package/sentry-cli: remove package

This package has had build failures for a very long time, and these
issues have not been fixed, and it is now the number 1 build failure
reason in our autobuilders. It is time to acknowledge that the package
needs to be removed.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/openvmtools: fix time_t build failure on 32-bit platforms

Add upstream pending patch[1] to fix time_t on 32-bit platform.

[1]: https://github.com/vmware/open-vm-tools/pull/387

Fixes:
http://autobuild.buildroot.net/results/eb3dfe679536b578a0f16762312a96ada7162095/

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/libvirt: fix NLS build

Fix the following build failure raised since the addition of the package
in commit ccfc90e1010e42e6529afae3a5ea8bf7226dabc1:

/tmp/instance-0/output-1/host/opt/ext-toolchain/bin/../lib/gcc/i686-buildroot-linux-uclibc/9.3.0/../../../../i686-buildroot-linux-uclibc/bin/ld: src/util/libvirt_util.a(viralloc.c.o): in function `virInsertElementsN':
viralloc.c:(.text+0x167): undefined reference to `libintl_dgettext'

Fixes:
 - http://autobuild.buildroot.net/results/2349c55a4a42f08ca52700c60cda3065b0c4bd88

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

boot/at91bootstrap3: update to final 4.0.0 version

Update to AT91Bootstrap 4.0.0 version.

This package is now released under MIT license, and a license file was
added.

Signed-off-by: Eugen Hristev <eugen.hristev@microchip.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/bullet: fix build failure due to gcc bug 101952

The bullet package exhibits gcc bug 101952 when built for the SH4
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_101952=y like we
already do for BR2_TOOLCHAIN_HAS_GCC_BUG_85180=y.

Fixes:
http://autobuild.buildroot.net/results/32b/32bfaf0aae57ed18c18e82a72a958af9b3e1b241/

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

toolchain: introduce BR2_TOOLCHAIN_HAS_GCC_BUG_101952

bullet package fails to build for the SH4 architecture with optimization
enabled with gcc = 11.1.0:
http://autobuild.buildroot.net/results/32b/32bfaf0aae57ed18c18e82a72a958af9b3e1b241/

It's been reported upstream:
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=101952

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

DEVELOPERS: add myself to toolchain topics/packages

Since I've dealt and deal with toolchain bugs and their work-around
very often add myself to toolchain topic(toolchain/) as well as
package/binutils and package/gcc.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/eigen: override Fortran path

The language detection is falling back to the host system
Fortran compiler. An example of this is in RHEL7.9
(gcc4.8.5 20150623 (Red Hat 4.8.5-44)).

This patch bypasses detection and points to the location
where the compiler would be installed (if present). In the
cases where it doesn't exist, the detection falls through
and leaves Fortran disabled.

Fixes:
http://autobuild.buildroot.net/results/8354da225d1e5e337aa7ea62a7e6524fb5f1135f/

Signed-off-by: Matthew Weber <matthew.weber@collins.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/rustc: fix conditions related to MIPS support

For MIPS64 architecture variants supported by Rust, we need to exclude
MIPS64R6, but due to a copy/paste mistake in commit
d69d40c029dc7d8199b745eaee759d92b66c5d17 ("package/rustc: add support
for Tier 1 and Tier 2 platform"), we used BR2_MIPS_CPU_MIPS32R6 for
both MIPS 32-bit and MIPS 64-bit configurations, while
BR2_MIPS_CPU_MIPS64R6 should be used on MIPS 64-bit.

Fixes:

  http://autobuild.buildroot.net/8bab232eb98b164df300581ae019254bde7c8ca3/

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/nvidia-driver: fix build without X.org drivers

Fix the following build failure without BR2_PACKAGE_NVIDIA_DRIVER_XORG
raised since commit 9cda982855315d683d4880e6bb2a129133ea7213:

ln: failed to create symbolic link '/tmp/instance-0/output-1/target/usr/lib/xorg/modules/extensions/libglx.so': No such file or directory

Fixes:
 - http://autobuild.buildroot.org/results/c2e7a826c105363309f70a3fcfe28bd53efcb94a

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Vincent Fazio <vfazio@xes-inc.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

Update for 2021.08-rc2

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/postgresql: security bump version to 13.4

Release notes:
https://www.postgresql.org/about/news/postgresql-134-128-1113-1018-9623-and-14-beta-3-released-2277/

Fixes CVE-2021-3677:
https://www.postgresql.org/support/security/CVE-2021-3677/

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

configs/{atmel, microchip}: bump at91bootstrap3 version

Signed-off-by: Ludovic Desroches <ludovic.desroches@microchip.com>
Tested-by: Pierre-Jean Texier <texier.pj2@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

configs/acmesystems_*: bump at91bootstrap3 version

Signed-off-by: Ludovic Desroches <ludovic.desroches@microchip.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

boot/at91bootstrap3: bump 3.x series to 3.10.3

Signed-off-by: Ludovic Desroches <ludovic.desroches@microchip.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/protobuf: fix build failure due to missing -mcmodel=large

When building protobuf for or1k -mcmodel=large is needed to link, so let's
add that gcc option in case we're building for or1k.

Upstream gcc doesn't have the -mcmodel=large option for or1k, but all
released Buildroot gcc versions have the patch to add it, so that's
fine.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Reviewed-by: Maxim Kochetkov <fido_max@inbox.ru>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/libgeos: fix build failure due to missing -mcmodel=large

When building libgeos for or1k -mcmodel=large is needed to link, so let's
add that gcc option in case we're building for or1k.

Upstream gcc doesn't have the -mcmodel=large option for or1k, but all
released Buildroot gcc versions have the patch to add it, so that's
fine.

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Reviewed-by: Maxim Kochetkov <fido_max@inbox.ru>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/gcc: add OpenRisc patches for gcc version 11.1.0

At the moment of gcc 11.1.0 release the OpenRisc patches for -mcmodel=large
were still pending. They have been upstreamed yesterday as pointed in gcc
bugzilla[1]. So they will be part of gcc 11.3.0 or maybe before on 11.2.
2. Anyway at the moment if we try to build packages libgeos and protobuf
with OpenRisc gcc 11.1.0 it fails due to missing -mcmodel=large. So let's
add OpenRisc patches for it as done for all the previous versions.

Fixes:
still not appeared on autobuilers

[1]: https://gcc.gnu.org/bugzilla/show_bug.cgi?id=99783

Signed-off-by: Giulio Benetti <giulio.benetti@benettiengineering.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/tor: security bump version to 0.4.6.7

Fixes CVE-2021-38385: https://blog.torproject.org/node/2062

Rebased patch 0001.

Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/python-treq: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't see the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/python-service-identity: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't see the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/python-secretstorage: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't see the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/python-pyopenssl: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't see the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/python-keyring: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't view the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/python-cryptography: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't see the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/libbpf: needs headers >= 4.13

Since its addition in commit f7b8508e4fe9c96bf679cc5423cc1500b6523e83,
libbpf unconditionally uses TCA_BPF_ID which is only available since
kernel 4.13 and
https://github.com/torvalds/linux/commit/e86283071fb0eed28136adb52997888f4beb202b:

netlink.c: In function '__get_tc_info':
netlink.c:515:11: error: 'TCA_BPF_ID' undeclared (first use in this function); did you mean 'TCA_BPF_FD'?
  515 |  if (!tbb[TCA_BPF_ID])
      |           ^~~~~~~~~~
      |           TCA_BPF_FD

Fixes:
 - http://autobuild.buildroot.org/results/14996a0b7b9b65d0d6d5717d6da8922752789749

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/fontconfig: fix build with NLS

Add TARGET_NLS_DEPENDENCIES and host-gettext dependency to avoid the
following build failure in a per-package-directorie build with
host-cairo raised because fontconfig installs its ITS files in the wrong
directory (i.e. outside of gettext-tiny symlink):

mkdir -p /tmp/instance-0/output-1/per-package/host-cairo/host
rsync -a --link-dest=/tmp/instance-0/output-1/per-package/host-fontconfig/host/ /tmp/instance-0/output-1/per-package/host-fontconfig/host/ /tmp/instance-0/output-1/per-package/host-cairo/host
rsync -a --link-dest=/tmp/instance-0/output-1/per-package/host-freetype/host/ /tmp/instance-0/output-1/per-package/host-freetype/host/ /tmp/instance-0/output-1/per-package/host-cairo/host
rsync -a --link-dest=/tmp/instance-0/output-1/per-package/host-libglib2/host/ /tmp/instance-0/output-1/per-package/host-libglib2/host/ /tmp/instance-0/output-1/per-package/host-cairo/host
cannot delete non-empty directory: share/gettext
could not make way for new symlink: share/gettext

This only happens with per-package directories because then the rsync is
done. Otherwise the fontconfig installation will simply follow the
symlink.

The error of course exists for target as well, but doesn't occur in
autobuilders since it already fails for host.

Fixes:
 - http://autobuild.buildroot.org/results/00e29958cecfffa4e994ab549637117dd8f55c30

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/fontconfig: break circular dependency

Build fails because of the following circular dependency:

fontconfig -> util-linux -> udev -> systemd -> polkit ->
gobject-introspection -> cairo -> fontconfig

which results in the following build failure:

checking for UUID... no
checking where uuid functions comes from... configure: error:
*** uuid is required. install util-linux.

To break it, apply the same ugly workaround that was applied for
libglib2 and cryptsetup until a better solution is found:
https://patchwork.ozlabs.org/project/buildroot/patch/20201101150619.1709959-1-fontaine.fabrice@gmail.com/

Fixes:
 - http://autobuild.buildroot.org/results/2c6ef073e7e98e13daa409e1ea6130e9abd32c87

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>

package/busybox/udhcpc.script: ensure action argument is correctly passed to hook scripts

commit f79a420825479c47d (package/busybox/udhcpc.script: support RFC3442
static routes) used 'set --' clobbering the positional arguments, causing
the action argument to not be correctly forwarded to hook scripts for the
renew / bound cases if static routes are provided by the server.

As a workaround, save the action argument at the beginning of the script and
use that when calling hook scripts.

Reported-by: 王琦 <wangwangqi2011@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/harfbuzz: fix build without threads

Fix the following build failures without threads by defining HB_NO_MT as
specified in https://github.com/harfbuzz/harfbuzz/blob/main/CONFIG.md:

In file included from ../src/hb.hh:458,
                 from ../src/hb-blob.cc:28:
../src/hb-mutex.hh:88:2: error: #error "Could not find any system to define mutex macros."
   88 | #error "Could not find any system to define mutex macros."
      |  ^~~~~

This build failure is raised since bump to version 2.8.1 in commit
e9ba1c80e8ebf5f5ad9d889d2dd37f27af87e811 and
https://github.com/harfbuzz/harfbuzz/commit/711c241f6c7e18c5403602375a733af74df76f83
which removed busyloop mutex implementation.

Fixes:
 - http://autobuild.buildroot.org/results/0c03aca5e3ade735b54b0b9233896b868aec1520

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libepoxy: bump to version 1.5.9

Update to libepoxy 1.5.9, which adds a fallback to use libOpenGL.so that
fixes a couple of regressions:

  https://github.com/anholt/libepoxy/releases/tag/1.5.9

Signed-off-by: Adrian Perez de Castro <aperez@igalia.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

boot/arm-trusted-firmware: don't force ENABLE_STACK_PROTECTOR

Commit 5f432df7e2d2 ("boot/arm-trusted-firmware: change
ENABLE_STACK_PROTECTOR value when disabled") set
ENABLE_STACK_PROTECTOR=0 when disabled. But since we pass this value as
MAKE_OPT, the internal ATF logic that sets ENABLE_STACK_PROTECTOR again
based on its initial value breaks. This leads to build failure:

make[1]: *** [/builds/buildroot.org/buildroot/output/build/arm-trusted-firmware-v2.4/build/a80x0_mcbin/release/libc/assert.o] Error 1
aarch64-buildroot-linux-uclibc-gcc.br_real: error: unrecognized command-line option ‘-fstack-protector-0’; did you mean ‘-fstack-protector’?

Move ENABLE_STACK_PROTECTOR to make environment instead to allow make to
change its value.

Fixes:
https://gitlab.com/buildroot.org/buildroot/-/jobs/1497663294

Cc: Dick Olsson <hi@senzilla.io>
Cc: Sergey Matyukevich <geomatsi@gmail.com>
Cc: Thomas Petazzoni <thomas.petazzoni@bootlin.com
Signed-off-by: Baruch Siach <baruch@tkos.co.il>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libargtable2: update LIBARGTABLE2_VERSION

Update LIBARGTABLE2_VERSION to reflect what is used by
https://release-monitoring.org

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/iozone: update IOZONE_VERSION

Update IOZONE_VERSION to reflect what is used by
https://release-monitoring.org

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/sqlite: add SQLITE_TAR_VERSION

3.35.0 is the version used by https://release-monitoring.org as well as
NVD NIST database so add SQLITE_TAR_VERSION and drop
SQLITE_CPE_ID_VERSION

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/unbound: bump to version 1.13.2

The release contains a bugfix to fix the make install of the python
module after build changes introduced in this release RC1.

This release contains a number of bug fixes. There is a crash fix for
broken internal structures in stream reuse, that is used when many TCP
or TLS upstream connections are made. Also a number of features are added.

https://github.com/NLnetLabs/unbound/releases/tag/release-1.13.2

Signed-off-by: Kyle Harding <kyle@balena.io>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/go: security bump to version 1.16.7

go1.16.7 (released 2021-08-05) includes a security fix to the
net/http/httputil package, as well as bug fixes to the compiler, the
linker, the runtime, the go command, and the net/http package.

https://golang.org/doc/devel/release#go1.16

Signed-off-by: Christian Stewart <christian@paral.in>
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

DEVELOPERS: drop Sven Fischer

Sven has privately asked to no longer receive notifications related to
this package.

