Richard Purdie [Tue, 10 Nov 2015 09:58:25 +0000 (09:58 +0000)]
insane: Don't depend on BB_TASKDEPDATA
The dependency data in BB_TASKDEPDATA is encoded into the sstate checksum
in a much more reliable format. This dependency runs the risk of depending
on the string representation of a dict which is a bad idea. Therefore
remove the dependency.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org> Signed-off-by: Ross Burton <ross.burton@intel.com>
Jens Rehsack [Thu, 12 Nov 2015 16:13:15 +0000 (17:13 +0100)]
kernel: fix race condition between compile_kernelmodules and shared_workdir
Fixes https://bugzilla.yoctoproject.org/show_bug.cgi?id=8127 which causes
[ 1225.089323] 8189es: Unknown symbol cfg80211_scan_done (err -22)
[ 1225.095916] 8189es: no symbol version for cfg80211_remain_on_channel_expired
when loading external compiled 8189es module.
Signed-off-by: Jens Rehsack <sno@netbsd.org> Signed-off-by: Ross Burton <ross.burton@intel.com>
Fixes the following warning sometimes appearing during image builds:
WARNING: The license listed ABC was not in the licenses collected for recipe xyz
The files being looked for here, which runs during do_rootfs,
are written out by the do_populate_lic task for each recipe. However,
there was no explicit dependency between do_rootfs and all of the
do_populate_lic tasks to ensure they had run - only an implicit link via
do_build, so it is possible that sometimes they had not depending on how
the tasks were scheduled. Add an explicit set of dependencies to fix
this.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Paul Eggleton [Mon, 9 Nov 2015 14:40:08 +0000 (14:40 +0000)]
classes/metadata_scm: fix git errors showing up on non-git repositories
Fixes the following error showing up for layers that aren't a git repo
(or aren't parented by one):
fatal: Not a git repository (or any of the parent directories): .git
This was because we weren't intercepting stderr. We might as well just
use bb.process.run() here which does that and returns stdout and stderr
separately.
Ross Burton [Sat, 24 Oct 2015 00:02:34 +0000 (01:02 +0100)]
sstate: respect GPG_BIN and GPG_HOME
The package feed signing code supports the user providing the path to the gpg
binary and an alternative gpg 'home' (usually ~/.gnupg), which are useful for
both deployment and QA purposes.
Factor out the gpg command line construction to a function which can fetch both
of these variables, and also use pipes.quote() to sanitise the arguments when
used in a shell context.
[ YOCTO #8559 ]
Signed-off-by: Ross Burton <ross.burton@intel.com>
archiver.bbclass: fix previous issue regarding work-shared for linux-yocto
A previous patch intended to fix and improve the archiver class, and while it
did for gcc packages, with it some issues ended up being hidden, the kernel
tasks taken from kernel.bbclass and kernel-yocto.bbclass specifically expect the
kernel to use work-shared, which either ended up causing issues or wasting time
doing unnecessary work, this patch fixes these issues by performing the right
tasks within the archiver in the right order
[YOCTO #8378]
Signed-off-by: Alejandro Hernandez <alejandro.hernandez@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Mon, 16 Nov 2015 23:18:21 +0000 (23:18 +0000)]
ptest-gnome: extend EXTRA_OECONF in all builds, not just target
This class was extending EXTRA_OECONF only in target builds with
--enable/--disable-installed-tests. However for native builds we don't care
about the test suite and should be explicitly disabling it.
This stops glib-2.0-native trying to build the test suite that we'll never
execute.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Richard Purdie [Mon, 16 Nov 2015 09:08:28 +0000 (09:08 +0000)]
layer.conf: Correct gcc-cross dependency
The dependency listed in layer.conf is incorrect, gcc-cross DEPENDS
on ${TARGET_PREFIX}libc-for-gcc, not virtual/libc. These happen to
resolve the same values however they may not always both be built.
The result of this was that gcc-cross gets a different task hash
depending on whether virtual/libc was included in the build.
Specifically "bitbake m4" and "bitbake virtual/kernel" would result
in different task checksums.
The fix is to use the correct dependency name.
