summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* bitbake: runqueue: Handle missing sstate dependencies betterRichard Purdie2016-09-211-5/+4
| | | | | | | | | | | | | | | | | | | | | | | If you "bitbake glibc-locale" then delete the libpcre-native sstate and "bitbake glibc-locale -C package_write_rpm", it will fail with rpmbuild missing the libprce library. The reason is that libpcre-native fails to install from sstate (since it isn't present) but doesn't get built and hence rpm-native tries to run without its dependencies. The simplest fix is not to add "covered" tasks which have failed to install sstate. I can't help feeling there is more to this issue but this does fix the current problem and shouldn't have adverse affects. It is an unusual situation to have missing dependencies in sstate since they're usually all present or not at all. I've taken the opportunity to remove some old cruft from when we had numeric task ids, the code can be simpler now. (Bitbake rev: ba566b46d530b495f12f3a74f76434717b22a020) Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: buildinfohelper local layer don't construct path using git ↵Michael Wood2016-09-211-3/+5
| | | | | | | | | | | | | | | info When the layer is local source don't try and work out the location of the layer by using the git information (getGitCloneDirectory) [YOCTO #10199] (Bitbake rev: 3dfea5214d4bd006e26630e5024774ecb84ea527) Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: bavery <brian.avery@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: Add tests to detect if we have missing db migrationsMichael Wood2016-09-212-0/+55
| | | | | | | | | | | Based on the same test as found in patchwork by Damien Lespiau https://github.com/dlespiau/patchwork/blob/master/patchwork/tests/test_db.py (Bitbake rev: 031cb194aaa1b6cc970fed3fa0d0dbd3ebac163f) Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: bavery <brian.avery@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: unlock BuildEnvirnoment when build is doneEd Bartosh2016-09-211-3/+5
| | | | | | | | | | | | | | | | There is no need to lock build environment before changing build status as this operation is very fast. However, there is a need to unlock it after changing build status. Explicitly unlocked BuildEnvironment after build reaches final status SUCCEEDED, FAILED or CANCELLED. This should allow runbuilds process to pickup next build faster. (Bitbake rev: faa88272d656640c039572c5c8f3e6c56535b6f7) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: runbuilds: code cleanup - remove unused importsEd Bartosh2016-09-211-5/+2
| | | | | | | | | | Fixed pylint warning: W0611(unused-import): Unused import (Bitbake rev: 49731a1a2b2b63c1a897d2e33bca4968524e8710) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: runbuilds: code cleanup - whitespaces, long linesEd Bartosh2016-09-211-19/+15
| | | | | | | | | | | | | | Fixed following pylint warnings: C0330(bad-continuation): Wrong hanging indentation before block. C0326(bad-whitespace): No space allowed around keyword argument assignment C0326(bad-whitespace): Exactly one space required before assignment C0301(line-too-long): Line too long (Bitbake rev: 0eecd660e374a4dbcefe4c59f4c8654bf3a0e937) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: runbuilds: process builds on startEd Bartosh2016-09-211-0/+2
| | | | | | | | | | | | If Toaster is stopped incorrectly there could be some build requests and builds in incorrect state left from the previous run. Running main processing function on start should take care of those. (Bitbake rev: 6b9f8f6bb51d1aa2ca4effc34e076e331d0cb8d1) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: runbuilds: process builds on SIGUSR1Ed Bartosh2016-09-211-2/+4
| | | | | | | | | | | | | | | | | Run main processing function 'runbuild' only if SIGUSR1 is received. This signal is sent by Toaster when build status is changed (either started, cancelled or finished). This should stop continuous database polling as run_builds function will be called only when needed, i.e. after build status is changed. [YOCTO #8918] (Bitbake rev: 62d598cc5aa01d23f1e9284e9e926bd55b1d1878) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: notify runbuilds when build status changesEd Bartosh2016-09-212-0/+6