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/refpolicy: fix build with samba

Fix the following build failure with samba raised since commit
49f6b2f39ee23e3e53c0b001894644734b8338b6:

 Compiling targeted policy.31
 env LD_LIBRARY_PATH="/tmp/instance-5/output-1/host/lib:/tmp/instance-5/output-1/host/usr/lib" /tmp/instance-5/output-1/host/usr/bin/checkpolicy -c 31 -U deny -S -O -E policy.conf -o policy.31
 policy/modules/services/samba.te:399:ERROR 'type crack_db_t is not within scope' at token ';' on line 360232:
  allow smbd_t crack_db_t:dir { getattr search open };
 #line 399
 checkpolicy:  error(s) encountered while parsing configuration

Fixes:
 - http://autobuild.buildroot.org/results/ab7098948d1920e42fa587e07f0513f23ba7fc74

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gd: fix CVE-2021-38115

read_header_tga in gd_tga.c in the GD Graphics Library (aka LibGD)
through 2.3.2 allows remote attackers to cause a denial of service
(out-of-bounds read) via a crafted TGA file.

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/python-txtorcon: depend comment on BR2_PACKAGE_PYTHON3

The user shouldn't view the comment on the python2 menu.

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/terminology: set eet path

Set eet path to avoid the following build failure raised since bump to
version 1.9.0 in commit 6ac557a54bb4613e79728ad96724e7681c9c3c18 and
https://git.enlightenment.org/apps/terminology.git/commit/?id=f15dd2881a5e87e7f3b03594983826f7480c12e7:

/home/buildroot/autobuild/instance-2/output-1/build/terminology-1.9.0/data/colorschemes/add_color_scheme.sh: 30: /home/buildroot/autobuild/instance-2/output-1/build/terminology-1.9.0/data/colorschemes/add_color_scheme.sh: /usr/bin/eet: not found

Fixes:
 - http://autobuild.buildroot.org/results/767acf73e4eade48cd3fdb43e9dd599ba7ef9661

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/python-matplotlib: merge dependency comments

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/jszip: add JSZIP_CPE_ID_VENDOR

cpe:2.3:a:jszip_project:jszip is a valid CPE identifier for this
package:

  https://nvd.nist.gov/products/cpe/search/results?namingFormat=2.3&keyword=cpe%3A2.3%3Aa%3Ajszip_project%3Ajszip

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/sylpheed: fix CVE-2021-37746

textview_uri_security_check in textview.c in Claws Mail before 3.18.0,
and Sylpheed through 3.7.0, does not have sufficient link checks before
accepting a click.

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/sylpheed: add SYLPHEED_CPE_ID_VENDOR

cpe:2.3:a:sylpheed_project:sylpheed is a valid CPE identifier for this
package:

  https://nvd.nist.gov/products/cpe/search/results?namingFormat=2.3&keyword=cpe%3A2.3%3Aa%3Asylpheed_project%3Asylpheed

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/nginx-naxsi: fix build without x_forwarded_for

Fix the following build failure without x_forwarded_for raised since
bump to version 1.3 in commit 3335c8d8685119ebb5b2a9a6e575f3a5337c4d90:

/home/buildroot/autobuild/instance-3/output-1/build/nginx-naxsi-1.3/naxsi_src/naxsi_runtime.c: In function 'ngx_http_naxsi_data_parse':
/home/buildroot/autobuild/instance-3/output-1/build/nginx-naxsi-1.3/naxsi_src/naxsi_runtime.c:2846:20: error: 'ngx_http_headers_in_t' has no member named 'x_forwarded_for'
   if (r->headers_in.x_forwarded_for.nelts >= 1) {
                    ^

Fixes:
 - http://autobuild.buildroot.org/results/cdbc1536f6b5de3d4c836efa2f0dcaf0cdbb1462

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/poppler: fix build with gcc 11

Fix the following build failure with gcc 11:

In file included from /tmp/instance-0/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/glib-2.0/glib/gthread.h:32,
                 from /tmp/instance-0/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/glib-2.0/glib/gasyncqueue.h:32,
                 from /tmp/instance-0/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/glib-2.0/glib.h:32,
                 from /tmp/instance-0/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/glib-2.0/gobject/gbinding.h:28,
                 from /tmp/instance-0/output-1/host/arm-buildroot-linux-gnueabihf/sysroot/usr/include/glib-2.0/glib-object.h:22,
                 from /tmp/instance-0/output-1/build/poppler-0.84.0/glib/poppler.h:22,
                 from /tmp/instance-0/output-1/build/poppler-0.84.0/glib/poppler-action.cc:19:
/tmp/instance-0/output-1/build/poppler-0.84.0/glib/poppler-action.cc: In function 'GType poppler_dest_get_type()':
/tmp/instance-0/output-1/build/poppler-0.84.0/glib/poppler-private.h:155:13: error: argument 2 of '__atomic_load' must not be a pointer to a 'volatile' type
  155 |         if (g_once_init_enter (&g_define_type_id__volatile)) {                        \
      |             ^~~~~~~~~~~~~~~~~
/tmp/instance-0/output-1/build/poppler-0.84.0/glib/poppler-action.cc:28:1: note: in expansion of macro 'POPPLER_DEFINE_BOXED_TYPE'
   28 | POPPLER_DEFINE_BOXED_TYPE (PopplerDest, poppler_dest, poppler_dest_copy, poppler_dest_free)
      | ^~~~~~~~~~~~~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/dfcaac6487aaeb10412c3fe72a23135f8a70fefe

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/postgis: bump version to 3.1.3

Release-notes: https://postgis.net/2021/07/02/postgis-3.1.3/

Signed-off-by: Maxim Kochetkov <fido_max@inbox.ru>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/python-pymupdf: depend on python3

According to the package documentation, python 3.6 or newer
is needed.

Fixes:
 - http://autobuild.buildroot.net/results/f857986b8c9b6d4a843f3694cae8527760540ed0/

Signed-off-by: Asaf Kahlon <asafka7@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gobject-introspection: fix gidatadir

Since commit 521d387640d704f24caee51841388d98bcac739b gidatadir is
wrongly set to ${libdir}/../share, set it back to
${datadir}/gobject-introspection-1.0

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/prosody: security bump version to 0.11.10

fixes CVE-2021-37601
see https://blog.prosody.im/prosody-0.11.10-released/

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/jszip: fix CVE-2021-23413

This affects the package jszip before 3.7.0. Crafting a new zip file
with filenames set to Object prototype values (e.g __proto__, toString,
etc) results in a returned object with a modified prototype instance.

