| Age | Commit message (Collapse) | Author |
|---|
|
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If nothing else is specified, 'git init' uses its default repository
template from the install location (e.g. /usr/share/git-core/templates),
which already includes an info/ subdirectory. However, when setting
init.templateDir to a different template path in ~/.gitconfig, this
isn't necessarily the case, and it can lead to setup_git_repo() failing
with stack traces like:
File: '.../scripts/lib/devtool/__init__.py', lineno: 234, function: setup_git_repo
0230: pass
0231: if 'singletask.lock\n' not in excludes:
0232: excludes.append('singletask.lock\n')
0233: bb.warn("try writing excludefile")
*** 0234: with open(excludefile, 'w') as f:
0235: for line in excludes:
0236: f.write(line)
0237:
0238: bb.process.run('git checkout -b %s' % devbranch, cwd=repodir)
Exception: FileNotFoundError: [Errno 2] No such file or directory: '.../devtooltmp-6m36b181/workdir/foobar-1.0.1/.git/info/exclude'
Fix this edge case by creating the .git/info/ directory first.
Fixes: 334ba846c795fc0d8c73 (2018-02-01, "devtool: set up git repos so that singletask.lock is ignored")
Signed-off-by: Roland Hieber <rhi@pengutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If --branch and --branch2 are the same, there is extra confusing data plotted
onto the graph. Only do that if the branches are actually different.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If buildstats is missing, avoid a backtrace.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This variable isn't used at all.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Adjust wic to correctly handle the new PSEUDO_IGNORE_PATH variable and avoid
inode corruption issues.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This is a pretty big change to the way pseudo operates when used in OpenEmbedded.
Normally, pseudo monitors and logs (adds to its database) any file created or
modified whilst in a fakeroot environment. There are large numbers of files
we simply don't care about the permissions of whilst in that fakeroot context,
for example ${S}, ${B}, ${T}, ${SSTATE_DIR}, the central sstate control
directories,
This change uses new functionality in pseudo to ignore these directory trees,
resulting in a cleaner database with less chance of "stray" mismatches if files
are modified outside pseudo context. It also should reduce some overhead from
pseudo as the interprocess round trip to the server is avoided.
There is a possible complication where some existing recipe may break, for
example, we found a recipe which was writing to "${B}/install" for
"make install" in do_install and since we listed ${B} as not to be tracked,
there were errors trying to chown root for files in this location.
This patch fixes a few corner cases in OE-Core when used with this new
ignore list:
* The archiver directory matched a "${WORKDIR}/deploy*" pattern so was renamed
to something else since that directory does need its root permissions
* The ${S} and ${B} ignoring is conditional on them being different to ${WORKDIR}
* package_write_* task output (the debs/rpms/ipks) are now owned by the build
user so we don't want the file ownership information in the hashequiv outhash
calculation even if they are built under pseudo.
* The fontcache postinstall intercept is run under qemu outside of pseudo context
so delete files it may delete up front where pseudo can see this.
* SSTATE_DIR is in PSEUDO_PATHS_IGNORE, which is in FAKEROOTENV which is cached
by bitbake. We therefore need to trigger reparsing if this changes, which means
SSTATE_DIR can be in BB_HASHBASE_WHITELIST but not BB_HASHCONFIG_WHITELIST.
Rework the variables to handle this. This otherwise breaks some of our sstate
tests in oe-selftest.
* Ignore the temp directory wic uses for rebuilding rootfs.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
RNG passthru has been enabled on all qemu machines but its being added
to each one of them, with this patch its turned into QB variables which
defaults to host passthru, yet it can be overridden if needed via
machine or config metadata if needed.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This means ot matches the rest of the system rather than causing confusing errors
where multiple python versions are mixed.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Instead of assuming that the buildhistory is stored in ./buildhistory
unless told otherwise, use the environment variable BUILDDIR if present
(which is set by oe-init-build-env) to locate the buildhistory
directory.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
and bootimg-partition
Due to recent changes in bootimg-efi to include IMAGE_BOOT_FILES,
when both bootimg-partition and bootimg-efi occur in a single .wks
and IMAGE_BOOT_FILES are defined, files listed in IMAGE_BOOT_FILES
will be duplicated in both partition.
Since IMAGE_BOOT_FILES are crucial for bootimg-partition, but
optional for bootimg-efi, hence allowing bootimg-efi to have the option
to ignore it.
The new variable, IMAGE_EFI_BOOT_FILES, was added to help handle this
issue. Its basic usage is the same as IMAGE_BOOT_FILES.
Usage example:
${IMGDEPLOYDIR}/${IMAGE_BASENAME}-${MACHINE}.ext4;rootfs.img \
This commit is also squashed with the updated testcase to cover for
this change.
