Changelog:
===========
- [bug] [lexer] Reverted the fix for #140 released in Mako 1.3.7 as it produced
regressions in existing user code.
- [bug] [codegen] [lexer] During the lexical analysis phase, add an additional prefix for undeclared
identifiers that have the same name as built-in flags, and determine the
final filter to be used during the code generation phase based on the
context provided by the user.
- [bug] [lexer] Support the direct passing of dictionary literals when calling functions
and fix the errors caused by nested braces. This revises the fix that was
released in 1.3.4 and then reverted in 1.3.5.
(From OE-Core rev: 76e1577b99b4319a9ea45fa4fb2b07d849f35f68)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
0001-python-module-do-not-manipulate-the-environment-when.patch
refreshed for 1.6.1
Changelog:
===========
- dependencies: support old vulkan SDK version
- Handle top level options set in subprojects. Closes#13847.
- ci: fix Ubuntu Bionic job
- modernize Rust template
- run_single_test.py: skip setup_symlinks() call on Windows
- tests: HDF5 no longer skips on MacOS
- mdevenv: exec directly into the program to run
- CMakeToolchain: Log output on compiler state failure
- msetup: Correction of the message text
- run_single_test.py: Fix for symlink changes
- libgcrypt-config is no more on ubuntu-rolling
- fix generate_gir with BothLibraries dependency
- env2mfile: add flag to use _FOR_BUILD envvars.
- environment: Never require an exe_wrapper for native builds
- rust: fix computation of library directory
- De-duplicate BuildTarget.sources
- Fix building on AIX when no project languages are used
- coverage.py: Guard use of --html-nested behind version check.
- Xcode backend: only use found appleframeworksn
- Enable GCC to find free-threaded python DLL library
- modules/rust: Add support for autolib field in the Cargo.toml
- env2mfile: Use a cross valac on Debian if possible
- env2mfile: Automatically set exe_wrapper on Debian if possible
- env2mfile: Use Debian cross-prefixed GObject-Introspection tools
- env2mfile: Generalize detection of pkg-config to have a list of tools
- Add GNU/Hurd kernel results
(From OE-Core rev: ca17f6c26c7f1fa93950143a41204daf57e5b4ed)
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Building weston with an SDK fails to find pkg-config-native:
```
Did not find pkg-config by name 'pkg-config-native'
Found pkg-config: NO
Found CMake: /opt/poky/5.1/sysroots/x86_64-pokysdk-linux/usr/bin/cmake (3.31.0)
Build-time dependency wayland-scanner found: NO (tried cmake)
protocol/meson.build:1:14: ERROR: Dependency lookup for wayland-scanner with method 'pkgconfig' failed: Pkg-config for machine build machine not found. Giving up.
```
Fix the problem by extending the pkg-config-native wrapper install to
class-nativesdk.
(From OE-Core rev: 065ab44943b78ccb806477684b1167d69d5d1d6d)
Signed-off-by: Tom Hochstein <tom.hochstein@oss.nxp.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Clean up the do_install append, and remove a long-standing unused
variable that appears to be intending to not install the scripts but
would have never actually done that as the relevant override since 2008
has been task-install. As we've been installing the scripts, keep
instaling them.
(From OE-Core rev: 3171dd16ec1a4a7461d29107b8c4bd29ff9acf42)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
expect has a hand-maintained aclocal.m4 so don't run aclocal, which has the
side effect of not deleting the aclocal.m4 file which pulls in macros.
The build works without this change more through luck and a combination
of behaviours than design.
(From OE-Core rev: 99fee9f51eb991207dea136465c18735ef833cca)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
tcl has a hand-maintained aclocal.m4 so don't run aclocal, which has the
side effect of not deleting the aclocal.m4 file which pulls in macros.
The build works without this change more through luck and a combination
of behaviours than design.
(From OE-Core rev: 34b354f82aa3d28820ecb29f463cbbb97c7c345a)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
tcl has a hand-maintained aclocal.m4 so don't run aclocal, which has the
side effect of not deleting the aclocal.m4 file which pulls in macros.
The build works without this change more through luck and a combination
of behaviours than design.
(From OE-Core rev: f2f0edb229a62e63807dd4b4ec2944e7cd06d364)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Upstream has an autogen.sh which constructs a hand-written aclocal.m4
and manually copies libtool into place. Instead of working around the
bad interaction between these expectations and our autotools class we
can just disable the execution of aclocal in autoreconf and copy files
as autogen.sh does.
