tools needs C++ since the addition of the package in commit
27ad470d7d resulting in the following
build failure:
no -DHAVE_CONFIG_H -I. -I.. -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -I../include -I../master -Wall -DREV=`if test -s ../revision; then cat ../revision; else hg id -i .. 2>/dev/null || echo "unknown"; fi` -fno-strict-aliasing -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -Ofast -g0 -c -o ethercat-Command.o `test -f 'Command.cpp' || echo './'`Command.cpp
/bin/bash: line 1: no: command not found
Fixes:
- http://autobuild.buildroot.org/results/89d096006839f32a3d03786e69e51ec3c5ea70f6
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr: move it before package's options]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 014ebc394d)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix CVE-2022-2652: Depending on the way the format strings in the card
label are crafted it's possible to leak kernel stack memory. There is
also the possibility for DoS due to the v4l2loopback kernel module
crashing when providing the card label on request (reproduce e.g. with
many %s modifiers in a row).
https://github.com/umlaeute/v4l2loopback/blob/v0.12.7/ChangeLog
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 922fb6ac85)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix CVE-2021-46784: In Squid 3.x through 3.5.28, 4.x through 4.17, and
5.x before 5.6, due to improper buffer management, a Denial of Service
can occur when processing long Gopher server responses.
https://github.com/squid-cache/squid/security/advisories/GHSA-f5cp-6rh3-284w
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit d3ef301f0c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix CVE-2021-46828: In libtirpc before 1.3.3rc1, remote attackers could
exhaust the file descriptors of a process that uses libtirpc because
idle TCP connections are mishandled. This can, in turn, lead to an
svc_run infinite loop without accepting new connections.
https://sourceforge.net/projects/libtirpc/files/libtirpc/1.3.3/Release-1.3.3.txt/download
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 408888a29b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Fix CVE-2022-29154: An issue was discovered in rsync before 3.2.5 that
allows malicious remote servers to write arbitrary files inside the
directories of connecting peers. The server chooses which
files/directories are sent to the client. However, the rsync client
performs insufficient validation of file names. A malicious rsync
server (or Man-in-The-Middle attacker) can overwrite arbitrary files
in the rsync client target directory and subdirectories (for example,
overwrite the .ssh/authorized_keys file).
- Drop patches (already in version)
- Update hash of COPYING (make openssl license exception clearer by
having it at the top and use modern links in COPYING:
dde4695136)
https://github.com/WayneD/rsync/blob/v3.2.5/NEWS.md
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit ae2807821d)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The upstream commit 7a01882eb37e7504e2450f474d0cc8db60ed26c2
("common: Kconfig.boot: Add FIT_PRINT config option") introduce
CONFIG_FIT_PRINT and make fit_print_contents() empty if it was
not enabled.
Adding CONFIG_FIT_PRINT=y to UBOOT_TOOLS_MAKE_OPTS does not help
while CONFIG_FIT_PRINT=y affects Makefiles only, not C sources.
Add "#define CONFIG_FIT_PRINT 1" to autoconf.h if FIT_SUPPORT enabled.
It would be better to convert uboot-tools to kconfig infrastructure so
we can use KCONFIG_ENABLE_OPT etc. However, that's a much bigger change
and not suitable for backporting to stable branches. Therefore, for now,
take the simple approach of updating autoconf.h.
Signed-off-by: Atsushi Nemoto <atsushi.nemoto@sord.co.jp>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 2ebf652589)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Add a custom case to make sure that a random configuration with an empty
version for aufs-util doesn't fail.
Fixes:
- http://autobuild.buildroot.org/results/e242cf66a02983bcf6e95b37f8e458bd18aee683
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit fee46b54e7)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
When enable DM for SPL binary, the DTB part of SPL may not 4 bytes aligned.
If u-boot-spl is not aligned, the offset of the DDR firmware is not 4
byte aligned when u-boot-spl-ddr.bin is created. This causes the ddr
firmware to not be loaded correctly at boot.
See imx-mkimage commit
https://source.codeaurora.org/external/imx/imx-mkimage/commit/?id=bba038d893046b44683182dba540f104dab80fe7
for the imx-mkimage details.
