Feature wise, this is a match for 4.19, with yaffs2 and aufs4 being
kept around for another release. Aufs is scheduled for removal in
the fall release.
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Wed, 6 Mar 2019 16:37:10 +0000 (11:37 -0500)]
linux-libc-headers: update to 5.x headers
Updating the linux-libc-headers to the 5.x kernel variant to match
the latest reference kernel in the 2.7 release.
We have two patches refreshed for context changes, and three patches
dropped since they have been merged to the mainline kernel and are
no longer necessary.
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Wed, 6 Mar 2019 16:37:09 +0000 (11:37 -0500)]
kernel-yocto: allow kernel-meta repositories to be patched
For testing purposes, it is often easier to patch a fragment
in a kernel-meta repository versus needing to make a copy or
modify the source repository.
We can allow this sort of patching when a patchdir of kernel-meta
is passed (to indicate the nested kernel-meta repository).
Also note that we must patch the meta data before they are
processed/gathered, since migrated copies to the kernel source
directory will be used later.
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Wed, 6 Mar 2019 00:19:19 +0000 (16:19 -0800)]
glibc: Drop upstream rejected patches
These patches were applied, hoping that they will eventually be accepted
upstream but they have been rejected, I think its best that they are
dropped so we can avoid novel unintended behaviours that no other
distros will be seeing
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Robert Yang [Tue, 5 Mar 2019 03:30:35 +0000 (11:30 +0800)]
send-error-report: Add --no-ssl to use http protocol
The script use https protocol by default, but the error-report-web server's
https connection may not work (e.g., doesn't work with python 2.7.6), so add an
option --no-ssl to make it use http.
Signed-off-by: Robert Yang <liezhi.yang@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
oe.scriptutils: enable color in a more flexible way
Rather than recreating handlers and forcing them, iterate over the handlers
and enable color on ones we can handle. This makes it easier to handle color
properly when we introduce the bb.msg default log filters.
Signed-off-by: Christopher Larson <chris_larson@mentor.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Jaewon Lee [Sat, 15 Dec 2018 00:39:22 +0000 (16:39 -0800)]
devtool: Support kmeta directory usage with devtool modify/finish
When using Kmeta directories, devtool finish will add every single file
in the directory to the bbappend. This is because in the current
implementation, the get_recipe_local_files function treats the kmeta
directory like a file. Modifying the function to loop through the
provided directories and return all included files instead of just the
top level directory. This will enable correct file to file comparison
when determing which files are new/changed and need to be added to the
bbappend.
Adding an extra check in devtool-source.bbclass to not copy the cfg file
if its already included somewhere in the kmeta directory
Also during 'modify', when moving necessary files in the kmeta directory
from the workdir to oe-local-files, the dangling parent directories are
left behind. This in itself is not an issue as the temporary devtool
workspace is automatically deleted, but this causes an incorrect include
directory to be added in kernel-yocto.bbclass. Changing the order of
the if statements to catch the correct conditional. This is safe to do
as when not in the devtool context, there will be no oe-local-files
directory.
Signed-off-by: Jaewon Lee <jaewon.lee@xilinx.com> Signed-off-by: Alejandro Enedino Hernandez Samaniego <alejandr@xilinx.com> Acked-by: Paul Eggleton <paul.eggleton@linux.intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This patch was to use BUILD_CC instead of CC when building mktables, a binary
used to generate tables.c.
However due to cross-compilation problems such as this, upstream since 4.12.1
(f929b62 specifically) avoids the entire issue and ships tables.c in the
repository, updating it as required.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Mon, 4 Mar 2019 17:31:41 +0000 (17:31 +0000)]
packagegroup-core-tools-testapps: remove lrzsz
We can assume that modern boards don't need X/Y/ZMODEM transfer tools for
testing purposes. The few users that need this tool for testing can add it
themselves.
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
initscripts: Add support for /etc/default/timestamp
Source /etc/default/timestamp if present. This allows the rootfs to be
read-only / store the file in a persistant location if the whole rootfs
is updated.
Signed-off-by: Jeroen Hofstee <jhofstee@victronenergy.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
André Draszik [Mon, 4 Mar 2019 09:07:24 +0000 (09:07 +0000)]
masterimage: remove code duplication
Two identical classes exist for SystemdbootTarget,
introduced in commit f9a61d3400ad
("gummiboot: Remove/change gummiboot references with systemd-boot")
poky commit 2dce2648e3d8
Remove the duplicated copy.