(From OE-Core rev: ce0865e9a79b1603a2553392f2cd3b6394239153)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
The configure script looks explicitly for expat, so add it to DEPENDS.
(From OE-Core rev: ab6b4f45da74727e7adae7ceab6f507b2b043e60)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
There is the following warning when building linux-yocto with
default configuration on x86-64 with gcc-14.2:
AR built-in.a
AR vmlinux.a
LD vmlinux.o
vmlinux.o: warning: objtool: .export_symbol+0x332a0: data relocation to !ENDBR: stpcpy+0x0
This change set removes the warning.
PR target/116174 [https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116174]
(From OE-Core rev: 30d4f18d1e11b3336c8668dccd96b9ff35c7bc76)
Signed-off-by: Bin Lan <bin.lan.cn@windriver.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
==========
- (mkfs.erofs) Fix multi-threaded compression with '-Eall-fragments';
- (mkfs.erofs) Fix large chunk-based image generation;
- (mkfs.erofs) Avoid large arrays on the stack
- (mkfs.erofs) Fix PAX format parsing in headerball mode
- (mkfs.erofs) Several fixes for incremental builds
- (mkfs.erofs) Fix reproducible builds due to 'i_ino'
- Use pkg-config for liblz4 configuration;
- Get rid of pthread_cancel() dependencies;
- (mkfs.erofs) Add '-U <clear|random>' support;
- (mkfs.erofs) Add '--hard-dereference' for NixOS reproducibility
- Several minor random fixes.
(From OE-Core rev: 4fb94c8d873e0eff878308deadf31cab413ed590)
Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Drop two patches merged upstream.
(From OE-Core rev: 540ba2ce2f6d138b386d0d7545c197fd7f54edc8)
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With the upstream check migrated to the simple repo API, a number of the
recipes required updates to:
1. Remove outdated UPSTREAM_CHECK_REGEX checks
2. Add recipe specific UPSTREAM_CHECK_PYPI_PACKAGE definitions for
packages that use '_', CamelCase, or other deviations from PEP625 in
the source archive
(From OE-Core rev: efbb98977664cd5392ec00d97d0eaf5374a32573)
Signed-off-by: Derek Straka <derek@asterius.io>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
A buffer overflow vulnerability exists in GNU Binutils’ objdump utility
when processing tekhex format files. The vulnerability occurs in the
Binary File Descriptor (BFD) library’s tekhex parser during format identification.
Specifically, the issue manifests when attempting to read 8 bytes at an address
that precedes the global variable ‘_bfd_std_section’, resulting in an out-of-bounds read.
Backport a patch from upstream to fix CVE-2024-53589.
Upstream-Status: Backport [https://sourceware.org/git/?p=binutils-gdb.git;a=commitdiff;h=e0323071916878e0634a6e24d8250e4faff67e88]
(From OE-Core rev: 7c9a9020d1e9204ba875ac10b20ab7ccabce82bc)
Signed-off-by: Yash Shinde <Yash.Shinde@windriver.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This commit updates the warning to use a check for "trivially constructible" instead of
"trivially copyable." The original check was incorrect, as "trivially copyable" only applies
to types that can be copied trivially, whereas "trivially constructible" is the correct check
for types that can be trivially default-constructed.
This change ensures the warning is more accurate and aligns with the proper type traits.
LLVM accepted a similar fix:
https://github.com/llvm/llvm-project/issues/47355
PR c++/116731 [https://gcc.gnu.org/bugzilla/show_bug.cgi?id=116731]
(From OE-Core rev: 7c9402953d1e4eed4bd4427171e96761812f7478)
Signed-off-by: Marek Polacek <polacek@redhat.com>
Signed-off-by: Sunil Dora <sunilkumar.dora@windriver.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This was prompted by sstatetests.SStateHashSameSigs2.test_sstate_allarch_samesigs
where adwaita-icon-theme would fail due to new librsvg introducing
cargo-c-native into dependency path.
Rather than look into why is that, I simply converted the recipe
into the standard 'target + BBCLASSEXTEND' approach.
(From OE-Core rev: 80a2673a3081e623f30f25cc773ac9217e717958)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
rpm2cpio has been deprecated upstream, so this prepares for its
eventual removal.
rpm2archive produces a tar archive which can be uncompressed
with tar executable from the host.
