See the release notes of intermediate versions:
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.28.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.29.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.30.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.30.1
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.30.2
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.31.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.32.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.33.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.33.1
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.34.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.35.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.36.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.37.0
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.37.1
- https://github.com/yhirose/cpp-httplib/releases/tag/v0.37.2
This fixes numerous vulnerabilities:
- CVE-2026-21428:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to version 0.30.0, the ``write_headers``
function does not check for CR & LF characters in user supplied
headers, allowing untrusted header value to escape header lines. This
vulnerability allows attackers to add extra headers, modify request
body unexpectedly & trigger an SSRF attack. When combined with a
server that supports http1.1 pipelining (springboot, python twisted
etc), this can be used for server side request forgery (SSRF). Version
0.30.0 fixes this issue.
https://www.cve.org/CVERecord?id=CVE-2026-21428
- CVE-2026-22776:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to version 0.30.1, a Denial of Service (DoS)
vulnerability exists in cpp-httplib due to the unsafe handling of
compressed HTTP request bodies (Content-Encoding: gzip, br, etc.). The
library validates the payload_max_length against the compressed data
size received from the network, but does not limit the size of the
decompressed data stored in memory.
https://www.cve.org/CVERecord?id=CVE-2026-22776
- CVE-2026-28434:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to 0.35.0, when a request handler throws a
C++ exception and the application has not registered a custom
exception handler via set_exception_handler(), the library catches the
exception and writes its message directly into the HTTP response as a
header named EXCEPTION_WHAT. This header is sent to whoever made the
request, with no authentication check and no special configuration
required to trigger it. The behavior is on by default. A developer who
does not know to opt in to set_exception_handler() will ship a server
that leaks internal exception messages to any client. This
vulnerability is fixed in 0.35.0.
https://www.cve.org/CVERecord?id=CVE-2026-28434
- CVE-2026-28435:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to 0.35.0, cpp-httplib (httplib.h) does not
enforce Server::set_payload_max_length() on the decompressed request
body when using HandlerWithContentReader (streaming ContentReader)
with Content-Encoding: gzip (or other supported encodings). A small
compressed payload can expand beyond the configured payload limit and
be processed by the application, enabling a payload size limit bypass
and potential denial of service (CPU/memory exhaustion). This
vulnerability is fixed in 0.35.0.
https://www.cve.org/CVERecord?id=CVE-2026-28435
- CVE-2026-29076:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to version 0.37.0, cpp-httplib uses
std::regex (libstdc++) to parse RFC 5987 encoded filename* values in
multipart Content-Disposition headers. The regex engine in libstdc++
implements backtracking via deep recursion, consuming one stack frame
per input character. An attacker can send a single HTTP POST request
with a crafted filename* parameter that causes uncontrolled stack
growth, resulting in a stack overflow (SIGSEGV) that crashes the
server process. This issue has been patched in version 0.37.0.
https://www.cve.org/CVERecord?id=CVE-2026-29076
- CVE-2026-31870:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to 0.37.1, when a cpp-httplib client uses
the streaming API (httplib::stream::Get, httplib::stream::Post, etc.),
the library calls std::stoull() directly on the Content-Length header
value received from the server with no input validation and no
exception handling. std::stoull throws std::invalid_argument for non-
numeric strings and std::out_of_range for values exceeding ULLONG_MAX.
Since nothing catches these exceptions, the C++ runtime calls
std::terminate(), which kills the process with SIGABRT. Any server the
client connects to — including servers reached via HTTP redirects,
third-party APIs, or man-in-the-middle positions can crash the client
application with a single HTTP response. No authentication is
required. No interaction from the end user is required. The crash is
deterministic and immediate. This vulnerability is fixed in 0.37.1.
https://www.cve.org/CVERecord?id=CVE-2026-31870
- CVE-2026-32627:
cpp-httplib is a C++11 single-file header-only cross platform
HTTP/HTTPS library. Prior to 0.37.2, when a cpp-httplib client is
configured with a proxy and set_follow_location(true), any HTTPS
redirect it follows will have TLS certificate and hostname
verification silently disabled on the new connection. The client will
accept any certificate presented by the redirect target — expired,
self-signed, or forged — without raising an error or notifying the
application. A network attacker in a position to return a redirect
response can fully intercept the follow-up HTTPS connection, including
any credentials or session tokens in flight. This vulnerability is
fixed in 0.37.2.
https://www.cve.org/CVERecord?id=CVE-2026-32627
Signed-off-by: Titouan Christophe <titouan.christophe@mind.be>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 8dad17ea06)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Require host-pkgconf only for libtirpc is not enough because libmnl
requires as well.
Upstream added it since 20200515 commit
553ca8ea3b ("net/route: Add netlink based route change tests")
using it in m4/ltp-libmnl.m4.