[YOCTO #14011]
Signed-off-by: Khairul Rohaizzat Jamaluddin <khairul.rohaizzat.jamaluddin@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Support publishing the sdk-update as well as the sdk installer in one
go.
Signed-off-by: Adrian Freihofer <adrian.freihofer@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Escaping does not work in my use case. It must be escaped for
python, ssh and shell as well as for different versions of echo.
Let's try it a little less elegant, but hopefully more reliable.
Signed-off-by: Adrian Freihofer <adrian.freihofer@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
We have an issue where qemu is being starved of resources on our autobuilders.
We can't raise its priority without special capacilties, therefore add a hook
which if present can allow this to happen using an executable
"~/runqemu-renice".
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If a package contains hard links to a file, the file size is added for
each hard link instead of once for the file. Therefore, the calculated
size may be much larger than the actual package size.
For example, the mesa-megadriver package contains several hard links to
the same library.
Keep track of the inode numbers when listing the files that are
installed and use the actual size only for the first occurrence of an
inode. All further hard links to the same inode are added to the file
list, but accounted with size 0.
All file names need to be added to the file list, because the list is
used for preserving the files/hard links on the target.
Signed-off-by: Michael Tretter <m.tretter@pengutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
/bin is also a valid path where one can find executables. Add
that to the search path.
Signed-off-by: Vijai Kumar K <vijaikumar.kanagarajan@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* I have recipes which use variable inside SRC_URI, e.g.:
ROS_BRANCH ?= "branch=release/melodic/swri_nodelet"
SRC_URI = "git://github.com/swri-robotics-gbp/marti_common-release;${ROS_BRANCH};protocol=https"
and devtool modify works fine, but devtool finish fails with:
$ devtool finish --force-patch-refresh swri-nodelet meta-ros/meta-ros1-melodic/
...
Traceback (most recent call last):
File "/OE/openembedded-core/scripts/devtool", line 334, in <module>
ret = main()
File "/OE/openembedded-core/scripts/devtool", line 321, in main
ret = args.func(args, config, basepath, workspace)
File "/OE/openembedded-core/scripts/lib/devtool/standard.py", line 2082, in finish
updated, appendfile, removed = _update_recipe(args.recipename, workspace, rd, args.mode, appendlayerdir, wildcard_version=True, no_remove=False, no_report_remove=removing_original, initial_rev=args.initial_rev, dry_run_outdir=dry_run_outdir, no_overrides=args.no_overrides, force_patch_refresh=args.force_patch_refresh)
File "/OE/openembedded-core/scripts/lib/devtool/standard.py", line 1737, in _update_recipe
mode = _guess_recipe_update_mode(srctree, rd)
File "/OE/openembedded-core/scripts/lib/devtool/standard.py", line 1721, in _guess_recipe_update_mode
params = bb.fetch.decodeurl(uri)[5]
File "/OE/bitbake/lib/bb/fetch2/__init__.py", line 390, in decodeurl
raise MalformedUrl(url, "The URL: '%s' is invalid: parameter %s does not specify a value (missing '=')" % (url, s))
bb.fetch2.MalformedUrl: The URL: 'git://github.com/swri-robotics-gbp/marti_common-release;${ROS_BRANCH};protocol=https' is invalid: parameter ${ROS_BRANCH} does not specify a value (missing '=')
let it expand the SRC_URI before trying to decode it.
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The autobuilder managed to pass the nographic option with "sdl gl" due
to a problem elsewhere. It would have been useful for runqemu to have
errored rather than passing conflicting options to qemu. Add an
error for this invalid usecase.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Allows the --offset argument to use the "s" or "S" suffix to specify
that it is reporting the number of 512 byte sectors.
This is required for some SoCs where the mask ROM looks for an item at a
sector that isn't aligned to a 1KB boundary.
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
runqemu help reports that gtk or sdl option is needed with gl or gl-es
option. But if user forgot to add gtk or sdl option, then gl or gl-es
options were silently skipped.
Signed-off-by: Vasyl Vavrychuk <vvavrychuk@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The code should check /dev/vhost-net instead of /dev/kvm. And
the permission check logically belongs to the part when kvm-vhost
is enabled. Also fix the URL to point to the correct wiki page.
Signed-off-by: Chen Qi <Qi.Chen@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Signed-off-by: Daniel Ammann <daniel.ammann@bytesatwork.ch>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
layer under test should absent from BBLAYERS when running
yocto-check-layer. This allow to get signatures before layer
under test. There are existing steps to add the layer under
test to BBLAYERS after getting initial signatures.
add steps to check for layer under test in BBLAYERS before
running any test, skip test for the layer if the layer under
test exist in BBLAYERS.