| | | | | | | | | | | | | Called signal_runbuilds API when build is scheduled, cancelled or finished to notify runbuilds process about builds status change. [YOCTO #8918] (Bitbake rev: fe08f0fa4b328908e73695ebbceca87bc86a49f9) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: implement signal_runbuilds functionEd Bartosh2016-09-211-1/+7
| | | | | | | | | | | | | This function reads pid of runbuilds process from BUILDDIR/.runbuilds.pid and sends SIGUSR1 to it. signal_runbuilds function will be used in Toaster code to notify runbuilds when build is scheduled, finished or cancelled. (Bitbake rev: 62955224a6d99e9f581d2bef924058070bfa4c43) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: fix checking of repository urlEd Bartosh2016-09-211-1/+1
| | | | | | | | | | | | | | | Toaster checks gir repository url is the same as locally cloned before checking it out to existing local clone. This check can be skipped if commit is 'HEAD' as in this case repository is not hard reset to commit, so the local clone won't be changed. [YOCTO #10163] (Bitbake rev: 7e9a89e3fde5e71cb859799635974ec41790c44d) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: cookerdata.py: Catch BBHandledException, preventing a backtrace in ↵Mark Hatle2016-09-211-1/+1
| | | | | | | | | | | | | an event The event handling 'Exception' was catching and triggering a backtrace. This trace was obscuring any errors from an event handler that had raised the BBHandledException, which should indicate do not print additional information. (Bitbake rev: 51ca5193a5674b27d816140b0254f485912177a2) Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: cooker.py: add LAYERRECOMMENDS processingJoe Slater2016-09-211-3/+34
| | | | | | | | | | | | | | | | | | | Add recommended layers to collection_depends[] so that dynamic priority assignment will work for both depends and recommends. Recommended layers do not cause an error or warning if they are not in the collection list, but debug messages are output for level 3 and above. explode_dep_versions2 returns a dictionary, so we change the variable deplist to depDict. The dictionary values are lists which are either empty or contain only one version specification. (Bitbake rev: 20cdc3d609f8aea992f97c3db336574d3a549973) Signed-off-by: Joe Slater <jslater@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: Remove duplicate layer informationBelen Barros Pena2016-09-211-2/+0
| | | | | | | | | | | | | | In the custom recipe details page, the layer information is displayed twice in the right hand column. Remove one of the layer entries, since showing the layer information once should be enough. [YOCTO #10037] (Bitbake rev: e2b5dc3732781dc933c6bb10482926335720d110) Signed-off-by: Belen Barros Pena <belen.barros.pena@linux.intel.com> Signed-off-by: bavery <brian.avery@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: Indicate active navigation elementBelen Barros Pena2016-09-211-2/+5
| | | | | | | | | | | | | | | The left navigation in the build history pages is not showing the active item when you navigate directly to the errors or warnings information in the build summary. Add a special case to make sure the "build summary" item is highlighted. [YOCTO #9864] (Bitbake rev: f236d9ca28e45a270f50bb3edcd466b1bc8d2960) Signed-off-by: Belen Barros Pena <belen.barros.pena@linux.intel.com> Signed-off-by: bavery <brian.avery@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: toaster: Fix links to tasks with specific outcomeBelen Barros Pena2016-09-211-3/+3
| | | | | | | | | | | | | | | | | The build dashboard provides a count of tasks that were executed and not executed, and of tasks that failed (if any). The number is a link to the list of tasks. Fix the links so that they filter the tasks table by the selected criteria (executed, not executed or failed). [YOCTO #9832] (Bitbake rev: a75e70bbc9081f77f1e4aeeee8222b06112e4406) Signed-off-by: Belen Barros Pena <belen.barros.pena@linux.intel.com> Signed-off-by: bavery <brian.avery@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* bitbake: uihelper: use elif instead of repeated ifRoss Burton2016-09-211-6/+5
