun 28 09:17:24 qschulz: you would read from there, yes, just not write to directly Jun 28 09:17:31 RP: understood! Jun 28 09:17:51 lukma: I'm pretty sure this is not possible. Jun 28 09:18:06 lukma: but let's say I've nothing to back it up :) Jun 28 09:20:12 lukma: for two different machines you need multiconfig Jun 28 09:21:18 lukma: you can then use mcdepends and either have a third config depending on the other two or a second image for MACHINE1/2 doing it Jun 28 09:24:50 rburton: a glance at the binconfig branch says it looks ok. I'm not sure the crossscrips native thing is a bug as such though, we just did it that way to avoid the temp file indirection Jun 28 09:25:34 shipping two copies in the sysroot is 'wrong' and leads to stuff like tcl-native poking around on its own doing stupid things Jun 28 09:27:13 RP: Ok I will investigate this issue (to avoid some "strange" dancing with MACHINE OVERRIDES) Jun 28 09:33:08 is setting PTEST_ENABLED_pn-glib-2.0 = "0" in the conf file the right way to disable glib's ptest? doesn't seem to work, the test is still running (and failing) Jun 28 09:39:15 is it perhaps libglib-2.0 rather than glib-2.0? the recipe uses the latter, but the package name is libglib-2.0-ptest-2.58.0-r0.aarch64.rpm Jun 28 09:53:40 did you rebuild glib? Jun 28 09:59:18 lukma: I'm not sure you could do what you want with MACHINE_OVERRIDES... Jun 28 10:10:12 yes, it's a build from scratch (CI server) Jun 28 13:31:12 RP, rburton I am going on holiday for the entire July :) Jun 28 13:31:20 back on the 28th or so Jun 28 13:31:48 a lot of it will be cabin by the finnish lake, so email will work if needed, but not irc I guess Jun 28 13:42:53 kanavin: sounds good, enjoy. Wish I could get away with doing that! :) Jun 28 13:46:35 Hmm, I think we might have to cut back the reproducible build test to comparing two clean builds without sstate instead of one with sstate and one without Jun 28 13:51:32 RP: you are working from a seafront home :) Jun 28 13:53:01 kanavin: true I guess :) Jun 28 14:15:53 kanavin: say hello to the mosquitos ;-) Jun 28 14:17:02 remember that God loved mosquitos so much, that he created men in order to feed them :-) Jun 28 14:17:37 mardy: I've covered in bites from a mountain biking trip last week :( Jun 28 14:56:25 mardy, I am already terrified by that :) Jun 28 14:56:37 anyway, I'm off, see you all later Jun 28 18:53:10 clear Jun 28 21:04:23 what is the name of the recipe for the wifi connection manager provided in matchbox? Jun 28 21:41:48 Has anyone successfully compiled a User-Mode Linux kernel using Yocto? Jun 28 21:42:14 i have compiled a loadable kernel module **** ENDING LOGGING AT Sat Jun 29 02:59:57 2019