[YOCTO #13176]
Signed-off-by: Lee Chee Yang <chee.yang.lee@intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
List of files defined using IMAGE_BOOT_FILES are installed into
the boot partition when preparing an image using the wic tool with
the bootimg-efi source plugin.
The code snippet introduced is taken as is from bootimg-partition.py
Change-Id: I8dbb6b4e7c24870f587a6f31e6ae4a87d7033782
Issue-Id: PLATFORMS-3134
Signed-off-by: Khasim Syed Mohammed <khasim.mohammed@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
For some filesystems, such as aufs which may be used by docker container,
don't support either the SEEK_DATA/HOLE or FIEMAP to get the block
map. So add a FileNobmap class to fall back to standard copy when there
is no way to get the block map.
[Yocto #12988]
Signed-off-by: Kevin Hao <kexin.hao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This method is not used by any code, so drop it.
Signed-off-by: Kevin Hao <kexin.hao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This method is not used by any code, so drop it.
Signed-off-by: Kevin Hao <kexin.hao@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Python now expects regex strings to be prepended with r.
Silence pylint/autopep8 and similar warnings by identifying
these regex patterns as... regex patterns.
Signed-off-by: Tim Orling <timothy.t.orling@linux.intel.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
[Bug 13971] https://bugzilla.yoctoproject.org/show_bug.cgi?id=13971
A recipe using LICENSE=CLOSED would normally not have LIC_FILES_CHKSUM
set.
If LIC_FILES_CHKSUM was not set, it would cause `devtool upgrade` to fail.
This fixes the issue by defaulting to an empty string if
LIC_FILES_CHKSUM is not found.
[YOCTO #13971]
Signed-off-by: Bjarne Michelsen <bjarne_michelsen@hotmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
LAYERDEPENDS is a string of this format:
"DEPEND1 (optional version) DEPEND2 (optional version) ..."
However when we parse LAYERDEPENDS in _get_layer_collections() we
parse it as a simple string, and if any optional versions are there the
'depends' field is wrong. For example, running yocto-check-layer
might result in such errors:
ERROR: Layer meta-python depends on (>= and isn't found.
ERROR: Layer meta-python depends on 12) and isn't found.
Let's use bb.utils.explode_dep_versions2() to parse LAYERDEPENDS, and
create a string that contains all dependencies, effectively
skipping/ignoring any optional versions.
[YOCTO #13957]
Signed-off-by: Nicolas Dechesne <nicolas.dechesne@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
There is no way to use runqemu and turn off the no-reboot flag with a
command line argument. If someone really wants it back it can be
added with the qemuparams="" argument.
Also if you use the "halt -p" from user space, the qemu will exit when
it is complete. It is impossible to test self deploying image with
runqemu if you cannot reboot the device. One might argue that you can
run runqemu in a loop, but that defeats the purpose of having a
wrapper around a tool that already does what is needed in the first
place.
For the vmtype images, the -no-reboot flag should not be added.
Signed-off-by: Jason Wessel <jason.wessel@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Autobuilder generated checksum files only have a single space between the sum and the
filename, tweak it to account for this.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
This fixes issues with openssl certs not working properly which meant error
reporting to an error report server was failing.
Also, all our downloads are now standarised on "sha256sum" so adjust for that.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Major update after v4.2.
Changes:
- os_find_datadir() was changed after the v4.2 release
causing v5.0 to not find the bios and not boot the
image. Fix is sent to upstream qemu.
See: qemu/find_datadir.patch
- v5.0 binary had host contamination for dynamically linked
libraries, "--extra-ldflags='${LDFLAGS}'" in EXTRA_OECONF
resolved the issue
- bluetooth code was removed: qemu.git$ git show 1d4ffe8dc7
hence removed PACKAGECONFIG[bluez]
- -show-cursor qemu option is now deprecated, updated
scripts/runqemu to use updated option instead
- added PACKAGECONFIG definitions
- added qemu-ptest to conf/distro/include/ptest-packagelists.inc
- increased support for ARM architecture, cpu and board
- removed patches merged upstream and refreshed
existing ones
Testing:
Build core-image-minimal against the machines in
openembedded-core/meta/conf/machine and succesfully
booted with qemu v5.0
Ran qemu-ptest on x86-64 and arm64 with identical results:
PASS: 1166
SKIP: 0
FAIL: 0
Signed-off-by: Sakib Sajal <sakib.sajal@windriver.com>
Signed-off-by: Joe Slater <joe.slater@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
specifiy -> specify
Signed-off-by: Paul Eggleton <paul.eggleton@linux.microsoft.com>
|
|
When explicitly passing a branch using --srcbranch in 'devtool add' or
'recipetool create', the branch name is not included in the params of
bb.fetch2.encodeurl and default 'master' branch is used instead.