| | | | | | (Bitbake rev: a1d6f6425cd9ef9e07344869817517172afd6e27) Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-yocto/4.8: introduce preempt-rtBruce Ashfield2016-09-211-1/+1
| | | | | | | | | | | | | | | Paul Gortmaker has made the preempt-rt patch available for the 4.8 kernel. This commit merges his queue to standard/preempt-rt/rebase, which will be kept up to date with a clean history, and it also makes it available in standard/preempt-rt/base, which will be kept fast forward for board support. (From OE-Core rev: b475bb88cc5d6700b4fadff1b2bb43a376492c7b) Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-yocto/4.8: integrate 4.8-rc7Bruce Ashfield2016-09-213-16/+16
| | | | | | | | | Updating to the latest release candidate. (From OE-Core rev: f15d5d8c09da9260a59ed396a4d5c9d009e5c751) Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-yocto/4.1: 4.1.32 content and configuration warning fixesBruce Ashfield2016-09-213-13/+13
| | | | | | | | | | | | | | | | | | The LINUX_VERSION was previously updated to 4.1.32, but the SRCREVs for the actual content were missed. This gets our actual version and the PV back in sync. We also update the meta data to fix configuration audit warnings from the beaglebone builds. (From OE-Core rev: 255adfaddec0dc44db619bf8726cd7804c09400d) Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> squash with 4.1 Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* kernel.bbclass: Add kernel_version_sanity_check functionCalifornia Sullivan2016-09-212-0/+31
| | | | | | | | | | | | | | | | | | The kernel being built should match what the recipe claims it is building. This function ensures that happens by comparing the version information in the kernel's Makefile to the PV the recipe is using. v2 changes: * Match against PV instead of LINUX_VERSION * Match against EXTRAVERSION as well (e.g., -rc4) * Cleaned up version string building Fixes [YOCTO #6767]. (From OE-Core rev: ec467cfaea5c8cf22c61daa8845c2e4e96449512) Signed-off-by: California Sullivan <california.l.sullivan@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* boost: Replace math::static_lcm with integerJackie Huang2016-09-211-1/+1
| | | | | | | | | | | Replace math with integer according to the commit upstream: (From OE-Core rev: 2e2ecd867d47ee24b37f61af6db5338df5059844) (From OE-Core rev: 4c9fe10f3aaa4ee6e8fee52816298896b18cdb60) Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* uninative-tarball: add SDKMACHINE to stamps-extra-infoJoshua Lock2016-09-211-1/+1
| | | | | | | | | | Otherwise the stamps for x86-64 and i686 uninative tarballs match and we can't deploy both to the DEPLOYDIR. (From OE-Core rev: 2a9603759fe87d6326c145f6213ffffeb6afc6ae) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* populate_sdk_base: fix support for changing SDKMACHINE settingsJoshua Lock2016-09-211-2/+2
| | | | | | | | | | | | | | | | | Include SDKMACHINE in the tasks stamp information and the name of the sstate-inputdirs so that changing SDKMACHINE doesn't result in valid output of the task being deleted when SDKMACHINE is changed. Without this patch changing SDKMACHINE and building an SDK resulted in toolchain installers for other SDKMACHINE's being deleted from the deploy directoy. [YOCTO #10275] (From OE-Core rev: d7a06b53af0066bd12f5f42e10e82b307fd069ce) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* oeqa/selftest/lic-checksum: don't report the expected failure to errors.ypRoss Burton2016-09-201-0/+1
| | | | | | | | | | This test has a bitbake invocation that is expected to fail, so inhibit report-error running. (From OE-Core rev: b2771e17a5f301423f65be9f93c9c1b1e7f8ab93) Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* openssl: fix do_configure error when cwd is not in @INCRobert Yang2016-09-202-0/+35
| | | | | | | | | | | Fixed when building on Debian-testing: | Can't locate find.pl in @INC (@INC contains: /etc/perl /usr/local/lib/x86_64-linux-gnu/perl/5.22.2 /usr/local/share/perl/5.22.2 /usr/lib/x86_64-linux-gnu/perl5/5.22 /usr/share/perl5 /usr/lib/x86_64-linux-gnu/perl/5.22 /usr/share/perl/5.22 /usr/local/lib/site_perl /usr/lib/x86_64-linux-gnu/perl-base) at perlpath.pl line 7. (From OE-Core rev: c28065671b582c140d5971c73791d2ac8bdebe69) Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* gcc-configure: Enable initfini-arraySaul Wold2016-09-201-0/+4