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/qemu: add libfuse3 optional dependency

libfuse3 is an optional dependency which is enabled by default since
version 6.0.0 and
https://gitlab.com/qemu-project/qemu/-/commit/0c9b70d5900a5108e899edfdd6f3790f8cb6bdc2

So add this dependency and fix the following build failures on musl
raised since bump to version 6.0.0 in commit
6b86c9335fc3ff381878156c6243454d4b688df9:

../block/export/fuse.c: In function 'fuse_fallocate':
../block/export/fuse.c:563:23: error: 'FALLOC_FL_ZERO_RANGE' undeclared (first use in this function)
  563 |     } else if (mode & FALLOC_FL_ZERO_RANGE) {
      |                       ^~~~~~~~~~~~~~~~~~~~
../block/export/fuse.c:563:23: note: each undeclared identifier is reported only once for each function it appears in
../block/export/fuse.c: In function 'fuse_lseek':
../block/export/fuse.c:639:19: error: 'SEEK_HOLE' undeclared (first use in this function)
  639 |     if (whence != SEEK_HOLE && whence != SEEK_DATA) {
      |                   ^~~~~~~~~
../block/export/fuse.c:639:42: error: 'SEEK_DATA' undeclared (first use in this function); did you mean 'SEEK_SET'?
  639 |     if (whence != SEEK_HOLE && whence != SEEK_DATA) {
      |                                          ^~~~~~~~~
      |                                          SEEK_SET

Fixes:
 - http://autobuild.buildroot.org/results/0196609043bd37543e6a0d17ffc2254302ea7ba3

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/tpm2-tools: fix build with gcc 4.8

Fix the following build failure with gcc 4.8 raised since bump to
version 4.3.2 in commit 91aa6efa8588bf7617cc4a640eb55052b524ceb7:

lib/tpm2_eventlog_yaml.c: In function 'yaml_uefi_var_unicodename':
lib/tpm2_eventlog_yaml.c:130:5: error: 'for' loop initial declarations are only allowed in C99 mode
     for(size_t i = 0; i < data->UnicodeNameLength; ++i, tmp += ret) {
     ^
lib/tpm2_eventlog_yaml.c:130:5: note: use option -std=c99 or -std=gnu99 to compile your code
lib/tpm2_eventlog_yaml.c: In function 'yaml_specid_algs':
lib/tpm2_eventlog_yaml.c:335:5: error: 'for' loop initial declarations are only allowed in C99 mode
     for (size_t i = 0; i < count; ++i, ++alg) {
     ^

Fixes:
 - http://autobuild.buildroot.org/results/018c75cfbb34006c0bca52dcc255ad6f9cc43b77

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/mongodb: fix build with gcc 11

Fix the following build failure with gcc 11:

In file included from src/mongo/db/query/plan_cache.h:36,
                 from src/mongo/db/catalog/collection_info_cache.h:33,
                 from src/mongo/db/catalog/collection.h:42,
                 from src/mongo/db/exec/requires_collection_stage.h:32,
                 from src/mongo/db/exec/delete.h:32,
                 from src/mongo/db/query/internal_plans.h:33,
                 from src/mongo/db/query/internal_plans.cpp:32:
src/mongo/db/exec/plan_stats.h:214:10: error: 'optional' in namespace 'std' does not name a template type
  214 |     std::optional<std::string> replanReason;
      |          ^~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/8c0875fa413923504515a83d8b679366418c2444

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gobject-introspection: fix exec_prefix

exec_prefix has been removed since version 1.66 and
https://gitlab.gnome.org/GNOME/gobject-introspection/-/commit/a88b1ac548516ab25140a2e9dea8cac894bbac93
resulting in the following build failure with rygel because libgee
installs Gee-0.8.gir in
output/host/aarch64-buildroot-linux-gnu/sysroot/share/gir-1.0/Gee-0.8.gir
instead of
output/host/aarch64-buildroot-linux-gnu/sysroot/usr/share/gir-1.0/Gee-0.8.gir:

Could not find GIR file 'Gee-0.8.gir'; check XDG_DATA_DIRS or use --includedir
error parsing file RygelCore-2.6.gir: Failed to parse included gir Gee-0.8
If the above error message is about missing .so libraries, then setting up GIR_EXTRA_LIBS_PATH in the .mk file should help.
Typically like this: PKG_MAKE_ENV += GIR_EXTRA_LIBS_PATH="$(@D)/.libs"

So replace exec_prefix by prefix

Fixes:
 - http://autobuild.buildroot.org/results/b04b27bbb84fa6ab9a8441e7bd94aad45226dce9

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/sconeserver: fix build with gcc 11

Fix the following build failure with gcc 11:

In file included from ../sconex/sconex.h:229,
                 from ../sconex/Descriptor.h:63,
                 from Descriptor.cpp:22:
Descriptor.cpp: In member function 'void scx::Descriptor::add_stream(scx::Stream*)':
Descriptor.cpp:150:22: error: ordered comparison of pointer with integer zero ('scx::Stream*' and 'int')
  150 |   DEBUG_ASSERT(stream>=0,"add_stream() Invalid stream");
      |                ~~~~~~^~~
Descriptor.cpp: In member function 'bool scx::Descriptor::remove_stream(scx::Stream*)':
Descriptor.cpp:204:22: error: ordered comparison of pointer with integer zero ('scx::Stream*' and 'int')
  204 |   DEBUG_ASSERT(stream>=0,"remove_stream() Invalid stream");
      |                ~~~~~~^~~

Fixes:
 - http://autobuild.buildroot.org/results/ccc9562e83fd2bd312d21b3124be42dfe4b7e850

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/usbguard: needs std::future

usbguard needs std::future to avoid the following build failure raised
since the addition of the package in commit
fbff7d7289cc95db991184f890f4ca1fcf8a101e:

In file included from src/Library/IPCClientPrivate.hpp:37:0,
                 from src/Library/IPCClientPrivate.cpp:24:
/home/buildroot/autobuild/run/instance-0/output-1/host/opt/ext-toolchain/arm-none-linux-gnueabi/include/c++/4.8.3/future:122:11: error: declaration of 'class std::promise<std::unique_ptr<google::protobuf::Message> >'
     class promise;
           ^
src/Library/IPCClientPrivate.cpp: In member function 'usbguard::IPC::MessagePointer usbguard::IPCClientPrivate::qbIPCSendRecvMessage(usbguard::IPC::MessageType&)':
src/Library/IPCClientPrivate.cpp:244:26: error: invalid use of incomplete type 'class std::promise<std::unique_ptr<google::protobuf::Message> >'
     auto future = promise.get_future();
                          ^
Fixes:
 - http://autobuild.buildroot.org/results/1c4e0dab5917e5559a97f7cf7a99e30a2b561fc9

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/flatbuffers: fix locale independent on musl

Fix locale indepdendent which is enabled on musl since version 1.11.0
and
https://github.com/google/flatbuffers/commit/5f32f948102e65eaeea461b44f3b43f96c7a7a5a
resulting in the following build failure on snort3 since bump to version
3.1.6 in commit e66f2fd310374779d415fa683813cc5f5ccf6be9:

In file included from /tmp/instance-0/output-1/host/x86_64-buildroot-linux-musl/sysroot/usr/include/flatbuffers/flexbuffers.h:24,
                 from /tmp/instance-0/output-1/host/x86_64-buildroot-linux-musl/sysroot/usr/include/flatbuffers/idl.h:26,
                 from /tmp/instance-0/output-1/build/snort3-3.1.6.0/src/network_inspectors/perf_monitor/fbs_formatter.cc:29:
/tmp/instance-0/output-1/host/x86_64-buildroot-linux-musl/sysroot/usr/include/flatbuffers/util.h: In function 'void flatbuffers::strtoval_impl(int64_t*, const char*, char**, int)':
/tmp/instance-0/output-1/host/x86_64-buildroot-linux-musl/sysroot/usr/include/flatbuffers/util.h:258:12: error: 'strtoll_l' was not declared in this scope; did you mean 'strcoll_l'?
  258 |     *val = __strtoll_impl(str, endptr, base);
      |            ^~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/68045b83e94f8caa337b1af7ed5f493ac1a55c47