Signed-off-by: Tuomas Salokanto <tuomas.salokanto@gmail.com>
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
From: Daniel McGregor <daniel.mcgregor@vecima.com>
In order to allow consistent output of buildhistory-collect-srcrevs sort
the list of directories returned by os.walk. Otherwise the list of
SRCREVs is returned in an unspecified order.
We save the output of this command on build, so it creates smaller diffs
between builds.
Signed-off-by: Daniel McGregor <daniel.mcgregor@vecima.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The error message was reporting the calculated offset instead of the
current offset, which made it confusing.
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add a filter subcommand to filter a task-depends.dot graph produced by
bitbake -g down to just a subset of targets/tasks.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.microsoft.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
argparse makes this a lot easier to extend.
Signed-off-by: Paul Eggleton <paul.eggleton@linux.microsoft.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
The --check option turns on checksum validation, but it defaults to 'on'
so is pointless. Add a corresponding --no-check option to turn off
validation.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Remove all instances of the hardcoded 'x86_64' and replace with the current
host platform.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Several failure paths were displaying an error message but not returning,
so the install process continued and failed further.
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
If there are multiple appends, ensure we concatenate compatible things:
Traceback (most recent call last):
File "/home/akiernan/poky/scripts/recipetool", line 111, in <module>
ret = main()
File "/home/akiernan/poky/scripts/recipetool", line 100, in main
ret = args.func(args)
File "/home/akiernan/poky/scripts/lib/recipetool/edit.py", line 38, in edit
return scriptutils.run_editor([recipe_path] + appends, logger)
TypeError: can only concatenate list (not "tuple") to list
Signed-off-by: Alex Kiernan <alex.kiernan@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
467f84e12b ("wic: Add --offset argument for partitions") broke the
--extra-space argument handling in wic. Fix the option and add a unit
test for the argument.
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Add support for an --offset argument when defining a partition. Many
SoCs require that boot partitions be located at specific offsets. Prior
to this argument, most WKS files were using the --align attribute to
specify the location of these fixed partitions but this is not ideal
because in the event that the partition couldn't be placed in the
specified location, wic would move it to the next sector with that
alignment, often preventing the device from booting. Unlike the --align
argument, wic will fail if a partition cannot be placed at the exact
offset specified with --offset.
Changes in V2:
* Fixed a small typo that prevented test_fixed_size_error from passing
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
* I see a case where a tarball contains .gitignore and bunch of files
which are normally ignored in git, but still included in the tarball
(e.g. configure script next to configure.ac)
* when devtool is creating a git repo in workspace it won't include these
files from tarball in the initial devtool-base commit, because
git ls-files won't list them
* but then the first .patch file (without git headers) when applied with
GitApplyTree._applypatch() will add all these still ignored files to a
commit which used to only modify some files, because it's using -f:
# Add all files
shellcmd = ["git", "add", "-f", "-A", "."]
output += runcmd(["sh", "-c", " ".join(shellcmd)], self.dir)
at least in this case it would be better to add all ignored files in
the initial devtool-base commit and then --force-patch-refresh will just
include the small modification as before instead of adding unrelated
files, just because they were initially ignored - this behavior will
also match with the do_patch task in the actual build where the
.gitignore is ignored when unpacking some tarball
* my use-case is fixed in setup_git_repo, but similar function is in
devtool upgrade, I've changed it there as well
Signed-off-by: Martin Jansa <Martin.Jansa@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
Currently only ptest logs are accessible with the log command, this
adds support so the ltp logs can be extracted too.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
I've long since wondered why there was some odd output in result reports,
remove the leftover debug which was causing it.
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
|
As in other places in the file, skip layers that don't define a
collection when searching for a layer to resolve a dependency. Fixes
KeyError exceptions when attempting to access the layer collections
later
Signed-off-by: Joshua Watt <JPEWhacker@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
|
Ross Burton
Roland Hieber
Richard Purdie
Khem Raj
Khairul Rohaizzat Jamaluddin
Adrian Freihofer
Michael Tretter
Vijai Kumar K
Martin Jansa
Joshua Watt
Vasyl Vavrychuk
Chen Qi
Daniel Ammann
Lee Chee Yang
Khasim Mohammed
Kevin Hao
Tim Orling
Bjarne Michelsen
Nicolas Dechesne
Jason Wessel
Sakib Sajal
Paul Eggleton
Tuomas Salokanto
Dan McGregor
Ross Burton
Alex Kiernan