**** BEGIN LOGGING AT Fri Feb 20 02:59:59 2015 Feb 20 07:59:09 good morning Feb 20 08:55:05 JaMa: hi. I plan to send 3.19 for meta-handheld this weekend Feb 20 08:55:55 JaMa: go ahead and fix meta-initramfs parse error Feb 20 08:56:13 thx Feb 20 09:00:54 ok, I've also renamed meta-handheld bbappends but still waiting for build test Feb 20 09:01:06 probably won't have time to update the patches.. will leave that to you Feb 20 09:01:21 and don't worry it's only in oe-core/master-next now Feb 20 09:05:12 imagine my laptop-build failed do_fetch yesterday night...I had seen it otherwise.. Feb 20 09:05:41 btw, do you think TMPDIR fits 13GB with rm_work nowadays? Feb 20 09:05:54 for these small builds I mean Feb 20 09:06:33 I gave 80% to tmpfs Feb 20 09:11:49 depends on how big are your images, you can still build in "steps" to force rm_work e.g. after building toolchain Feb 20 09:12:58 main target is core-image-base Feb 20 09:13:12 right, I'll build tc as first Feb 20 09:15:28 is the BB_SCHEDULER issue still actual? Feb 20 09:29:30 what issue? Feb 20 09:30:10 "rm_work.bbclass sets BB_SCHEDULER ?= "completion", but it doesn't Feb 20 09:30:11 work as I would expect," Feb 20 09:30:25 "it still keeps many build directories around, Feb 20 09:30:25 before rm_work task is executed, simple solution for lazy people is to Feb 20 09:30:25 build in "steps" Feb 20 09:31:11 that was 2yrs ago but I don't see related patches for rm_work.bbclass Feb 20 09:32:44 or maybe "rm_work: Speed up rootfs/populate_sdk removal" had some impact?? Feb 20 09:33:12 [YOCTO #6413] Feb 20 09:49:27 ah right, it's still an issue Feb 20 09:54:57 whereever the problem is, I don't think it is in rm_work Feb 20 09:55:32 yes it's in bitbake scheduler, that it should schedule more agressively for completion Feb 20 10:10:24 hi guys. I have a device where there are two PCIe slots. I connected my mini PCIe wireless card into one of the slot but it doesn't seem to be recognized. Feb 20 10:10:35 here is the result of lspci: http://paste.ofcode.org/3amFtuvums2Qp8uvLQxyQma Feb 20 10:10:52 could anyone tell me if this could be a hardware failure? Feb 20 10:11:05 or how can I be sure that it is a hardware problem? Feb 20 10:12:40 and here is the -vvv output of lspci: http://paste.ofcode.org/wFHdpMAw5JdkxK78aycUMM Feb 20 10:30:55 dev1: you'd better ask to any AMD channel or ML Feb 20 10:31:07 mckoan. what is ML? Feb 20 10:31:48 dev1: Mailing List Feb 20 10:31:56 dev1: are you using OE? Feb 20 10:32:03 ahh ok. Thanks for your suggestion. Feb 20 10:32:07 yep Feb 20 10:59:18 when building sdk packages, how come it's not separated per machine? for example, if I build for 2 arm-based machines: MACHINE=m1 bitbake -c populate_sdk myimg && MACHINE=m2 bitbake -c populate_sdk myimg, shouldn't there be 2 sdks coming out? I just see 1 since m1 and m2 both are ARMv7-based. Shouldn't there be 1 SDK for each target, in case one of the machines add extra packages to the image (e.g m1 version of Feb 20 10:59:19 myimg might be different from m2 version of myimg)? Feb 20 11:08:15 these are Application Development SDKs Feb 20 11:08:20 not Platform SDKs Feb 20 11:08:46 but why are the generated SDKs different then? Feb 20 11:11:39 they should be names by architecture IIRC Feb 20 22:18:09 fyi line now Feb 20 22:18:10 https://www.youtube.com/watch?v=rzKI0ASB_Js&feature=youtu.be Feb 20 22:30:11 is base_bindir /bin ? What is /usr/bin ? Feb 20 22:30:58 bindir Feb 20 22:31:43 kergoth: thanks! Feb 20 22:31:48 np **** ENDING LOGGING AT Sat Feb 21 02:59:59 2015