Uses TEST_SERIALCONTROL_CMD to open a serial connection to the target
and execute commands. This is a drop in replacement for the ssh target,
fully supporting the same API. Supported with testexport.
To use, set the following in local.conf:
- TEST_TARGET to "serial"
- TEST_SERIALCONTROL_CMD to a shell command or script which connects to
the serial console of the target and forwards that connection to
standard input/output.
- TEST_SERIALCONTROL_EXTRA_ARGS (optional) any parameters that must be
passed to the serial control command.
- TEST_SERIALCONTROL_PS1 (optional) A regex string representing an empty
prompt on the target terminal. Example: "root@target:.*# ". This is
used to find an empty shell after each command is run. This field is
optional and will default to "root@{MACHINE}:.*# " if no other value is
given.
- TEST_SERIALCONTROL_CONNECT_TIMEOUT (optional) Specifies the timeout in
seconds for the initial connection to the target. Defaults to 10 if no
other value is given.
The serial target does have some additional limitations over the ssh
target.
1. Only supports one "run" command at a time. If two threads attempt to
call "run", one will block until it finishes. This is a limitation of
the serial link, since two connections cannot be opened at once.
2. For file transfer, the target needs a shell and the base32 program.
The file transfer implementation was chosen to be as generic as
possible, so it could support as many targets as possible.
3. Transferring files is significantly slower. On a 115200 baud serial
connection, the fastest observed speed was 30kbps. This is due to
overhead in the implementation due to decisions documented in #2
above.
(From OE-Core rev: d817b27d73d29ba2beffa2e0a4e31a14dbe0f1bf)
Signed-off-by: Andrew Oppelt <andrew.j.oppelt@boeing.com>
Signed-off-by: Matthew Weber <matthew.l.weber3@boeing.com>
Signed-off-by: Chuck Wolber <chuck.wolber@boeing.com>
--
Tested with core-image-sato on real hardware. TEST_SERIALCONTROL_CMD
was set to a bash script which connected with telnet to the target.
Additionally tested with QEMU by setting TEST_SERIALCONTROL_CMD to
"ssh -o StrictHostKeyChecking=no root@192.168.7.2". This imitates
a serial connection to the QEMU instance.
Steps:
1) Set the following in local.conf:
- IMAGE_CLASSES += "testexport"
- TEST_TARGET = "serial"
- TEST_SERIALCONTROL_CMD="ssh -o StrictHostKeyChecking=no root@192.168.7.2"
2) Build an image
- bitbake core-image-sato
3) Run the test export
- bitbake -c testexport core-image-sato
4) Run the image in qemu
- runqemu nographic core-image-sato
5) Navigate to the test export directory
6) Run the exported tests with target-type set to serial
- ./oe-test runtime --test-data-file ./data/testdata.json --packages-manifest ./data/manifest --debug --target-type serial
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The Valgrind recipe has had support for aarch64 for 9 years but the
packagegroup-core-tools-profile recipe still excludes valgrind on
aarch64 builds. This patch adds Valgrind when building the
package-group-core-tools-profile for aarch64 systems.
(From OE-Core rev: 2f8f6d722b39d2d1080367bf780dead4ed4ed781)
Signed-off-by: John Ripple <john.ripple@keysight.com>
Signed-off-by: Alexandre Belloni <alexandre.belloni@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Unfortunately some recent patches caused non-deterministic output.
One input hash lead to both good and bad output and whilst that patch
has been fixed, the problematic hash 'cross' linkage remains. Bump to
a new sstate and hash equivalence version to avoid this and work from
a clean slate.
(From OE-Core rev: 639e42b9c14dff516688964dba4ab25bba7b8a55)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If we are using 'ld-is-gold' as DISTRO_FEATURE, the U-Boot Makefile has
mechanisms to ensure that we use BFD as linker. However, this does not
work for the GCC when it is used as linking tool.
To support that, we added explicit BFD use whenever we have the
DISTRO_FEATURE feature enabled.
