**** BEGIN LOGGING AT Tue Jan 18 02:59:56 2022 Jan 18 08:32:18 Tofe: setting PARALLEL_MAKE seems to work for qtwebengine builds for me, I've used PARALLEL_MAKE:pn-qtwebengine = "-j 10" in local.conf and see just 10 processes running (instead of 70) Jan 18 08:44:58 martin 2205582 0.6 0.1 208800 205640 ? S 09:26 0:07 ninja -j 10 -C /OE/build/luneos-honister/webos-ports/tmp-glibc/work/core2-64-webos-linux/qtwebengine/5.15.4+gitAUTOINC+dd7f7a9166_7aad021f2e-r0/build/src/core/release QtWebEngineCore Jan 18 08:49:35 JaMa: yes, after some investigation, this is a pebcak issue: I didn't re-run do_configure of qtwebengine after changing PARALLEL_MAKE ... Jan 18 08:50:33 Herrie: I now have issues on pinephone when building luneos-dev-image.bb:do_image_wic : install: cannot stat '/media/LuneOS/testing/webos-ports/tmp-glibc/deploy/images/pinephone/initramfs-uboot-image-pinephone.rootfs.cpio.gz.u-boot': No such file or directory Jan 18 08:50:55 something isn't deployed anymore, or something like that Jan 18 08:51:19 and that was the issue? strange I would expect NINJAFLAGS to be respected in do_compile (not persisted with whatever value was there in do_configure), but maybe qmake generates the value to some file Jan 18 08:51:55 also it's a bit more than 10 processes as each x86_64-webos-linux-g++ spawns cc1plus and as Jan 18 08:52:03 JaMa: looks like it was the issue yes Jan 18 08:52:29 but I'm not going to restart a qtwebengine build just to double-check :p Jan 18 08:52:30 ok, will grep for -j value after running just do_configure Jan 18 08:53:35 Herrie: there is no ".rootfs" in the initramfs u-boot image generated on pinephone Jan 18 08:57:19 fix PR'ed Jan 18 09:08:08 strange, didn't fail for me, I guess I have the fix already in my branch (oe-core/jansa/master) as https://github.com/webOS-ports/meta-webos-ports/commit/4e809241816354edd4ede4a2b1fa430eadc9539c Jan 18 09:11:14 PARALLEL_MAKE http://dpaste.com//C4KWHTYV3 doesn't seem to be persisted, not sure what's going on Jan 18 09:11:44 JaMa: ah, do you mean my PR is incorrect? Jan 18 09:12:50 I don't know, I'm just surprised I haven't seen it in local build, but I do have some extra changes in this area in oe-core branch I'm using Jan 18 09:13:41 https://git.openembedded.org/openembedded-core-contrib/log/?h=jansa/artifacts (the top 2 commits) Jan 18 09:15:08 JaMa: ok, yes hard to say :) Jan 18 10:58:20 Tofe: Well my PR was different, JaMa changed something that he said worked in his local build ;) Jan 18 10:58:36 Because I PR-ed without rootfs Jan 18 10:59:32 It would be good to decide what is the right solution :) Should the suffix be there? Jan 18 11:58:41 I was changing just the PR for meta-webos-ports, not meta-pine64 Jan 18 11:59:42 I guess I'm not buiding wic IMAGE_FSTYPE for pinephone as I don't have any initramfs images at all Jan 18 12:00:01 webos-ports$ grep FSTYPE conf/local.conf Jan 18 12:00:01 IMAGE_FSTYPES_forcevariable = "tar.gz" Jan 18 12:00:01 IMAGE_FSTYPES_forcevariable_qemuall = "wic.vmdk ext4" Jan 18 12:00:01 IMAGE_FSTYPES_forcevariable_rpi = "rpi-sdimg.gz" Jan 18 12:00:01 #IMAGE_FSTYPES_DEBUGFS = "tar.gz" Jan 18 12:00:08 that explains why I wasn't seeing the issue Jan 18 12:01:03 conf/machine/pinephone.conf:IMAGE_FSTYPES += "wic.gz wic.bmap" Jan 18 12:01:03 conf/machine/pinephonepro.conf:IMAGE_FSTYPES += "wic.gz wic.bmap" Jan 18 12:01:03 recipes-core/images/initramfs-uboot-image.bb:IMAGE_FSTYPES:forcevariable = "cpio.gz.u-boot" Jan 18 12:05:56 still strange as local.conf had old override syntax, so the forcevariable shouldn't work Jan 18 12:08:11 -rw-r--r-- 4 martin martin 3.3M Jan 14 20:41 tmp-glibc/deploy/images/pinephonepro/initramfs-uboot-image-pinephonepro-kirkstone-jama.cpio.gz.u-boot Jan 18 12:08:14 -rw-r--r-- 4 martin martin 625 Jan 14 20:29 tmp-glibc/deploy/images/pinephonepro/initramfs-uboot-image-pinephonepro-kirkstone-jama.manifest Jan 18 12:08:17 -rw-r--r-- 4 martin martin 399K Jan 14 20:29 tmp-glibc/deploy/images/pinephonepro/initramfs-uboot-image-pinephonepro-kirkstone-jama.testdata.json Jan 18 12:08:20 -rw-r--r-- 4 martin martin 3.3M Jan 14 20:41 tmp-glibc/deploy/images/pinephonepro/initramfs-uboot-image-pinephonepro.rootfs.cpio.gz.u-boot Jan 18 12:08:23 -rw-r--r-- 4 martin martin 625 Jan 14 20:29 tmp-glibc/deploy/images/pinephonepro/initramfs-uboot-image-pinephonepro.rootfs.manifest Jan 18 12:08:26 -rw-r--r-- 4 martin martin 399K Jan 14 20:29 tmp-glibc/deploy/images/pinephonepro/initramfs-uboot-image-pinephonepro.rootfs.testdata.json Jan 18 13:03:07 JaMa: you seem to have both variants ? Jan 18 13:08:26 the 2nd set are hardlinks without the version suffix (set to "kirkstone-jama" for me) Jan 18 13:09:54 from IMAGE_VERSION_SUFFIX Jan 18 13:10:52 WEBOS_DISTRO_BUILD_ID="kirkstone-jama" Jan 18 13:12:46 you probably have a datetime there instead Jan 18 13:12:56 and jenkins jobs used to have BUILD_NUMBER Jan 18 19:57:00 Tofe: cannot reproduce your initramfs issue, I've updated https://github.com/webOS-ports/meta-pine64-luneos/pull/28 Jan 18 20:29:59 JaMa: strange. I began from a clean tmp-work folder, with just sstate already there **** ENDING LOGGING AT Wed Jan 19 02:59:57 2022