The latest 6.5 kernels do not appear to create the source file in
${D}${nonarch_base_libdir}/modules/${KERNEL_VERSION}/source so the
recipe errors out when trying to remove it. Simple fix is to add the
-f (force) flag to the call.
(From OE-Core rev: 2e669bf797b15d803e7d6a700e449bdc467a4bcc)
(From OE-Core rev: 844faa7c51ae8ec0966e9c5c3f70a1dbf2222c21)
Signed-off-by: Ryan Eatmon <reatmon@ti.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Backported from kirkstone commit 7e177848f97e.
Signed-off-by: Paul Barker <paul.barker.ct@bp.renesas.com>
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Due to an oversight in the do_symlink_kernsrc function, the path
comparison between "S" and "STAGING_KERNEL_DIR" is broken. The code
obtains both variables, but modifies the local copy of "S" before
comparing them, causing the comparison to always return false.
This can cause the build to fail when the EXTERNALSRC flag is enabled,
since the code will try to create a symlink even if one already exists.
This patch resolves the issue by comparing the variables before they are
modified.
(From OE-Core rev: 27a982807caa7ffbdf2d4ef02bc0b037150b1b3b)
Signed-off-by: Staffan Rydén <staffan.ryden@axis.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
(cherry picked from commit afd2038ef8a66a5e6433be31a14e1eb0d9f9a1d3)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
When including an initramfs bundle inside a FIT image, the fitImage is created
after the install task by do_assemble_fitimage_initramfs.
This happens after the generation of the initramfs bundle
(done by do_bundle_initramfs).
So, at the level of the install task we should not try to install the fitImage.
The fitImage is still not generated yet.
After the generation of the fitImage, the deploy task copies the fitImage from
the build directory to the deploy folder.
Change-Id: I3eaa6bba1412f388f710fa0f389f66631c1c4826
(From OE-Core rev: b70a8333a7467162b9d148b99f5970c0af2a531f)
Signed-off-by: Abdellatif El Khlifi <abdellatif.elkhlifi@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 1b67fd9ac74935fa41e960478c54e45422339138)
Signed-off-by: Frederic Martinsons <frederic.martinsons@gmail.com>
Signed-off-by: Steve Sakoman <steve@sakoman.com>
pkgconfig is being required to find dependencies for building kernel
native tools, move "inherit pkgconfig" to kernel.bbclass so BSP kernel
recipes can also benefit from it.
(From OE-Core rev: a5e62888768be641d5da00fd847e0cac96820c75)
Signed-off-by: Ming Liu <liu.ming50@gmail.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
(cherry picked from commit 8a84bd98e3fbc16c782f83064801e469d086911e)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently, the KERNEL_DEBUG_TIMESTAMPS is not working as expected
at rebuild. That is, even if we set it to "1", the kernel build time
is not changed. The problem could be reproduced by the following steps.
1. bitbake core-image-minimal; start image and check `uname -a` output.
2. set in local.conf: KERNEL_DEBUG_TIMESTAMPS = "1"
3. bitbake core-image-minimal; start image and check `uname -a` output.
It's expected that after enabling KERNEL_DEBUG_TIMESTAMPS, the kernel
build time will be set to current date. But it's not. This is because
the compile.h was not re-generated when do_compile task was re-executed.
In mkcompile_h, we have:
"""
# Only replace the real compile.h if the new one is different,
# in order to preserve the timestamp and avoid unnecessary
# recompilations.
# We don't consider the file changed if only the date/time changed,
# unless KBUILD_BUILD_TIMESTAMP was explicitly set (e.g. for
# reproducible builds with that value referring to a commit timestamp).
# A kernel config change will increase the generation number, thus
# causing compile.h to be updated (including date/time) due to the
# changed comment in the
# first line.
"""
It has made it very clear that it will not be re-generated unless
we have KBUILD_BUILD_TIMESTAMP set explicitly. So we set this variable
explicitly in do_compile to fix this issue.
(From OE-Core rev: e44f0cda8176186d42a752631810c1cb5f1971eb)
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
(cherry picked from commit 1b68c2d2d385013a1c535ef81172494302a36d74)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
In 526bdd88ccd758204452579333ba188e29270bde the imageType loop in
kernel_do_deploy was changed to use KERNEL_IMAGETYPE_FOR_MAKE rather
than KERNEL_IMAGETYPES. This broke the special handling for fitImage
immediately below because KERNEL_IMAGETYPE_FOR_MAKE never contains
fitImage.
