Commit a379499f73 (package/glibc: bump to version 2.37) forgot to
update the pach for localedef.
There is no version check for 'as' now, so drop the related hunk.
Fixes: a379499f73
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Romain Naour <romain.naour@smile.fr>
Cc: Sebastian Weyer <sebastian.weyer@smile.fr>
Cc: Bagas Sanjaya <bagasdotme@gmail.com>
Tested-by: Sebastian Weyer <sebastian.weyer@smile.fr>
Last-minute changes to d9b244d8a3 (board/qemu: define start qemu
script outside of post-image script) were not tested before being
pushed, so they introduced a bug.
As Arnout put it:
[...] multiple expressions need a -e to interpret them as
expressions (now they're interpreted as file names).
Fixes: d9b244d8a3
Reported-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: James Knight <james.d.knight@live.com>
In Cargo, it is quite typical for "build scripts" to be written in Rust
and therefore they need to be compiled as part of the overall build. In
cross-compilation, that means a mixed host and target build.
Unfortunately, by default Cargo makes no distinction between the
RUSTFLAGS used for the host and the target. There is, however, an
unstable feature to make this distinction [1][2].
We already have CARGO_TARGET_APPLIES_TO_HOST="false". This makes sure
that any configuration that we make for the target doesn't automatically
apply to the host as well. However, this only applies for per-target
configuration, for example the setting of "cc" in the config.toml
generated by package/rust/rust.mk. Flags that are passed with RUSTFLAGS
still apply to both host and target. Therefore, we need to use the
CARGO_TARGET_<tuple>_RUSTFLAGS environment variable instead of plain
RUSTFLAGS.
This, however, doesn't allow us to specify flags that apply only to the
host. We could use CARGO_TARGET_<hosttuple>_RUSTFLAGS for that, but that
doesn't work in case the host and target tuple are the same. For this,
we need another unstable feature, enabled with
CARGO_UNSTABLE_HOST_CONFIG="true". With this enabled, we can specify
flags that apply only for the host build using CARGO_HOST_RUSTFLAGS.
Currently, we don't have any such flags, but we really should: we should
pass the proper link flags to point to $(HOST_DIR)/lib. Therefore, add
CARGO_HOST_RUSTFLAGS doing exactly that.
[1] https://doc.rust-lang.org/nightly/cargo/reference/unstable.html#host-config
[2] https://github.com/rust-lang/cargo/pull/10395
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Commit b32d7c99 fixed the shellcheck errors in board/qemu/post-image.sh,
but forgot to remove it from .checkpackageignore, leading to "Shellcheck
was expected to fail" errors.
Re-generate .checkpackageignore.
Fixes: b32d7c99c6
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Provides the ability to use a host system's QEMU. While a Buildroot
generated QEMU should work for most cases, a developer may wish to use
the system's QEMU for options which may not have been configured in the
Buildroot's QEMU build (e.g. configuring a different display mode).
Signed-off-by: James Knight <james.d.knight@live.com>
[yann.morin.1998@free.fr:
- use false/true instead of 0/1
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Provides the ability to forward command line options directly to QEMU.
When invoking `start-qemu.sh`, users can forward arguments by adding a
double dash (`--`) into the argument set, and any trailing arguments
will be forwarded into QEMU. For example, `start-qemu.sh -- --help`.
The original implementation supported a "serial-only" command line
argument to help run in a non-graphical mode for some use cases. These
changes try to promote a newly added `--serial-only` argument to drive
this mode; that being said, a `serial-only` argument will still be
accepted for backwards compatibility.
Signed-off-by: James Knight <james.d.knight@live.com>
[yann.morin.1998@free.fr:
- drop the warning: unknown options cause 'exit 1' already, and any
leftover is explicitly for qemu.
- use false/true instead of 0/1
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
The following moves the definition of the QEMU board's `start-qemu.sh`
helper script from being inlined in the post-image script into its own
file. This should, in theory, make it easier to maintain the script in
the future.
Signed-off-by: James Knight <james.d.knight@live.com>
[yann.morin.1998@free.fr:
- don't sub-shell in the script
- merge all 3 sed calls into one
- create dest file with the sed, don't cp first
- also substitute HOST_DIR
- fix shellcheck
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
For some documents, we may want a terse or deeper TOC depth. For
example, short documents may want just the level-0 in the TOC, while
longer documents may want depth 1 or 2, or even deeper; also, some
documents may not use the document-title levels [0], only section
levels [1], and so may want to increase the TOC depth.
Additionally, allow per-format depth. For example, split-html has a
single page dedicated to the TOC, so there we may want a deeper TOC,
while on the html output, where the TOC is on the same page as the
whole document, a shorter TOC is preferred.
[0] https://docs.asciidoctor.org/asciidoc/latest/syntax-quick-reference/#document-header
[1] https://docs.asciidoctor.org/asciidoc/latest/syntax-quick-reference/#section-titles
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Commit 17feaf0016 (manual: set toc depth to 4 for html outputs) did
not explain why it set the depth of the table of content (TOC) only for
the html and split-html formats.
