5390 2013-10-28 11:20:15 +0000 linux-yocto_3.10.bb do_populate_sysroot_setscene fails and retriggers a re-run of most tasks 2013-12-05 16:00:45 +0000 1 1 7 Build System, Metadata & Runtime OE-Core kernel 1.6 x86 Multiple RESOLVED FIXED Medium normal 1.5.1 1 sstncr bruce.ashfield bruce.ashfield richard.purdie tom.zanussi --- --- No (bug/feature does not impact docs) oldest_to_newest 38197 0 sstncr 2013-10-28 11:20:15 +0000 Lately I've been getting this warning: (which I've completely ignored until I noticed that almost all tasks get re-run instead of being used from sstate): WARNING: Setscene task 95 (/mnt/back/yocto/poky/meta/recipes-kernel/linux/linux-yocto_3.10.bb, do_populate_sysroot_setscene) failed with exit code '1' - real task will be run instead Log shows: DEBUG: Shell function sstate_unpack_package finished DEBUG: Staging files from /mnt/back/yocto/poky/build/tmp/work/qemux86_64-poky-linux/linux-yocto/3.10.11+gitAUTOINC+452f0679ea_3f6c8243ed-r0/sysroot-destdir to /mnt/back/yocto/poky/build/tmp/sysroots/qemux86-64 DEBUG: Executing shell function kernelscripts_sstate_postinst NOTE: make scripts make: x86_64-poky-linux-gcc: Command not found HOSTCC scripts/basic/fixdep HOSTCC scripts/kconfig/conf.o SHIPPED scripts/kconfig/zconf.tab.c SHIPPED scripts/kconfig/zconf.lex.c SHIPPED scripts/kconfig/zconf.hash.c HOSTCC scripts/kconfig/zconf.tab.o HOSTLD scripts/kconfig/conf scripts/kconfig/conf --silentoldconfig Kconfig make: x86_64-poky-linux-gcc: Command not found CC scripts/mod/empty.o /bin/sh: x86_64-poky-linux-gcc: command not found make[2]: *** [scripts/mod/empty.o] Error 127 make[1]: *** [scripts/mod] Error 2 make: *** [scripts] Error 2 ERROR: oe_runmake failed WARNING: /mnt/back/yocto/poky/build/tmp/work/qemux86_64-poky-linux/linux-yocto/3.10.11+gitAUTOINC+452f0679ea_3f6c8243ed-r0/temp/run.kernelscripts_sstate_postinst.21098:1 exit 1 from [ "populate_sysroot_setscene" = "populate_sysroot" -o "populate_sysroot_setscene" = "populate_sysroot_setscene" ] DEBUG: Python function do_populate_sysroot_setscene finished There's something horribly broken here. Build Configuration: BB_VERSION = "1.21.0" BUILD_SYS = "x86_64-linux" NATIVELSBSTRING = "Fedora-19" TARGET_SYS = "x86_64-poky-linux" MACHINE = "qemux86-64" DISTRO = "poky" DISTRO_VERSION = "1.5+snapshot-20131028" TUNE_FEATURES = "m64" TARGET_FPU = "" meta meta-yocto meta-yocto-bsp = "master:78b91ab23d9856525fc7ac1bb8da2975382813bf" 38198 1 dvhart 2013-10-28 11:24:26 +0000 What was the bitbake target here? 38204 2 bruce.ashfield 2013-10-28 12:23:28 +0000 This is likely due to the modpost / scripts re-creation that I did just recently. Clearly the compiler isn't available in all cases, which is caused the problem. 38205 3 sstncr 2013-10-28 12:43:41 +0000 Ooups sorry about that, target was core-image-sato. Just to be sure it wasn't something with my build dir/sstate I did: bitbake virtual/kernel -ccleansstate && bitbake virtual-kernel && rm -rf tmp/ && bitbake core-image-sato and I got the same thing. 38275 4 bruce.ashfield 2013-10-29 18:02:29 +0000 (In reply to comment #3) > Ooups sorry about that, target was core-image-sato. > Just to be sure it wasn't something with my build dir/sstate I did: > bitbake virtual/kernel -ccleansstate && bitbake virtual-kernel && rm -rf > tmp/ && bitbake core-image-sato and I got the same thing. Which exactly log are you seeing the error ? I just built and rebuilt my qemux86-64 kernel and image here and I'm not seeing message/error in my logs. 38276 5 bruce.ashfield 2013-10-29 18:02:55 +0000 sending bug back for exact location of the log file. 38295 6 sstncr 2013-10-30 10:04:40 +0000 The warning shows up on bitbake's output. The log I pasted is from (in my case): tmp/work/qemux86_64-poky-linux/linux-yocto/3.10.11+gitAUTOINC+452f0679ea_3f6c8243ed-r0/temp/log.do_populate_sysroot_setscene 38301 7 bruce.ashfield 2013-10-30 12:31:35 +0000 (In reply to comment #6) > The warning shows up on bitbake's output. > The log I pasted is from (in my case): > tmp/work/qemux86_64-poky-linux/linux-yocto/3.10. > 11+gitAUTOINC+452f0679ea_3f6c8243ed-r0/temp/log.do_populate_sysroot_setscene I'm definitely not seeing this .. I would have noticed it in my output. So clearly, there's still something different between our configurations. 38304 8 sstncr 2013-10-30 13:16:36 +0000 Hmm, is there anything else you need, besides what's in the first comment? The contents of the log file don't appear in the output, only the WARNING Also AB has the same warning message: WARNING: Setscene task 168 (/home/pokybuild/yocto-autobuilder/yocto-slave/nightly-arm/build/meta/recipes-kernel/linux/linux-yocto_3.10.bb, do_populate_sysroot_setscene) failed with exit code '1' - real task will be run instead Build Configuration: BB_VERSION = "1.21.0" BUILD_SYS = "x86_64-linux" NATIVELSBSTRING = "Fedora-19" TARGET_SYS = "arm-poky-linux-gnueabi" MACHINE = "qemuarm" DISTRO = "poky" DISTRO_VERSION = "1.5+snapshot-20131030" TUNE_FEATURES = "armv5 thumb dsp" TARGET_FPU = "soft" meta meta-yocto meta-yocto-bsp = "stage/master_under_test:121c8d23a9322e3a8d1c36e25f8fa56f9ab08033" meta-qt3 = "master:b73552fb998fd30a01dbee5a172e432a16078222" http://autobuilder.yoctoproject.org:8011/builders/nightly-arm/builds/333/steps/Building%20Images/logs/stdio 39227 9 sstncr 2013-12-04 14:33:24 +0000 This seems to be fixed now in 1.6_M1.rc1 ( c51730aaa91066cfb51ad6ec2920fb4f9d9ea94c ) I think http://git.yoctoproject.org/cgit/cgit.cgi/poky/commit/?id=3f5e982107ce464ebc24ff4698f1e1863417427c was the commit that fixed it (with other commits before). 39251 10 richard.purdie 2013-12-05 16:00:45 +0000 This was resolved in master as Stefan mentions.