It has always been my understanding that KERNEL_IMAGETYPE_FOR_MAKE
controlled what was passed to make, but KERNEL_IMAGETYPE controlled what
was installed/deployed. When the two are different then it's the
responsibility of whoever set KERNEL_IMAGETYPE_FOR_MAKE to ensure that
whatever comes out of the kernel build system has been transformed in to
the requested form by the time of installation. This is what happens for
kernel.bbclass's own support for vmlinux.gz.
I think this means that for KERNEL_IMAGETYPE vmlinux.gz, kernel.bbclass
is responsible for generating vmlinux.gz.initramfs[1] so that
kernel_do_deploy can deploy it. This means that the change in
526bdd88ccd758204452579333ba188e29270bde can be reverted, fixing
KERNEL_IMAGETYPE = "fitImage".
In addition, it ought to be possible for recipes and other classes that
use kernel.bbclass to hook into this mechanism by setting
KERNEL_IMAGETYPE_FOR_MAKE and performing their own transformations.
do_bundle_initramfs calls kernel_do_compile and we don't want it to
transform vmlinux to vmlinux.gz at that point, since it will fight
against the careful renaming and preserving that do_bundle_initramfs
does. Let's separate the transformation out of kernel_do_compile to a
new do_transform_kernel task that can be run at the right time. This
means that it's also logical to perform the equivalent translation for
the kernel with the initramfs in a separate
do_transform_bundled_initramfs task too.
This leaves two clear customisation points for recipes and other classes
to hook into the process and perform their transformations:
do_transform_kernel and do_transform_bundled_initramfs.
(I care about this because our recipes that use kernel.bbclass also set
KERNEL_IMAGETYPE_FOR_MAKE and transform vmlinux into a form suitable for
our bootloader after do_compile and do_bundle_initramfs into the format
matching KERNEL_IMAGETYPE. I'm unable to successfully bundle an
initramfs after 526bdd88ccd758204452579333ba188e29270bde, but I didn't
want to just revert that change to reintroduce the bug that it was
fixing.)
I can't say that I'm entirely happy with this change, but I'm unsure
what to do to improve it. I find the way that both the bare kernel and
the one with the initramfs both get deployed to be confusing, and a
waste of build time. I would like to not actually generate a publishable
kernel image at all during do_compile when an initramfs is in use, but I
suspect that this would affect valid use cases that I'm not aware of.
(From OE-Core rev: d9e9653616638f2b187d5e04540071ac34d99f56)
Signed-off-by: Mike Crowe <mac@mcrowe.com>
[1] It could be argued that this should be vmlinux.initramfs.gz, but
that would require another special case in kernel_do_deploy and the
filename is only visible within this class and the recipes that use it
anyway.
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
(cherry picked from commit 10a4a132e87e835726bf5da81a60f6f509b90765)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When packaging is disabled using the nopackages class, ensure we don't
add to PACKAGES. This fixes builds where we have an unpackaged kernel
alongside a packaged kernel.
(From OE-Core rev: 6683b784d3258672c8d56c945db02ba37379cbf6)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 2522daf22e2c27dd9c7926feda0345978217c6c3)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The routine do_sizecheck() was historically needed by legacy
devices with limited flash memory.
The lowest extreme is probably with Zaurus collie having exactly
1024*1024 = 1048576 bytes for the kernel partition.
In the years the KERNEL_IMAGE_MAXSIZE has been converted to kilobytes
thus rounded so we have now KERNEL_IMAGE_MAXSIZE_collie = "1024".
The effect is that now the check fails because we hit curiously this
| WARNING: This kernel zImage (size=1024(K) > 1024(K)) is too big for...
even though zImage is 1047288 bytes (kernel + kexecboot-klibc-initramfs).
Fix this case using test -gt (greater) instead of -ge (greater or equal).
(From OE-Core rev: f5fc716d744745d5c2ea83f0b1d63907cfe04c06)
Signed-off-by: Andrea Adami <andrea.adami@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 254ca956d63b4ce6aa294213b60bb943f9f3a9e6)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Prior work has refactored the do_install task multiple times, and any
references to PKGV and PKGR (even indirect ones) have been removed.
(From OE-Core rev: d2fe8c5861283ff4a4bd18239a61eb6745b7f696)
Signed-off-by: Mark Hatle <mark.hatle@kernel.crashing.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 36fe4db8ae827a93abe9fce6740459d215411965)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When compiling xilinx-zynq board linux-kernel-dev(v5.8) if
"GCC_PLUGINS=y", The following error will appear:
"HOSTCXX -fPIC scripts/gcc-plugins/arm_ssp_per_task_plugin.o
fatal error: gmp.h: No such file or directory"
the GCC_PLUGINS depend on return result of gcc-plugin.sh execution
however in gcc-plugin.sh use HOSTCC to detect the feature of GNU
extension of gcc, this will result that HOSTCC can compile the file
successfully, but HOSTCXX is used in the actual compilation process.