Signed-off-by: Bram Vlerick <bram.vlerick@openpixelsystems.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 81aa9e7b8b)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
SIMD needs VSX with little endian to avoid the following build failure:
In file included from /nvmedata/autobuild/instance-12/output-1/build/jpeg-turbo-2.1.3/simd/powerpc/jccolor-altivec.c:25:
/nvmedata/autobuild/instance-12/output-1/build/jpeg-turbo-2.1.3/simd/powerpc/jccolext-altivec.c: In function 'jsimd_rgb_ycc_convert_altivec':
/nvmedata/autobuild/instance-12/output-1/build/jpeg-turbo-2.1.3/simd/powerpc/jsimd_altivec.h:93:26: warning: implicit declaration of function 'vec_vsx_ld'; did you mean 'vec_vsl'? [-Wimplicit-function-declaration]
93 | #define VEC_LD(a, b) vec_vsx_ld(a, b)
| ^~~~~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/be6d5ad0cee4ee19eb25e595d44555a1af6e073b
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle <arnout@mind.be>
(cherry picked from commit 701e6f34e0)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In case of an unexpected error, we currently only print the exception as
an str(). For example, the recent issue with the glibc version check
only reported:
TypeError: cannot use a string pattern on a bytes-like object
That does not help in fixing the issue; the exception text is also not
usually very user-friendly either anyway.
We change the reporting to print the traceback, which in the glibc
version check mentioned above, the error is reported as:
Traceback (most recent call last):
File "./utils/genrandconfig", line 740, in <module>
ret = gen_config(args)
File "./utils/genrandconfig", line 676, in gen_config
if not is_toolchain_usable(configfile, toolchainconfig):
File "./utils/genrandconfig", line 186, in is_toolchain_usable
if StrictVersion('2.14') > StrictVersion(glibc_version):
File "/usr/lib/python3.8/distutils/version.py", line 40, in __init__
self.parse(vstring)
File "/usr/lib/python3.8/distutils/version.py", line 135, in parse
match = self.version_re.match(vstring)
TypeError: cannot use a string pattern on a bytes-like object
With this, the error is much easier to pinpoint (it's the last one that
is not in a system module).
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit b6bfa3f744)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Unless explicitly told otherwise, subprocess.check_output() returns
bytes objects [0].
When we try to check the C library version (to check the Linaro
toolchain is usable), genrandconfig currently fails with:
TypeError: cannot use a string pattern on a bytes-like object
So, as suggested in the python documentation, decocde() the output of
subprocess.check_output() before we can use it.
[0] https://docs.python.org/3/library/subprocess.html#subprocess.check_output
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 12e4f7c5c4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Upstream dropped in:
aa6631a618,
which is present since webkitgtk 2.36.4.
Signed-off-by: Thomas Devoogdt <thomas.devoogdt@barco.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit a0690a8e24)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit fda53f0791 ("package/Makefile.in:
add detection for the lack of C library") added an $(error ...)
message when no C library is available for the currently selected
architecture.
However, this error message pops up not just when building, so for
example, the command:
make BR2_HAVE_DOT_CONFIG=y VARS=%_LICENSE printvars
no longer works (this command is used by the pkg-stats script).
We restore a functional behavior by doing the check only when
BR_BUILDING=y.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit d349d50dac)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 7652817c93 updated the
documentation but forgot to update support/dependencies
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit ba2659401f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following aarch64_be build failure probably raised since the
addition of the package:
/home/buildroot/autobuild/instance-3/output-1/host/opt/ext-toolchain/bin/../lib/gcc/aarch64_be-none-linux-gnu/10.3.1/../../../../aarch64_be-none-linux-gnu/bin/ld: ./.libs/libtesseract.so: undefined reference to `tesseract::IntSimdMatrix::intSimdMatrixNEON'
Fixes:
- http://autobuild.buildroot.org/results/b9246a37fcf6be4fabfc491daddadfb09e0a320a
Update the comment about _AUTORECONF=YES, list the two patches since
both touch configure.ac
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr: upsdate comment about _AUTORECONF=YES]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit f0a96f7364)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure raised since the addition of the package
in commit fbff7d7289 and
4b4551023e:
checking whether we need to link to -lstdc++fs for PEGTL explicitly... ERROR
configure: error: Link test failed both with and without -lstdc++fs; something is broken, please check file config.log for details.
