Since commit 972b4fc (feature-arm-neon.inc: restore vfpv3-d16 support)
we're replacing _all_ dashes (-) in ARMPKGSFX_FPU, which is causing
problems for all legitimate uses of the dash as TUNE_PKGARCH doesn't
have the right value anymore:
E.g. on raspberrypi2:
ERROR: OE-core's config sanity checker detected a potential misconfiguration.
Either fix the cause of this error or at your own risk disable the checker (see sanity.conf).
Following is the list of potential problems / advisories:
Error, the PACKAGE_ARCHS variable (all any noarch armv5hf-vfp armv5thf-vfp
armv5ehf-vfp armv5tehf-vfp armv6hf-vfp armv6thf-vfp armv7ahf-vfp
armv7at2hf-vfp armv7vehf-vfp armv7vet2hf-vfp armv7vehf-neon armv7vet2hf-neon
armv7vehf-neon-vfpv4 armv7vet2hf-neon-vfpv4 cortexa7hf-vfp cortexa7hf-neon
cortexa7hf-neon-vfpv4 cortexa7t2hf-vfp cortexa7t2hf-neon
cortexa7t2hf-neon-vfpv4 raspberrypi3) for DEFAULTTUNE (cortexa7thf-neon-vfpv4)
does not contain TUNE_PKGARCH (cortexa7hf-neonvfpv4).
Fix this by being more explicit about what we're modifying.
Reported-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: André Draszik <git@andred.net> Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Dan McGregor [Tue, 10 May 2016 16:04:47 +0000 (10:04 -0600)]
openjade-native: work around bug exposed by GCC 6
From: Daniel McGregor <daniel.mcgregor@vecima.com>
Simply turn off the optimzation that is causing this breakage. I had
originally used -fno-lifetime-dse, but -fno-tree-dse works at least
going back as far as gcc 4.8.
This isn't a real fix, but it allows openjade to work enough to complete
a build.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Michael Wood [Thu, 12 May 2016 13:31:58 +0000 (14:31 +0100)]
report-error: Replace the build directory path in the error text
Replace the TOPDIR in the output error file so that the error report once
submitted can then be more easily matched to find duplicate error
reports. This also reduces the need to manually redact any information that
might be in the error log path such as hostnames or home directories.
Signed-off-by: Michael Wood <michael.g.wood@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Maxin B. John [Thu, 12 May 2016 13:16:52 +0000 (16:16 +0300)]
connman: update to version 1.32
Removed following upstreamed/backported patches:
a) 0001-Detect-backtrace-API-availability-before-using-it.patch
b) 0001-iptables-Add-missing-function-item-of-xtables-to-mat.patch
Rearranged musl related patches.
Signed-off-by: Maxin B. John <maxin.john@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mariano Lopez [Thu, 12 May 2016 11:28:13 +0000 (11:28 +0000)]
package_manager.py: Add extract() method for opkg and dpkg
Sometimes it is needed to have the content of a package outside
the recipe context. This new method extract the content of an
IPK/DEB file to a tmpdir, without actually installing the package.
A new OpkgDpkgPM class was added to share the code for opkg and dpkg.
There were need some changes to opkg_query() in order to use it
with apt-cache output. Also set default values to avoid UnboundLocalError
[YOCTO #9569]
Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
image_types: adjust default level of XZ compression
XZ extreme compression method usually uses a lot more CPU time with not
that often big saving on space. Same goes with -6 level of compression.
Compression level -3 usually the best balance for time/size, especially
on big images.
Signed-off-by: Alexander D. Kanevskiy <kad@kad.name> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:45 +0000 (10:35 -0700)]
gcc-runtime, libgcc: Symlink c++ header and startup files in target_triplet for SDK use
We build SDKs such that gcc-cross-candian is built for only one
target *-*-linux and then use -muclibc or -mmusl to let it compile
code for other libc variants. This works fine when libc = glibc
however it does not work for c++ programs when libc != glibc since
there are c++ headers installed under ${includedir}/c++/${BINV}/${TARGET_SYS}
which is fine when gcc-runtime and gcc-cross-candian uses same --target options
gxx includedir searches in right triplet, but it fails with musl/uclibc
since gcc will look for glibc based triplet but gcc-runtime will install
them under musl/uclibc triplet.
This patch symlinks the musl/uclibc triplet to glibc triplet when libc != glibc
This fixes SDKs for musl/uclibc
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:40 +0000 (10:35 -0700)]
musl: Create symlinks for stub libraries
Some libraries e.g. libm.so are needed to be
created so that SDKs built with distros which
disable static librararies can have the stubs
and since default linker script requires -lm
this helps in compiling applications with SDK
there are .a equivalents for these libraries
but they do not land in SDKs when static libs
are disabled distrowide
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Marek Vasut [Wed, 11 May 2016 17:35:39 +0000 (10:35 -0700)]
bitbake: Oldest kernel for nios2 is 3.19
The oldest kernel version for nios2 is 3.19.0, set it so, otherwise
qemu-nios2 -r ${OLDEST_KERNEL} fails with "FATAL: kernel too old"
message from glibc.