From memory, but it is a bit fuzzy after all those years, the TOC for
HTML documentation was by default deeper than for the other formats, so
it was really useless to expand it more than one-level deep, while for
other formats, the default was acceptable.
However, it does not make much sense to limit/enforce the depth of the
TOC only for specific formats, and keep the default for the others;
indeed, there is nothing that prevents the default to change with
various versions of asciidoc and the rendering backends.
Render all the formats with the same, explicit level of TOC.
We move the assignment of _A2X_OPTS above the comment, because the
comment does not apply to the TOC setting. Also note that the previous
assignment to an empty value was completely superfluous...
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
- disable new v4l2-tracer option (fails to compile with rpi4_64 configuration)
For details see [1].
[1] https://git.linuxtv.org/v4l-utils.git/tree/ChangeLog?h=stable-1.24
Signed-off-by: Peter Seiderer <ps.report@gmx.net>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
It has been years that we removed $(HOST_DIR)/usr. Because of this
symlink, however, there are still packages that install things in
$(HOST_DIR)/usr. Remove the symlink so those packages will start to
fail.
Note that there's now a check-package check to detect such incorrectly
instaled things, so this should only affect external packages.
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
See:
https://sourceware.org/glibc/wiki/Release/2.37https://lists.gnu.org/archive/html/info-gnu/2023-02/msg00000.html
Security related changes:
CVE-2022-39046: When the syslog function is passed a crafted input
string larger than 1024 bytes, it reads uninitialized memory from the
heap and prints it to the target log file, potentially revealing a
portion of the contents of the heap
Remove patch 0001-Revert-Linux-Implement-a-useful-version-of-_startup_.patch fixed
in 2.37 release by [1].
[1] https://sourceware.org/bugzilla/show_bug.cgi?id=29249
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Sebastian Weyer <sebastian.weyer@smile.fr>
Tested-by: Bagas Sanjaya <bagasdotme@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
The shebang in check-package now defines python3. There is no longer a
need to maintain support with python 2.x.
See-also: 02b165dc71 (check-package: fix Python3 support)
Signed-off-by: Vincent Fazio <vfazio@gmail.com>
Tested-by: James Knight <james.d.knight@live.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Since kernel version 6.2, the minimum GNU Make version is 3.82 [1]. We
have an optional host-make 4.0 minimum dependency, so we can use it as
is. It's a bit unfortunate that we have to apply this even to older
kernel versions, but make itself builds fairly fast compared to the
kernel.
Use BR2_MAKE and BR2_MAKE1 for linux, and depend on
BR2_MAKE_HOST_DEPENDENCY. In addition, we need to set LINUX_MAKE to
BR2_MAKE for use in the kconfig infrastructure.
Signed-off-by: Oleg Lyovin <ovlevin@sberdevices.ru>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Since kernel version 6.2, the minimum GNU Make version is 3.82 [1]. We
have an optional host-make 4.0 minimum dependency, so we can use it as
is. It's a bit unfortunate that we have to apply this even to older
kernel versions, but make itself builds fairly fast compared to the
toolchain (which is the only reason to build linux-headers).
Use BR2_MAKE for linux-headers, and depend on BR2_MAKE_HOST_DEPENDENCY.
Signed-off-by: Oleg Lyovin <ovlevin@sberdevices.ru>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Excerpt of https://github.com/troglobit/mg/releases/tag/v3.6:
- Sync with mg upstream, OpenBSD 7.2, as of April 10 2023
- Drop trailing whitespace on RET (c-mode)
- Add zap-to-char and zap-up-to-char, binding the former to M-z
- Fix dobeep_msgs() usage, does not support format strings, this may
have caused unexpected crashes for some operations, e.g. goto line
that does not exist
- Fix memleaks and possible crashes in ctags support
- Fall back to /bin/sh if $SHELL is undefined
- Fix typos and duplicate undo in *quick* help buffer
- Add support for exuberant/universal Ctags tags file format
- Add support for M-, to pop-tag-mark, like GNU Emacs
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
This patch bumps version of the sunxi-tools to latest commit on master branch.
There are no new tags\releases since 2016 so let's use commit hash as version.
New version brings support for the following SOCs: A63, V5, R528/T113s, V853,
F1C100s, R329, H616, H6
Signed-off-by: Maksim Kiselev <bigunclemax@gmail.com>
[Arnout: use full hash in VERSION]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
When we updated .checkpackageignore in commit 29d6f319a, this was done
using a different version of shellcheck, that apparently detects some
additional issues. Thus,
board/terasic/de10nano_cyclone5/barebox-env/boot/mmc was detected as
containing shellcheck errors. However this doesn't happen with the
version in CI, leading to "Shellcheck was expected to fail" errors.