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/osm2pgsql: needs std::future

osm2pgsql needs std::future since its addition in commit
46c42042867e672eb21b386eb5e29747c30e80ca to avoid the following build
failure:

In file included from /home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/db-copy.cpp:12:0:
/home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/db-copy.hpp:218:24: error: field 'barrier' has incomplete type
     std::promise<void> barrier;
                        ^
/home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/db-copy.hpp: In constructor 'db_cmd_sync_t::db_cmd_sync_t(std::promise<void>&&)':
/home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/db-copy.hpp:221:37: error: class 'db_cmd_sync_t' does not have any field named 'barrier'
     : db_cmd_t(db_cmd_t::Cmd_sync), barrier(std::move(b))
                                     ^
In file included from /home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/middle.hpp:18:0,
                 from /home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/dependency-manager.cpp:11:
/home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/thread-pool.hpp: In member function 'void thread_pool_t::check_for_exceptions()':
/home/buildroot/autobuild/run/instance-3/output-1/build/osm2pgsql-1.4.2/src/thread-pool.hpp:49:19: error: invalid use of incomplete type 'class std::future<void>'
             result.get();
                   ^

Fixes:
 - http://autobuild.buildroot.org/results/69528232ceba128a2e60aa778bb0943fdefc2ed1

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Maxim Kochetkov <fido_max@inbox.ru>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/perl-net-ssh2: fix gcrypt build

Pass crypto_backend option to avoid the following build failure raised
since bump to version 0.72 in commit
cd1d56bcde2a571062a1943c88506a8d9d2db357 and
https://github.com/rkitover/net-ssh2/commit/86a1274534dd7e0da03d98fb88c9c7ad4b1adb9b:

/tmp/instance-7/output-1/per-package/perl-net-ssh2/host/opt/ext-toolchain/bin/../lib/gcc/arm-buildroot-linux-uclibcgnueabi/9.3.0/../../../../arm-buildroot-linux-uclibcgnueabi/bin/ld: cannot find -lssl
/tmp/instance-7/output-1/per-package/perl-net-ssh2/host/opt/ext-toolchain/bin/../lib/gcc/arm-buildroot-linux-uclibcgnueabi/9.3.0/../../../../arm-buildroot-linux-uclibcgnueabi/bin/ld: cannot find -lcrypto

Fixes:
 - http://autobuild.buildroot.org/results/25747ec239e0b92775aa883e4f531f77d45f352e

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/polkit: fix install of polkit.its

Install polkit.its in $(HOST_DIR)/share/gettext/its as suggested by
https://gitlab.gnome.org/GNOME/gvfs/-/issues/508 to avoid the following
build failure with gvfs because polkit installs polkit.its in
$(STAGING_DIR)/usr/share/gettext:

FAILED: daemon/org.gtk.vfs.file-operations.policy
/tmp/instance-3/output-1/host/bin/meson --internal msgfmthelper daemon/org.gtk.vfs.file-operations.policy.in daemon/org.gtk.vfs.file-operations.policy xml /tmp/instance-3/output-1/build/gvfs-1.44.1/po
msgfmt: cannot locate ITS rules for daemon/org.gtk.vfs.file-operations.policy.in

Fixes:
 - http://autobuild.buildroot.org/results/ba9a8b31e6e275d4db1905241a9153a0513eb335

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/xenomai: fix build on arc and riscv32

Fix the following build failures on arc and riscv32:

latency.c: In function 'display':
latency.c:326:21: error: format '%ld' expects argument of type 'long int', but argument 2 has type 'time_t' {aka 'long long int'} [-Werror=format=]
  326 |         ("RTT|  %.2ld:%.2ld:%.2ld  (%s, %Ld us period, "
      |                 ~~~~^
      |                     |
      |                     long int
      |                 %.2lld
  327 |          "priority %d)\n", dt / 3600,
      |                            ~~~~~~~~~
      |                               |
      |                               time_t {aka long long int}

altency.c: In function ‘display’:
altency.c:262:21: error: format ‘%ld’ expects argument of type ‘long int’, but argument 2 has type ‘time_t’ {aka ‘long long int’} [-Werror=format=]
  262 |         ("RTT|  %.2ld:%.2ld:%.2ld  (%s, %Ld us period, "
      |                 ~~~~^
      |                     |
      |                     long int
      |                 %.2lld
  263 |          "priority %d)\n", dt / 3600,
      |                            ~~~~~~~~~
      |                               |
      |                               time_t {aka long long int}

Fixes:
 - http://autobuild.buildroot.org/results/448efe22e8fe058a1b354a3c124874e30b9ce138

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

docs/website: update for 2021.02.4

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

Update for 2021.02.4

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit 004a792dcf10e6c474070c9571f7504411e786cc)
[Peter: drop Makefile change]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

docs/website: update for 2021.05.1

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

Update for 2021.05.1

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>
(cherry picked from commit d5127a4de714d6608e897b4585e5ab5d3d5bd965)
[Peter: drop Makefile change]
Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/ruby: fix RUBY_VERSION_EXT

Correct RUBY_VERSION_EXT after commit be9783951d1d44 (package/ruby: security
bump to version 3.0.1):

ls output/target/usr/lib/ruby/
3.0.0  site_ruby  vendor_ruby

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/rust-bin: fix armv7-unknown-linux-gnueabi

Commit d69d40c029dc7d8199b745eaee759d92b66c5d17 forgot to add hash for
armv7-unknown-linux-gnueabi resulting in the following build failure:

ERROR: No hash found for rust-std-1.54.0-armv7-unknown-linux-gnueabi.tar.xz

Fixes:
 - http://autobuild.buildroot.org/results/443f94bb9ae0addb5130f96fb363ec9651e8af3d

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

Partial revert "configs/olimex_a20_olinuxino: switch to extlinux.conf"

Old scripts are still used by a10_olinuxino & a20_olinuxino_micro

Fixes olimex_a20_olinuxino_micro
see https://gitlab.com/buildroot.org/buildroot/-/jobs/1478738568

Signed-off-by: Francois Perrad <francois.perrad@gadz.org>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/efl: fix build with tslib

Build with tslib is broken since bump to version 1.25.0 in commit
cf57eceabcca134197fed79c043587439b3d4e39:

/data/buildroot-autobuilder/instance-0/output-1/host/opt/ext-toolchain/bin/../lib/gcc/powerpc-buildroot-linux-uclibc/9.3.0/../../../../powerpc-buildroot-linux-uclibc/bin/ld: src/lib/ecore_fb/libecore_fb.so.1.25.1.p/ecore_fb_ts.c.o: in function `_ecore_fb_ts_fd_handler':
ecore_fb_ts.c:(.text+0x4c): undefined reference to `ts_read_raw'

Fixes:
 - http://autobuild.buildroot.org/results/fa111306b95316eed54771550474dae804fa261e

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/ffmpeg: add -latomic to pkg-config files

-latomic was added to extralibs to fix static build of ffmpeg in commit
fc8798197b57488a852e2c14b4677b5eb0114a0b. However, extralibs is not
added to libavformat.pc resulting in the following static build failure
of motion:

/home/buildroot/autobuild/instance-1/output-1/host/opt/ext-toolchain/bin/../lib/gcc/arm-buildroot-linux-uclibcgnueabi/9.3.0/../../../../arm-buildroot-linux-uclibcgnueabi/bin/ld: /home/buildroot/autobuild/instance-1/output-1/host/bin/../arm-buildroot-linux-uclibcgnueabi/sysroot/usr/lib/libavformat.a(fifo.o): in function `fifo_init':
/home/buildroot/autobuild/instance-1/output-1/build/ffmpeg-4.4/libavformat/fifo.c:519: undefined reference to `__atomic_store_8'

So add a patch to add extralibs (and so -latomic) to all pkg-config
files

Fixes:
 - http://autobuild.buildroot.org/results/62ec618e40081a250b8129ec6f5a178eb06fba1d

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/bullet: fix musl build

Fix the following build failure on musl raised since bump to version
3.0.9 in commit 28b4947ed8f53c4edfbf8fef9304dc76480c01ca:

In file included from /tmp/instance-5/output-1/build/bullet-3.09/src/LinearMath/btScalar.h:289,
                 from /tmp/instance-5/output-1/build/bullet-3.09/src/LinearMath/btVector3.h:19,
                 from /tmp/instance-5/output-1/build/bullet-3.09/src/LinearMath/btConvexHullComputer.h:18,
                 from /tmp/instance-5/output-1/build/bullet-3.09/Extras/VHACD/src/VHACD.cpp:28:
/tmp/instance-5/output-1/build/bullet-3.09/Extras/BulletRobotics/../../Extras/VHACD/inc/vhacdMutex.h: In constructor 'VHACD::Mutex::Mutex()':
/tmp/instance-5/output-1/build/bullet-3.09/Extras/BulletRobotics/../../Extras/VHACD/inc/vhacdMutex.h:97:54: error: 'PTHREAD_MUTEX_RECURSIVE_NP' was not declared in this scope; did you mean 'PTHREAD_MUTEX_RECURSIVE'?
   97 |   VHACD_VERIFY(pthread_mutexattr_settype(&mutexAttr, PTHREAD_MUTEX_RECURSIVE_NP) == 0);
      |                                                      ^~~~~~~~~~~~~~~~~~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/79cd2024b3dfc8d3e896cdacf67fb891df81ca6e

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/ogre: fix build on musl

Fix the following build failure on musl raised since the addition of the
package in commit eb91fa730c5d92202c38514345e86315e138944c:

/tmp/instance-1/output-1/build/ogre-1.12.0/OgreMain/src/OgreStringConverter.cpp: In static member function 'static bool Ogre::StringConverter::parse(const String&, Ogre::int32&)':
/tmp/instance-1/output-1/build/ogre-1.12.0/OgreMain/src/OgreStringConverter.cpp:253:22: error: 'strtol_l' was not declared in this scope; did you mean 'strtold_l'?
  253 |         ret = (int32)strtol_l(val.c_str(), &end, 0, _numLocale);
      |                      ^~~~~~~~
      |                      strtold_l

Fixes:
 - http://autobuild.buildroot.org/results/491f89e45610a7752c0700ac02b80a92b7876ec3

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/libvirt: nss needs yajl

nss can't be built without yajl since the additoon of the package in
commit ccfc90e1010e42e6529afae3a5ea8bf7226dabc1:

../output-1/build/libvirt-7.4.0/meson.build:1986:6: ERROR: Problem encountered: Can't build nss plugin without yajl

Fixes:
 - http://autobuild.buildroot.org/results/18db9ec30d5e60fbe9f5096c3fd1a30396e11856

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/prelink-cross: fix handling of TLS offsets for s390 arch

/tmp/instance-1/output-1/host/sbin/prelink-rtld: error while loading
shared libraries: /lib64/libc.so.6: cannot handle TLS data
ERROR: can't resolve libraries to shared libraries: glib-2.0, gobject-2.0
ninja: build stopped: subcommand failed.
make: *** [package/pkg-generic.mk:270:
/tmp/instance-1/output-1/build/gobject-introspection-1.68.0/.stamp_built]
Error 1
make: Leaving directory '/tmp/instance-1/buildroot'

Fixes:
- http://autobuild.buildroot.org/results/7614f5928e845331d7295349da891a76f1786b79/

Signed-off-by: Alexander Egorenkov <egorenar@linux.ibm.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/qt5/qt5base: fix libatomic handling to not affect host builds

qt5declarative builds qmltyperegistrar for the host as part of its build
process.
When building qt target packages (which is the case for qt5declarative),
-spec devices/linux-buildroot-g++ is passed to qmake in QT5_QMAKE
variable and this spec currently has -latomic in its LIBS.

This -latomic makes it to the build of the host build of
qmltyperegistrar which is not useful.

This was discovered on Fedora 34 where libatomic is not pulled with gcc
package, therefore was missing on the host machine.

This makes sure that -latomic is not added for host build of qt
packages.

Fixes: 7d286be4f9 ("package/qt5base: link with -latomic when needed")
Cc: Quentin Schulz <foss@0leil.net>
Suggested-by: Arnout Vandecappelle <arnout@mind.be>
Signed-off-by: Quentin Schulz <quentin.schulz@theobroma-systems.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/poke: fix build with uclibc-ng < 1.0.35

Fix the folllowing build failure with uclibc-ng which is raised since
activation of uclibc-ng in commit
a4dc754178b7c24ed129e7e6f8effd555249c941:

In file included from ./sys/random.h:40,
                 from getrandom.c:22:
/tmp/instance-0/output-1/host/x86_64-buildroot-linux-uclibc/sysroot/usr/include/sys/random.h:27:35: error: unknown type name 'size_t'
   27 | extern int getrandom(void *__buf, size_t count, unsigned int flags)
      |                                   ^~~~~~
/tmp/instance-0/output-1/host/x86_64-buildroot-linux-uclibc/sysroot/usr/include/sys/random.h:8:1: note: 'size_t' is defined in header '<stddef.h>'; did you forget to '#include <stddef.h>'?
    7 | #include <features.h>
  +++ |+#include <stddef.h>
    8 |

Fixes:
 - http://autobuild.buildroot.org/results/30105094e39374ec6d4e47e2fb5e99696f7f7981

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/seatd: fix build with gcc < 7

Since its addition in commit daae31149017cbf8b03634cfc56da132f7e1ad83,
seatd fails to build with gcc < 7 because it unconditonally uses
-Wimplicit-fallthrough which is only available since gcc 7.1 and
https://github.com/gcc-mirror/gcc/commit/81fea426da8c4687bb32e6894dc26f00ae211822
resulting in the following build failure:

arm-none-linux-gnueabi-gcc: error: unrecognized command line option '-Wimplicit-fallthrough'

Fixes:
 - http://autobuild.buildroot.org/results/0ee6816a7cceebdafd07612677a594bdf68e0790

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/iputils: fix rdisc server option

Commit 95b0078cc089b48be8973b9507599cc836397869 wrongly added
BUILD_RDISC_SERVER instead of ENABLE_RDISC_SERVER:

WARNING: Unknown options: "BUILD_RDISC_SERVER"

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/iputils: disable tests

Disable tests to avoid the following build failure raised since bump to
version 20210722 in commit b984cba490248e6636b53b4b26b959cda5ed6f6c:

../output-1/build/iputils-20210722/ping/meson.build:38:0: ERROR: Program or command 'ip' not found or not executable

Fixes:
 - http://autobuild.buildroot.org/results/af79d5a8a84f439f4d1d720fad2f3e197923cd4a

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Petr Vorel <petr.vorel@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/gobject-introspection: fix datarootdir

datarootdir has been removed since version 1.66 and
https://gitlab.gnome.org/GNOME/gobject-introspection/-/commit/a88b1ac548516ab25140a2e9dea8cac894bbac93
resulting in the following build failures with autotools-based
applications using introspection (e.g. libmediaart or libsecret) if
gobject-introspection is not installed on host since commit
abc110e3624aa4e8a9bd8595b3fdc0aec60845e0:

make[3]: *** No rule to make target 'MediaArt-2.0.typelib', needed by 'all'.  Stop.

or

Makefile:2917: /usr/share/gobject-introspection-1.0/Makefile.introspection: No such file or directory

Fixes:
 - http://autobuild.buildroot.org/results/70585b81fd6aa78e2aa6e5173f35c0f26a1408a9
 - http://autobuild.buildroot.org/results/95aa8e189f3326ecae6133e298c56ffcbbc3198f

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Reviewed-by: Yegor Yefremov <yegorslists@googlemail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/network-manager: needs headers >= 4.6

network-manager uncondtionally uses ETHTOOL_GLINKSETTINGS since version
1.31.3 and
https://github.com/NetworkManager/NetworkManager/commit/4a81fe13ae18cb6d157d7cb3eec300f3061263fe
which is only available since kernel 4.6 and
https://github.com/torvalds/linux/commit/3f1ac7a700d039c61d8d8b99f28d605d489a60cf
resulting in the following build failure since commit
360d1aad84449c6a1993d8c7e469dc0cd35339a2:

In file included from ./src/libnm-glib-aux/nm-default-glib.h:69:0,
                 from ./src/libnm-glib-aux/nm-default-glib-i18n-lib.h:13,
                 from src/libnm-platform/nm-platform-utils.c:6:
src/libnm-platform/nm-platform-utils.c: In function '_ethtool_cmd_to_string':
src/libnm-platform/nm-platform-utils.c:280:51: error: 'ETHTOOL_GLINKSETTINGS' undeclared (first use in this function)
                                 NM_UTILS_ENUM2STR(ETHTOOL_GLINKSETTINGS, "ETHTOOL_GLINKSETTINGS"),
                                                   ^

Fixes:
 - http://autobuild.buildroot.org/results/10b9b68ab6987b99b3e8dfb2d651808008ad0bb7

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/libnss: fix build on ARM big endian

Fix the follownig build failure on ARM big endian raised since bump to
version 3.68 in commit 0a73b9b962421ce7a72c88e5f5bbd8849f21b041:

Linux2.6_aarch64_aarch64_be-linux-gnu-gcc_glibc_PTH_64_DBG.OBJ/Linux_SINGLE_SHLIB/sha256-armv8.o: In function `SHA256_Compress_Native':
sha256-armv8.c:(.text.SHA256_Compress_Native+0x0): multiple definition of `SHA256_Compress_Native'
Linux2.6_aarch64_aarch64_be-linux-gnu-gcc_glibc_PTH_64_DBG.OBJ/Linux_SINGLE_SHLIB/sha512.o:sha512.c:(.text.SHA256_Compress_Native+0x0): first defined here
Linux2.6_aarch64_aarch64_be-linux-gnu-gcc_glibc_PTH_64_DBG.OBJ/Linux_SINGLE_SHLIB/sha256-armv8.o: In function `SHA256_Update_Native':
sha256-armv8.c:(.text.SHA256_Update_Native+0x0): multiple definition of `SHA256_Update_Native'
Linux2.6_aarch64_aarch64_be-linux-gnu-gcc_glibc_PTH_64_DBG.OBJ/Linux_SINGLE_SHLIB/sha512.o:sha512.c:(.text.SHA256_Update_Native+0x0): first defined here

Fixes:
 - http://autobuild.buildroot.org/results/237aba0c16a34fec1b0fe50fe08cace438eda1bf

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/tar: fix build with uclibc-ng < 1.0.35

Fix the folllowing build failure with uclibc-ng which is raised since
bump to version 1.34 in commit 27fffea6db7358af20f5a3a8faa174d3b782d61f:

In file included from ./sys/random.h:40,
                 from getrandom.c:22:
/tmp/instance-0/output-1/per-package/tar/host/x86_64-buildroot-linux-uclibc/sysroot/usr/include/sys/random.h:27:35: error: unknown type name 'size_t'
   27 | extern int getrandom(void *__buf, size_t count, unsigned int flags)
      |                                   ^~~~~~
/tmp/instance-0/output-1/per-package/tar/host/x86_64-buildroot-linux-uclibc/sysroot/usr/include/sys/random.h:8:1: note: 'size_t' is defined in header '<stddef.h>'; did you forget to '#include <stddef.h>'?
    7 | #include <features.h>
  +++ |+#include <stddef.h>
    8 |

Fixes:
 - http://autobuild.buildroot.org/results/f40e09d621ab5ba66dd97138dec174acfb7fda2a

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/pulseview: fix build with libglib >= 2.68

Fix the following build failure with libglib >= 2.68 raised since commit
c72524fb1b1c6b7dd1b1f511eea0fdf5a39dd3a0:

In file included from /tmp/instance-0/output-1/host/mips64el-buildroot-linux-uclibc/sysroot/usr/include/glib-2.0/glib/gatomic.h:31,
                 from /tmp/instance-0/output-1/host/mips64el-buildroot-linux-uclibc/sysroot/usr/include/glib-2.0/glib/gthread.h:32,
                 from /tmp/instance-0/output-1/host/mips64el-buildroot-linux-uclibc/sysroot/usr/include/glib-2.0/glib/gasyncqueue.h:32,
                 from /tmp/instance-0/output-1/host/mips64el-buildroot-linux-uclibc/sysroot/usr/include/glib-2.0/glib.h:32,
                 from /tmp/instance-0/output-1/host/mips64el-buildroot-linux-uclibc/sysroot/usr/include/libsigrokdecode/libsigrokdecode.h:25,
                 from /tmp/instance-0/output-1/build/pulseview-0.4.2/pv/data/decode/annotation.cpp:21:
/tmp/instance-0/output-1/host/opt/ext-toolchain/mips64el-buildroot-linux-uclibc/include/c++/8.4.0/type_traits:56:3: error: template with C linkage
   template<typename _Tp, _Tp __v>
   ^~~~~~~~
/tmp/instance-0/output-1/build/pulseview-0.4.2/pv/data/decode/annotation.cpp:20:1: note: 'extern "C"' linkage started here
 extern "C" {
 ^~~~~~~~~~

Fixes:
 - http://autobuild.buildroot.org/results/33cbaaeefd8637bac4ad0fbbfae7c369de4875d2

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

boot/arm-trusted-firmware: change ENABLE_STACK_PROTECTOR value when disabled

Since commit
cf176128ec4ad60c29c8828cab1887f69edb9473 ("boot/arm-trusted-firmware:
add SSP option"), we are passing ENABLE_STACK_PROTECTOR=none when we
want to disable SSP usage in TF-A. While this works fine in recent
versions of TF-A, older versions such as TF-A will end up passing
-fstack-protector-none in this situation, which fails as this is not a
valid gcc option (the valid gcc option is -fno-stack-protector).