| | | | | | | | | | | | | | | | This adds the correct support for initfini-array which replaces .init and .fini with .init-array and .fini-array. There is no appreciable size difference with this change. The change is needed since configure will not correctly detect support when building cross-compilers. (From OE-Core rev: 9f8a075ab46691534e2b22d0a363b3c847394215) Signed-off-by: Haitao Huang <haitao.huang@intel.com> Signed-off-by: Saul Wold <sgw@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* oeqa/runtime/parselogs.py: Add ignore of tsc calibration fail in x86Aníbal Limón2016-09-201-0/+1
| | | | | | | | | | | | | | | | | | | We are experimenting failures to calibrate CPU's using TSC in x86 VM's due to usage of nested KVM [1], this is a known issue [2][3] in virtualization environments, for detail explnation see [4]. Also we already have an ignore for 'TSC Fast calibration fail'. [1] http://errors.yoctoproject.org/Errors/Details/83684/ [2] https://bugzilla.redhat.com/show_bug.cgi?id=814231 [3] https://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01547.html [4] https://www.kernel.org/doc/Documentation/virtual/kvm/timekeeping.txt (From OE-Core rev: 2271f59a0f506f89f9fea6777701c4b40790ddd9) Signed-off-by: Aníbal Limón <anibal.limon@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* enchant: add PACKAGECONFIG for *spellMartin Jansa2016-09-201-7/+9
| | | | | | | | | | | | | | | * use PACKAGECONFIG for various *spell implementations * keep aspell enabled by default * prevent hunspell/myspell being autodetected as reported in "bitbake world status" e-mails: WARN: enchant: enchant rdepends on hunspell, but it isn't a build dependency? (From OE-Core rev: d7df8092c2be4c984d9292749f6cfa7ce41cbb53) Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* boost: switch to using C++11 atomics when availableAndré Draszik2016-09-202-0/+60
| | | | | | | | | | | | The alleviates the need to use and maintain hand-written assembly, just relying on the compiler to do the right thing. (From OE-Core rev: 0e683664dfbe2b71288445a86e31fbea062ed564) Signed-off-by: André Draszik <git@andred.net> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* runqemu: improve finding of rootfs, kernel and dtbRobert Yang2016-09-201-31/+42
| | | | | | | | | | | | | | | | | | | | | | | | | | * Search rootfs in the following order: - IMAGE_NAME*.FSTYPE - IMAGE_LINK_NAME*.FSTYPE * Search kernel in the following order: - QB_DEFAULT_KERNEL - KERNEL_IMAGETYPE - KERNEL_IMAGETYPE* * Search dtb in the following order: - QB_DTB - QB_DTB* - *.dtb * Fix DTB, it should only work with "-kernel" option. [YOCTO #10265] (From OE-Core rev: 32ff0974ed06f797c6b7d9092a8dc9ae50e9a572) Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* linux-firmware: package carl9170 separatelyIoan-Adrian Ratiu2016-09-201-1/+5
| | | | | | | | | | | | | | | | | | | | | Package the carl9170 binary firmware separately because it is needed by various usb dongles and installing whole linux-firmware is overkill. carl9170 is an atheros-based firmware, the succesor of ar9170 which is deprecated and its driver (ar9170) was removed from the kernel tree. carl9170 comes both as a binary blob and also with free sources; the sources are deleted in linux-firmware.bb to avoid depeding on bash/etc. Also we should keep ar9170 as is because OOT drivers might still use it. The license for carl9170 is GPLv2 (sources are beside the bin in the linux-firmware git repo). (From OE-Core rev: 907e6596275850cc31b143a4c3094029b0cd078e) Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* runqemu-gen-tapdevs: remove /etc/runqemu-nosudo when remove tapsRobert Yang2016-09-201-20/+26
| | | | | | | | | | | It creats /etc/runqemu-nosudo when creats taps, so should remove it when remove taps. (From OE-Core rev: 3d4bf5b0ea581e3e9b388328b086d03f9174fd61) Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* default-distrovars.inc: remove libidn from LGPLv2_WHITELIST_GPL-3.0Jackie Huang2016-09-201-1/+1
