Age | Commit message (Collapse) | Author |
|
The "sdkbasepath + '/conf/local.conf.bak" doesn't exist when
"oe.copy_buildsystem.check_sstate_task_list()" fails, then os.replace() would
raise FileNotFoundError, which overcomes the real error. Keep the error status
makes debug easier, so remove the try..finally.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
The "bitbake image -cpopulate_sdk/ext" generates SDK/eSDK for all multilib
variants, so "bitbake lib32-image -cpopulate_sdk/ext" is not needed, and it
doesn't work well, for example:
MACHINE ?= "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
$ bitbake lib32-core-image-minimal -cpopulate_sdk_ext
[snip]
Exception: FileExistsError: [Errno 17] File exists: '/buildarea/lyang1/test_q64/tmp/sysroots-components/core2-64/openssl/sysroot-providers/openssl10' -> '/buildarea/lyang1/test_q64/tmp/work/qemux86_64-pokymllib32-linux/lib32-core-image-minimal/1.0-r0/lib32-recipe-sysroot/sysroot-providers/openssl10'
[snip]
The problem is populate_sdk_ext installs all multilib variants, and
extend_recipe_sysroot() handles foo-image depends lib32-foo-image, but doesn't
handle lib32-foo-image depends foo-image, we can use a lot of trick ways to make
it work:
1) Get foo-image's RECIPE_SYSROOT when build lib32-foo-image
2) Handle conflicts with foo-image.do_rootfs
3) Handle conflicts when "bitbake lib32-foo-image foo-image -cpopulate_sdk_ext"
And maybe other potential problems, this looks painful, so just delete the task.
[YOCTO #12210]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
Install all MULTILIB_VARIANTS to sysroot and pack them only makes sense to
image recipe, so limit it to image recipe only, and it can fix:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "core2-32"
$ bitbake lib32-meta-ide-support
WARNING: lib32-meta-ide-support-1.0-r3 do_configure: Manifest for ncurses not found, searched manifests:
/workspace2/lyang1/test_mlwarn/tmp-glibc/sstate-control/manifest-qemux86_64-ncurses.populate_sysroot
/workspace2/lyang1/test_mlwarn/tmp-glibc/sstate-control/manifest-core2-32-ncurses.populate_sysroot
/workspace2/lyang1/test_mlwarn/tmp-glibc/sstate-control/manifest-i586-ncurses.populate_sysroot
/workspace2/lyang1/test_mlwarn/tmp-glibc/sstate-control/manifest-x86-ncurses.populate_sysroot
/workspace2/lyang1/test_mlwarn/tmp-glibc/sstate-control/manifest-allarch-ncurses.populate_sysroot
This is because core2-64 is not in lib32-meta-ide-support's
PACKAGE_ARCHS which doesn't like image recipe (lib32-image recipe's
DEFAULTTUNE is not overridded)
[YOCTO #12298]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
This can fix do_rootfs and depends chaos when multilib, for example
ca-certificates is an allarch recipe:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "core2-32"
IMAGE_INSTALL_append = " ca-certificates"
$ bitbake ca-certificates
$ rpm -qpR tmp-glibc/deploy/rpm/noarch/ca-certificates-20170717-r0.noarch.rpm
[snip]
openssl
[snip]
It requires openssl, but lib32-openssl doesn't proivide it:
$ rpm -qp --provides tmp-glibc/deploy/rpm/core2_32/lib32-openssl-1.0.2l-r0.core2_32.rpm
lib32-openssl = 1.0.2l-r0
So openssl-1.0.2l-r0.core2_64.rpm would be installed when install
ca-certificates, e.g.:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "core2-32"
IMAGE_INSTALL_append = " ca-certificates"
$ bitbake lib32-core-image-minimal
The openssl-1.0.2l-r0.core2_64.rpm and other 64 bit packages are installed,
this is incorrect.
We have two solutions on this:
1) Let lib32-openssl provide openssl, but this would cause more problems when
building since more than one providers for openssl.
Or
2) Extend allarch recipes, there would be ca-certificates-20170717-r0.noarch.rpm
and lib32-ca-certificates-20170717-r0.noarch.rpm, then ca-certificates
requires openssl, and lib32-ca-certificates requires lib32-openssl, the
packages will be installed correctly, this is the best solution that I can
find.
The problem has existed for years for deb and ipk (rpm did work since it didn't
use lib32-BPN in rpm filename before changed to dnf, for example, there was no
lib32-openssl.core2_32.rpm, but openssl.core2_32.rpm), this patch can fix the
problem for rpm, ipk and deb.