(From OE-Core rev: 9a14fb4aefb3a0bad1d1a98f44eaa10177737e04)
Signed-off-by: Zhang Qiang <qiang.zhang@windriver.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Cc: Armin Kuster <akuster808@gmail.com>
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The do_bundle_initramfs() only processes kernel image
types that are found in KERNEL_IMAGETYPE_FOR_MAKE whereas
the build system can generate other types that are not
directly supported by the kernel build system. In which
case when we come to the deploy phase not all the images
mentioned in KERNEL_IMAGETYPES would have a respective
initramfs bundled image. An example is using vmlinux.gz
in KERNEL_IMAGETYPES and enabling initramfs and then we
see
install: cannot stat 'arch/arm64/boot/vmlinux.gz.initramfs': No such file or directory
So we align the deploy phase with bundle initramfs phase
and pick up relevant initramfs bundled images using
KERNEL_IMAGETYPE_FOR_MAKE instead of KERNEL_IMAGETYPES.
(From OE-Core rev: 352cecca98a0a845e1dacd10e399e26cdf8c7a93)
Signed-off-by: Awais Belal <awais_belal@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 526bdd88ccd758204452579333ba188e29270bde)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The target system triple contains whether the target is Linux or not,
so use it to avoid situations where you can attempt to build a kernel
for systems which don't support Linux.
(From OE-Core rev: 3f08f5b7068ecd3a59fc30dd90515df314a94f84)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit b1eeeab21a81990321468ddbdd1745ea24d1828d)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With 'ld-is-gold' and linux kernel 5.4 or later, menuconfig
task for kernel recipes will fail with:
$ bitbake -c menuconfig virtual/kernel
...
scripts/kconfig/mconf Kconfig
scripts/Kconfig.include:43: gold linker 'x86_64-poky-linux-ld' not supported
/OE/build/tmp/work-shared/qemux86-64/kernel-source/scripts/kconfig/Makefile:29:
recipe for target 'menuconfig' failed
make[2]: *** [menuconfig] Error 1
/OE/build/tmp/work-shared/qemux86-64/kernel-source/Makefile:606:
recipe for target 'menuconfig' failed
make[1]: *** [menuconfig] Error 2
/OE/build/tmp/work-shared/qemux86-64/kernel-source/Makefile:185:
recipe for target '__sub-make' failed
make: *** [__sub-make] Error 2
Command failed.
This is because that the KERNEL_LD variable already set in
kernel-arch.bbclass isn't used by do_menuconfig function of
cml1.bbclass.
To fix this issue specify LD variable while calling the kernel
menuconfig command through KERNEL_KCONFIG_COMMAND.
(From OE-Core rev: 5f85104363cf496d76c3d29a05ba4406701d2878)
Signed-off-by: Wonmin Jung <wonmin82@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 1faf66ce0b1f8f5165277161e07e25e672370c3f)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There were two copies of this patch floating around, and the merged
variant has the copy in the wrong place.
module.lds is only created during modules_prepare, and that target is
not invoked during our main build of the kernel. We aren't about to
change the kernel build (there's no need), so we move the copy into
the compile_kernelmodules task. After that runs, we have module.lds
availble to copy.
This has been tested against clean kernel + out of tree module
builds, and the dependencies are correct that the file is copied
before the out of tree module build starts.
(From OE-Core rev: 38c6c6f4a460e9c0593e2530ffeaf20b24fd83a9)
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 7d94f9209ebaaf59ea001239a889dd7f928a0e7c)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The upstream commit 596b0474d3d [kbuild: preprocess module linker
script], adds a dependency on module.lds for external module
building.
Since module.lds is generated as part of 'modules_prepare', we
must make it available with the other kernel artifacts in the
kernel shared workdir, otherwise out of tree builds fail.
This fixes errors like:
| make[4]: *** No rule to make target 'scripts/module.lds', needed by
'build/tmp/work/qemuarm64-poky-linux/cryptodev-module/1.11-r0/git/cryptodev.ko'.
Stop.
| make[4]: *** Waiting for unfinished jobs....
We also ensure that kernel-devsrc has a copy to support on
target module builds that are often prepared with 'make scripts
prepare'. Those targets won't regenerate it, so the build fails.