(From OE-Core rev: d36ee7a87c0d89b2820eea4bfcdae342adc75019)
Signed-off-by: Otavio Salvador <otavio@ossystems.com.br>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
For postmortem analysis it's helpful to know if the build environment was
clean or contained any modifications.
(From OE-Core rev: edaaa2ad311663beabd2416037de00d82fca5fba)
Signed-off-by: Jörg Sommer <joerg.sommer@navimatix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Some redundant return statements were left over from
insane: Drop oe.qa.add_message usage
(From OE-Core rev: 1e49635f802b04acad14115640ce9fcd63cc32a7)
Signed-off-by: Ola x Nilsson <olani@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
These tests verify that the correct files are left behind when systemd
is inherited and depending on whether the systemd and/or sysvinit distro
features are enabled.
(From OE-Core rev: 4ef25caa00ad08046567d0e7f4523486c3a256d0)
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Previously, rm_systemd_unitdir() would remove one parent directory of
${systemd_unitdir} if it was empty after removing ${systemd_unitdir}.
rm_sysvinit_initddir() would not remove any parent directory. Thus, if
the only directory created in /etc was /etc/init.d, an empty /etc would
remain after the cleanup and would be packaged.
Simplify rm_systemd_unitdir() and rm_sysvinit_initddir() by rewriting
them in shell, and use rmdir -p to remove all empty parent directories.
(From OE-Core rev: 73159c0bbc636a08934f47690885d75fd37b701a)
Signed-off-by: Peter Kjellerstedt <peter.kjellerstedt@axis.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
After the recent piglit update it fails X11 builds reporting that it can
not find wayland-client package. Commit 9662bdabf3f6 ("piglit: upgrade
22eaf6a91c -> c11c9374c1") has added wayland-related dependencies to the
recipe's DEPENDS variable, but it seems it missed the target wayland
pacakge, making cross-compilation of piglit fail when building
X11-targeted configurations.
Pushing more and more wayland dependencies to a generic build doesn't
seem to be a good idea, so introduce "wayland" PACKAGECONFIG option,
move existing wayland-related dependencies to be guarded by it and add
dependency on the wayland package.
Suggested-by: Alexander Kanavin <alex.kanavin@gmail.com>
Fixes: 9662bdabf3f6 ("piglit: upgrade 22eaf6a91c -> c11c9374c1")
Cc: Trevor Gamblin <tgamblin@baylibre.com>
(From OE-Core rev: 6c504d03c8bd8e2a9c8d0b946dbf28e94d13b5d6)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
runqemu currently only supports scsi, ide and virtio drive types.
Implement QB_DRIVE_TYPE=/dev/mmcblk which adds an sdhci-pci device and
mounts the rootfs file as an sd card.
(From OE-Core rev: 0740896afbe13f37058e990a0a13b2b51bba70fd)
Signed-off-by: Alessandro Pecugi <alessandro.pecugi@seco.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Without unsetting `j` and `k` for each `UBOOT_MACHINE`, `j` and `k`
are incremented in the same frequency as `i` and therefore `$j -eq $i`
and `$k -eq $i` is always true for the first `type` from `UBOOT_CONFIG`
and the first `binary` from `UBOOT_BINARIES`.
(From OE-Core rev: 3aef55c7ceb654b0012f20618bfd6ead1ef578b6)
Signed-off-by: Paul Gerber <paul.gerber@ew.tq-group.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Separate 300KB wfx firmware from the rest of linux-firmware package.
With this change the wfx/LICENCE.wf200 is really packaged, before the
change it didn't fall into copying wildcard.
(From OE-Core rev: 48e14565abcfe0ec8eb21b1c3e6108bb5b0a0b7c)
Signed-off-by: Alexander Sverdlin <alexander.sverdlin@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The pattern `-serial` matches also `-device usb-serial` and `virtio-serial`
which are not the desired parameter. This causes the serial console ttyS1 is
missing and Systemd's getty@ttyS1 fails constantly.