Re-generate .checkpackageignore under docker-run, like it should have
been done to begin with.
Fixes: 29d6f319a0
Fixes: https://gitlab.com/buildroot.org/buildroot/-/jobs/4084485150
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
This package is no longer used as python-pypa-build now uses
python-pyproject-hooks instead. In fact, pyproject-hooks is simply the
new name of pep517 [1].
[1] https://github.com/pypa/pyproject-hooks/issues/136
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
It's been ages (5 years at the next release) that we've not installed
host packages in $(HOST_DIR)/usr, but we still have a few packages that
reference it or install things in there. See [1]
Add a new check_function that warns when a file is added installing to
or referencing $(HOST_DIR)/usr .
[1] "d9ff62c4cd pacakge: drop remnants of $(HOST_DIR)/usr"
Cc: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
[Arnout: exclude skeleton.mk with disable comment instead of explicit
code]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
For years already, $(HOST_DIR)/usr is deprecated. Install directly in
$(HOST_DIR). This will trigger a check-package error in the following
commit.
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Teach check-package to detect python files by type and check them using
flake8.
Do not use subprocess to call 'python3 -m flake8' in order to avoid too
many spawned shells, which in its turn would slow down the check for
multiple files. (make check-package takes twice the time using a shell
for each flake8 call, when compared of importing the main application)
Expand the runtime test and the unit tests for check-package.
Remove check-flake8 from the makefile and also from the GitLab CI
because the exact same checks become part of check-package.
Suggested-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
[Arnout: add a comment to x-python to explain its purpose]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
The .mk files inside both support/dependencies and support/misc are not
package recipes, similar to package/pkg-*.mk. The check-package don't
apply to them. Therefore ignore such files.
In the test infra, some br2-externals are used as fixtures to provide
(sometimes) failure cases, so ignore files in these directories.
Files inside support/kconfig are files copied from linux upstream, so do
not generate warnings for them.
support/gnuconfig contains auto-generated config.{guess,sub} files,
so do not generate shellcheck warnings for them.
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
When a SysV init script is inside package/ it doesn't need to be
executable. However, when an init script is inside a fs_overlay, it
*does* need to be executable. Therefore, skip the NotExecutable test for
init scripts. We detect them based on the directory /etc/init.d
Signed-off-by: Ricardo Martincoski <ricardo.martincoski@gmail.com>
[Arnout: update .checkpackageignore]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Move RUSTFLAGS for host builds to HOST_PKG_CARGO_ENV so that they
are set for host cargo builds which don't use the cargo infrastructure,
e.g. python packages that include rust code.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Bump U-Boot to version 2023.04 and remove the patch that has already
been upstreamed.
Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
https://github.com/stephane/libmodbus/releases/tag/v3.1.10
Also update URL in Config.in to use https instead of http.
Also remove --without-documentation configure option, which already
didn't exist in 3.1.8.
Signed-off-by: Daniel Lang <dalang@gmx.at>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
mdadm contains various udev rules to manage RAID assembly and create
symlinks, install them if udev (or equivalent) is enabled.
Signed-off-by: Simon Rowe <simon.rowe@nutanix.com>
[Arnout: use MDADM_INSTALL_TARGET_OPTS for install-bin as well]
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
Fixes, from https://github.com/troglobit/libite/releases/tag/v2.5.3
- rsync() does not copy single files correctly
- copyfile() does not handle empty source files correctly (error)
- copyfile() does not properly create destination dir
Signed-off-by: Joachim Wiberg <troglobit@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
The version of U-Boot must be explicitly set because if you keep the
default setting, the bootloader is continuously updated in buildroot with
the risk of building a non-working image.
Signed-off-by: Dario Binacchi <dario.binacchi@amarulasolutions.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
The matio package exhibits gcc bug 99410 when built for the Nios2
architecture with optimization enabled, which causes a build failure.
As done for other packages in Buildroot work around this gcc bug by
setting optimization to -O0 if BR2_TOOLCHAIN_HAS_GCC_BUG_99410=y.
Fixes:
http://autobuild.buildroot.net/results/971/9714160b80a0d1a01983d29af38a95049c07b4d3/
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
go1.19.8 (released 2023-04-04) includes security fixes to the go/parser,
html/template, mime/multipart, net/http, and net/textproto packages, as well as
bug fixes to the compiler, the linker, the runtime, and the time package.
Fixes security vulnerabilities:
go/parser: infinite loop in parsing (CVE-2023-24537)
html/template: backticks not treated as string delimiters (CVE-2023-24538)
net/http, net/textproto: denial of service from excessive memory
allocation (CVE-2023-24534)
net/http, net/textproto, mime/multipart: denial of service from excessive
resource consumption (CVE-2023-24536)
https://go.dev/doc/devel/release#go1.19.8https://github.com/golang/go/issues?q=milestone%3AGo1.19.8+label%3ACherryPickApproved
Signed-off-by: Christian Stewart <christian@paral.in>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