If 'make modules_prepare' is used, the file will be regenerated
and overwrite our copy (as expected).
(From OE-Core rev: 46bc7719ea5ddf4342921c1b2721c73c1a3694a6)
Signed-off-by: Pan, Kris <kris.pan@intel.com>
Signed-off-by: Lili Li <lili.li@intel.com>
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 0fc66a0b64953aae38d0124b57615fffaec8de52)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There's a race between do_symlink_kernsrc and do_populate_lic, since
the latter is ordered "after do_patch"; so the two may run in
parallel. In some cases, that actually causes do_populate_lic to fail
if it happens to look for a license file somewhere under ${S} in the
short window after shutil.move and before the symlink has been
created.
Fix that by simply ordering symlink_kernsrc before do_patch. Any task
that pokes around in ${S} looking for files should be ordered after
do_patch, so this should also fix similar latent races with other ad
hoc tasks.
(From OE-Core rev: 914d8a00e34a8edf75a349da6f5978e4ed60e6d5)
Signed-off-by: Rasmus Villemoes <rasmus.villemoes@prevas.dk>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit c5dfc2586b4135cc86e91bb04fed837daf505676)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If the initramfs image is type lz4, then a native lz4 is needed.
Additionally an output filename needs to be specified when calling lz4,
otherwise STDOUT will be used implicitly.
(From OE-Core rev: 589b2d7e902125199377d5f1550597bcb38a1bf6)
Signed-off-by: Timon Ulrich <t.ulrich@anapur.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit c83a7aec59defbbc049d44ba34a7a409ada4fa19)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Starting from v5.8-rc1 commit 269a535ca931 (modpost: generate
vmlinux.symvers and reuse it for the second modpost"), kernel will
generate new vmlinux.symvers instead of dumping all the vmlinux symbols
into Module.symvers in the first pass.
Error log:
'run.do_shared_workdir.16614' failed with exit code 1:
DEBUG: cp: cannot stat 'Module.symvers': No such file or directory
This change will check the file Module.symvers existence before copying it.
(From OE-Core rev: e96d99824f085644162fe0599988a3408eab90f8)
Signed-off-by: Lili Li <lili.li@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit cd2d62a08a1dfcd890a03ee55132b6d6c65f5ab7)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If CONFIG_IKHEADERS is set to =m, then reproducibility issues creep
into the modules build, since the variables we are setting for the
main kernel build are not present.
Since the source code must be available for a possibly git query
on the timestamp, there didn't seem to be an easy way to move the
environment variable setting to a common routine. As such, we
duplicate the block of code that exports the required variables for
reproducible builds. There is a maintenance risk to this, but any
issues should be easy enough to catch.
(From OE-Core rev: f511d78164581f80e7b8c592fe88ffbf38738150)
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
(cherry picked from commit 82cdfcdccfedd320ebc0cdc778c7d4966198b96f)
Signed-off-by: Steve Sakoman <steve@sakoman.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The source directory is not always a git repository, so when querying
git for data to set SOURCE_DATE_EPOCH, specify ${S}/.git as the git
directory to prevent retrieving incorrect data from any parent directory.
Fixes the following errors with the prior logic when building a kernel
that is not obtained from a git repository:
1. With TMPDIR set to a directory outside any git repository on a
mounted filesystem, reproducible builds fail in do_compile with this git
error:
fatal: not a git repository (or any parent up to mount point <abspath>)
Stopping at filesystem boundary (GIT_DISCOVERY_ACROSS_FILESYSTEM not set).
aborting before the error handling logic.
2. With TMPDIR located within a subdirectory of a git repository, the
SOURCE_DATE_EPOCH timestamp would be that of said repository rather than
that of the kernel.
(From OE-Core rev: 270ae94fe345b9ce98d822034cbfad7e24c5f393)
Signed-off-by: Christopher Clark <christopher.w.clark@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
legitimize_package_name wants the actual value of KERNEL_REVISION, so
use d.getVar() to fetch it as is done elsewhere in the file. Failing to
do so can result it weird errors at parsing time.