(From OE-Core rev: b6d035894120b45b42f146ab5b3110522c58d178)
Signed-off-by: Jörg Sommer <joerg.sommer@navimatix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog (https://github.com/pyparsing/pyparsing/releases):
3.1.4
- Fixed a regression introduced in pyparsing 3.1.3, addition of a type
annotation that referenced re.Pattern. Since this type was introduced
in Python 3.7, using this type definition broke Python 3.6 installs of
pyparsing 3.1.3. PR submitted by Felix Fontein, nice work!
3.1.3
- Added new Tag ParserElement, for inserting metadata into the parsed
results. This allows a parser to add metadata or annotations to the
parsed tokens. The Tag element also accepts an optional value
parameter, defaulting to True. See the new tag_metadata.py example in
the examples directory.
Example:
# add tag indicating mood
end_punc = "." | ("!" + Tag("enthusiastic")))
greeting = "Hello" + Word(alphas) + end_punc
result = greeting.parse_string("Hello World.")
print(result.dump())
result = greeting.parse_string("Hello World!")
print(result.dump())
prints:
['Hello', 'World', '.']
['Hello', 'World', '!']
- enthusiastic: True
- Added example mongodb_query_expression.py, to convert human-readable
infix query expressions (such as a==100 and b>=200) and transform them
into the equivalent query argument for the pymongo package ({'$and':
[{'a': 100}, {'b': {'$gte': 200}}]}). Supports many equality and
inequality operators - see the docstring for the transform_query
function for more examples.
- Fixed issue where PEP8 compatibility names for ParserElement static
methods were not themselves defined as staticmethods. When called
using a ParserElement instance, this resulted in a TypeError
exception. Reported by eylenburg (#548). To address a compatibility
issue in RDFLib, added a property setter for the ParserElement.name
property, to call ParserElement.set_name. Modified
ParserElement.set_name() to accept a None value, to clear the defined
name and corresponding error message for a ParserElement. Updated
railroad diagram generation for ZeroOrMore and OneOrMore expressions
with stop_on expressions, while investigating #558, reported by user
Gu_f. Added <META> tag to HTML generated for railroad diagrams to
force UTF-8 encoding with older browsers, to better display Unicode
parser characters.
- Fixed some cosmetics/bugs in railroad diagrams: fixed groups being
shown even when show_groups=False show results names as quoted strings
when show_results_names=True only use integer loop counter if
repetition > 2
- Some type annotations added for parse action related methods, thanks
August Karlstedt (#551).
- Added exception type to trace_parse_action exception output, while
investigating SO question posted by medihack.
- Added set_name calls to internal expressions generated in
infix_notation, for improved railroad diagramming.
- delta_time, lua_parser, decaf_parser, and roman_numerals examples
cleaned up to use latest PEP8 names and add minor enhancements.
- Fixed bug (and corresponding test code) in delta_time example that did
not handle weekday references in time expressions (like "Monday at
4pm") when the weekday was the same as the current weekday.
- Minor performance speedup in trim_arity, to benefit any parsers using
parse actions.
- Added early testing support for Python 3.13 with JIT enabled.
(From OE-Core rev: 320d54ab73071fd58667e672a3d5415fc8bd5e29)
Signed-off-by: Trevor Gamblin <tgamblin@baylibre.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This option was incorrectly removed in the recent upgrade. We don't use/need
the static library so disable the build of it to save time/disk usage.
(From OE-Core rev: 2ba36d98bb0fcdd8d918da8fb69f09a06b75fd46)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The autobuilder was seeing an intermittent reproducbility issue in lz4 with
regard to symbol ordering in the static library. Add a patch to fix this which
has been submitted upstream.
Fix the SRC_URI whitespace and cleanup the patch directory naming whilst here.
(From OE-Core rev: 79d6d0fa239ab0cf271c01d8169a7d41811f8b4e)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* Sort the exclude list in alphabetical order.
* Add some of the upsupported/failing tests to the exclude list
and ignore the failing unit tests.
* Remove duplicated tests from the exclude list.