Fixes:
- http://autobuild.buildroot.org/results/511c47802ce171caeeb9919371c58e6ad2d11a78
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 442cbf4691)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following build failure with BR2_PACKAGE_WOLFSSL_ALL raised
since bump to version 5.9.0 in commit
da912a3d2a:
In file included from ../../../../src/libstrongswan/utils/utils.h:59,
from ../../../../src/libstrongswan/library.h:101,
from wolfssl_common.h:29,
from wolfssl_aead.c:23:
wolfssl_aead.c:90:16: error: conflicting types for 'encrypt'; have '_Bool(union <anonymous>, chunk_t, chunk_t, chunk_t, chunk_t *)'
90 | METHOD(aead_t, encrypt, bool,
| ^~~~~~~
../../../../src/libstrongswan/utils/utils/object.h:99:20: note: in definition of macro 'METHOD'
99 | static ret name(union {iface *_public; this;} \
| ^~~~
In file included from /home/autobuild/autobuild/instance-5/output-1/host/powerpc64le-buildroot-linux-musl/sysroot/usr/include/wolfssl/wolfcrypt/wc_port.h:573,
from /home/autobuild/autobuild/instance-5/output-1/host/powerpc64le-buildroot-linux-musl/sysroot/usr/include/wolfssl/wolfcrypt/types.h:35,
from /home/autobuild/autobuild/instance-5/output-1/host/powerpc64le-buildroot-linux-musl/sysroot/usr/include/wolfssl/wolfcrypt/logging.h:33,
from /home/autobuild/autobuild/instance-5/output-1/host/powerpc64le-buildroot-linux-musl/sysroot/usr/include/wolfssl/ssl.h:35,
from wolfssl_common.h:64,
from wolfssl_aead.c:23:
/home/autobuild/autobuild/instance-5/output-1/host/powerpc64le-buildroot-linux-musl/sysroot/usr/include/unistd.h:149:6: note: previous declaration of 'encrypt' with type 'void(char *, int)'
149 | void encrypt(char *, int);
| ^~~~~~~
Fixes:
- http://autobuild.buildroot.org/results/02f080c2f6d8272cb8cc1de66e058d66fb7499bc
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 528155f23a)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In addition to --enable-video-opengles, SDL2 configure script also
looks at --enable-video-opengles1 and --enable-video-opengles2. Since
all OpenGL ES providers in Buildroot provide at least up to OpenGL ES
2, enable both options when BR2_PACKAGE_SDL2_OPENGLES=y.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr: split long lines]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit e48121750f)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
We extend the modules search path to be able to load the package
metadata. Currently, it is only restored when loading those
succeeded, not when it failed.
Restore it to its previous state also in case of error, to avoid
leaking the path further.
Signed-off-by: James Hilliard <james.hilliard1@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 69400611b2)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The CPIO filesystem generated by the test_python_botocore test is too
large, and doesn't fit as an initramfs in the 256MB of RAM available
in the versatilepb machine. This causes a "Initramfs unpacking failed:
write error" when booting, and many files being missing from the root
filesystem, ultimately causing the test to fail.
It would make sense to switch all test cases to use ext2 + a
hard-drive, but for now, let's fix the few test cases that are causing
problems.
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/2884635042
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr:
- drop superfluous# BR2_TARGET_ROOTFS_TAR is not set
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 0813ec1aa0)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The CPIO filesystem generated by the test_python_boto3 test is too
large, and doesn't fit as an initramfs in the 256MB of RAM available
in the versatilepb machine. This causes a "Initramfs unpacking failed:
write error" when booting, and many files being missing from the root
filesystem, ultimately causing the test to fail.
It would make sense to switch all test cases to use ext2 + a
hard-drive, but for now, let's fix the few test cases that are causing
problems.
Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/2884635041
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
[yann.morin.1998@free.fr:
- drop superfluous# BR2_TARGET_ROOTFS_TAR is not set
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit a9df206190)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The "find" and "xargs" commands, from the "findutils" package are used
during the build process. See for example [1].
Even if it's a quite common package which is almost sure to be present
on the host, it should be listed here. When writing new recipes, hooks
and scripts, it is generally safe and portable to restrict to the
host dependencies listed in those prerequisites.
