Changelog:
============
* Add NULL check to cJSON_SetValuestring()(CVE-2024-31755)
* Remove non-functional list handling of compiler flags
* Fix heap buffer overflow
* remove misused optimization flag -01
* Set free'd pointers to NULL whenever they are not reassigned immediately after
Signed-off-by: Wang Mingyu <wangmy@fujitsu.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
(From meta-openembedded rev: 535822eff7)
Signed-off-by: Peter Marko <peter.marko@siemens.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Location of the file that systemd uses to check whether to
start adbd or not has been updated from /var to /etc in
android-tools-adbd.service. This change changes the path
of creation of usb-debugging-enabled flag file in
android-tools recipes from /var/usb-debugging-enabled to
/etc/usb-debugging-enabled
Backport-of: 2a3d4be999 ("android-tools: create flag flag file for adbd at a proper location")
Fixes: a29c6386d5 ("android-toold-adbd: Fix inconsistency between selinux configurations")
Fixes: 8106cfe769 ("android-tools-adbd.service: Change /var to /etc in ConditionPathExists")
Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Raghuvarya S <quic_raghuvar@quicinc.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
To ensure android-tools-adbd.service starts at boot, the path
for ConditionPathExists must be present at build time. /etc is
more suitable for build-time files than /var, which is for
runtime files. Changed ConditionPathExists from
/var/usb-debugging-enabled to /etc/usb-debugging-enabled
Backport-of: 8106cfe769 ("android-tools-adbd.service: Change /var to /etc in ConditionPathExists")
CC: Khem Raj <raj.khem@gmail.com>
CC: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Signed-off-by: Raghuvarya S <quic_raghuvar@quicinc.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The expiration date of the "NameConstraints.*.cert" test certificate in
the nss package is Sep 4 2023 and causing a test failure.
This commit regenerate NameConstraints test certificates and changes the
validity period of test certs generated by `make-nc` from ~10 years to
~20 years.
regenerate_NameConstrain_test_certificates.tar.gz is a snapshot of certs
files based on the commit which update them. It fails to apply binary
commit, so create a tarball as part of SRC_URI rather than a .patch
file.
Upstream-Status: Backport [https://hg.mozilla.org/projects/nss/rev/1d565dc7e17dad6d2851b2d6ff522c5d6345ae26]
Signed-off-by: Haixiao Yan <haixiao.yan.cn@windriver.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
2848cc99a1 ("php-fpm: Add support for systemd") introduced a systemd
service file, where ExecStart and ExecStop uses /etc/init.d/php-fpm,
which does not exist if systemd is enabled. Consequently, the php-fpm
service fails to start even though it is correctly installed. This is
fixed by this commit in which the service file is identical to the one
from the PHP source code except for the use of BitBake variables. Also,
use ${systemd_system_unitdir} instead of ${systemd_unitdir}/system.
Signed-off-by: Emil Kronborg <emil.kronborg@protonmail.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
The first version of this patch introduced a problem with python3-pyqt5.
Python emitted the following error message when one attempted to import
PyQt5.Qt:
ImportError: dynamic module does not define module export function (PyInit_Qt)
This came about due to segfault in sip when executed in do_configure of
python3-pyqt5. This resulted in a zero-length sipQtcmodule.c file being
produced. This compiled successfully which meant no build failure was
observed.
The segfault was caused by a mistake in backporting the patch from SIP 6.
The generateCompositeCpp() function uses the generate_include_sip_h()
helper function in later versions which doesn't exist in SIP 4.
We must replace the first parameter passed to isPY_SSIZE_T_CLEAN() from
mod to pt->module to account for this. The change is not necessary for
generateInternalAPIHeader()
To simplify the patch we can remove the generated lexer and parser files
and run flex and bison in do_configure instead.
Signed-off-by: Rob Woolley <rob.woolley@windriver.com>
Tested-by: Toby Flynn <campingandskiing@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
A flaw was found in libssh. By utilizing the
ProxyCommand or ProxyJump feature, users can exploit
unchecked hostname syntax on the client. This issue
may allow an attacker to inject malicious code into
the command of the features mentioned through the
hostname parameter
Signed-off-by: Nikhil R <nikhil.r@kpit.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
There is a heap buffer overflow in DumpScreen2RGB() in gif2rgb.c. This
occurs when a crafted gif file, where size of color table is < 256 but
image data contains pixels with color code highier than size of color
table. This causes oferflow of ColorMap->Colors array.
Fix the issue by checking if value of each pixel is within bounds of
given color table. If the value is out of color table, print error
message and exit.
Signed-off-by: Nikhil R <nikhil.r@kpit.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
bat-format-pretty hardcoded the lib folder that cause it reports
missing formatter.bash error when multilib is enabled.
Signed-off-by: Xiangyu Chen <xiangyu.chen@windriver.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
CVE fixed:
- CVE-2023-40359 xterm: ReGIS reporting for character-set names containing characters other than alphanumerics or underscore
Upstream-Status: Backport from https://github.com/ThomasDickey/xterm-snapshots/commit/41ba5cf31da5e43477811b28009d64d3f643fd29
Note: The CVE patch is part of minor version-up and is extracted from the snapshot of xterm-379c.
Documentation of the commit shows 2 different overflows being fixed and hence the fix was extracted from the commit.
Signed-off-by: Rohini Sangam <rsangam@mvista.com>
Signed-off-by: Siddharth Doshi <sdoshi@mvista.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
An out-of-bounds stack write flaw was found in unixODBC on 64-bit
architectures where the caller has 4 bytes and callee writes 8 bytes.