* Testing summary
Target PASS SKIPPED
ARM 17241 599
ARM64 17279 561
MIPS64 17228 612
PPC 17194 629
X86 17257 583
X86-64 17416 424
(From OE-Core rev: 61ec0f0f78d9db4c3fc02365713d5fd77b78a7ea)
Signed-off-by: Yash Shinde <Yash.Shinde@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The zlib crate in rust uses libz.a which comes from the zlib-native build.
Some distros like alma9, fedora etc. do not have PIE enabled by default for system compiler.
This leads to target-rust-ccld linking error for cargo-native as (line no 22936):
error: linking with `/home/pokybuild/yocto-worker/qemuarm64/build/build/tmp/work/x86_64-linux/cargo-native/1.79.0/wrapper/target-rust-ccld` failed: exit status: 1
https://autobuilder.yoctoproject.org/typhoon/#/builders/42/builds/9385/steps/13/logs/stdio
Hence, enable PIE option to CFLAGS for native builds.
(From OE-Core rev: 7146d260f655fa924461333c8c2944ebb93b2b3c)
Signed-off-by: Yash Shinde <Yash.Shinde@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This patch is already proposed upstream and perhaps landing
soon in gdb master.
(From OE-Core rev: 6721de5a049b245f274081b9b474e81761ea40fd)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This adds the functionality to convert the results of the
testresults.json file to a unit test report in JUnit XML format. The
unit test report can be used in the CI/CD pipeline to display the test
results.
To use the resulttool scripts, first source oe environment, then run the
entry point script to look for help.
$ resulttool
To generate the unit test report, execute the below
$ resulttool junit <json_file>
By default the unit test report is stored as
<build_dir>/tmp/log/oeqa/junit.xml.
(From OE-Core rev: 3f9be03946243feaa09b908d7010899769091fe6)
Signed-off-by: Clara Kowalsky <clara.kowalsky@siemens.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Now that adding new values to ERROR_QA doesn't invalidate tasks, add this
check to the selftests so that this functionality doens't regress in future.
(From OE-Core rev: 823b5f6cf64ad37ee68e6274c58823f7cc75ddf0)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
We want ERROR_QA to operate using the "contains" optimizations which means
accessing the variable only using the contains function.
To do this, remove usage of ALL_QA, open coding the few references to check
both WARN_QA and ERROR_QA.
Move the function table generation to a separate function where we can exclude
the ERROR_QA and WARN_QA variables since they are handled by the handle_error()
function calls.
Ensure all the chain of functions to the handle_error calls is correctly
recognised in the variable dependencies.
(From OE-Core rev: 384e9a6b2e7943b6a3ade1215ed79351c78a0b0d)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Now handle_error is used, we can further simplify the QA test execution
as we don't need seperate function lists for warnings and errors.
(From OE-Core rev: 6896c9fcfc57f007c0ce15f7804e79b6b88f5ded)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
If we improve the function dependency visibility in insane.bbclass, it
exposes some dependencies which were previously not seen causing variances
in the do_package_qa task checksums. Update vardepsexclude in a couple of
test cases to ensure the sstate hash selftests pass and the taskhashes
don't vary when we don't expect them too.
(From OE-Core rev: 9b6dae2771ed86bd2946548004f4da58e8c0b44c)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop the oe.qa.add_message() usage in favour of oe.qa.handle_error() which has
code allowing it to be optimised with contains usage.
The patch also drops unused return values which we stopped using a while ago
and drops the now unneeded function parameters, generally leading to cleaner
code.
The code should be functionally equivalent.
(From OE-Core rev: 9b2eea9fd4eab4f5e12e955738db22091b91f698)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The new contains code can't inspect variable references in handle_error()
calls. Expand what is effectively a hardcoded reference anyway so the code
can optimise it.
(From OE-Core rev: 51f767d92efb3daeb4aa3b91d72e6d2993cb0f46)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Early functions like do_recipe_qa (which do_fetch depends upon) reference
oe.qa.handle_error() which in turn adds dependencies on ERROR_QA and
WARN_QA. This means that ERROR_QA:append = " nothing" will cause
literally everything to rebuild and break sstate reuse.