Signed-off-by: Marek Vasut <marex@denx.de> Cc: Richard Purdie <richard.purdie@linuxfoundation.org> Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:37 +0000 (10:35 -0700)]
gdb: Disable binutils components
when building from git we also get the binutils
components since the tree for gdb and binutils is
same, this can then casue building binutils with
gdb as well and cause packaging conflicts, hence
disable the binutils pieces in configure itself
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:33 +0000 (10:35 -0700)]
libgcc: Ensure that gcc configure options are passed to libgcc too
libgcc uses certain options from EXTRA_OECONF as well, curently we are
ignoring them, as a result we do not configure libgcc to match cross gcc
in some cases e.g. ppc/musl should have used 64bit long doubles but
it went for 128-ldbls which is default, works on glibc but not on musl
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:31 +0000 (10:35 -0700)]
libgcc: Ensure that gcc configure options are passed to libgcc too
libgcc uses certain options from EXTRA_OECONF as well, curently we are
ignoring them, as a result we do not configure libgcc to match cross gcc
in some cases e.g. ppc/musl should have used 64bit long doubles but
it went for 128-ldbls which is default, works on glibc but not on musl
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Andrew Kelley (1):
fix incorrect protocol name and number for egp
Bobby Bingham (1):
add powerpc64 port
LeMay, Michael (1):
fix redundant processing of --build flag in configure script
Petr Vaněk (1):
remove dead store in res_msend
Rich Felker (10):
fix undefined pointer comparison in stdio-internal __toread
fix regression disabling use of pause instruction for x86 a_spin
fix read past end of haystack buffer for short needles in memmem
add support for mips and mips64 r6 isa
add mips n32 port (ILP32 ABI for mips64)
fix thread structure/dtv-pointer corruption on powerpc
fix FILE buffer underflow in ungetwc
update COPYRIGHT file to clarify that permissions apply for all files
follow standard configure behavior for cross compile prefix
fix spurious trailing whitespace in powerpc & powerpc64 bits/errno.h
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:28 +0000 (10:35 -0700)]
bitbake.conf: Empty out BUILDSDK_CPPFLAGS
The nativesdk sysroot is already being passed in
via --sysroot option, adding -isystem is redundant
and more so it causes build failures with gcc-6
since it messed with internal search order for system
includedirs
see
https://lists.fedoraproject.org/archives/list/devel@lists.fedoraproject.org/thread/Q5SWCUUMWQ4EMS7CU2CBOZHV3WZYOOTT/
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:14 +0000 (10:35 -0700)]
nss: Upgrade to 3.23
Disable Werror on native builds. This helps
in building nss-native on build hosts which have
gcc < 4.9 eg. ubuntu 14.04
The real issue is that we use headers from native
staging sysroot and it has the updated glibc headers
which then ends up with errors e.g.
| In function 'memset',
| inlined from 'sec_PKCS7Encrypt' at p7local.c:715:14:
| /usr/include/x86_64-linux-gnu/bits/string3.h:81:30: error: call to '__warn_memset_zero_len' declared with attribute warning: memset used with constant zero length parameter; this could be due to transposed parameters [-Werror]
| __warn_memset_zero_len ();
| ^
| cc1: all warnings being treated as errors
| make[2]: *** [Linux3.4_x86_64_glibc_PTH_64_OPT.OBJ/p7local.o] Error 1
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:11 +0000 (10:35 -0700)]
webkitgtk: Upgrade to 2.12.1
Fix build with gcc-6
Do not tinker with -isystem, this causes issues
where gcc does not find standard C++ headers since
the default search ordered for system headers is
changed and is not correct anymore. Errorr like
WebCore/editing/gtk/EditorGtk.cpp:27:
/mnt/oe/poky/build/tmp/sysroots/qemuarm/usr/include/c++/6.0.1/cstdlib:75:25: fatal error: stdlib.h: No such file or directory
#include_next <stdlib.h>
It brings a patch to fix typename mismatch
for pow()
see
https://bugs.webkit.org/show_bug.cgi?id=153071
Drop upstreamed patches
User -fPIC on aarch64 to fix
relocation truncated to fit: R_AARCH64_LD64_GOTPAGE_LO15 against symbol `__stack_chk_guard@@GLIBC_2.17' defined in .data.rel.ro section
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 11 May 2016 17:35:06 +0000 (10:35 -0700)]
glib-2.0: Ignore useless warning found with gcc-6
../../glib-2.46.2/glib/gdate.c:2497:7: error: format not a string literal, format string not checked [-Werror=format-nonliteral]
tmplen = strftime (tmpbuf, tmpbufsize, locale_format, &tm);
^~~~~~
| ../../../../../../../../workspace/sources/glib-2.0/glib/tests/gdatetime.c: In function 'test_strftime':
| ../../../../../../../../workspace/sources/glib-2.0/glib/tests/gdatetime.c:1338:3: error: '%c' yields only last 2 digits of year in some locales [-Werror=format-y2k]
| "a%a A%A b%b B%B c%c C%C d%d e%e F%F g%g G%G h%h H%H I%I j%j m%m M%M " \
Additionally fix the problem seen where write() return code is ignored
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
staging.bbclass: Make it easier to define which dirs to stage
The directories that should be staged in the sysroot are now specified
in the SYSROOT_DIRS variable. Extra directories that should be staged
for native are specified in SYSROOT_DIRS_NATIVE. Finally, directories
that should not be staged are specified in SYSROOT_DIRS_BLACKLIST.