(From OE-Core rev: ed824d3fb23f0c89d8dfdacb2c4ef0b7c21a5144)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
With this release, as expected, deprecated openpgp support
has been fully removed and replaced with sequoia:
https://github.com/rpm-software-management/rpm-sequoia/
Alas, it's written in rust, and has no recipe, so for now
signing rpm packages has to be disabled.
Remove package name parameter to %prep in source archiving, as it's
been dropped in the new rpm, and wasn't needed to begin with.
Drop 0001-perl-disable-auto-reqs.patch
(files removed upstream).
(From OE-Core rev: 8c15b4577d5e554cc2dd5adfb88b816894b05a9a)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Changelog:
============
* subvolume delete: add new option to do recursive subvolume deletion (for
regular user delete only accessible subvolumes)
* mkfs:
* new option --subvol to create subvolumes in given paths, read-write,
read-only and default
* add hard link detection support for --rootdir option
* fixes:
* receive: message verbosity fixes
* check: fix false positive report of missing checksum for extent holes
* check: handle compressed extents when checking tree log
* when asking Y/N user questions, flush the terminal so the question is
displayed (e.g. btrfstune -S)
* other
* code refactoring, error handling
* python packaging fixes
* documentation updates
* new tests
(From OE-Core rev: ef59c676cf0aa9acb6180cdc74c6404d859b4880)
Signed-off-by: Wang Mingyu <wangmy@cn.fujitsu.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
2.5.2 was a beta, upstream made two stable releases since:
https://savannah.gnu.org/news/?id=10676https://savannah.gnu.org/news/?id=10693
Drop libtool/0001-ltmain.in-Handle-trailing-slashes-on-install-command.patch
(merged upstream)
Apply the multilib patch unconditionally as it could be applied to add variants.
For the other it was pointed out that it:
"potentially has side effects for native builds. It breaks the ability
to run files in the .libs directory which in the cross case we don't
care about but some natives did once used to do that. The risk is they
find host things instead."
License-Update: change of FSF address to a URI
(From OE-Core rev: 9a672faf21648c60d7ff29439bb236f3ac87100d)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove patch, as it is a leftover from meta-virtualization times
where python3-dtc was ahead of dtc in oe-core and had to be tweaked.
(From OE-Core rev: 0706f0e3a60713bffad219c0658b15020010eb5e)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Remove 0002-sepdebugcrcfix.c-do-not-use-64bit-variants.patch
as the issues were fixed upstream.
(From OE-Core rev: d4cbc3fe8354ce2bbcee90c5bd9e6b821b488900)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
This was prompted by working on librsvg update: the new meson-driven
version wants to query values from .pc files residing in its own
build directory, and modifies PKG_CONFIG_PATH accordingly.
When using the pkg-config-native wrapper such modifications
have no effect, and we have to pass them in manually
from the recipe via EXTRA_NATIVE_PKGCONFIG_PATH variable.
This variable is already defined (with an empty value) and
appended to PKG_CONFIG_PATH export in the native class, so this
simply extends its use to the wrapper.
(Appending to PKG_CONFIG_PATH in the wrapper, instead of resetting it,
is not an option as that can lead to contamination with the cross values).
(From OE-Core rev: 2bc050146d47b14d890a1b0db2b55f9057a08b65)
Signed-off-by: Alexander Kanavin <alex@linutronix.de>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
Fixed:
bitbake automake-native failed with error:
| :; HELP2MAN_NAME="Generate Makefile.in files for configure from Makefile.am"; export HELP2MAN_NAME; LONGPATH/tmp/hosttools/mkdir -p doc && AUTOMAKE_HELP2MAN=true ./pre-inst-env LONGPATH/tmp/hosttools/perl ../automake-1.17/doc/help2man --output=doc/automake-1.17.1 --info-page=automake --name="${HELP2MAN_NAME}" bin/automake-1.17
| help2man: can't get `--help' info from bin/aclocal-1.17
| Try `--no-discard-stderr' if option outputs to stderr
| make: *** [Makefile:3818: doc/aclocal-1.17.1] Error 2
In above command, `bin/aclocal-1.17 --help` will be runned, since
@PERL@ is replaced as the full path of perl, it is possible that the
shebang line exceeds the max limit and causes above error, because the
kernel imposes a maximum length on the text that follows the "#!"
characters at the start of a script; characters beyond the limit are
ignored. Before Linux 5.1, the limit is 127 characters. Since Linux
5.1, the limit is 255 characters.