[YOCTO #8692]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ruslan Bilovol [Fri, 6 Nov 2015 14:09:06 +0000 (16:09 +0200)]
scripts: runqemu: remove QEMUARCH from help message
The QEMUARCH env variable is not used since commit
"d469c92 classes/imagetest-qemu: remove old image
testing class". Remove it from help message so
it will not confuse other people
Signed-off-by: Ruslan Bilovol <ruslan.bilovol@gmail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Andre McCurdy [Fri, 6 Nov 2015 08:52:27 +0000 (00:52 -0800)]
cairo: update 1.14.2 -> 1.14.4
Release 1.14.4 (2015-10-28 Bryce Harrington <bryce@osg.samsung.com>)
========================================================================
Just in time for Halloween we see another bug-fix release for Cairo.
This brings a few dozen straightforward bug fixes with no API changes.
In addition, this includes a typical assortment of fixes to tests,
cleanup of warnings and memory leaks, correction of misspellings,
updates to documentation, etc.
For a complete log of changes since 1.14.2, please see:
Paul Eggleton [Fri, 6 Nov 2015 09:34:05 +0000 (09:34 +0000)]
classes/distrodata: split SRC_URI properly before determining type
We weren't splitting SRC_URI values containing multiple URIs here; this
didn't cause any errors except when a trailing ; was left on a URI, in
which case the next URI was considered part of the parameter, which
didn't contain a = and therefore was considered invalid.
We only care about the first URI in SRC_URI in this context (since
that's the upstream URI by convention) so split it as we should and take
the first item.
Fixes [YOCTO #8645].
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Paul Eggleton [Sun, 4 Oct 2015 07:38:16 +0000 (08:38 +0100)]
classes/buildhistory: split package history values only once
We don't actually use values we read in here that are likely to
contain = characters but we might as well split the value properly in
case we do in future.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Paul Eggleton [Thu, 3 Sep 2015 11:03:21 +0000 (12:03 +0100)]
gitignore: fix overzealous exclusion
This was excluding any subdirectory anywhere in the tree named build*,
rather than just at the root - thus anything in
meta/recipes-devtools/build-compare had to be forcibly added. Change the
line so that it only operates at the root of the repo.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Paul Eggleton [Thu, 23 Jul 2015 11:04:23 +0000 (12:04 +0100)]
verify-homepage: fix recipe file selection
* We need to check all recipe files, not just the preferred ones
(i.e. we have multiple recipes for different versions of the same
piece of software). Print the recipe file name (without path) so we
can tell the difference between them.
* We can skip BBCLASSEXTENDed variants of recipes
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Fri, 6 Nov 2015 15:30:58 +0000 (15:30 +0000)]
libaio: don't disable linking to the system libraries
For some reason that I don't understand (a decade-old attempt at optimisation?)
libaio disables linkage to the system libraries. Enabling fortify means linking
to the system libraries, so remove the existing addition of -lc for x86 (the
problem also happens on at least PPC) and just link to the system libraries on
all platforms.
Also remove the sed of src/Makefile as the build not respecting LDFLAGS has been
fixed upstream.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Thu, 5 Nov 2015 17:13:22 +0000 (17:13 +0000)]
runqemu: don't specify IP when starting a VNC server
Whilst qemu doesn't appear to support opening sockets on IPv6 yet, future-proof
the script by just specifying a port and letting qemu work out the rest.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Mariano Lopez [Thu, 5 Nov 2015 09:08:58 +0000 (09:08 +0000)]
qemurunner: Remove the timeout in run_serial
Sometmes when there is high load in the server the
commands executed in the target take a lot of time
to complete and this lead to incorrect dump files or
empty files. This is caused because run_serial has a
timeout of five seconds when running the commands in
the target.
This change removes the timeout and just keep reading
the socket until it finds the prompt from the target
or when the socket is not ready to be read.
[YOCTO #8510]
Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Armin Kuster [Thu, 29 Oct 2015 23:22:41 +0000 (16:22 -0700)]
libxslt: CVE-2015-7995
This is a is being give a High rating so please consider it for
all 1.1.28 versions.
A type confusion error within the libxslt "xsltStylePreCompute()"
function in preproc.c can lead to a DoS. Confirmed in version 1.1.28,
other versions may also be affected.