(From OE-Core rev: 9bff4d14693c1890fc181ec68c9f883dc4e4accf)
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* with 5.4 kernel the do_configure fails when gold is enabled (with ld-is-gold in DISTRO_FEATURES)
* bfd is already being set in KERNEL_LD:
kernel-arch.bbclass:KERNEL_LD = "${CCACHE}${HOST_PREFIX}ld.bfd ${HOST_LD_KERNEL_ARCH}"
but KERNEL_LD is currently respected only by do_compile and do_compile_kernel modules
and new kernel which contains a check for gold in Kbuild:
https://lore.kernel.org/lkml/alpine.DEB.2.21.1907161434260.1767@nanos.tec.linutronix.de/
more details:
https://lore.kernel.org/r/CAMe9rOqMqkQ0LNpm25yE_Yt0FKp05WmHOrwc0aRDb53miFKM+w@mail.gmail.com
will fail during the configuration even when gold wouldn't be used to
build it in the end, add LD setting to KERNEL_CONFIG_COMMAND to prevent
premature error when configuring
scripts/kconfig/conf --olddefconfig Kconfig
scripts/Kconfig.include:39: gold linker 'x86_64-oe-linux-ld' not supported
make[2]: *** [/OE/build/oe-core/tmp-glibc/work-shared/qemux86-64/kernel-source/scripts/kconfig/Makefile:73: olddefconfig] Error 1
(From OE-Core rev: 66c1f4b0abd3483759f9e4141a80f982eaf8ccfd)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some mips 64 bit platforms use vmlinux.64 image name
Make a symbol link to vmlinux.64 in arch/mips/boot/
(From OE-Core rev: 70626b52e5e61c80018e9f1f85bc169d0434b986)
Signed-off-by: Vasyl Gomonovych <gomonovych@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This reverts commit 6676411fccff2d331878e4ca1f9411aafb056a80.
This revert restores the original code and adds a comment. The commit
that was reverted broke a number of wic templates and tools which rely
on the initramfs creation dependency and the case where the
INITRAMFS_IMAGE_BUNDLE is not set.
If an end user does not want the INITRAMFS_IMAGE generated, it should
be set to "".
[ Issue: LIN1019-1791 ]
(From OE-Core rev: ab61a11623ac0a25ba1c98d686c79815abab573f)
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If one has provided external key/certificate for modules signing, Kbuild
will skip creating signing_key.pem and will write only signing_key.x509
certificate. Thus we have to check for .x509 file existence rather than
.pem one.
(From OE-Core rev: 2527e731eba43bd36d0ea268aca6b03155376134)
Signed-off-by: Dmitry Eremin-Solenikov <dmitry_eremin-solenikov@mentor.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The package 'make-mod-scripts' creates files in 'kernel-build-artifacts/include/config'
which are removed by 'cleanall/cleansstate' of 'virtual/kernel'. And this causes the
below error while building out of tree kernel module:
ERROR: Kernel configuration is invalid.
include/generated/autoconf.h or include/config/auto.conf are missing.
Run 'make oldconfig && make prepare' on kernel src to fix it.
Suggested-by: Jun Nie <jun.nie@linaro.org>
(From OE-Core rev: f79c95f6a883e999e0c2ecfd60b6b696b7595497)
Signed-off-by: Haiqing Bai <Haiqing.Bai@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
As of commit 898490c010b [moduleparam: Save information about built-in
modules in separate file] (kernels v5.2-rc1+), modules.builtin.modinfo
is generated as part of the kernel build process.
We package it along with the other module artifacts, so it can be used
by scripts/other build steps.
(From OE-Core rev: 112a4d3b2b4a829dd5657b6533b1c1bb589d6c8e)
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
do_bundle_initramfs does not have to depend on
${INITRAMFS_IMAGE}:do_image_complete if INITRAMFS_IMAGE_BUNDLE is not
set.
(From OE-Core rev: 6676411fccff2d331878e4ca1f9411aafb056a80)
Signed-off-by: Ming Liu <liu.ming50@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
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 ...
(From OE-Core rev: 7d4fdd3b95c949277f77df90bb2e8120b8d167ae)
Signed-off-by: Bruce Ashfield <bruce.ashfield@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If ${DEPLOY_DIR_IMAGE}/${INITRAMFS_IMAGE_NAME}.cpio does not exist,
nor any of the compressed variants, nothing is copied to kernel build's
./usr directory.
The code does not fail, but silently proceeds without a bundled initramfs.
Change to fail and tell the user something is wrong.
Also, if an initramfs is found, contrary to the comments, it does not stop
at the first uncompressed/compressed cpio image found. Instead it keeps
processing all so the last is used. Fix this to behave as per the comments.
[YOCTO #12909]
(Patch by Leon Woestenberg)
(From OE-Core rev: 5b5604e288af755eb5553a97d26533445b2cf94b)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fix handling of escape characters in regexs and hence fix python
Deprecation warnings which will be problematic in python 3.8.