Add back the forced perl settings and add a patch for workaround the
fatal error caused by space in perl path
(From OE-Core rev: 0061f441501be923ddf326dcc5b3f25b59ac86d4)
Signed-off-by: Changqing Li <changqing.li@windriver.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
1.26.3:
- Support an old import path that is still used by some consumers like
Hatch
1.26.2:
- Back-populate string license fields (License-Expression) for core
metadata versions prior to 2.4
- Remove the License-Expression and License-Files core metadata from
version 2.2 that was missed in the previous minor release
1.26.1:
- Add backward compatibility for the old license-files metadata field
- Support an old import path that is still used by some consumers like
Hatch
1.26.0:
- The license-files metadata field has been updated to the latest spec
and is now just an array of glob patterns
- Support version 2.4 of core metadata for the wheel and sdist targets
- Add HATCH_METADATA_CLASSIFIERS_NO_VERIFY environment variable to
disable trove classifier verification
- Add .pixi to the list of directories that cannot be traversed
- Bump the minimum supported version of packaging to 24.2
- No longer write package metadata for license expressions and files for
versions of core metadata prior to 2.4
- Properly enable Zip64 support for the wheel target
- Properly ignore parent .gitingore files when the project root matches
one of the patterns
(From OE-Core rev: 2c09f03e62f96016c2a3713362f83a7591bd0f9b)
Signed-off-by: Ross Burton <ross.burton@arm.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
RISC-V offers several virtual memory address schemes (Sv39, Sv48, and Sv57),
but ASan currently supports only Sv39 on RISC-V64.
For RISC-V64 Sv39, ASan uses custom allocator configurations
tuned to manage large allocations efficiently.
These tunings are incompatible with larger address spaces like
Sv48/Sv57 due to differences in region sizes and alignment.
For riscv64, Asan's tuning for Sv39 can be enabled in qemu
by using the appropriate flag in the command line as shown below:
runqemu nographic qemuparams="-cpu rv64,sv39=true"
Additionally, the COMPATIBLE_HOST list in gcc-sanitizers has been
updated to include riscv64. All necessary tests were successfully
conducted on both hardware(Microchip PolarFire SoC)
and the qemurisv64 environment, with ASan effectively
detecting memory errors in both scenarios.
(From OE-Core rev: 4b4450ff695ef73bf7a2437e142d2e0730d6a547)
Signed-off-by: Deepthi Hemraj <Deepthi.Hemraj@windriver.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
* The package now requires python3-hatch-vcs-native at build-time
* Updated HOMEPAGE to point to the correct Github repository. The old link re-directs to the same link.
2.2.3 (2024-09-12)
==================
Features
--------
- Added support for Python 3.13.
Bugfixes
--------
- Fixed the default encoding of chunked request bodies to be UTF-8 instead of ISO-8859-1. All other methods of supplying a request body already use UTF-8 starting in urllib3 v2.0.
- Fixed ResourceWarning on CONNECT with Python < 3.11.4 by backporting https://github.com/python/cpython/issues/103472
- Adjust tolerance for floating-point comparison on Windows to avoid flakiness in CI
- Fixed a crash where certain standard library hash functions were absent in restricted environments.
- Fixed mypy error when adding to ``HTTPConnection.default_socket_options``.
HTTP/2 (experimental)
---------------------
See details: https://github.com/urllib3/urllib3/releases/tag/2.2.3
Deprecations and Removals
-------------------------
- Note for downstream distributors: the ``_version.py`` file has been removed and is now created at build time by hatch-vcs.
- Drop support for end-of-life PyPy3.8 and PyPy3.9
(From OE-Core rev: 61331e291a2f96c7ac7efccd7d95de68432ea992)
Signed-off-by: Guðni Már Gilbert <gudni.m.g@gmail.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
python3-iniparse dependency was dropped 2019, see the
following commit as reference:
d7d0e0e2f9
When looking at the Git history, this happened around tag 4.2.1
(From OE-Core rev: 05220b4907ab654e829d97ab68d48f76330937dd)
Signed-off-by: Guðni Már Gilbert <gudni.m.g@gmail.com>
Signed-off-by: Mathieu Dubois-Briand <mathieu.dubois-briand@bootlin.com>
Signed-off-by: Richard Purdie <richard.purdie@linuxfoundation.org>