Take advantage of new bitbake functionality to add a custom visitorcode
function to handle_error which optimises the references into contains
expressions which means the ERROR_QA and WARN_QA references are optmised
to containing specific strings. This dramatically improves sstate reuse.
The qa module has to be imported first since other code in later modules
references it and bitbake can't handle the dependency ordering internally
without a lot of unwanted complexity.
(From OE-Core rev: a911ea9659503e9442a183f366e4545a5efe246e)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Bring following changes on top of 1.2.5
* dd1e63c3 syslog: revert LOG_FAC/LOG_FACMASK changes
* 008f737d siglongjmp: document why this function just calls longjmp
* 947b4574 inet_ntop: fix the IPv6 leading zero sequence compression
* 50ab8306 dynlink: avoid copying to temp buffer in get_lfs64
* 1b97d006 sys/epoll.h: add epoll ioctls
* ab31e9d6 getusershell: skip blank lines and comments
* 53ac44ff dynlink: fix get_lfs64() with posix_fallocate64
* 895736d4 syslog: fix incorrect LOG_MAKEPRI and LOG_FAC[MASK] macros
* 05ce67fe add renameat2 linux syscall wrapper
* 00799729 fix mismatched type in posix_getdents definition
* cbf59dd6 aarch64 crti.o: fix alignment of _init/_fini
* 84015cee fix typo that broke sys/reg.h and sys/user.h
* 1b0d4851 implement posix_getdents adopted for next issue of POSIX
* 2c124e13 stdint.h: derive limits from __LONG_MAX, use common fast16 types
* 7019fbe1 sys/user.h: derive __WORDSIZE from __LONG_MAX
* e709a6f0 sys/reg.h: derive __WORDSIZE from __LONG_MAX
* 29b216b2 unistd.h: derive ILP32/LP64 macros from __LONG_MAX instead of arch bits
* 0dfa1d8c unify bits/stat.h for all archs sharing a common definition
* ef600888 align aarch64, riscv64, loongarch64 stat structure padding type
* 6f666231 ldso: fix non-functional fix to early dynamic PAGE_SIZE access
* fced99e9 strptime: implement conversion specifiers adopted for next POSIX issue
* 3f9d4224 printf decimal integer formatting: shave off one division
* a23cf8f9 riscv mcontext_t/sigcontext: use __aligned__ instead of aligned
* cbf1c7b6 add missing STATX_ATTR_* macros omitted when statx was added
* 3f49203c initgroups: do not artificially limit number of supplementary groups
* 24ebbbde printf: fix edge case where hex float precision was not honored
* e3b0ace5 complex: fix comment in cacosh
* 9683bd62 math: fix fma(x,y,0) when x*y rounds to -0
* 5370070f fix pwrite/pwritev handling of O_APPEND files
* bdc9a9ff uio.h: add RWF_NOAPPEND flag for pwritev2
* 7ada6dde iconv: fix missing bounds checking for shift_jis decoding
* fd7d0185 add missing inline keyword on default a_barrier definition
* b5121e2e iconv: add aliases for GBK
* ca6f46af iconv: add euro symbol to GBK as single byte 0x80
(From OE-Core rev: 70179bc94c90ba1f33a3fff8f3019cb96fcdbaef)
(From OE-Core rev: bcfaa9542fac82e90fbb8bdf67e3ade2697fbca4)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This release adds extension functions statx and preadv2/pwritev2, with
fallback implementations for older kernels, and adds two new ports:
loongarch64 and riscv32. Minor changes to the printf family of
functions have been made for conformance to new standards
interpretations/requirements. TLSDESC support for riscv64 has also
been added.
