The host mkfs.xfs is needed to create XFS root filesystems
in Buildroot.
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Remove incorrect $BINARIES_DIR reference from the symbolic link creation to
make a relative path and not an absolute path.
Signed-off-by: Neal Frager <neal.frager@amd.com>
Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Backport upstream patches to fix a silent crash in U-Boot on
STM32MP135F-DK.
Fixes:
799d184e89
Signed-off-by: Thomas Richard <thomas.richard@bootlin.com>
[Romain: add Fixes link]
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Specifying a list of br2-external trees is poorly documented, and the
only example uses a colon to separate the br2-external paths.
Adding the support for colon-separated list is the biggest mistake that
was made when introducing support for multiple br2-external [0]. Indeed,
both space and colon can be used to separate entries in the list, and it
is also possible to mix the two. However, internally, the list is stored
as a space-separated list, and all the code will split on spaces.
Besides, all other lists in Buildroot are a space-separated:
BR2_ROOTFS_DEVICE_TABLE
BR2_ROOTFS_STATIC_DEVICE_TABLE
BR2_TARGET_TZ_ZONELIST
BR2_ROOTFS_USERS_TABLES
BR2_ROOTFS_OVERLAY
BR2_ROOTFS_PRE_BUILD_SCRIPT
BR2_ROOTFS_POST_BUILD_SCRIPT
BR2_ROOTFS_POST_FAKEROOT_SCRIPT
BR2_ROOTFS_POST_IMAGE_SCRIPT
...
So, using colons is odd.
The fact that BR2_EXTERNAL is passed on the command line rather than
being a Kconfig item is not a reason enough to justify that it be
colon-separated.
Change the documentation to only mention using a space-separated list.
Of course, for backward compatibility, we keep the code as-is to accept
a colon-separated list, but we just do not advertise it.
Note that keeping the split on colons means that colons are not accepted
in pathnames of br2-external trees; in practice, this is not a new
restriction, or one that could lift as usign colons in Makefiles are
problematic anyway.
[0] in 20cd497387 core: add support for multiple br2-external trees
Reported-by: Fiona Klute (WIWA) <fiona.klute@gmx.de>
Reported-by: Brandon Maier <Brandon.Maier@collins.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Cc: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Boring changes: either do what shellcheck suggested, or comment why we
don't want to fix the code.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
The trap was initially introduced in c5fa9308ea (core/br2-external:
properly report unexpected errors), in 2017, to catch all unexpected
errors, back when a single file was generated, and errors emitted to
stderr.
Since commit d027cd75d0 (core: generate all br2-external files in
one go), in 2019 the single output file 'ofile' is no longer created,
as multiple output files were then introduced, while messages for
*expected errors* were redirected to a Makefile variable assignment
emitted on stdout, at which point the script just exits (in error);
expected failures only occur in do_validate().
Unexpected errors can only occur on failure to create, or write to,
output files, either '.br2-external.mk' in do_validate() or do_mk(),
or any of the kconfig fragments in do_kconfig(). Cause for failure to
create those can only be a no-space-left-on-device condition, as they
are created in a directory that was just created by the script earlier
in main(), and thus has the necessary mode; failure to create that
directory is now caught explicitly.
A trap on ERR is not called when the shell exits explicitly with a call
to 'exit', thus, only failures to create or write to output file would
be caught. In that case, we are better off not trying to write to those
files anyway: failure to create the file would already be reported by
the shell on stderr, while disk-full would not allow to store the output
anyway...
In any case, the script exits in error, which is going to be caught by
the caller, which will terminate.
So, drop the trap altogether.
As a side effect, that squelches a shellcheck error.
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
That plugin can be used instead of the builtin support to persist the
mosquitto state across restarts.
Note that this plugin has a dependency, so we don't use the $(if)
one-liner.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Until version 2.0.x, support for using static password or ACL files, was
builtin to the broker. With version 2.1.x, two new plugins have been
introduced to replace the builtin support, which is now deprecated and
will get removed in the next version.
Add two new configuration options for those plugins.
We decided to do a single commit, rather than one per option, because
they are relatively tied together (ACL needs passwords, at least).
We also choose to make those options enabled by default, because the
traditional way to configure mosquitto is to use static files for
authentication and authorization, and the builtin support if now
deprecated in favour for the plugins.
The usual ifeq-else-endif conditional block is a bit verbose when just
setting an option ON or OFF, when no additional dependency is needed.