Signed-off-by: André Draszik <andre.draszik@jci.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ovidiu Panait [Mon, 4 Mar 2019 11:38:45 +0000 (13:38 +0200)]
openssl: Add cryptodev-linux PACKAGECONFIG
The old bsd cryptodev engine was removed in
https://github.com/openssl/openssl/pull/3699
and the new one added in:
https://github.com/openssl/openssl/pull/3744
It can be enabled by configuring with "enable-devcryptoeng".
Signed-off-by: Ovidiu Panait <ovidiu.panait@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Luca Boccassi [Mon, 4 Mar 2019 12:42:08 +0000 (12:42 +0000)]
swig: fix build with musl
Building with musl currently fails. The fix has been committed
upstream, so backport it.
Note that it requires running yacc again to regenerate the parser file,
so a bison-native dependency is needed. Also the upstream's configure
step creates the directory where the parser file is generated at
make-dist time, and for some reason running through bitbake does not
create it, so do it manually. All of this can be reverted once a new
swig version that includes the fix is released and rebased onto.
Signed-off-by: Luca Boccassi <luca.boccassi@microsoft.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
do_create_extlinux_config performs its own override processing
for several variables, so we have to explicitly add the label-
suffixed variable names to its vardeps to make sure that changes
get detected.
Signed-off-by: Matt Madison <matt@madison.systems> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ming Liu [Sun, 3 Mar 2019 09:08:41 +0000 (10:08 +0100)]
gdk-pixbuf: export XDG_DATA_DIRS in wrappers
We need export XDG_DATA_DIRS to be ${STAGING_DATADIR} for gdk-pixbuf*
wrappers or else they will try to load mime from /usr/share/mime of the
host machine.
Signed-off-by: Ming Liu <liu.ming50@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 2 Mar 2019 15:00:43 +0000 (15:00 +0000)]
selftest: Add multiconfig test
Add a test for a multiconfig build which mixes tiny and musl builds
along with using the mcextend class to combine and package multiple images
into another image. This gives the multiconfig a decent test in a scenario
users may use.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Sat, 2 Mar 2019 14:34:01 +0000 (14:34 +0000)]
mcextend: Add helper class useful for multiconfig
This class allows users to easily create multiconfig BBCLASSEXTEND recipes
and is generally useful. It will be used by new tests so add as a general
feature.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
PATCH DROPPED:
==============
0005-Make-root-s-home-directory-configurable.patch
systemd has its hardcoded assumption about /home and /, and it also respects
$HOME environment var, so this patch is somehow useless. This patch was originally
added but in fact had no real runtime effect except messing up some hardcoded assumptions,
and it was accidently manipulated during systemd upgrade. We have in fact not
used the orignal patch for more than two releases and things were working out
well.
0006-remove-nobody-user-group-checking.patch
The issue has been fixed upstream by the following commit.
"check nobody user/group validity only when not cross compiling"
0008-Do-not-enable-nss-tests-if-nss-systemd-is-not-enable.patch
0009-nss-mymachines-Build-conditionally-when-ENABLE_MYHOS.patch
The issue has been fixed upstream by the following commit.