This commit just add the missing "findutils" package in this list.
[1] https://git.buildroot.org/buildroot/tree/Makefile?h=2022.05.1#n737
Signed-off-by: Julien Olivain <ju.o@free.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 7652817c93)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
We are trying to not build the perf documentation. However, the hook
being used to do so was named incorrectly. As a result, the build steps
to disable the documentation were never executed.
Rename the hook from
LINUX_POST_PATCH_HOOKS
to
LINUX_TOOLS_POST_PATCH_HOOKS
to fix the issue.
Fixes: 20b1446669 ("linux/tools: make it a real, separate package")
Signed-off-by: Markus Mayer <mmayer@broadcom.com>
Tested-by: Florian Fainelli <f.fainelli@gmail.com>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 612ae4bd18)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
- Fix CVE-2022-1114: A heap-use-after-free flaw was found in
ImageMagick's RelinquishDCMInfo() function of dcm.c file. This
vulnerability is triggered when an attacker passes a specially crafted
DICOM image file to ImageMagick for conversion, potentially leading to
information disclosure and a denial of service.
- Fix CVE-2022-32545: A vulnerability was found in ImageMagick, causing
an outside the range of representable values of type 'unsigned char'
at coders/psd.c, when crafted or untrusted input is processed. This
leads to a negative impact to application availability or other
problems related to undefined behavior.
- Fix CVE-2022-32546: A vulnerability was found in ImageMagick, causing
an outside the range of representable values of type 'unsigned long'
at coders/pcl.c, when crafted or untrusted input is processed. This
leads to a negative impact to application availability or other
problems related to undefined behavior.
- Fix CVE-2022-32547: In ImageMagick, there is load of misaligned
address for type 'double', which requires 8 byte alignment and for
type 'float', which requires 4 byte alignment at
MagickCore/property.c. Whenever crafted or untrusted input is
processed by ImageMagick, this causes a negative impact to application
availability or other problems related to undefined behavior.
- Update hash of LICENSE (year updated with
80629dfb3f)
https://github.com/ImageMagick/Website/blob/main/ChangeLog.md
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 685100fe85)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
ualpn is not compatible with libressl as stated by upstream in
32546c7caa
resulting in the following build failure:
ualpn.c: In function 'ssl_client_hello_cb':
ualpn.c:2038:16: error: 'SSL_CLIENT_HELLO_RETRY' undeclared (first use in this function); did you mean 'SSL_F_CLIENT_HELLO'?
2038 | return SSL_CLIENT_HELLO_RETRY;
| ^~~~~~~~~~~~~~~~~~~~~~
| SSL_F_CLIENT_HELLO
Fixes:
- http://autobuild.buildroot.org/results/d7d49cfce6f99c59e99c8e15399164fd5ecacc21
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit ac64086ce5)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
In 96c3b52132 (package/uacme: don't allow ualpn with mbedTLS), the
preference order moved openssl before mbedtls, because ualpn was not
compatible with mbedtls. That caused the preference order in the .mk to
diverge semantically from the preference order in the Config.in.
Indeed, openssl is only selected when neither gnutls nor mbedtls are
enabled, so openssl is clearly leastpreferred crypto backend. But when
both openssl and mbedtls were enabled, then uacme would use opensslC
because of ualpn.
The ualpn limitation was lifted in 6c7b46945e (package/uacme: allow
ualpn with mbedTLS), but the preference order in the .mk was not
restored to match that of the Config.in.
Restore the order in the .mk so that openssl is again treated as the
least-preferred crypto backend.
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr:
- split off to its own patch
- write the full commit log
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 192e047fda)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The gcc man page states that specifying Neon as part of the fpu setting
has no effect, unless the -funsafe-math-optimizations is also specified,
because Neon is not compliant with IEEE 754:
```
If the selected floating-point hardware includes the NEON extension
(e.g. -mfpu=neon), note that floating-point operations are not
generated by GCC's auto-vectorization pass unless
-funsafe-math-optimizations is also specified. This is because NEON
hardware does not fully implement the IEEE 754 standard for
floating-point arithmetic (in particular denormal values are treated
as zero), so the use of NEON instructions may lead to a loss of
precision.