Later, in 20210524 in commit
6e17e2ba13 ("configure: Improve error message on missing pkg-config")
configure.ac required it explicitly:
m4_ifndef([PKG_CHECK_EXISTS],
[m4_fatal([must install pkg-config or pkgconfig and pkg.m4 macro (usual dependency), see INSTALL])])
This fixes error:
checking pkg-config is at least version 0.9.0... ./configure: line 7419: br-mips64r6-el-hf-glibc/host/bin/pkg-config: No such file or directory
Signed-off-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit d421a5d278)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Fixes the following vulnerability:
- CVE-2026-27135:
nghttp2 is an implementation of the Hypertext Transfer Protocol
version 2 in C. Prior to version 1.68.1, the nghttp2 library stops
reading the incoming data when user facing public API
`nghttp2_session_terminate_session` or
`nghttp2_session_terminate_session2` is called by the application.
They might be called internally by the library when it detects the
situation that is subject to connection error. Due to the missing
internal state validation, the library keeps reading the rest of the
data after one of those APIs is called. Then receiving a malformed
frame that causes FRAME_SIZE_ERROR causes assertion failure. nghttp2
v1.68.1 adds missing state validation to avoid assertion failure. No
known workarounds are available.
For more information, see:
- https://www.cve.org/CVERecord?id=CVE-2026-27135
- 5c7df8fa81
(cherry picked from commit 7d26ff6c14)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Add a basic runtime test for memcached. The test starts memcached and
checks if it responds to a basic set/get request.
Signed-off-by: Franciszek Stachura <fbstachura@gmail.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit aa2d71ac38)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Fixes the following security issue:
CVE-2026-34155: Improper Signing of Plain Bundles Exceeding 2 GiB
RAUC bundles using the 'plain' format exceeding a payload size of 2 GiB
cause an integer overflow which results in a signature which covers only the
first few bytes of the payload. Given such a bundle with a legitimate
signature, an attacker can modify the part of the payload which is not
covered by the signature.
Bundles using the recommended 'verity' or 'crypt' formats are not affected.
For more details, see the advisory:
https://github.com/rauc/rauc/security/advisories/GHSA-6hj7-q844-m2hxhttps://github.com/rauc/rauc/releases/tag/v1.15.2
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 6e4a136363)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The edk2 package fails to build with gcc 15, with the following error:
EfiRom.c:78:17: error: assignment discards 'const' qualifier from pointer target type [-Werror=discarded-qualifiers]
Cherry-pick one patch from edk2-stable202602 to fix this.
This can be verified with the tests.boot.test_edk2 python tests.
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Dick Olsson <hi@senzilla.io>
[Julien: add "host" gcc 15 and glibc 2.43 in commit title]
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit d18f7e0916)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Config.in selects libiconv in certain situations (uClibc with locales
not enabled). This may lead to a build error with utils/test-pkg -p mpd.
Add missing optional dependency and restore sorting order.
Signed-off-by: Andreas Ziegler <br025@umbiko.net>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 83e637d7cc)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
The documentation feature was introduced with version 0.22, but is not
used. Instead two other options (also introduced with 0.22) control
building HTML manual and man pages. If the tools needed are not found,
the documentation build finishes silently.
Save Meson some effort by setting both options to 'false'.
Signed-off-by: Andreas Ziegler <br025@umbiko.net>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit e7e2050d54)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Fix a silent build issue with the host-bootgen package if host-flex is already
built. The host-flex package will install a version of the flexlexer.h header
file that is not compatible with the version embedded in bootgen.
While the build will still 'succeed', the binary will not be correct. This
can be seen by just running the binary that is built. By default, running
bootgen without any parameters should print the help menu, but if the wrong
version of flexlexer.h is used, it will print the following instead:
$ output/host/bin/bootgen
****** Bootgen v2025.1-Merged
**** Build date : Mar 18 2026-07:08:01
** Copyright 1986-2022 Xilinx, Inc. All Rights Reserved.
** Copyright 2022-2025 Advanced Micro Devices, Inc. All Rights Reserved.
ERROR: syntax error
-h
Add a patch to the bootgen package that corrects the issue.
Upstream: CR to AMD jira
Signed-off-by: Neal Frager <neal.frager@amd.com>
Tested-by: Steven J. Hill <steven.hill@collins.com>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 75f62dc669)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Backport a patch from Xen 4.21 to fix the following build errors for Arm v7
with gcc-15:
xen-4.14.6/xen/include/asm/platforms/midway.h:1: error: header guard '__ASM_ARM_PLATFORMS_MIDWAY_H' followed by '#define' of a different macro [-Werror=header-guard]
xen-4.14.6/xen/include/asm/platforms/omap5.h:1: error: header guard '__ASM_ARM_PLATFORMS_OMAP5_H' followed by '#define' of a different macro [-Werror=header-guard]
Since the external 32b arm toolchain has been updated to a version based on
gcc-15, the tests.package.test_xen.TestXenArmv7 python test does not build
anymore.