"meson: allow building resolved and machined without nss modules"
0001-login-use-parse_uid-when-unmounting-user-runtime-dir.patch
0001-sd-bus-make-BUS_DEFAULT_TIMEOUT-configurable.patch
Backport
0022-build-sys-Detect-whether-struct-statx-is-defined-in-.patch
Merged
0023-resolvconf-fixes-for-the-compatibility-interface.patch
0001-core-when-deserializing-state-always-use-read_line-L.patch
0001-chown-recursive-let-s-rework-the-recursive-logic-to-.patch
0001-dhcp6-make-sure-we-have-enough-space-for-the-DHCP6-o.patch
0001-Revert-sysctl.d-request-ECN-on-both-in-and-outgoing-.patch
0001-timesync-changes-type-of-drift_freq-to-int64_t.patch
Backport
0001-sysctl-Don-t-pass-null-directive-argument-to-s.patch
0002-core-Fix-use-after-free-case-in-load_from_path.patch
Merged
0001-meson-rename-Ddebug-to-Ddebug-extra.patch
0024-journald-do-not-store-the-iovec-entry-for-process-co.patch
0025-journald-set-a-limit-on-the-number-of-fields.patch
0026-journal-fix-out-of-bounds-read-CVE-2018-16866.patch
CVE-2019-6454.patch
sd-bus-if-we-receive-an-invalid-dbus-message-ignore-.patch
0005-basic-user-util-properly-protect-use-of-gshadow.patch
0022-Use-if-instead-of-ifdef-for-ENABLE_GSHADOW.patch
Backport
0001-Remove-fstack-protector-flags-to-workaround-musl-bui.patch
No build failure for qemux86/qemuppc + musl
PATCH OTHERS:
=============
0003-comparison_fn_t-is-glibc-specific-use-raw-signature-.patch
0011-src-basic-missing.h-check-for-missing-__compar_fn_t-.patch
are combined into one patch:
0003-missing_type.h-add-__compare_fn_t-and-comparison_fn_.patch
Add two more PACKAGECONFIG, nss-mymachines and nss-resolve which are introduced
by the following commit.
meson: allow building resolved and machined without nss modules
Signed-off-by: Chen Qi <Qi.Chen@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
openssl10: Fix mutliple include assumptions for des.h in opensslconf.h
The fix is heavily based on Khem's previous fix for bn.h/BN_LLONG breakage:
https://git.openembedded.org/openembedded-core/commit/?id=f787b0bb9b0626ddbf2ac94cb206c76716a3773d
Khem Raj [Fri, 1 Mar 2019 03:21:09 +0000 (19:21 -0800)]
glibc: Disable Werror when building with debug options
Since compiler does not optimize away a lot of stuff we end up with
Werrors e.g.
./sysdeps/ieee754/flt-32/s_log1pf.c: In function '__log1pf':
../sysdeps/ieee754/flt-32/s_log1pf.c:114:22: error: 'c' may be used uninitialized in this function [-Werror=maybe-uninitialized]
114 | + (k * ln2_lo + c))) - f);
| ~~~~~~~~~~~~^~~~
which otherwise wont happen, so lets build with warnings-as-errors
disabled in debug mode
given we disable werror, now we don't have to restrict user to compile
without -O0
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Fri, 1 Mar 2019 03:21:08 +0000 (19:21 -0800)]
bitbake.conf: Use -Og in DEBUG_OPTIMIZATION
-Og is for optimized debugging experience.
this makes this consistent across different compilers especially gcc and
clang, -O in clang is equal to -O2 where as in gcc its similar to -O1
so it was not giving consistent debugging experience across compilers
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Justin Bronder [Fri, 1 Mar 2019 05:34:26 +0000 (00:34 -0500)]
kernel: add objtool to shared workdir when CONFIG_UNWINDER_ORC is set
Based on: https://patchwork.openembedded.org/patch/148047/
When x86_64 enables CONFIG_UNWINDER_ORC [ORC (Oops Rewind Capability)
unwinder for unwinding kernel stack traces], objtool is required to
generate the required information.
If we don't copy objtool to the shared workdir, out-of-tree modules
cannot generate object files.
For instance, meta-skeleton/hello-mod fails with:
| make[3]: *** No rule to make target 'tools/objtool/objtool', needed by ...
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bruce Ashfield [Fri, 1 Mar 2019 05:34:24 +0000 (00:34 -0500)]
linux-yocto/4.19: configuration cleanups
Integrating the following configuration tweaks (which were already
on yocto-4.18):
9bddc4aa1eb bsp/intel-x86: Rename CONFIG_R8723BE to CONFIG_RTL8723BE 6504b0ed616 features: drop the obsolete kernel option 5d18756a7fc features/hostapd: drop obsolete configs bdf91f835b9 bsp/intel-x86: Drop configs that has been removed by kernel 0f8bb536c22 intel-x86: Drop configs which are not defined in kernel
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
[
In order for logging to work, as for example seen with the default
configuration of 'ufw' we need to have logging support enabled. This is
currently gated on the CONFIG_NETFILTER_XT_TARGET_LOG option, so enable
it here.