| | | | | | | | | | | | The libidn recipe is now buildable in distros which blacklist GPL-3.0 without needing to be explicitly whitelisted (since it provides at least one non GPLv3 package). (From OE-Core rev: 63d6d014a0a3da8bf5689b27d1155492e2fdb0f2) Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* openssh: fix potential signed overflow to enable compilation with -ftrapvYuanjie Huang2016-09-202-0/+100
| | | | | | | | | | | | | | | Pointer arithmatic results in implementation defined signed integer type, so that 's - src' in strlcpy and others may trigger signed overflow. In case of compilation by gcc or clang with -ftrapv option, the overflow would lead to program abort. Upstream-status: Submitted [https://bugzilla.mindrot.org/show_bug.cgi?id=2608] (From OE-Core rev: 2ce02941300aa3e826df0c59fd8d4ce19950028e) Signed-off-by: Yuanjie Huang <yuanjie.huang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* image_types: add parted-native to do_image_wic dependsEd Bartosh2016-09-202-1/+2
| | | | | | | | | | | | As parted is always used by wic it makes sense to make do_image_wic dependent on parted-native:do_populate_sysroot. This should help to avoid adding it to all wic image recipes. (From OE-Core rev: c687c9fcc010947f52e1cd153ea74ae5f6343ca4) Signed-off-by: Ed Bartosh <ed.bartosh@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* gawk-3.1.5: fix non-glibc & gcc >= v4 buildsAndré Draszik2016-09-202-2/+72
| | | | | | | | | | See commit message of patch added (From OE-Core rev: bdfce776843bf4287251923c79bc15365c20b4a1) Signed-off-by: André Draszik <git@andred.net> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* package_manager.py: Change diagnostic messages per IRCMark Hatle2016-09-201-8/+12
| | | | | | | | | | | | | Based on a discussion with IRC user: Ulfalizer It was suggested that removing the diagnostic list, and replacing it with a simple hint to what might be causing the problem was a better solution. (From OE-Core rev: ca78313665b23bd7fee85f034acfe1eb1009bd65) Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* package_manager.py: Adjust error message orderMark Hatle2016-09-201-4/+4
| | | | | | | | | | | | Move the debug before the error (as it can take many pages.) This makes it much easier for the user to see the actual error message as it is still on the screen. (From OE-Core rev: d643fb2a9cb5bd0d8b0105e9d44b989a49ffa963) Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* qemuboot.bbclass: don't fail on very first buildAndré Draszik2016-09-201-0/+1
| | | | | | | | | | | | | | | | | | During the very first build, the DEPLOY_DIR_IMAGE directory might not have been created yet, causing the creation of the qemuboot.conf config file to fail. This is because write_qemuboot_conf() runs at rootfs creation time, i.e. before deploy. So let's create the directory if necessary before trying to write the config file. (From OE-Core rev: ee4697350a553a36ca17b9376911e56eee43a465) Signed-off-by: André Draszik <git@andred.net> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* externalsrc.bbclass: Make reparsing work for BBCLASSEXTENDed recipesPeter Kjellerstedt2016-09-201-3/+17
| | | | | | | | | | | | | | | | | To make sure changes to any source files are detected when externalsrc is used, it sets BB_DONT_CACHE to force the recipe to be reparsed every time. Previously, this was done conditionally based on whether EXTERNALSRC was set. This worked fine for building the base recipe. But if one tried to build, e.g., a native version of it (provided via BBCLASSEXTEND), the recipe would not be reparsed as expected. To solve the above problem, BB_DONT_CACHE is now set for the base recipe if EXTERNALSRC is set for it or any of it derivatives. (From OE-Core rev: 449a0b21255d895e8620383ce76a9d7ea41b5cc6) Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* kernel.bbclass: assign INITRAMFS_BASE_NAME using ?=Andre McCurdy2016-09-201-1/+1