Instead, use the not-unusual $(if)-inline one-liner. For consistency,
switch the existing dynamic-security plugin to use that one-liner too.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
The mosquitto broker (not the library) can have listeners configured to
serve HTTP requests; it can optionally accept (some) HTTP API requests
on such listeners.
Add a new option to enable the availability of http_pi.
Note that we do not just depend on libmicrohttpd to be enabled, because
the HTTP API is a security boundary, and enabling it must be an explicit
decision.
Co-developped-by: Titouan Christophe <titouan.christophe@mind.be>
[yann.morin@orange.com: make it an explicit option]
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Mosquitto 2.1.x adds the possibility to use a builtin websocket
implementation, as an alternative to using libwebsockets.
When using libwebsockets as the implementation, only the broker supports
websockets, and CLI tools do not; only when using the builtin one are
websockets usable with CLI tools (and the broker, of course).
Add a choice to select what type of websockets support to enable, if
any. Since the builtin implementation is still new, we keep the
libwebsockets one available.
Since this inverts the dependency logic to libwebsockets, we can't
provide a backward compatibility with existing (def)config files.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
mosquitto_{ctrl,db_dump,passwd,signal} can be handy to interact with
the mosquitto broker during development, but are usually unnecessary
on the target.
Add an option to enable or disable them. Make that new option enabled
by default when the broker is enabled, to keep backward compatibility
with previous (def)config files.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Fiona Klute <fiona.klute@gmx.de>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
mosquitto_pub/sub/rr can be handy to test a broker from the command
line, but they can get superfluous when only the broker is required
on the target.
Add an option to enable or disable them. Make that new option enabled
by default to keep backward compatibility with previous (def)config
files.
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Version 2.1.x has introduced a cmake-based build, and upstream strongly
recomends using it over the legacy Makefiles, which will ultimately be
retired.
So we do the switch, which cause quite some noise in the .mk file, but
at the same time allows for a bit of cleanup in the build process, as we
can now use the cmake-package infra.
Mosquitto now wants to peek into the malloc() internals for memory
tracking, and that only works on systems with an MMU (uClibc-ng does not
expose it for noMMU builds, as it's part of its malloc-standard
implementation).
Static-only builds are broken, even when only building the library. This
seems beyond a simple repair, so just require shared libs now (since
we're requiring an MMU as well, requiring shared libs is not too much of
an additional burden).
cJSON is now a required dependency, used in common parts of the code
(not just for the plugins).
There are a few options that we forcibly disable; they'll get addressed
in followup patches.
There are still a few build failures that are difficult to account for
(except):
$ printf 'BR2_PACKAGE_MOSQUITTO=y\n' >mosq.cfg
$ ./utils/docker-run ./utils/test-pkg -d $(pwd)/run-tests -c mosq.cfg -p mosquitto
br-arm-full-static [5/6]: FAILED
=> old uClibc-ng, would need __GNU_SOURCE (with dunder) to define
getrandom(); no longer needed since uClibc-ng 1.0.50; would need
openssl otherwise
bootlin-aarch64-glibc-old [6/6]: FAILED
=> really old glibc, missing getrandom(); would need openssl
Drop our existing patches, they've either been applied upstream, or are
no longer needed. Add new patches to fix various build issues
(submission upstream pending the signature of the CLA..,).
Thanks a lot to Titouan for providing his initial work on the update!
Some of his findings ended up in this patch. 👍
Note: by lack of a trusted path back to the PGP key that signed the
archive, the comment was dropped, as checking a signature without a
trust-chain does not make much sense...
Co-developped-by: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Yann E. MORIN <yann.morin@orange.com>
Cc: Peter Korsgaard <peter@korsgaard.com>
Cc: Titouan Christophe <titouan.christophe@mind.be>
[Romain: remove "mosquitto broker" comment for static builds]
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Updating the hash of the WHENCE file, due to firmware additions and
firmware changes, but no changes to the redistribution/licensing
conditions.
Signed-off-by: Bernd Kuhls <bernd@kuhls.net>
Signed-off-by: Julien Olivain <ju.o@free.fr>
https://curl.se/ch/8.19.0.htmlhttps://curl.se/docs/security.html
Fixes the following CVEs:
CVE-2026-3805: use after free in SMB connection reuse
CVE-2026-3784: wrong proxy connection reuse with credentials
CVE-2026-3783: token leak with redirect and netrc
CVE-2026-1965: bad reuse of HTTP Negotiate connection
Switch to sha256 tarball hash provided by upstream.