This issue may go unnoticed on little-endian architectures, while
big-endian architectures can be broken.
References:
https://nvd.nist.gov/vuln/detail/CVE-2024-1013
Signed-off-by: Soumya Sambu <soumya.sambu@windriver.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Download and unpack contrib modules and other repositories in S instead
of WORKDIR so they don't escape file-prefix-map substitutions.
Fixes a number of reproducibility problems because of OpenCV check
macros that were embedding path to files in contrib/.
MJ: this is backport from langdale and fixes not only reproducibility
issues but also pseudo aborts in do_install when do_install
is executed again after do_package (in incremental builds):
| DEBUG: Executing shell function do_install
| NOTE: DESTDIR=opencv/4.5.5-r0/image VERBOSE=1 cmake --build opencv/4.5.5-r0/build --target install --
| abort()ing pseudo client by server request. See https://wiki.yoctoproject.org/wiki/Pseudo_Abort for more details on this.
| Check logfile: opencv/4.5.5-r0/pseudo//pseudo.log
| Subprocess aborted
and pseudo.log file shows:
path mismatch [3 links]: ino 214373575 db 'opencv/4.5.5-r0/package/usr/src/debug/lib32-opencv/4.5.5-r0/contrib/modules/intensity_transform/src/bimef.cpp' req 'opencv/4.5.5-r0/contrib/modules/intensity_transform/src/bimef.cpp'.
easily reproducible with:
bitbake -c cleansstate opencv; bitbake -c package opencv; bitbake -c install -f opencv
unlike ${S} ${WORKDIR}/contrib isn't in default PSEUDO_IGNORE_PATHS
Signed-off-by: Anuj Mittal <anuj.mittal@intel.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Martin Jansa <martin.jansa@gmail.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Graphviz 2.36 before 10.0.0 has an out-of-bounds read via a crafted config6a file.
NOTE: exploitability may be uncommon because this file is typically owned by root.
Signed-off-by: Meenali Gupta <meenali.gupta@windriver.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Vulnerability in the MySQL Server product of Oracle MySQL
(component: InnoDB). Supported versions that are affected
are 5.7.43 and prior, 8.0.34 and prior and 8.1.0. Easily
exploitable vulnerability allows high privileged attacker
with network access via multiple protocols to compromise
MySQL Server. Successful attacks of this vulnerability can
result in unauthorized ability to cause a hang or frequently
repeatable crash (complete DOS) of MySQL Server. CVSS 3.1
Base Score 4.9 (Availability impacts).
CVSS Vector: (CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:N/I:N/A:H).
References:
https://nvd.nist.gov/vuln/detail/CVE-2023-22084https://security-tracker.debian.org/tracker/CVE-2023-22084
Signed-off-by: Yogita Urade <yogita.urade@windriver.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
with make 4.4, linuxptp do_compile will failed with error:
In file included from clock.c:35:
missing.h:61:9: error: redeclaration of enumerator 'HWTSTAMP_TX_ONESTEP_P2P'
61 | HWTSTAMP_TX_ONESTEP_P2P = 3,
| ^~~~~~~~~~~~~~~~~~~~~~~
In file included from clock.c:21:
/buildarea2/WRLCD_Regression/Rerun/build_dir/11201532-build_scp_world_Feature_Test/qemux86-64-standard-std-OE/build/tmp-glibc/work/core2-64-wrs-linux/linuxptp/3.1.1-r0/recipe-sysroot/usr/include/linux/net_tstamp.h:128:9: note: previous definition of 'HWTSTAMP_TX_ONESTEP_P2P' with type 'enum hwtstamp_tx_types'
128 | HWTSTAMP_TX_ONESTEP_P2P,
|
Following change of make 4.4 changes behavior of shell function:
* WARNING: Backward-incompatibility!
Previously makefile variables marked as export were not exported to commands
started by the $(shell ...) function. Now, all exported variables are
exported to $(shell ...).
Signed-off-by: Changqing Li <changqing.li@windriver.com>
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Backport of commit 05c1003c4 ("linuxptp: fix do_compile error").
This is present in dunfell/kirkstone as well. If net_tstamp.h of the
build host disagrees with net_tstamp.h of the OE kernel or I remove
the build host's net_tstamp.h do_compile fails.
Changed Upstream Status to Backport with the git sha as the commit is
now applied upstream.
Signed-off-by: Max Krummenacher <max.krummenacher@toradex.com>
Signed-off-by: Armin Kuster <akuster808@gmail.com>
old tarballs disappear from main location, use a backup location to
fetch it in such cases.
Signed-off-by: Khem Raj <raj.khem@gmail.com>
Signed-off-by: Joao Marcos Costa <joaomarcos.costa@bootlin.com>
Backported from Nanbield
(cherry-picked from commit 8d34444c74)
Signed-off-by: Armin Kuster <akuster808@gmail.com>
Guocai He
Mingli Yu
Peter Marko
Dmitry Baryshkov
Raghuvarya S
Peter Kjellerstedt
Niko Mauno
Vijay Anusuri
Hitendra Prajapati
Haixiao Yan
Wentao Zhang
Emil Kronborg
Wang Mingyu
Martin Jansa
Soumya Sambu
Jiaqing Zhao
nikhil
Siddharth Doshi
Rob Woolley
Archana Polampalli
Priyal Doshi
Vivek Kumbhar
akash hadke
Xiangyu Chen
Rohini Sangam
Yogita Urade
Anuj Mittal
Meenali Gupta
virendra thakur
Changqing Li
Joao Marcos Costa