Note that some show up as:
"""
meta/classes/package.bbclass:1293: DeprecationWarning: invalid escape sequence \.
"""
where the problem isn't on 1293 in package.bbclass but in some _prepend to a
package.bbclass function in a different file like mesa.inc, often from
do_package_split() calls.
(From OE-Core rev: 4b1c0c7d5525fc4cea9e0f02ec54e92a6fbc6199)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When kernel-fitimage and initramfs is enabled using
INITRAMFS_IMAGE_BUNDLE = "1", kernel do_deploy tries to deploy
fitImage.initramfs with following error
| install: cannot stat 'arch/arm64/boot/fitImage.initramfs': No such
file or directory
Skip deploying fitimage.initramfs, since fitimage does not
create fitimage.initramfs
(From OE-Core rev: 0f87f8522a2c1b5a7042738887952b98060fbd9d)
Signed-off-by: Manjukumar Matha <manjukumar.harthikote-matha@xilinx.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The KBUILD_BUILD_USER and KBUILD_BUILD_HOST variables were
assigned at the kernel class level, which made it impossible
to override them in the local configuration.
By setting only the default values of those variables in the
kernel class, it is now possible to override them as expected.
(From OE-Core rev: a3e8cdf9c3ba966fa4b5a21235540eb0b00fb487)
Signed-off-by: Federico Sauter <federico.sauter@ableton.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
As was warned by commit 312ee68752fa [kconfig: announce removal of
oldnoconfig if used], oldnoconfig has been removed from the 4.20 kernel.
So we switch our default mode to olddefconfig.
commit fb16d8912 [kconfig: replace 'oldnoconfig' with 'olddefconfig',
and keep the old name as an alias] introduced olddefconfig in the 3.10
kernel, we shuffle oldnoconfig to the fallback target.
The fallback mode allows kernels between 3.10 and the currently listed
oldest kernel of 3.2 to continue to configure.
(From OE-Core rev: 8593dcb7e8c938530ff00ffedf7f3d02d26c3bad)
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some distros want to install the kernel in a custom location other
than /boot and have it properly packaged, so it's useful to use a
weaker assignment.
(From OE-Core rev: cadd91655a336eb6d65ca43dc1231bbe5271e1cf)
Signed-off-by: Ioan-Adrian Ratiu <adrian.ratiu@ni.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
During deploy we iterate over KERNEL_IMAGETYPES twice, with no consumer
of our changes in the interim. Collapse these two loops into one.
(From OE-Core rev: 8d3139640ed4cd4e196dc2d1f6ae02d612fb9564)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
When usrmerge is enabled the kernel modules aren't under /lib, but /usr/lib;
ensure we fetch the modules from the correct location so we don't have a
tarball which consists of a single symlink.
(From OE-Core rev: 62d2d7d12dda69d6019a1a2d5492450487e41bfd)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This reverts commit 65cded55fa96b1ba72149a31689c081ac27af7f7.
The fitImage deployed here is the one generated in kernel-fitimage.bbclass,
whereas the one deployed by kernel-fitimage.bbclass is a raw image, so we
need the one deployed here keeping.
(From OE-Core rev: 798d2acc0141b13da7ddd18875a417b2df9d57b3)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
kernel-fitimage.bbclass replaces an occurance of "fitImage" in
KERNEL_IMAGETYPE_FOR_MAKE by an image type that is buildable for the
architecture (such as zImage). The kernel-fitimage.bbclass packs that
image as sub-image in a flattened image tree image (fitImage) and
deploys this fitImage along with the image tree source file (.its).
kernel-fitimage.bbclass does not alter KERNEL_IMAGETYPES, which thus
also contains "fitImage", which kernel.bbclass will also deploy
redundantly with different naming.
The result is a dual deployment with slightly different naming,
each with a set of symlinks.
The solution chosen is to have fitImage deployment be handled by
kernel-fitimage.bbclass, and have kernel.bbclass ignore fitImage
types during deployment.