[YOCTO #12288]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
The PACKAGE_INSTALL is only used by image recipe, the previous code had
handled it in "if bb.data.inherits_class('image', d)", handle it again
doesn't make any sense (there is no PACKAGE_INSTALL for non-image
recipe), so remove it.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
It had reversed all the archs, which mixed multilib and common archs, e.g.:
"all any noarch x86_64 core2-64 qemux86_64 x86 i586 core2-32"
After reversed:
"core2-32 i586 x86 qemux86_64 core2-64 x86_64 noarch any all"
The core2-32 has a higher priority than core2-64 after reversed which is
incorrect. Don't mix with mulitlib when reverse can fix the problem, and let
multilib archs have a higher priority for multilib image.
Fixed:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "core2-32"
IMAGE_INSTALL_append = " ca-certificates"
$ bitbake wrlinux-image-glibc-small
Check rootfs.manifest, no core_32 packages is installed, this is
correct, but after we build lib32-bash, it will be incorrect:
$ bitbake lib32-bash
$ bitbake wrlinux-image-glibc-small
Check rootfs.manifest, a few lib32 packages are installed, such as
lib32-bash, this is incorrect. It was because ca-certificates is
allarch, and requires /bin/sh, which is provided by both bash and
lib32-bash, and lib32-bash has a higher priority than bash, so it would
be installed.
[YOCTO #12288]
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
Fixed:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "core2-32"
$ bitbake lib32-wrlinux-image-glibc-small
WARNING: lib32-wrlinux-image-glibc-small-1.0-r1 do_rootfs: Manifest for lib32-qemuwrapper-cross not found, searched manifests:
/path/to/tmp/sstate-control/manifest-qemux86_64-lib32-qemuwrapper-cross.populate_sysroot
/path/to/tmp/sstate-control/manifest-core2-64-lib32-qemuwrapper-cross.populate_sysroot
/path/to/tmp/sstate-control/manifest-x86_64-lib32-qemuwrapper-cross.populate_sysroot
/path/to/tmp/sstate-control/manifest-allarch-lib32-qemuwrapper-cross.populate_sysroot
It was because image recipes' multilib PACKAGE_EXTRA_ARCHS are not changed by
mutlilib.bbclass, so extend_recipe_sysroot() didn't know anything about
mutlilib's PACKAGE_EXTRA_ARCHS, now fix it.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
The old warning was:
WARNING: lib32-wrlinux-image-glibc-small-1.0-r1 do_prepare_recipe_sysroot: Manifest /path/to/tmp/sstate-control/manifest-allarch-lib32-qemuwrapper-cross.populate_sysroot not found?
The message wasn't clear enough, it searched a few manifests, but only
reported the last one, which confused user.
Now the warning is:
WARNING: lib32-wrlinux-image-glibc-small-1.0-r1 do_rootfs: Manifest for lib32-qemuwrapper-cross not found, searched manifests:
/path/to/tmp/sstate-control/manifest-qemux86_64-lib32-qemuwrapper-cross.populate_sysroot
/path/to/tmp/sstate-control/manifest-core2-64-lib32-qemuwrapper-cross.populate_sysroot
/path/to/tmp/sstate-control/manifest-x86_64-lib32-qemuwrapper-cross.populate_sysroot
/path/to/tmp/sstate-control/manifest-allarch-lib32-qemuwrapper-cross.populate_sysroot
Whick makes debug easier.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
Fixed:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
$ bitbake core-image-sato lib32-core-image-sato
$ bitbake build-sysroots
All lib32 manifests which had been built should be installed, but only a few
such as qemux86_64-lib32-base-files were installed (it was installed because
MACHINE_ARCH was still qemux86-64 when multilib), but others such as
x86-lib32-zlib were not installed, this was incorrect. For multilib builds,
fix-up overrides to prepend :virtclass-multilib- and then again append
additional multilib arches from the PACKAGE_EXTRA_ARCHS list if needed can fix
the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
It only substituted staging_target for target recipe which didn't work
for multilib, for example, postinst-useradd-lib32-polkit:
* No multilib:
PATH=/path/to/tmp-glibc/work/core2-64-wrs-linux/polkit/0.113-r0/recipe-sysroot-native/bin
staging_target=/path/to/tmp-glibc/work/core2-64-wrs-linux/polkit/0.113-r0/recipe-sysroot
The PATH would be substituted to:
FIXMESTAGINGDIRTARGET-native/bin
Not the funny "-native/bin", this works well.