Signed-off-by: Armin Kuster <akuster@mvista.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
The lgpl PACKAGECONFIG is replaced by a gpl one,
since the LGPL build is the default now. There is no
--enable-lgpl switch anymore, it got replaced by
--enable-gpl.
Signed-off-by: Carlos Rafael Giani <dv@pseudoterminal.org> Signed-off-by: Ross Burton <ross.burton@intel.com>
gstreamer1.0-plugins-base: upgrade to version 1.6.1
A number of patches had to be deleted in order for the recipe to work.
Patches which cannot be applied cleanly to 1.6.1 and are nontrivial to fix:
* 0001-basetextoverlay-make-memory-copy-when-video-buffer-s.patch
* handle-audio-video-decoder-error.patch
* taglist-not-send-to-down-stream-if-all-the-frame-cor.patch
Patches which were rejected by upstream:
* 0002-gstplaysink-don-t-set-async-of-custom-text-sink-to-f.patch
Patches which either were backports or accepted and integrated:
* 0001-video-frame-Don-t-ref-buffers-twice-when-mapping.patch
* 0002-video-frame-Add-GST_VIDEO_FRAME_MAP_FLAG_NO_REF.patch
* 0003-videofilter-Use-new-GST_VIDEO_FRAME_MAP_FLAG_NO_REF.patch
* do-not-change-eos-event-to-gap-event2.patch
* do-not-change-eos-event-to-gap-event3.patch
* do-not-change-eos-event-to-gap-event-if.patch
* fix-id3demux-utf16-to-utf8-issue.patch
* videobuffer_updata_alignment_update.patch
* videoencoder-Keep-sticky-events-around-when-doing-a-soft-.patch
Signed-off-by: Carlos Rafael Giani <dv@pseudoterminal.org> Signed-off-by: Ross Burton <ross.burton@intel.com>
Mark Hatle [Mon, 2 Nov 2015 14:52:28 +0000 (08:52 -0600)]
gcc: Update default Power GCC settings to use secure-plt
The gcc default, bss-plt, will cause errors when using the prelinker. All
other distributions that I am aware of are using the the secure-plt. For an
explanation of the differences, the gcc docs:
Current PowerPC GCC accepts a `-msecure-plt' option that generates code
capable of using a newer PLT and GOT layout that has the security
advantage of no executable section ever needing to be writable and no
writable section ever being executable. PowerPC ld will generate this
layout, including stubs to access the PLT, if all input files (including
startup and static libraries) were compiled with `-msecure-plt'.
`--bss-plt' forces the old BSS PLT (and GOT layout) which can give
slightly better performance.
The security of the new PLT and ability to run the prelinker outweigh
any performance penalty.
The secure-plt is enabled by default. The old bss-plt can be enabled by
selecting 'bssplt' in the DISTRO_FEATURES.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Mark Hatle [Fri, 30 Oct 2015 14:48:27 +0000 (09:48 -0500)]
glibc: Fix ld.so / prelink interface for ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA
A bug in glibc 2.22's ld.so interface for the prelink support causes
the displayed values to be incorrect. The included path fixes this
issue.
Clear ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA for prelink
prelink runs ld.so with the environment variable LD_TRACE_PRELINKING
set to dump the relocation type class from _dl_debug_bindings. prelink
has the following relocation type classes:
where ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA has a conflict with
RTYPE_CLASS_TLS.
Since prelink doesn't use ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA, we
should clear the ELF_RTYPE_CLASS_EXTERN_PROTECTED_DATA bit when the
DL_DEBUG_PRELINK bit is set.
Signed-off-by: Mark Hatle <mark.hatle@windriver.com>
Having the timestamps in the log is useful when debugging those cases when
the booting time takes longer than expected. printk.times=1 enables stamping
into every printk Kernel call.