| | | | | | | | | | | | Default values for KERNEL_IMAGE_BASE_NAME and MODULE_IMAGE_BASE_NAME are already assigned using ?= and anyone wanting to over-ride one is likely to want to over-ride them all. Make the three consistent with each other. (From OE-Core rev: e30c6c93bb70d17244c90c2be12229148f8f6314) Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* dropbear: deterministic selection of system -vs- bundled libtom libsAndre McCurdy2016-09-201-0/+4
| | | | | | | | | | | | | | Dropbear will use system versions of libtommath and libtomcrypt if available. To make builds deterministic, add a PACKAGECONFIG option to choose system libs or force use of the bundled versions. Note that currently there are no libtommath or libtomcrypt recipes in oe-core, so default to using the bundled versions. (From OE-Core rev: b7c2edd2d6ded287d8b34dd047ae84d3fd69d4c6) Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* dropbear: fix -ltomcrypt -ltommath order when using system libtom libsAndre McCurdy2016-09-202-0/+49
| | | | | | | | | | | | | | | To prevent build failures when using system libtom libraries and linking with --as-needed, LIBTOM_LIBS should be in the order -ltomcrypt -ltommath, not the other way around, ie libs should be prepended to LIBTOM_LIBS as they are found, not appended. Note that LIBTOM_LIBS is not used when linking with the bundled libtom libs. (From OE-Core rev: 62e96283fe77469e24e8df86c6c037c92009b00a) Signed-off-by: Andre McCurdy <armccurdy@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* runqemu: use OECORE_NATIVE_SYSROOT from sdkRobert Yang2016-09-201-4/+11
| | | | | | | | | | There is no STAGING_DIR_NATIVE or bitbake in a extracted sdk, so check OECORE_NATIVE_SYSROOT and use it. (From OE-Core rev: 93649edc034f2540ff55dc9b41638797209cfb9c) Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* runqemu: work even if a *.qemuboot.conf isn't foundJoshua Lock2016-09-201-1/+7
| | | | | | | | | | | | | | | A qemuboot conf file is a convenience but it should still be possible to invoke runqemu without them, especially for examples such as using the SDK with an extracted rootfs via NFS. As read_qemuboot() is always called we need to be sure that function can return cleanly, without throwing Exceptions, even if a qemuboot conf file isn't found. (From OE-Core rev: 3541c21f1976b517b79a19882240a8f36b970292) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* runqemu: try symlinks when kernel or rootfs can't be foundJoshua Lock2016-09-201-3/+13
| | | | | | | | | | | | | | | If the kernel or rootfs names written to the qemuboot.conf can't be found, try and find the symlinked variant of the filename. This will help usability of runqemu, for example where a user downloads an image and associated files as the symlinked names yet the qemuboot.conf variables point to the full, non-linked, file names. (From OE-Core rev: ca5a686c6e165a51f95cb6a834cd53f6f66d42d4) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* qemuboot: also write the kernel link name to the conf fileJoshua Lock2016-09-201-2/+3
| | | | | | | | | | This will allow runqemu to fall back to trying the link name when a file matching the full name can't be found. (From OE-Core rev: 3ccbaaad75f0a53d8bcf6a5c748ec80c96a383bd) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* runqemu: clarify an INFO messageJoshua Lock2016-09-201-1/+1
| | | | | | | | | | Make it clearer that we are looking for a file which ends with qemuboot.conf (From OE-Core rev: 2579e05269a14b53a54232a8bf4414ac2dfe6472) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* qemuboot: write the full kernel filename, not the link nameJoshua Lock2016-09-201-0/+8
| | | | | | | | | | | | | | | | | | | KERNEL_IMAGETYPE gives the filename of a symlink to the kernel, which may not be available i.e. if the user downloads some build artefacts to run on a local machine. It's also possible that the link will point to a newer kernel than was intended for use with the rootfs in the qemuboot.conf. It's much more reliable to read the name of the file KERNEL_IMAGETYPE is linking to and assign the full filename to QB_DEFAULT_KERNEL. [YOCTO #10285] (From OE-Core rev: d57bdacab13605ada4cd9e9159c18fdcd6eeacbc) Signed-off-by: Joshua Lock <joshua.g.lock@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
generated by cgit v1.2.3-54-g00ecf (git 2.39.0) at 2024-11-13 20:18:46 +0000