Adding the patch repairs it.
Link: https://gitlab.com/buildroot.org/buildroot/-/jobs/13518318473
Fixes: 86d453a7dc ("toolchain/toolchain-external/toolchain-external-arm-arm: bump to 15.2.rel1")
Signed-off-by: Vincent Stehlé <vincent.stehle@arm.com>
Cc: Alistair Francis <alistair@alistair23.me>
Cc: Dowan Gullient <dowan.gullient@smile.fr>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 0e8739fab2)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
This feature was made optional in mesa3d 25.2.0 as it is deprecated,
however some packages still require it so lets add a new config
option that those packages can select until they no longer require
this feature.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Reviewed-by: Adrian Perez de Castro <aperez@igalia.com>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
(cherry picked from commit 0260741639)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Add 'source' attribute to each CVE in vulnerabilities node, including NVD
URL reference to enable proper import into Dependency-Track.
Dependency-Track's VEX importer requires the source attribute to
properly process vulnerability entries. Without it, vulnerabilities are
skipped during import with "does not have an ID and / or source" warnings.
Include the full NVD URL following the CycloneDX 1.6 documentation format:
https://nvd.nist.gov/vuln/detail/{CVE-ID}
Test Environment:
- Buildroot: 2025.02.11 (or master)
- Dependency-Track: v4.13.6
Test Results - BEFORE (without source attribute):
apiserver_1 | 2026-02-23 16:05:40,890 INFO [VexUploadProcessingTask] Processing CycloneDX VEX uploaded to project: e43fe185-c0a3-4e3a-a908-667344a66a9c
apiserver_1 | 2026-02-23 16:05:40,941 WARN [CycloneDXVexImporter] VEX vulnerability at position #0 does not have an ID and / or source; Skipping it
apiserver_1 | 2026-02-23 16:05:40,941 WARN [CycloneDXVexImporter] VEX vulnerability at position #1 does not have an ID and / or source; Skipping it
...
apiserver_1 | 2026-02-23 16:05:40,941 WARN [CycloneDXVexImporter] VEX vulnerability at position #19 does not have an ID and / or source; Skipping it
apiserver_1 | 2026-02-23 16:05:40,941 INFO [CycloneDXVexImporter] The uploaded VEX does not contain any applicable vulnerabilities; Skipping VEX import
Test Results - AFTER (with source):
apiserver_1 | 2026-02-23 16:17:13,492 INFO [VexUploadProcessingTask] Processing CycloneDX VEX uploaded to project: e43fe185-c0a3-4e3a-a908-667344a66a9c
apiserver_1 | 2026-02-23 16:17:14,054 INFO [VexUploadProcessingTask] Completed processing of CycloneDX VEX for project: e43fe185-c0a3-4e3a-a908-667344a66a9c
CVEs are correctly imported in Dependency-Track
Signed-off-by: Fabien Lehoussel <fabien.lehoussel@smile.fr>
Acked-By: Thomas Perale <thomas.perale@mind.be>
Signed-off-by: Romain Naour <romain.naour@smile.fr>
(cherry picked from commit 3a5e70d1d1)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
Following changes made it into the release:
Joris van Rantwijk (1):
Fix adjtimex() with TIME64
Waldemar Brodkorb (8):
sys/stat.h: remove _STAT_VER/_MKNOD_VER
fix gettid() declaration
add statx syscall wrapper
sparc: add optimize build support for leon3
sparc: sync with Linux kernel definition, fixes gdb compile
fix compile error on earlier Kernels predating statx
select: fix compilation failure with very old kernel
bump version for 1.0.57 release
Signed-off-by: Waldemar Brodkorb <wbx@openadk.org>
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit 84bc50c82d)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
uclibc 1.0.57 added a statx() syscall wrapper in upstream commit [1].
zfs fail to build with uclibc 1.0.57 (not yet in Buildroot), because:
1. uclibc <fcntl.h> internally includes <sys/stat.h>, and
2. a zfs test redefines a statx() wrapper with a slightly different
prototype.
In that case, zfs fails to compile with error:
tests/zfs-tests/cmd/statx.c:58:1: error: conflicting types for 'statx'; have 'int(int, const char *, int, unsigned int, void *)'
Issue has been reported upstream at [2].
This commit adds a package patch to fix that issue.
[1] d3a819aff2
[2] https://github.com/openzfs/zfs/pull/18316
Signed-off-by: Julien Olivain <ju.o@free.fr>
(cherry picked from commit b6884913ae)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
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>
(cherry picked from commit f9cdca48a5)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
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>
(cherry picked from commit 244e4283a9)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
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>
(cherry picked from commit 3a5e071e4f)
Signed-off-by: Thomas Perale <thomas.perale@mind.be>