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Ross Burton [Wed, 4 Nov 2015 19:45:39 +0000 (19:45 +0000)]
at: modify sources in do_patch
Instead of copying the forked source files into ${S} in do_compile_prepend(), do
it in a do_patch() postfunc. This ensures that the source archiver sees the
changes for compliance and legal verification.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Hongxu Jia [Mon, 2 Nov 2015 07:12:50 +0000 (15:12 +0800)]
default-distrovars: remove less from WHITELIST_GPL-3.0
As oe-core commit 37c8c2c5f6937520eaf9f5d981f9315e36eba4bb and fa7bdf96b8eb8b5078db38249c5c60ec511c35c6 said, Mark Nudelman
(author of less) has given permission to utilize a generic 2-clause
BSD, so we remove it from WHITELIST_GPL-3.0 which caused
a QA Warning while INCOMPATIBLE_LICENSE += "GPLv3 LGPLv3 GPLv3+ LGPLv3+"
...
|NOTE: INCLUDING less as buildable despite INCOMPATIBLE_LICENSE
because it has been whitelisted
...
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Andre McCurdy [Thu, 29 Oct 2015 21:00:22 +0000 (14:00 -0700)]
cmake.bbclass: don't duplicate CMAKE_C_FLAGS in CMAKE_C_FLAGS_RELEASE
The base CMAKE_<LANG>_FLAGS are included for all build types,
therefore the CMAKE_<LANG>_FLAGS_RELEASE should contain only
additional flags or over-rides.
Kai Kang [Fri, 16 Oct 2015 06:28:11 +0000 (14:28 +0800)]
sqlite: 3.8.10.2 -> 3.9.0
Upgrade sqlite from 3.8.10.2 to 3.9.0.
* update python function to get right SRC_URI
* drop 0001-using-the-dynamic-library.patch which use dynamic library
that it is done that way in new version
Signed-off-by: Kai Kang <kai.kang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
All of the rt-tests patches that OE has been carrying have been
upstreamed or superceded by changes in the v0.94 release.
Adjust SRC_URI to point to canonical upstream git repo, instead of a
development tree.
There was a notable change upstream that required slight reworking of
the recipe. rt-tests now joins other kbuild-inspired projects by making
use of a CROSS_COMPILE flag to indicate the compiler prefix.
Previously TOOLCHAIN_OPTIONS were conveyed via $CC directly, however,
this does not work with CROSS_COMPILE. Workaround this by both
specifying CROSS_COMPILE, and feeding the rt-tests build system the
proper $(HOST_CC_ARCH)$(TOOLCHAIN_OPTIONS) via $CFLAGS.
Signed-off-by: Josh Cartwright <joshc@ni.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
When building mesa without X11 enabled, and disabling dri package config,
the enforced "--enable-gbm" fails building mesa, because this feature
requires "--enable-dri", too - which has been disabled by package config.
Consequently, when gbm requires dri, --disable-dri causes in --disable-gbm,
which makes explicit --disable-gbm in case of --disable-dri redundant.
Signed-off-by: Jens Rehsack <sno@netbsd.org> Signed-off-by: Ross Burton <ross.burton@intel.com>
combo-layer: introduce ability to exclude component from mass update
There is no ability at the moment for situations where users would like
to keep section in combo-layer.conf but don't update it, unless explicitly
specified.
Now, by adding "update = no" to desired section would exclude
that repository from "combo-layer update" command.
It is still possible to explicitly update it by "combo-layer update $section".
By default, all repositories are assumed as "update = yes"
Signed-off-by: Alexander D. Kanevskiy <kad@kad.name> Signed-off-by: Ross Burton <ross.burton@intel.com>
By default ptest-runner executes all ptests found in a system.
With this change, ptest-runner can be given a list of ptest
package names to run (the default is still available).
For example, to run only the zlib and rt-tests ptests:
"ptest-runner zlib rt-tests"
Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Lucian Musat [Fri, 9 Oct 2015 17:16:26 +0000 (20:16 +0300)]
oeqa/testimage: Add support for test folder export.
If tests are organized in folders then it will copy the whole
folder to the export location. This should keep the original
directory structure for exported oeqa/runtime.
Signed-off-by: Lucian Musat <george.l.musat@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Jackie Huang [Wed, 28 Oct 2015 04:34:55 +0000 (12:34 +0800)]
qemu: enable user mode for mips64 and mips64el
- remove mips64 and mips64el from softmmuonly list
to enable user mode, they have been supported
since 2012.
- keep the softmmuonly list and for loop although
there is only one for now in case more supported
arches added.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Wenzong Fan [Wed, 28 Oct 2015 05:53:37 +0000 (01:53 -0400)]
gnupg: fix find-version for beta checking
find-version always assumes that gnupg is beta if autogen.sh is run
out of git-repo. This doesn't work for users whom just take release
tarball and re-run autoconf in their local build dir.