```
-funsafe-math-optimizations must be explictly specified per package to
really use NEON as FPU, but it's something that is left to the user as
well as setting BR2_ARM_FPU_NEON_VFPV4. This way the default
BR2_ARM_FPU_VFPV4D16 is used as previously. So let's revert the
offending patch.
This reverts commit aaced92e8c.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 25ecee3f17)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The gcc man page states that specifying Neon as part of the fpu setting
has no effect, unless the -funsafe-math-optimizations is also specified,
because Neon is not compliant with IEEE 754:
```
If the selected floating-point hardware includes the NEON extension
(e.g. -mfpu=neon), note that floating-point operations are not
generated by GCC's auto-vectorization pass unless
-funsafe-math-optimizations is also specified. This is because NEON
hardware does not fully implement the IEEE 754 standard for
floating-point arithmetic (in particular denormal values are treated
as zero), so the use of NEON instructions may lead to a loss of
precision.
```
-funsafe-math-optimizations must be explictly specified per package to
really use NEON as FPU, but it's something that is left to the user as
well as setting BR2_ARM_FPU_NEON_VFPV4. This way the default
BR2_ARM_FPU_VFPV4D16 is used as previously. So let's revert the
offending patch.
This reverts commit 115ee05214.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit d5c1e67d3a)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The gcc man page states that specifying Neon as part of the fpu setting
has no effect, unless the -funsafe-math-optimizations is also specified,
because Neon is not compliant with IEEE 754:
```
If the selected floating-point hardware includes the NEON extension
(e.g. -mfpu=neon), note that floating-point operations are not
generated by GCC's auto-vectorization pass unless
-funsafe-math-optimizations is also specified. This is because NEON
hardware does not fully implement the IEEE 754 standard for
floating-point arithmetic (in particular denormal values are treated
as zero), so the use of NEON instructions may lead to a loss of
precision.
```
-funsafe-math-optimizations must be explictly specified per package to
really use NEON as FPU, but it's something that is left to the user as
well as setting BR2_ARM_FPU_NEON_VFPV4. This way the default
BR2_ARM_FPU_VFPV4D16 is used as previously. So let's revert the
offending patch.
This reverts commit f8528acdfd.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 869fe1fbab)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
The gcc man page states that specifying Neon as part of the fpu setting
has no effect, unless the -funsafe-math-optimizations is also specified,
because Neon is not compliant with IEEE 754:
```
If the selected floating-point hardware includes the NEON extension
(e.g. -mfpu=neon), note that floating-point operations are not
generated by GCC's auto-vectorization pass unless
-funsafe-math-optimizations is also specified. This is because NEON
hardware does not fully implement the IEEE 754 standard for
floating-point arithmetic (in particular denormal values are treated
as zero), so the use of NEON instructions may lead to a loss of
precision.
```
-funsafe-math-optimizations must be explictly specified per package to
really use NEON as FPU, but it's something that is left to the user as
well as setting BR2_ARM_FPU_NEON_VFPV4. This way the default
BR2_ARM_FPU_VFPV4D16 is used as previously. So let's revert the
offending patch.
This reverts commit 23329364e2.
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 68d0385533)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
netsniff-ng unconditionally uses pthread_spin_lock since its addition in
commit 500d287b07 and
1a9fbac03c
resulting in the following build failure:
/home/autobuild/autobuild/instance-1/output-1/per-package/netsniff-ng/host/bin/../lib/gcc/x86_64-buildroot-linux-uclibc/10.3.0/../../../../x86_64-buildroot-linux-uclibc/bin/ld: netsniff-ng/tprintf.o: in function `tprintf_flush':
tprintf.c:(.text+0x42c): undefined reference to `pthread_spin_lock'
Fixes:
- http://autobuild.buildroot.org/results/ceadbdea8cc35bfd7d601a6d4b18137f81f61406
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit a969e0f63c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following uclibc build failure on aarch64 raised since bump to
version 6.14 in commit 5292d1cf9a and
9070a04adf:
rngd_rndr.c:34:10: fatal error: sys/auxv.h: No such file or directory
34 | #include <sys/auxv.h>
| ^~~~~~~~~~~~
Strangely enough, there is no autobuilder failure for powerpc64le raised
since version bump to version 6.11 in commit
da83261c9b
Fixes:
- http://autobuild.buildroot.org/results/41d5ab9e67eb0d8af8d789fc94d4366f130a7fb2
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit 5874667922)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Fix the following sparc build failure with BR2_OPTIMIZE_FAST raised
since bump to version 2.9.19 in commit
65ed981ce0:
cc1: error: argument to '-O' should be a non-negative integer, 'g', 's' or 'fast'
Fixes:
- http://autobuild.buildroot.org/results/e1a330e1a899fcdf4900e9156d62c90813321e30
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 5ea5ac0c60)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Github repository mripard/sunxi-mali states to use Lima in place of
sunxi-mali because it's deprecated, but this package is still useful in
Buildroot so I want to move the SITE to my Github fork of the original
repository that already contains a patch to fix a build failure showing
up with Linux version >= 5.15.