Updated license hash due to copyright year bump:
e83c82f05f
Signed-off-by: Bernd Kuhls <bernd@kuhls.net>
[Julien: add back pgp signature info in hash file]
Signed-off-by: Julien Olivain <ju.o@free.fr>
Building Go 1.26 and later requires Go 1.24.6 or later for bootstrap.
To support this we use Go version 1.25.8 as the version for
go-bootstrap-stage5 and have the build for Go 1.26.1 depend on
go-bootstrap-stage5.
Go version 1.25.8 is the latest Go version we can build using
go-bootstrap-stage4.
The package build for go-bootstrap-stage5 is effectively identical to
go-bootstrap-stage4 with only the Go version and stage number changed.
Go 1.28 is expected to require a minor release of Go 1.26 for bootstrap.
Fixes the following security vulnerabilities:
- CVE-2026-25679: net/url: reject IPv6 literal not at start of host
- CVE-2026-27142: html/template: URLs in meta attribute actions not escaped
- CVE-2026-27137: crypto/x509: incorrect enforcement of email constraints
- CVE-2026-27138: crypto/x509: panic in name constraint checking: certificates
- CVE-2026-27139: os: FileInfo can escape from a Root
For full release notes, see:
https://go.dev/doc/devel/release#go1.26.0
Signed-off-by: Christian Stewart <christian@aperture.us>
Signed-off-by: Julien Olivain <ju.o@free.fr>
This commit switches the aarch64_efi_defconfig to an external glibc
stable Bootlin toolchain, to follow recommendation from [1].
Since an external toolchain is used, the linux-headers.hash custom
hash file is no longer needed and is removed.
[1] https://elinux.org/Buildroot:DeveloperDaysELCE2024#Rules_for_defconfigs
Signed-off-by: Julien Olivain <ju.o@free.fr>
This commit:
- updates the Kernel from 6.10 to 6.18.16
- updates U-Boot from 2024.07 to 2026.01
- updates arm-trusted-firmware from v2.11 to v2.12 LTS
- switches to Bootlin external glibc stable toolchain
Replace OP-TEE TEE binary file tee-pager_v2.bin by tee-raw.bin using
the new option BR2_TARGET_UBOOT_NEEDS_OPTEE_TEE_RAW_BIN.
Remove unsued linux-headers.hash since we now use a prebuilt toolchain.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Replace OP-TEE TEE binary file tee-pager_v2.bin by tee-raw.bin using
the new option BR2_TARGET_UBOOT_NEEDS_OPTEE_TEE_RAW_BIN.
Remove unsued linux-headers.hash since we use a prebuilt toolchain.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
There was a previous attempt in commit [1] to bump the kernel to 6.18
but the SD card failed at boot. It seems the actual issue is related to
the "cheap" (or fake?) SD card with SD UHS SDR50 speed.
This offending SD card come from a lot recently purchased in order to
fix my “Too many boards, not enough SD cards” issue.
[1] 4abb8a98b2
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
The commit [1] update arm-trusted-firmware to v2.12 LTS without
removing the custom file hash.
While at it, remove unused linux-headers.hash since we already use a
prebuilt toolchain.
[1] 4abb8a98b2
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
This defconfig use BR2_TARGET_UBOOT_CUSTOM_MAKEOPTS to provide the path
to the TEE in raw binary format but use the legacy file name
"tee-pager_v2.bin" rather than the recommended "tee-raw.bin" [1][2].
Instead of just replacing the file name, use the newly introduced
BR2_TARGET_UBOOT_NEEDS_OPTEE_TEE_RAW_BIN option. Since this option
needs BR2_TARGET_UBOOT_NEEDS_OPTEE_TEE, we now have an explicit
dependency between u-boot and optee-os package.
Previously we had an indirect dependency: optee-os <- ATF <- u-boot
with both BR2_TARGET_ARM_TRUSTED_FIRMWARE_BL32_OPTEE and
BR2_TARGET_UBOOT_NEEDS_ATF_BL31 options enabled at the same time.
[1] 376cb124dd
[2] 4e0b8238ee
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
OP-TEE generates a few binaries that can be used by u-boot on some
platforms (Rockchip, Texas Instruments K3) using TEE variable.
Add a link to the OP-TEE documentation were we can find easily the
desciption for each TEE format that can be used by u-boot.
For convenience, copy tee.elf and tee.bin description in each option
choice help text.
Signed-off-by: Romain Naour <romain.naour@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
Remove myself from packages that I'm no longer personally interested in.
Signed-off-by: Marcus Hoffmann <buildroot@bubu1.eu>
Signed-off-by: Julien Olivain <ju.o@free.fr>