This fixes runtime issue:
$gpg --list-sigs
gpg: NOTE: THIS IS A DEVELOPMENT VERSION!
gpg: It is only intended for test purposes and should NOT be
gpg: used in a production environment or with production keys!
Signed-off-by: Wenzong Fan <wenzong.fan@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Roy Li [Wed, 28 Oct 2015 09:52:43 +0000 (17:52 +0800)]
rpm: define EM_AARCH64 for debugedit
EM_AARCH64 maybe not be defined due to old version elf.h when compile
rpm-native, and lead to that debugedit can not work on aarch64 elf object
files, since there is no other dependence, except these two macro,
define them to make debugedit work on aarch64 elf files.
debugedit: /bitbake_build/tmp/work/aarch64-wrs-linux/libvirt/1.2.19-r0/
package/usr/lib64/libvirt/ptest/daemon/libvirtd_admin_la-admin_server.o:
Unhandled relocation 258 in .debug_info section
Signed-off-by: Roy Li <rongqing.li@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Joshua Lock [Tue, 20 Oct 2015 09:33:05 +0000 (10:33 +0100)]
weston: add PACKAGECONFIG to build with systemd-login support
Automatically enable this support for distributions which have
systemd in DISTRO_FEATURES.
We add additional patches to weston, backported from Weston
git, to support the newer single libsystemd pkg-config file
and to make the enabling of systemd-login support explicit.
Signed-off-by: Joshua Lock <joshua.lock@collabora.co.uk> Signed-off-by: Ross Burton <ross.burton@intel.com>
Craig McQueen [Thu, 15 Oct 2015 05:13:09 +0000 (16:13 +1100)]
os-release: put double-quotes around variable contents
This makes the resulting /etc/os-release file have valid shell
assignment syntax. This makes it loadable by a shell script, using the
'source' command:
source /etc/os-release
Signed-off-by: Ross Burton <ross.burton@intel.com>
Robert Yang [Sat, 10 Oct 2015 02:51:33 +0000 (19:51 -0700)]
coreutils: fix ALTERNATIVE_PRIORITY conflict with procps and mktemp
Signed-off-by: Ross Burton <ross.burton@intel.com>
Conflicts:
coreutils and procps have the same priority 100 for kill.1
coreutils and procps have the same priority 100 for uptime.1
coreutils and procps have the same priority 100 for uptime
coreutils and procps have the same priority 100 for kill
coreutils and mktemp have the same priority 100 for mktemp
Turn down coreutils' priority to fix the conflict, the larger set
of tools should use a lower priority.
[YOCTO #8477]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Robert Yang [Sat, 10 Oct 2015 02:51:33 +0000 (19:51 -0700)]
util-linux: fix ALTERNATIVE_PRIORITY conflict with ncurses procps and e2fsprogs
Signed-off-by: Ross Burton <ross.burton@intel.com>
Conflicts:
util-linux and ncurses have the same priority 100 for reset
util-linux and procps have the same priority 100 for kill.1
util-linux and procps have the same priority 100 for kill
e2fsprogs and util-linux have the same priority 100 for libblkid.3
e2fsprogs and util-linux have the same priority 100 for blkid.8
e2fsprogs and util-linux have the same priority 100 for findfs.8
e2fsprogs and util-linux have the same priority 100 for fsck.8
Turn down util-linux' priority to fix the conflict, the larger set
of tools should use a lower priority.
[YOCTO #8477]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Robert Yang [Sat, 10 Oct 2015 02:51:33 +0000 (19:51 -0700)]
console-tools: fix ALTERNATIVE_PRIORITY conflict with kbd
Signed-off-by: Ross Burton <ross.burton@intel.com>
Conflicts:
console-tools and kbd have the same priority 100 for chvt
console-tools and kbd have the same priority 100 for deallocvt
console-tools and kbd have the same priority 100 for fgconsole
console-tools and kbd have the same priority 100 for openvt
Turn down console-tools' priority to fix the conflict, the larger set
of tools should use a lower priority.
[YOCTO #8477]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