This also removes the sysroot_stage_libdir() function as it is no
longer used (it was just a wrapper for sysroot_stage_dir()).
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Randy Witt [Mon, 9 May 2016 22:46:27 +0000 (10:46 +1200)]
devtool: Fix build-sdk when pn doesn't match filename
If an image with the filename foo.bb could be built using the name "bar"
instead, then build-sdk would fail to create the derivative sdk.
This was because the code assumed that the file name matched the target,
which is not necessarily the case.
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com> Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Paul Eggleton [Mon, 9 May 2016 22:25:05 +0000 (10:25 +1200)]
classes/insane: do not force LIC_FILES_CHKSUM unless SRC_URI is set
Recipes such as images and packagegroups don't actually fetch or build
any source, so there's really no point having LIC_FILES_CHKSUM set.
Forcing users to set it (as we have done for images inheriting
image.bbclass) just makes things more difficult for the user for no
discernable benefit. The easiest way to adjust this check is simply to
skip it if SRC_URI is not set since this is a pretty good indicator that
no source is being pulled in.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 9 May 2016 13:12:16 +0000 (14:12 +0100)]
update-alternatives: Fix determinism issue
getVarFlags returns a dict and there is therefore no sort order. This
means the order of the X_VARDEPS_X variables can change and hence the
task checksums can change. This can lead to rebuilds of any parts of
the system using update-alternatives and their dependees. This is a
particular issue under python v3.
Add in a sort to make the order of the variables deterministic.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Mon, 9 May 2016 13:09:37 +0000 (14:09 +0100)]
image: Fix IMAGE_FEATURES determinism issue
remain_features uses a dict which means the order is not deterministic. This
can lead to the task hash changing depending on the state of the memory at
parse time. This is particularly noticeable under python v3.
Since the dict is helpful in constructing the data, pass the data through
sort() so the order is always deterministic.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
metadata_scm.bbclass: Do not assume ${COREBASE} is a Git repo
The functions base_detect_revision() and base_detect_branch() try to
extract SCM meta information from the path returned by
base_get_scmbasepath(), which currently returns ${COREBASE}. However,
making the assumption that ${COREBASE} contains SCM meta information
can be false. It is true for Poky, but not necessarily other
environments. A better option is to look for the SCM meta information
based on the meta layer.
Since this works as expected for Git but not SVN, the call to
base_get_metadata_svn_revision() from base_detect_revision() was also
removed. This is not expected to affect anyone (partly based on the
comment in base_get_metadata_svn_revision()).
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Dengke Du [Fri, 29 Apr 2016 08:36:45 +0000 (04:36 -0400)]
lttng-tools: filter random filename of ptest output
When run the ptest of lttng-tools, it produced many random filename
when the tests passed, the output confused QA analysis, so we need
to filter the ptest output if tests passed and add up the passed and
failed tests.
NOTE:The tests invoked the run.sh twice, so it output like this:
...
FAIL:...
unit_tests statistics
total pass: 133 tests passed!
total fail: 5 tests failed!
...
FAIL:...
fast_regression statistics
total pass: 1904 tests passed!
total fail: 202 tests failed!
Signed-off-by: Dengke Du <dengke.du@windriver.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
André Draszik [Tue, 26 Apr 2016 09:27:59 +0000 (10:27 +0100)]
feature-arm-neon.inc: restore vfpv3-d16 support
Commit 6661718 (feature-arm-{neon,vfp}.inc: refactor and fix issues)
effectively changed the gcc -mfpu= option from -mfpu=vfpv3-d16 to
-mfpu=vfpv3d16, which gcc doesn't understand.
Restore the original value.
After doing that, we also need to adjust ARMPKGSFX_FPU which should
contain the same value without dash '-' as it is used that way
throughout.