* When multilib:
PATH=/path/to/tmp-glibc/work/core2-32-wrsmllib32-linux/lib32-polkit/0.113-r0/recipe-sysroot-native/bin
staging_target=/path/to/tmp-glibc/work/core2-32-wrsmllib32-linux/lib32-polkit/0.113-r0/lib32-recipe-sysroot
Now staging_target endswith "/lib32-recipe-sysroot", so it can't
replace '/recipe-sysroot-native' in PATH , so PATH can't be fixed, and
there would be build errors when building multilib + rm_work, for
example:
chown: invalid user: ‘polkitd:root’
Substitute staging_host for target recipe can fix the problem, now all
of native, cross and target need substitute staging_host, so we can
simply the code a little.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
Fixed:
MACHINE = "qemux86-64"
require conf/multilib.conf
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
$ bitbake <image> -cpopulate_sdk_ext
$ bitbake <image> -ctestsdkext
[snip]
Standard Output: /bin/sh: 1: i686-wrsmllib32-linux-gcc: not found
[snip]
It was failed because no lib32 toolchain or lib installed.
This patch fixes:
* Set SDK_TARGETS correctly
* Return multilib depends in get_ext_sdk_depends()
* Write information to all environment-setup-* scripts.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
Fixed:
MACHINE = "qemux86-64"
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
$ bitbake <image> -cpopulate_sdk_ext
[snip]
Exception: subprocess.CalledProcessError: Command 'sed -e [snip]'
[snip]
Subprocess output:
[snip]
sed: can't read /path/to/work/qemux86_64-wrs-linux/wrlinux-image-glibc-small/1.0-r1/recipe-sysroot/usr/lib/perl/5.24.1/ExtUtils/Liblist/Kid.pm: No such file or directory
[snip]
It was failed because "/usr/lib" is in qemux86_64-wrsmllib32-linux/, not in
qemux86_64-wrs-linux. The code has considered mutitlib, but seems not
completed, the multilib "variant" was not in fixme, so it wasn't handled
correctly, this patch fixes the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
|
|
Make sure build host references do not end up being compiled in the image.
This only affects libicutu and icu-dbg.
Signed-off-by: Juro Bystricky <juro.bystricky@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Improve reproducibility: Remove all build host references from
distributed files.
Do not package non-linux OS related files.
Also remove some additional files not needed by run-ptest. (There are
probably still more files that can be removed, but as long as they
don't leak build host references they are harmless).
Fix the expected checksum of MakeMaker/lib/ExtUtils/Liblist/Kid.pm
to match the one expected: We modified the file, but did not
recalculate/update the file checksum accordingly.
(This fixes the only failing test.)
Signed-off-by: Juro Bystricky <juro.bystricky@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Remove various build host references scattered within
comments in numerous files that are distributed in the package.
Signed-off-by: Juro Bystricky <juro.bystricky@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We don't have latex2man in HOSTTOOLs so documentaion is never built but this
dependency does cause problems on older releases like morty, pre-HOSTTOOLS.
Document the configuration explicitly in master.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
seds can get stale without warning and repeated application can cause problems,
so move the gtk-doc seds into a patch.
(From OE-Core rev: a704411ab0f1e5f8cbf57ff54b36d60ccaf0d223)
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
cmake.bbclass now ensures that ${datadir}/cmake ends up in the dev package,
so this recipe doesn't need to do so itself.
Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
cmake.bbclass now ensures that ${libdir}/cmake ends up in the dev package,
so this recipe doesn't need to do so itself.
Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
cmake.bbclass now ensures that ${libdir}/cmake and ${datadir}/cmake end up
in the dev package, so this recipe doesn't need to do so itself.
Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
cmake.bbclass now ensures that ${libdir}/cmake ends up in the dev package,
so this recipe doesn't need to do so itself.
Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
cmake.bbclass now ensures that ${datadir}/cmake ends up in the dev package,
so this recipe doesn't need to do so itself.
Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Various recipes that inherit cmake contain FILES_${PN}-dev magic to add the
generated package files to their -dev packages. Since this is a standard
feature of cmake, we might as well teach cmake.bbclass to do this itself so
those recipes can be simpler.
Signed-off-by: Mike Crowe <mac@mcrowe.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
It had been removed since 2011:
commit b774bf44ef004276da12a83ebd69715c00b596ac
Author: Lianhao Lu <lianhao.lu@intel.com>
Date: Tue Aug 16 16:26:49 2011 +0800
package(_ipk).bbclass: opkg using ALL_MULTILIB_PACKAGE_ARCHS
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This can make debug easier, for example, makes it easy to run the
command mannually.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
It is duplicated to previous.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
It had been dropped by:
commit 65581c68d130fa74d703f6c3c92560e053857ac7
Author: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Date: Mon Feb 13 16:44:48 2017 +0200
rootfs_rpm.bbclass: migrate image creation to dnf
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Otherwise it doesn't work since SYSTEMD_SERVICE_volatile-binds is not defined
when multilib.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
It had a problem when nested layer before, e.g.:
layer_a/layer_b/
And when layer_b is handled before layer_a, then layer_a dir existed, so
it would be treated as already handled, which was wrong, check
conf/layer.conf can fix the problem.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The previous code:
os.rename(sdkbasepath, temp_sdkbasepath)
try:
foo
finally:
os.rename(temp_sdkbasepath, sdkbasepath)
always renamed the path, it made the debug harder when error happened.