Richard Purdie [Thu, 28 Feb 2019 12:24:49 +0000 (12:24 +0000)]
scripts/wic: Be consistent about how we call bitbake
Whilst its tempting to call bitbake_main(), the script also directly executes
bitbake. By doing both this creates all kinds of races. Therefore
change to use subprocess execution in all cases.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mingli Yu [Tue, 26 Feb 2019 02:50:25 +0000 (10:50 +0800)]
libdazzle: add check for GTK3DISTROFEATURES
When do world buid, there comes below error:
| ERROR: Nothing PROVIDES 'gtk+3' (but /build/layers/oe-core/meta/recipes-gnome/libdazzle/libdazzle_3.30.2.bb DEPENDS on or otherwise requires it)
| gtk+3 was skipped: one of 'x11 wayland' needs to be in DISTRO_FEATURES
Add the check for GTK3DISTROFEATURES to make
the world build work even without GTK3DISTROFEATURES.
Signed-off-by: Mingli Yu <Mingli.Yu@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Thu, 28 Feb 2019 00:37:24 +0000 (16:37 -0800)]
glibc: Move common bits to glibc-common.inc
We have been duplicating few variables in glibc recipes which could
actually be defined once, therefore move them to glibc-common.inc which is
included by all glibc family of recipes
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Ross Burton [Wed, 27 Feb 2019 18:33:54 +0000 (18:33 +0000)]
libpcre: recipe cleanup
The following options are the defaults, so remove them:
--enable-newline-is-lf
--with-match-size=2
--with-match-limit=10000000
We don't appear to need to pass -D_REENTRANT anymore (added with no explanation
to oe-classic in 2006).
Explicitly adding -lstdc++ doesn't appear to be required anymore (added for
PowerPC in 2008).
This recipe has always rebuilt the character tables but back in PCRE 4.4 (first
added to OE) a copy of the tables wasn't distributed with the tarball so this
was required. Since 2007 the tarball includes the tables for ASCII and
regeneration is only required if we wish to use EBCDIC, which we do not. Drop
the patch adding CC_FOR_BUILD support and remove --enable-rebuild-chartables
Signed-off-by: Ross Burton <ross.burton@intel.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Khem Raj [Mon, 25 Feb 2019 23:17:42 +0000 (15:17 -0800)]
mesa-gl: Set PACKAGECONFIG for target
mesa.inc uses PACKAGECONFIG_class-target override to setup default
packageconfigs, which means that when we build mesa-gl for target then
it does no honor the ??= setting we have in mesa-gl recipe, and ends up
compiling egl, gles2 as well, which is not intended.
Signed-off-by: Khem Raj <raj.khem@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
qemu: split the native version into usermode and system parts
The rationale is to streamline the overall build.
The system parts are only needed to run target images, and so can be
built towards the end of the build process. At the same time, the
system parts may need gtk+-native and mesa-native which add significantly
to the build time.
On the other hand, the usermode parts have almost no dependencies
and can be built quickly. They are needed at recipes build time to
run target binaries, and so are required quite early in the typical
build process.
Signed-off-by: Alexander Kanavin <alex.kanavin@gmail.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Richard Purdie [Wed, 27 Feb 2019 17:05:29 +0000 (17:05 +0000)]
resulttool/report: Ensure test suites with no results show up on the report
ptest suites with no results don't show up on the reports even though we have
a duration for them. Fix this so the fact they report no tests is visible.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mark Asselstine [Tue, 26 Feb 2019 17:10:35 +0000 (12:10 -0500)]
cpio: avoid conflict with tar-doc and rmt.8
This fixes:
Error: Transaction check error:
file /usr/share/man/man8/rmt.8 conflicts between attempted installs of
cpio-doc-2.12-r0.core2_64 and tar-doc-1.31-r0.core2_64
Prior to commit 348a96a5b401 [tar: upgrade to 1.31] the copies of
rmt.8 found in the tar(-doc) and cpio(-doc) packages were the same and
thus no conflict was seen. After the upgrade there were small changes
in the manpage header which results in the conflict quoted above. The
applications themselves make use of the 'update-alternatives'
mechanism to allow a user to select which version of 'rmt' to use but
since the man pages are essentially the same we disambiguate the
source of the man pages and make them both available should both
cpio-doc and tar-doc are both installed. And as such we avoid the
conflict.