Bugs fixed include some DNS issues related to new TCP fallback
functionality, several rare race conditions, potentially incorrect
return value when glob aborts, and several signifiant arch-specific
bugs affecting TLSDESC on arm, riscv64 icache flushing, and sh
sigsetjmp and dlsym RTLD_NEXT. [1]
Do not use https protocol for fetching
Musl author confirms that https protocol is not well supported yet on
musl git host, currently we experience this problem intermittently on
some build hosts where the fetching fails.
fatal: protocol error: bad line length character: erro
WARNING: Failed to fetch URL git://git.etalabs.net/git/musl;branch=master;protocol=https
[1] https://www.openwall.com/lists/musl/2024/03/01/2
(From OE-Core rev: c6c79477209f5e7e1a0206942de9603a7accec67)
(From OE-Core rev: 0d0a2d62810bfa7ea51d536c4e43c2edae823a6b)
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The commit [1] has dropped the qcom/vpu-1.0/venus.mdt symlink which is
required for the kernels before 6.6. Pick up a patch restoring this
symlink.
[1] 36db650dae
(From OE-Core rev: e84b673b5788d1eeadf49c009cf58feebfcedede)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Release 20240811 has restructured the locations of Qualcomm VPU
firmware (see commit [1]). Follow those changes and implement a single
linux-firmware-qcom-vpu package holding all VPU firmware files.
[1] 36db650dae
(From OE-Core rev: 3a4204b1393b46c1ee14fa9e546e19e3f250c002)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
It makes little sense to keep board-specific firmware-5.bin in
linux-firmware-ath10k package if the wlanmdsp.mbn file is a part of the
linux-firmware-qcom-qcm2290 / -qrb4210 packages. Use wildcards to pick
those files too.
(From OE-Core rev: 0b179622bdff4b26d14f949595e9772629e5f0ac)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The linux-firmware-qcom-qcm2290-wifi package needs to steal some files
from the linux-firmware-ath10k (like qcom-qrb4210-wifi and
qcom-sdm845-modem packages do). Move the package name before the ath10k
in the list so that thieft is successful.
Fixes: f65e01a0c7eb ("linux-firmware: update to 20240513")
Cc: Quentin Schulz <quentin.schulz@cherry.de>
(From OE-Core rev: b5fb1fd145288d80fb89badb3533a3b7d1e47cf1)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Split audio topology files to separate packages so that users of the
corresponding boards don't have to pull in the whole linux-firmware
package.
(From OE-Core rev: b8a35c6eddf9ca47587d9e526b8bd345aad64cb4)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The linux-firmware-qcom-sc8280xp-lenovo-x13s-audio package includes
audio topology file which has separate licensing terms. Make this
package RDEPEND on the corresponding licence package.
Fixes: 958d6a209edf ("linux-firmware: package audio topology for Lenovo X13s")
(From OE-Core rev: 6c2ec809edd1606ec55fbd7d15e4cbea8e2752d3)
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Allow the metadata to define code visitor expressions which mean that
custom dependencies can be handled in function libraries.
An example is the qa.handle_error function in OE which can set something
like:
"""
def handle_error_visitorcode(name, args):
execs = set()
contains = {}
warn = None
if isinstance(args[0], ast.Constant) and isinstance(args[0].value, str):
for i in ["ERROR_QA", "WARN_QA"]:
if i not in contains:
contains[i] = set()
contains[i].add(args[0].value)
else:
warn = args[0]
execs.add(name)
return contains, execs, warn
handle_error.visitorcode = handle_error_visitorcode
"""
Meaning that it can have contains optimisations on ERROR and WARN_QA
instead of hard dependencies.
One drawback to this solution is the parsing order. Functions with
visitorcode need to be defined before anything else references them
or the visitor code will not function for the earlier references.
(Bitbake rev: 5bd0c65c217394cde4c8e382eba6cf7f4b909c97)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
lz4 is a new required dependency.
(From OE-Core rev: 74690600e2d5037a45e754ae28cd6afb39cf2c5c)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop most EXTRA_OEMAKE settings as no longer necessary.
(makefiles use weak assignments and shell variables then
take precedence).
License-Update: clarification that it's gpl2-or-later.
(From OE-Core rev: a4dde4e0dcc2cc75c4eec5d78fbf2a1e47401050)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