drop the "try: finally" makes the debug easier.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Otherwise it would generate lines like the following when multilib:
INHERIT += "own-mirrors"SSTATE_MIRRORS += " \n file://.* file:///path/to/../share/sstate-cache/PATH"
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This is very useful for debugging. The similar to testsdk.bbclass.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
pulse is added as a system user, so the group 'pulse' is
meant to be a system group as well, which is the same with
other distros like ubuntu/centos.
Signed-off-by: Jackie Huang <jackie.huang@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The packages loosely follow the debian package names. In that way more
projects, e.g. Qt5 for 16-bit, are able use system libraries. This does
not change the existing default package.
Signed-off-by: Stefan Müller-Klieser <s.mueller-klieser@phytec.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If you have a recipe which does not include SRCPV in PV but does set
SRCREV = "${AUTOREV}" and you run do_fetch, then change the repo to a
new commit then run do_unpack, do_unpack will fail since the new commit
doesn't exist in the repo that was fetched.
The problem is the revision chosen is not represented in the do_fetch
task hash. It if were, the fetch would rerun first and the commit would be
present. It works when PV includes SRCPV since that does contain the chosen
commit from the AUTOREV.
The solution is to include the SRCPV value into the representation of AUTOREV
used for checksum calculation purposes.
Add a selftest for this issue.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
To make it easier to debug problems with renaming caused by debian.bbclass,
explicitly log when packages are renamed.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This recipe installs the test suite by copying the entire build tree into
/usr/share/ptest, which is both wasteful and breaks packaging as lz4-ptest then
gets renamed by debian.bbclass to liblz4.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
When the gtk-doc API documentation is enabled (via the api-documentation DISTRO_FEATURE, typically) this recipe fails to build:
WORKDIR/build/docs/validate/gst-validate-scan: line 117:
WORKDIR/build/docs/validate/.libs/lt-gst-validate-scan:
No such file or directory
Forcibly disable the gtk-doc for now.
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We see occasional network glitches which break this test. Use our own
mirror (which has a .gz instead of .bz2) to avoid the errors, we're not
trying to test network connectivity.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
For various reasons we need to be able to set and override this from
auto.conf on our test infrastructure. We have tried forcing the variable
but this then breaks other selftests. In the interests of not complicating
things further and needing to modify the tests across releases, weaken
the default assignment.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Switch to using current repository, switch to meson (following
upstream), rename the recipe like upstream.
Add a patch to install the player binaries.
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Drop autotools-specific patches.
Rename polkit packageconfig option to sysprofd as 'polkit' does not
at all match what is happening.
Remove --enable-compiler-warnings, as the equivalent in meson
could not be found.
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Drop libunwind option, as it is only used if tests are enabled
(and they're unconditionally not enabled).
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add a patch to work around absence of dlvsym() on musl
(wasn't previously a problem as autotools weren't building tests by default)
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Note that meson flags for gobject introspection and gtk-doc
appear to be non-standardized; going forward we should devise
a common way to deal with it.
gettext inherit is removed, as there is no equivalent functionality
in meson; NLS bits are always built and installed.
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The update to 0.44.0 did not add this patch required for qt builds.
Signed-off-by: Ricardo Ribalda Delgado <ricardo.ribalda@gmail.com>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Although the meson crossfile should take care of setting the right cross
environment for a target build, meson slurps any set CFLAGS, CXXFLAGS,
LDFLAGS, and CPPFLAGS from the environment and injects them into the
build (see mesonbuild/environment.py:get_args_from_envvars for details).
This means that we are seeing native CFLAGS, CXXFLAGS, LDFLAGS, and
CPPFLAGS in the target build, which is wrong and causes build failures
when target and native have libraries in common (the linker gets
confused and bails).
That said, we *do* need to set certain vars for all builds so that meson
can find the right build tools. Without this, meson will fail during its
sanity checking step because it will determine the build tools to be
unrunnable since they output target instead of native artifacts.
The solution to all of this is to set CC, CXX, LD, and AR globally to
the native tools while setting the other native vars *only* for the
native build. For target builds, these vars will get overridden by the
cross file as we expect.
Signed-off-by: Martin Kelly <mkelly@xevo.com>
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
gnomebase.bbclass hardcodes the autotools inherit, so make it
configurable and allow meson to be specified instead.
Signed-off-by: Alexander Kanavin <alexander.kanavin@linux.intel.com>
[RP: patches tweaked to only need the one class]
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|