(From OE-Core rev: 65cded55fa96b1ba72149a31689c081ac27af7f7)
Signed-off-by: Leon Woestenberg <leon@sidebranch.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* to avoid confusion with "type" command in shell
(From OE-Core rev: 8d454ea754c96561257b1cc011fa638ceaa771db)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* for consistency with other artifacts variables, include only the version string, not the actual name or extension
* changing .tgz to something else in the MODULE_TARBALL_NAME variable only wouldn't make much sense
because then kernel.bbclass still calls "tar -cvzf" to create it
(From OE-Core rev: 43bd7f39157da49f7fb0c2d6d9751059471c8d53)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* for consistency with IMAGE_NAME and IMAGE_LINK_NAME
and to avoid confusion with IMAGE_BASENAME (which is the
actual name of the artifact, e.g. PN while KERNEL_IMAGE_BASE_NAME
was only the version suffix)
(From OE-Core rev: f952c8e08b4798aa0f8bf764cfd70bda0eae9b8b)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* this makes it easier to access these variables from some other bbclass
e.g. sdcard_image-rpi.bbclass in meta-raspberry where we need to know
how some files in deploy are named, but we cannot inherit kernel.bbclass
as it's used in image recipe not kernel recipe
* alternatively we can move these to bitbake.conf like similar image variables are:
meta/conf/bitbake.conf:IMAGE_BASENAME = "${PN}"
meta/conf/bitbake.conf:IMAGE_NAME = "${IMAGE_BASENAME}-${MACHINE}-${DATETIME}"
meta/conf/bitbake.conf:IMAGE_LINK_NAME = "${IMAGE_BASENAME}-${MACHINE}"
(From OE-Core rev: 7d0ef0eaa1bfe97015a774c26f5791622e7e8b12)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* use INITRAMFS_BASE_NAME and INITRAMFS_SYMLINK_NAME variables, like
other kernel artifacts are using
* use "${PKGE}-${PKGV}-${PKGR}-${MACHINE}-${DATETIME}" instead of
"${PV}-${PR}-${MACHINE}-${DATETIME}" to be consistent with other files
* allow to modify default symlink name with INITRAMFS_SYMLINK_NAME
instead of currently used:
initramfs_symlink_name=${type}-initramfs-${MACHINE}
(From OE-Core rev: 935b9d5a2bd12effad65f69a631ecff96b8bb553)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* add 2 new variables:
KERNEL_DTB_BASE_NAME
KERNEL_DTB_SYMLINK_NAME
instead of reusing KERNEL_IMAGE_SYMLINK_NAME and than expecting that
default value ${MACHINE} was being used in e.g.:
DTB_SYMLINK_NAME=`echo ${symlink_name} | sed "s/${MACHINE}/${DTB_BASE_NAME}/g"`
* install normal DTB files only once even if there is multiple entries
in KERNEL_IMAGETYPE_FOR_MAKE and don't prefix them with the type of
the kernel image, use the KERNEL_IMAGETYPE_FOR_MAKE as a prefix only
when installing them bundled with kernel or initramfs image.
* deploy the files from ${D}/${KERNEL_IMAGEDEST}/ instead of kernel
build directory, so that we don't need to call
DTB_PATH=`get_real_dtb_path_in_kernel "${DTB}"`
again in do_deploy
* create all links in do_deploy task, because default KERNEL_DTB_BASE_NAME
like KERNEL_IMAGE_BASE_NAME contains PKGR and PKGR is different in
do_install and do_deploy, because kernel.bbclass calls
meta/classes/kernel.bbclass:do_install[prefuncs] += "package_get_auto_pr"
meta/classes/kernel.bbclass:do_deploy[prefuncs] += "package_get_auto_pr"
* the filenames are a bit different, but with separate variable it
should be easier for other bbclasses which use these DTB files to
find them correctly, just use either the cannonical name
$dtb_base_name.$dtb_ext or $dtb_base_name-${KERNEL_DTB_SYMLINK_NAME}.$dtb_ext
because PKGR (and other PKG* variables) might be different in your
task and kernel.do_deploy task.
