2503 2012-05-25 19:36:37 +0000 task-base-extended depends on libc6 >=2.15, 2.13 is built 2012-06-18 22:02:55 +0000 1 1 7 Build System, Metadata & Runtime OE-Core core 1.2 x86 Multiple RESOLVED WORKSFORME Proposed patch invalid, need method to reproduce High major 1.2.1 1 dvhart richard.purdie meta.mr.watcher meta.watcher --- --- --- oldest_to_newest 21625 0 dvhart 2012-05-25 19:36:37 +0000 When building core-image-base for fri2 using denzil-next dvhart@rage:~/source/poky [denzil-next] $ git rev-parse HEAD 7d7b305293d52a7f6ec732b83d4da627fb11b417 with the following additional packages: EXTRA_IMAGE_FEATURES = "debug-tweaks" CORE_IMAGE_EXTRA_INSTALL += "tcpdump ethtool minicom setserial parted strace ofono-tests i2c-tools alsa-utils" CORE_IMAGE_EXTRA_INSTALL += "python-pygobject" I get the following error: | Collected errors: | * satisfy_dependencies_for: Cannot satisfy the following dependencies for task-base-extended: | * libc6 (>= 2.15) * libc6 (>= 2.15) * | * opkg_install_cmd: Cannot install package task-base-extended. NOTE: package core-image-base-1.0-r0: task do_rootfs: Failed ERROR: Task 8 (/home/dvhart/source/poky/meta/recipes-core/images/core-image-base.bb, do_rootfs) failed with exit code '1' Cleaning eglibc results in: $ bitbake -c cleansstate eglibc Loading cache: 100% |###############################################################################| ETA: 00:00:00 Loaded 1118 entries from dependency cache. Parsing recipes: 100% |#############################################################################| Time: 00:00:00 Parsing of 846 .bb files complete (841 cached, 5 parsed). 1122 targets, 23 skipped, 0 masked, 0 errors. WARNING: No recipes available for: /home/dvhart/source/poky/layers/poky-extras/meta-kernel-dev/recipes-kernel/linux/linux-yocto-rt_2.6.34.bbappend OE Build Configuration: BB_VERSION = "1.15.2" TARGET_ARCH = "i586" TARGET_OS = "linux" MACHINE = "fri2" DISTRO = "poky" DISTRO_VERSION = "1.2" TUNE_FEATURES = "m32 core2" TARGET_FPU = "" meta meta-yocto meta-skeleton = "denzil-next:7d7b305293d52a7f6ec732b83d4da627fb11b417" meta-intel meta-fri2 = "dvhart/fri2:25231166c509747acbf21fda68d1259917b1bfe8" meta-kernel-dev = "master:b05f0787abd3c32eb64c6e4215d0f4f34c16f1d0" NOTE: Resolving any missing task queue dependencies NOTE: Preparing runqueue NOTE: Executing RunQueue Tasks NOTE: Running task 1 of 2 (ID: 0, /home/dvhart/source/poky/meta/recipes-core/eglibc/eglibc_2.13.bb, do_clean) NOTE: package eglibc-2.13-r26+svnr15508: task do_clean: Started NOTE: package eglibc-2.13-r26+svnr15508: task do_clean: Succeeded NOTE: Running task 2 of 2 (ID: 1, /home/dvhart/source/poky/meta/recipes-core/eglibc/eglibc_2.13.bb, do_cleansstate) NOTE: package eglibc-2.13-r26+svnr15508: task do_cleansstate: Started NOTE: package eglibc-2.13-r26+svnr15508: task do_cleansstate: Succeeded NOTE: Tasks Summary: Attempted 2 tasks of which 0 didn't need to be rerun and all succeeded. Summary: There was 1 WARNING message shown. Which shows eglibc is building version 2.13 21626 1 dvhart 2012-05-25 19:43:17 +0000 Something appears to be changing the PREFERRED_VERSION, and I'm trying to dig up what that might be... $ bitbake eglibc -e | grep -e "^PREFERRED_VERSION_eglibc.*=" PREFERRED_VERSION_eglibc-nativesdk="2.15" PREFERRED_VERSION_eglibc-initial-nativesdk="2.15" PREFERRED_VERSION_eglibc="2.13" PREFERRED_VERSION_eglibc-locale="2.13" PREFERRED_VERSION_eglibc-initial="2.13" 21627 2 dvhart 2012-05-25 19:44:43 +0000 dvhart@rage:~/source/poky [denzil-next] $ git grep PREFERRED_VERSION_eglibc meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc ?= "${EGLIBCVERSION}" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc-locale ?= "${EGLIBCVERSION}" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc-nativesdk ?= "2.15" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc-initial ?= "${EGLIBCVERSION}" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc-initial-nativesdk ?= "2.15" dvhart@rage:~/source/poky [denzil-next] $ git grep EGLIBCVERSION meta/conf/distro/include/tcmode-default.inc:EGLIBCVERSION ?= "2.13" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc ?= "${EGLIBCVERSION}" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc-locale ?= "${EGLIBCVERSION}" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_eglibc-initial ?= "${EGLIBCVERSION}" meta/conf/distro/include/tcmode-default.inc:PREFERRED_VERSION_cross-localedef-native ?= "${EGLIBCVERSION}" 21629 3 dvhart 2012-05-25 20:16:42 +0000 core-image-base builds successfully with the following change: $ git diff diff --git a/meta/conf/distro/include/tcmode-default.inc b/meta/conf/distro/include/tcmode-default.inc index 3dc0932..c85553b 100644 --- a/meta/conf/distro/include/tcmode-default.inc +++ b/meta/conf/distro/include/tcmode-default.inc @@ -20,7 +20,7 @@ PREFERRED_PROVIDER_virtual/gettext ??= "gettext" GCCVERSION ?= "4.6%" SDKGCCVERSION ?= "4.6%" BINUVERSION ?= "2.22" -EGLIBCVERSION ?= "2.13" +EGLIBCVERSION ?= "2.15" UCLIBCVERSION ?= "0.9.33" LINUXLIBCVERSION ?= "3.2" 22118 4 richard.purdie 2012-06-13 13:04:36 +0000 We can't change the eglibc version in a point release. It is meant to be building 2.15 for nativesdk and 2.13 for everything else. Somehow it sounds like there is corruption creeping from the nativesdk into the target build :(. I thought I might be able to reproduce this locally but so far I've been unable to. 22119 5 richard.purdie 2012-06-13 13:08:43 +0000 Darren, had you previously build a meta-toolchain or similar maybe prior to the core-image-base? 22135 6 dvhart 2012-06-13 15:10:02 +0000 (In reply to comment #5) > Darren, had you previously build a meta-toolchain or similar maybe prior to > the core-image-base? There is evidence that I have built meta-toolchain in the same build directory based on a filename search in tmp of "*meta-toolchain*". ./work/core2-poky-linux/meta-toolchain-1.0-r7... among others 22179 7 richard.purdie 2012-06-14 13:07:30 +0000 Need help in reproducing this as I've been unable to do so 22184 8 dvhart 2012-06-14 13:47:12 +0000 (In reply to comment #7) > Need help in reproducing this as I've been unable to do so I've tried various things and cannot reproduce either :( 22293 9 richard.purdie 2012-06-18 22:02:55 +0000 Please reopen if steps to reproduce the issue can be figured out. If anyone has a broken build I'd be interested to inspect it.