The upstream patch fixes missing DMA_BUF module inclusion that leads to
build failure. The patch includes DMA_BUF by using:
MODULE_IMPORT_NS(DMA_BUF);
My idea is to continue to maintain this package in parallel to Lima since
it seems to be still useful.
Fixes:
http://autobuild.buildroot.net/results/8f25c26de737c358b3b43a10737609465b4e1398/
Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit 6dcaa20ca4)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Set CROSS variable otherwise makedumpfile will use it to undefine the
host architecture through -U__$(HOST_ARCH)__ if $(TARGET) is not equal
to $(HOST_ARCH). This will result in the following build failure since
the addition of the package in commit
adb64a97e7 if aarch64_be is cross-compiled
on a aarch64 host for example:
/home/autobuild/autobuild/instance-5/output-1/host/bin/aarch64_be-buildroot-linux-uclibc-gcc -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -D_FILE_OFFSET_BITS=64 -O3 -g0 -g -O2 -Wall -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -D_LARGEFILE64_SOURCE -DVERSION='"1.7.0"' -DRELEASE_DATE='"8 Nov 2021"' -D__aarch64_be__ -U__aarch64__ -DUSELZO -c -o ./print_info.o print_info.c
[...]
makedumpfile.c: In function 'is_kvaddr':
makedumpfile.c:1547:46: error: 'KVBASE' undeclared (first use in this function)
1547 | return (addr >= (unsigned long long)(KVBASE));
| ^~~~~~
Fixes:
- http://autobuild.buildroot.org/results/e4e10364e1a24099ce31bf20eacf5adedf93e5a7
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
(cherry picked from commit b8665e39f1)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
Commit 4dff1be05e (package/libvirt: libvirtd needs C++ for nmap-ncat)
introduce a recursive dependency (really: a circular dependency):
package/busybox/Config.in:33:error: recursive dependency detected!
package/busybox/Config.in:33: symbol BR2_PACKAGE_BUSYBOX_SHOW_OTHERS is selected by BR2_PACKAGE_EBTABLES_UTILS_SAVE
package/ebtables/Config.in:11: symbol BR2_PACKAGE_EBTABLES_UTILS_SAVE depends on BR2_PACKAGE_EBTABLES
package/ebtables/Config.in:1: symbol BR2_PACKAGE_EBTABLES is selected by BR2_PACKAGE_LIBVIRT_DAEMON
package/libvirt/Config.in:44: symbol BR2_PACKAGE_LIBVIRT_DAEMON depends on BR2_PACKAGE_NETCAT_OPENBSD
package/netcat-openbsd/Config.in:1: symbol BR2_PACKAGE_NETCAT_OPENBSD depends on BR2_PACKAGE_BUSYBOX_SHOW_OTHERS
We can't drop the C++ dependency and switch the netcat-openbsd and
nmap-ncat dependencies conditions without adding a glibc dependency.
So always mandate C++ even if is only needed by nmap and not
netcat-openbsd
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit a17c456b2c)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
This reverts commit f81242ae4f to avoid
the following build failure:
Makefile:575: *** libbsd is in the dependency chain of netcat-openbsd that has added it to its _DEPENDENCIES variable without selecting it or depending on it from Config.in. Stop.
Fixes:
- http://autobuild.buildroot.org/results/aada1d92df6cab0d01e27431b7b7483e3d165e79
Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>
(cherry picked from commit ecf49374ee)
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