To solve this, we pass ENABLE_STACK_PROTECTOR=0 which was in older
TF-A versions used to say "don't do anything with SSP", and is also
still supported in newer versions of TF-A.

Fixes:

 https://gitlab.com/buildroot.org/buildroot/-/jobs/1478738580

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/ruby: security bump to version 3.0.2

Fixes the following security issues:

- CVE-2021-31810: Trusting FTP PASV responses vulnerability in Net::FTP
- CVE-2021-32066: A StartTLS stripping vulnerability in Net::IMAP
- CVE-2021-31799: A command injection vulnerability in RDoc

For more details, see the announcement:
https://www.ruby-lang.org/en/news/2021/07/07/ruby-3-0-2-released/

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/nodejs: security bump to version 12.22.4

Fixes the following security issues:

- CVE-2021-22930: Use after free on close http2 on stream canceling (High)

  Node.js is vulnerable to a use after free attack where an attacker might
  be able to exploit the memory corruption, to change process behavior.

Drop 0002-Fix-build-with-ICU-68.patch as this is now fixed upstream since
https://github.com/nodejs/node/commit/e459c79b02

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

{linux, linux-headers}: bump 4.{4, 9, 14, 19}.x / 5.{4, 10, 12, 13}.x series

Includes fixes for the recent "Sequoia" seq_file vulnerability
(CVE-2021-33909):

https://lwn.net/Articles/863729/

Signed-off-by: Peter Korsgaard <peter@korsgaard.com>

package/libvirt: needs xsltproc

libvirt needs xsltproc since its addition in commit
ccfc90e1010e42e6529afae3a5ea8bf7226dabc1:

../output-1/build/libvirt-7.4.0/meson.build:807:2: ERROR: Program 'xsltproc' not found

Fixes:
 - http://autobuild.buildroot.org/results/0341be10c4c36f37b5b6486428e55170a42b7692

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/guile: fix build on riscv32

Fix the following build failure on riscv32:

system/base/target.scm:132:16: In procedure triplet-pointer-size:
unknown CPU word size "riscv32"

Fixes:
 - http://autobuild.buildroot.org/results/6705630c1484239ec8b73d57ebc2e2570fbfc8f8

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/libvirt: disable stack-protector

Set warning_level to 1 instead of the default value of 2 to disable
stack-protector and avoid the following build failure since the addition
of the package in commit ccfc90e1010e42e6529afae3a5ea8bf7226dabc1:

/tmp/instance-5/output-1/host/opt/ext-toolchain/bin/../lib/gcc/i686-buildroot-linux-uclibc/9.3.0/../../../../i686-buildroot-linux-uclibc/bin/ld: tests/commandhelper.p/commandhelper.c.o: in function `printInput':
commandhelper.c:(.text+0xad): undefined reference to `__stack_chk_guard'

Fixes:
 - http://autobuild.buildroot.org/results/a2657eb18b579752c8217ab356adcb52931f6785

Signed-off-by: Fabrice Fontaine <fontaine.fabrice@gmail.com>
[yann.morin.1998@free.fr:
  - move to a separate assignment
  - add a big fat comment to explain it
]
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

package/qemu: filter out long make options

Since qemu started using ninja, they have the following fragment in the
Makefile:

MAKE.n = $(findstring n,$(firstword $(MAKEFLAGS)))
NINJAFLAGS = ... $(if $(MAKE.n), -n) ...

Buildroot's generated makefile in the O= directory invokes make in the
base buildroot with --no-print-directory. However, make's placement of
the --no-print-directory in MAKEFLAGS varies between the versions of
the host make; make 4.3 places that at the end while make 3.82 places
it at the beginning. As a result, if building on a system with an older
host make, qemu's makefile invokes `ninja -n` which does not generate
any outputs.

To reproduce, on a CentOS 7 machine or docker image:
  mkdir /tmp/br-build && cd /tmp/br-build
  make -C ~/buildroot pc_x86_64_bios_defconfig O=`pwd`
  make menuconfig # Switch to glibc, enable "QEMU" and "QEMU tools"
  make all # Build succeeds
  find target -name qemu-img # No binary has been built

Pick up the fix commited in Qemu upstream.

Signed-off-by: Alexey Neyman <stilor@att.net>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>

configs/acmesystems_acqua_a5_*: bump at91bootstrap3 version

AT91Bootstrap 3.10.2 has a bug in its Makefile:[1] it gives the
-nostartfiles flag to the ld command, although this is a gcc flag, not
an ld flag. The bug happens to be harmless when using binutils 2.35.
However, the ld shipped with binutils 2.36 is more strict in its option
parsing, and fails with the following error:

    .../output/host/bin/arm-buildroot-linux-uclibcgnueabihf-ld:
    Error: unable to disambiguate: -nostartfiles
    (did you mean --nostartfiles ?)

The bug was fixed upstream in commit 3207586a.[2] However, since there
has been no tagged 3.x release since then, we just upgrade to the
current head of the branch at91bootstrap-3.x.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/1478738145
  https://gitlab.com/buildroot.org/buildroot/-/jobs/1478738152

[1] https://github.com/linux4sam/at91bootstrap/issues/127
[2] https://github.com/linux4sam/at91bootstrap/commit/3207586a79aba527bbf20560930089b800f191ac

Signed-off-by: Edgar Bonet <bonet@grenoble.cnrs.fr>
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

configs/microchip_sama5d2_icp_mmc_dev_defconfig: drop BR2_TARGET_AT91BOOTSTRAP3_CUSTOM_GIT=y

This defconfig is download at91bootstrap3 using a tarball, not over
git.

Fixes:

  configs/microchip_sama5d2_icp_mmc_dev_defconfig:63:warning: override: BR2_TARGET_AT91BOOTSTRAP3_CUSTOM_TARBALL changes choice state

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

configs/microchip_sama5d2_icp_mmc_defconfig: drop BR2_TARGET_AT91BOOTSTRAP3_CUSTOM_GIT=y

This defconfig is download at91bootstrap3 using a tarball, not over
git.

Fixes:

  configs/microchip_sama5d2_icp_mmc_defconfig:20:warning: override: BR2_TARGET_AT91BOOTSTRAP3_CUSTOM_TARBALL changes choice state

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

configs/microchip_sam9x60ek_mmc_dev_defconfig: add missing options

This defconfig needs wchar, thread debugging, and udev support to be
able to use all the packages it enables.

Fixes:

  https://gitlab.com/buildroot.org/buildroot/-/jobs/1478738516

Signed-off-by: Thomas Petazzoni <thomas.petazzoni@bootlin.com>

package/glibc: Use 2.33 for RISC-V 32-bit

RISC-V 32-bit was special cased to use glibc master as support was not
in a released version.

Recently 2.33 was released, so RV32 can now use the same glibc version
as other platforms.

Signed-off-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Yann E. MORIN <yann.morin.1998@free.fr>