* fix DTB files being deployed with incorrect filenames when
KERNEL_IMAGE_SYMLINK_NAME isn't set to ${MACHINE}, e.g. instead of
the default:
-rw-r--r-- 2 bitbake bitbake 1.4K Nov 20 07:41 deploy/images/raspberrypi3-64/Image-1-4.9.59+git0+e7976b2aff-r0.2-lirc-rpi-20171120043031.dtbo
lrwxrwxrwx 2 bitbake bitbake 64 Nov 20 07:41 deploy/images/raspberrypi3-64/Image-lirc-rpi.dtbo -> Image-1-4.9.59+git0+e7976b2aff-r0.2-lirc-rpi-20171120043031.dtbo
lrwxrwxrwx 2 bitbake bitbake 64 Nov 20 07:41 deploy/images/raspberrypi3-64/lirc-rpi.dtbo -> Image-1-4.9.59+git0+e7976b2aff-r0.2-lirc-rpi-20171120043031.dtbo
I was getting:
-rw-r--r-- 2 bitbake bitbake 1348 Nov 20 10:28 deploy/images/raspberrypi3-64/Image-linux-raspberrypi-lirc-rpi.dtbo
lrwxrwxrwx 2 bitbake bitbake 37 Nov 20 10:28 deploy/images/raspberrypi3-64/Image-linux-raspberrypi-lirc-rpi-master-20171120102653.dtbo -> Image-linux-raspberrypi-lirc-rpi.dtbo
lrwxrwxrwx 2 bitbake bitbake 37 Nov 20 10:28 deploy/images/raspberrypi3-64/lirc-rpi.dtbo -> Image-linux-raspberrypi-lirc-rpi.dtbo
and e.g. sdcard_image-rpi.bbclass from meta-raspberrypi:
37e4e18f4a/classes/sdcard_image-rpi.bbclass
was failing in:
mcopy -i ${WORKDIR}/boot.img -s ${DEPLOY_DIR_IMAGE}/${KERNEL_IMAGETYPE}-${DTB_BASE_NAME}.dtb ::${DTB_BASE_NAME}.dtb
because ${KERNEL_IMAGETYPE}-${DTB_BASE_NAME}.dtb doesn't exist in my
build, due to
DTB_SYMLINK_NAME=`echo ${symlink_name} | sed "s/${MACHINE}/${DTB_BASE_NAME}/g"`
not replacing whole "${KERNEL_IMAGE_SYMLINK_NAME}" (read ${MACHINE})
with just ${DTB_BASE_NAME}
* with this change applied the deploy dir looks like this:
-rw-r--r-- 2 bitbake bitbake 1.4K Nov 20 15:49 deploy/images/raspberrypi3-64/lirc-rpi-1-4.9.59+git0+e7976b2aff-r0.8-raspberrypi3-64-20171120154716.dtbo
lrwxrwxrwx 2 bitbake bitbake 74 Nov 20 15:49 deploy/images/raspberrypi3-64/lirc-rpi.dtbo -> lirc-rpi-1-4.9.59+git0+e7976b2aff-r0.8-raspberrypi3-64-20171120154716.dtbo
lrwxrwxrwx 2 bitbake bitbake 74 Nov 20 15:49 deploy/images/raspberrypi3-64/lirc-rpi-raspberrypi3-64.dtbo -> lirc-rpi-1-4.9.59+git0+e7976b2aff-r0.8-raspberrypi3-64-20171120154716.dtbo
and works correctly even with DISTRO using different naming scheme
* the sdcard_image-rpi.bbclass still needs to be modified, I've provided
updated version here:
https://github.com/agherzan/meta-raspberrypi/pull/159
* mpc8315e-rdb.conf MACHINE in meta-yocto-bsp also needs small fix:
https://lists.yoctoproject.org/pipermail/poky/2018-July/011436.html
(From OE-Core rev: 1860d9d3c62e2e94cd68a809385873ffd8270b6d)
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If SOURCE_DATE_EPOCH is unset (in addition to the existing "0" behaviour)
parse out the top most commit timestamp from the kernel tree to use as the
timestamp.
(From OE-Core rev: 2f0dd67a5a8d4269f5155004d532d8fa972b3223)
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With recent kernels (i.e. 4.17+) the configuration phase of the kernel
will check for capabilities/options of the compiler for CVE and other
mitigation support.
For a general kernel, we want to ensure that CC is fully defined when
the config targets are invoked (so the proper compiler will be checked).
For linux-yocto, we also need to specify the compiler/tools dependencies
for the configme task since it executes before configure and hence the
main kernel build DEPENDS will not always be in the sysroot before it
executes. Without those dependencies the kernel will be incorrectly
configured (i.e. bison is missing) or the configuration will fail the
mitigation tests.
[YOCTO #12757]
(From OE-Core rev: ff1bdd75d50f0ebac3d599e461685ace29559a82)
Signed-off-by: Bruce Ashfield <bruce.ashfield@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Currently if KERNEL_IMAGETYPES contains both vmlinux and vmlinux.gz,
KERNEL_IMAGETYPE_FOR_MAKE will end up containing two copies of
vmlinux, which will result in two calls to "make vmlinux" from
kernel_do_compile().
Avoid duplicating vmlinux in KERNEL_IMAGETYPE_FOR_MAKE plus some
minor non-functional updates to formatting and comments.
(From OE-Core rev: 80455a0b6cce6d12a5b32194d0cad2e4c7f71599)
Signed-off-by: Andre McCurdy <armccurdy@gmail.com>
Signed-off-by: Ross Burton <ross.burton@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