Signed-off-by: Mark Asselstine <mark.asselstine@windriver.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Manjukumar Matha [Wed, 20 Feb 2019 18:29:06 +0000 (10:29 -0800)]
kernel-fitimage.bbclass: Introduce a way to provide external dtb
Currently kernel-fitimage gets the dtb from KERNEL_DEVICETREE, however
there are instances when the DTB file can be generated using other
recipes, provide a way to include the DTB instead of inkernel device
tree.
Add support for external device tree during the creation of fitimage. If
virtual/dtb is set using PREFERRED_PROVIDER_virtual/dtb = "devicetree", which
inherits the devicetree.bbclass then use the path provided in
EXTERNAL_KERNEL_DEVICETREE else use KERNEL_DEVICETREE during fitimage process
Signed-off-by: Manjukumar Matha <manjukumar.harthikote-matha@xilinx.com> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Mike Crowe [Sun, 24 Feb 2019 21:00:52 +0000 (21:00 +0000)]
glibc: Install AArch64 loader link correctly for usrmerge+multilib
The AArch64 little-endian ABI requires that the dynamic loader is always
available at /lib/ld-linux-aarch64.so.1. Similarly, the big-endian ABI
requires that the dynamic loader is always available at
/lib/ld-linux-aarch64_be.so.1.
glibc-package.inc contains code that tries to ensure this, but
unfortunately it is defeated by the combination of multilib and usrmerge
because it does not take into account that /lib is the same as /usr/lib
with usrmerge when it adds the loader path to libc_baselibs and when it
attempts to show that /usr/lib is empty in do_poststash_install_cleanup.
This results in the symlink not being included in the package and a build
failure due to rmdir failing.
Richard Purdie also suggested[1] that ${nonarch_base_libdir} should not be
used as a synonym for /lib in this case.
This hopefully-fixed version always sets ARCH_DYNAMIC_LOADER and then uses
${root_prefix}/lib/${ARCH_DYNAMIC_LOADER} to refer to the dynamic loader
which works with both multilib and usrmerge. Since ARCH_DYNAMIC_LOADER is
only non-empty if the symlink is required, the code to create it can move
to do_install_append. Then do_poststash_install_cleanup needs to be taught
that ${exec_prefix}/lib may not be empty if the dynamic loader symlink is
there.
It appears not to be possible to specify the name of the loader via a
variable with an override, since the _aarch64 override is applied even for
_aarch64-be, so I've set the loader name using ${TARGET_ARCH} instead.
Build-tested and inspected core-image-minimal rootfs with:
* AArch64 no multilib (real loader in correct place)
MACHINE = "qemuarm64"
* big-endian versions of all of the above by also setting
DEFAULTTUNE = "aarch64_be".
(building glibc only.)
* x86_64 (real loader in /lib as before)[2]
MACHINE = "qemux86"
* x86_64 multilib (real loader in /lib64 as before)
MACHINE="qemux86-64"
MULTILIBS = "multilib:lib32"
DEFAULTTUNE_virtclass-multilib-lib32 = "x86"
require conf/multilib.conf
I also tested leaving an unwanted file in ${exec_prefix}/lib for
do_poststash_install_cleanup to detect, and I believe the detection always
worked correctly.
The two patches have now been upstreamed. Apart from those changes, the
only thing that changed was the hwmixvolume tool, which got ported from
PyGTK to PyGObject, and from GTK 2 to GTK 3.
When testing hwmixvolume, I found that it depends on pyalsa, which has
not been packaged for OE. I believe hwmixvolume has never worked on OE.
It certainly didn't work before this patch, trying to build it failed
due to python-pygtk not being available. Even if python-pygtk was
available at some point in the past, hwmixvolume has always used pyalsa,
but the alsa-tools recipe has never had that dependency declared.
Signed-off-by: Tanu Kaskinen <tanuk@iki.fi> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Daniel Díaz [Mon, 25 Feb 2019 22:02:52 +0000 (16:02 -0600)]
ltp: Add patch for string formatting security warning
A revamp of the syscalls/setregid tests made to use a new
library [LTP:7a1b4427] required a conversion to test messages.
This test in particular did not include a formatted string,
but a just a string element:
testcases/kernel/syscalls/setregid/setregid01.c
This patch allows LTP to be built with -Wformat,
-Wformat-security, and -Werror=format-security.
Signed-off-by: Daniel Díaz <daniel.diaz@linaro.org> Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