Signed-off-by: André Draszik <git@andred.net> Signed-off-by: Ross Burton <ross.burton@intel.com>
Randy Witt [Mon, 2 May 2016 00:13:02 +0000 (17:13 -0700)]
populate_sdk_ext: Change lockedsigs task mismatch to a warning
It has been determined that it is highly likely that users might get
signatures that don't match in an extensible sdk. This doesn't
necessarily happen with oe-core, so we can set the mismatch to an error
during testing if we like.
However, for the case where users are creating their own sdks, we don't
need an error halting their progress. locked-sigs will still function as
it should.
Signed-off-by: Randy Witt <randy.e.witt@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Andreas Müller [Fri, 29 Apr 2016 20:59:46 +0000 (22:59 +0200)]
xserver-xorg: update 1.18.3 and add PACKAGECONFIG for glamor
* drm-version patch was applied mainline
* libsystemd was solved similar to patch upstream
* glamor was enhanced twice by Eric Anholt and makes vc4 work on Raspberry Pi
* glamor does not link against libegl but won't work without libegl
Signed-off-by: Andreas Müller <schnitzeltony@googlemail.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
populate_sdk_ext.bbclass : Show logfile in case the SDK EXT installation failed
To avoid lots of output in the SDK EXT installation phase, system redirects
it to a logfile ($target_sdk_dir/preparing_build_system.log) but in case of error,
the contents should be shown so debugging could be faster.
[YOCTO #9576]
Signed-off-by: Leonardo Sandoval <leonardo.sandoval.gonzalez@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Patrick Ohly [Mon, 2 May 2016 13:27:28 +0000 (15:27 +0200)]
combo-layer: implement "update with history"
The core idea is that all commits get imported, including merge
commits, and joined into one big merge commit that imports the changes
from the individual components into the main branch of the combined
repository.
This is done by copying the files in each commit and removing deleted
ones, instead of trying to patch the combined repository.
The advantages of doing updates in this mode are:
- works for arbitrary upstream repos, not just those which
support conversion into a linear set of patches
- listing history shows that commits where developed
independently in the different components, instead of
artificially showing them as if they had been developed
one after the after (component "aaaa" before "bbbb", then "ccc", ...)
- bisecting becomes easier: when upstream repos only ensure consistency
when merging into their "master" branches, then those merge
commits are good candidates for test builds also in the combined
repo
- more data mining can be done, for example showing who merged a commit
and when
Selecting a subset of the files is supported, albeit with a slight
different semantic for wild card matching compared to other code paths
(/ is matched by * and ?). Empty commits get skipped because typically
they are a result of filtering (but that is not checked, so
intentionally empty commits also get skipped).
Other code paths are intentionally left unchanged, to avoid
regressions. However, the downside is that some opportunities for
refactoring (in particular regarding file filtering) were ignored.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Patrick Ohly [Mon, 2 May 2016 13:27:27 +0000 (15:27 +0200)]
combo-layer: runcmd() enhancements
Allow setting the environment. Due to a subprocess quirk, it must
always be set explicitly (reuses the one from the previous call if not
set, instead of falling back to os.environ).
Embedding nul characters will be useful for parsing git output more
reliably; support dumping such output a bit better.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Patrick Ohly [Mon, 2 May 2016 13:27:26 +0000 (15:27 +0200)]
combo-layer: dummy "update with history"
When setting "history = True" in combo-layer.conf consistently for the
components involved in an update or using "update" together with the
"--history" command line flag, a new mode for updating will be used
that does not rely on exporting/importing patches.
A config setting is used because it should be used consistently by
everyone using the same config, without having to remember to use an
additional command line parameter.
There are no real global settings, so the setting is checked
separately for each component although the setting has to be set
consistently. This restriction could be removed later.
In practice, putting "history" into the "[DEFAULT]" section is the
easiest approach for configuring it.
The actual code changes split up action_update and the
combo-layer.conf handling in preparation for this new mode, without
implementing the mode itself.
Signed-off-by: Patrick Ohly <patrick.ohly@intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
Mariano Lopez [Mon, 2 May 2016 11:27:10 +0000 (11:27 +0000)]
sshcontrol.py: Add methods to copy dirs and delete files
This patch add new methods to SSHControl class. These methods
include:
- Copy a dir to DUT
- Delete a file in the DUT
- Delete a directory in the DUT (if empty)
- Delete a directory structure in the DUT
[YOCTO #9565]
Signed-off-by: Mariano Lopez <mariano.lopez@linux.intel.com> Signed-off-by: Ross Burton <ross.burton@intel.com>
classes/image_vm: allow different filesystems to be used for VM images.
This allows for things like btrfs to be used vs just ext4.
The default value of ext4 is kept so there is no functional
change unless VM_ROOTFS_TYPE is set in the inherting recipe.
Signed-off-by: Ian Reinhart Geiser <geiseri@geekcentral.pub> Signed-off-by: Ross Burton <ross.burton@intel.com>
